From ssklar at stanford.edu Fri Feb 1 01:06:58 2008 From: ssklar at stanford.edu (Sandor W. Sklar) Date: Thu, 31 Jan 2008 17:06:58 -0800 Subject: [et-mgmt-tools] Creating a yum repo (via cobbler) from "scratch"? Message-ID: I apologize in advance if my questions here are "dumb", but I'm kind of new to this, and am trying to wrap my head around a lot of information. I've successfully set up a basic cobbler (0.6.5) on an RHEL 5 server, have imported a distro from the rhel-5.1-server-x86_64-dvd iso, and have PXE booted and built a basic system from this cobbler server. That I was able to do all of that in a day of work was amazing, and a testament to the quality of all of the software involved. My question is this: I'd like to set up a yum repository, available from the cobbler server, containing a few RPMs that aren't available from another repository (meaning, I'm not "mirroring" another repository.) One example would be that I want to have my own repository of the IBM TSM client RPMs, which one only gets by FTPing them from IBM's site: $ ls -1 *.rpm TIVsm-API64.i386.rpm TIVsm-API.i386.rpm TIVsm-BA.i386.rpm TIVsm-HSM.i386.rpm I'd like to have it so I could simply run "yum install tsm-client" on a system built from my cobbler server, and have the above RPMs and their dependencies (all included in rhel5 core) installed. This question might be best for a yum-specific list, but since cobbler has repository mirroring integrated, I'd give it a shot here. Is there an easy way to accomplish this? Thanks, -s- From thestrider at gmail.com Fri Feb 1 02:16:42 2008 From: thestrider at gmail.com (Adam Rosenwald) Date: Thu, 31 Jan 2008 21:16:42 -0500 Subject: [et-mgmt-tools] Creating a yum repo (via cobbler) from "scratch"? In-Reply-To: References: Message-ID: <47A2810A.2040905@gmail.com> Sandor, * Set "yum_post_install_mirror" to "1" in /*/var/lib/cobbler/settings*/. * Create a directory somewhere for your repository. Create a base directory: *mkdir -p /my/repo/base* * Copy any and all custom or third-party RPMs to the base directory. * Create the metadata for the repository ("yum install createrepo" if you don't have the utility). This may not be necessary, but this will ensure that the metadata is available: *createrepo* */my/repo *# note that the target is /not/ the base directory * Add the repo to cobbler. *cobbler repo add --mirror=/my/repo --name=name_of_your_repo* * Sync repositories: *cobbler reposync* * Add repos to your cobbler profiles/distros. *cobbler distro edit --name=my_distro --repos='name_of_your_repo ...' cobbler profile edit --name=my_profile --repos='name_of_your_repo...' *# Modify only the item that most closely meets your scope; e.g. if all profiles under "my_distro" should include 'name_of_your_repo', you can ignore the '--repos' field whenever creating profiles that exist under "my_distro". * You are done. Just *cobbler sync* and your systems' kickstart files will include YUM repo stanzas provided by Cheetah template expansions. ---- Note that adding RPMS to existing repositories may require re-running '*createrepo /my/repo*' and *'cobbler reposync*' to be seen on your deployed systems. You will also need to execute '*yum clean all*' (or some variant of 'yum clean') on your client to force it to redownload the repository metadata (which describe the new RPM set). I hope this helps. Regards, -Adam. -A. Sandor W. Sklar wrote: > > I apologize in advance if my questions here are "dumb", but I'm kind > of new to this, and am trying to wrap my head around a lot of > information. > > I've successfully set up a basic cobbler (0.6.5) on an RHEL 5 server, > have imported a distro from the rhel-5.1-server-x86_64-dvd iso, and > have PXE booted and built a basic system from this cobbler server. > That I was able to do all of that in a day of work was amazing, and a > testament to the quality of all of the software involved. > > My question is this: I'd like to set up a yum repository, available > from the cobbler server, containing a few RPMs that aren't available > from another repository (meaning, I'm not "mirroring" another > repository.) One example would be that I want to have my own > repository of the IBM TSM client RPMs, which one only gets by FTPing > them from IBM's site: > > $ ls -1 *.rpm > TIVsm-API64.i386.rpm > TIVsm-API.i386.rpm > TIVsm-BA.i386.rpm > TIVsm-HSM.i386.rpm > > I'd like to have it so I could simply run "yum install tsm-client" on > a system built from my cobbler server, and have the above RPMs and > their dependencies (all included in rhel5 core) installed. This > question might be best for a yum-specific list, but since cobbler has > repository mirroring integrated, I'd give it a shot here. Is there an > easy way to accomplish this? > > Thanks, > -s- > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ssklar at stanford.edu Fri Feb 1 04:09:06 2008 From: ssklar at stanford.edu (Sandor W. Sklar) Date: Thu, 31 Jan 2008 20:09:06 -0800 Subject: [et-mgmt-tools] Creating a yum repo (via cobbler) from "scratch"? In-Reply-To: <47A2810A.2040905@gmail.com> References: <47A2810A.2040905@gmail.com> Message-ID: <6EE805A9-840B-42BF-B574-08E87FD488CE@stanford.edu> Your instructions were perfect! Thanks so much ... I really appreciate the help! -s- On Jan 31, 2008, at 6:16 PM, Adam Rosenwald wrote: > Sandor, > > * Set "yum_post_install_mirror" to "1" in /var/lib/cobbler/settings. > > * Create a directory somewhere for your repository. Create a base > directory: > > mkdir -p /my/repo/base > > * Copy any and all custom or third-party RPMs to the base directory. > > * Create the metadata for the repository ("yum install createrepo" > if you don't have the utility). This may not be necessary, but this > will ensure that the metadata is available: > > createrepo /my/repo # note that the target is not the base > directory > > * Add the repo to cobbler. > > cobbler repo add --mirror=/my/repo --name=name_of_your_repo > > * Sync repositories: > > cobbler reposync > > * Add repos to your cobbler profiles/distros. > > cobbler distro edit --name=my_distro -- > repos='name_of_your_repo ...' > cobbler profile edit --name=my_profile -- > repos='name_of_your_repo...' > > # Modify only the item that most closely meets your scope; e.g. > if all profiles under "my_distro" should include > 'name_of_your_repo', you can ignore the '--repos' field whenever > creating profiles that exist under "my_distro". > > * You are done. Just cobbler sync and your systems' kickstart files > will include YUM repo stanzas provided by Cheetah template expansions. > > ---- > > Note that adding RPMS to existing repositories may require re- > running 'createrepo /my/repo' and 'cobbler reposync' to be seen on > your deployed systems. You will also need to execute 'yum clean > all' (or some variant of 'yum clean') on your client to force it to > redownload the repository metadata (which describe the new RPM set). > > I hope this helps. > > Regards, > > -Adam. > > -A. > > Sandor W. Sklar wrote: >> >> >> I apologize in advance if my questions here are "dumb", but I'm >> kind of new to this, and am trying to wrap my head around a lot of >> information. >> >> I've successfully set up a basic cobbler (0.6.5) on an RHEL 5 >> server, have imported a distro from the rhel-5.1-server-x86_64-dvd >> iso, and have PXE booted and built a basic system from this cobbler >> server. That I was able to do all of that in a day of work was >> amazing, and a testament to the quality of all of the software >> involved. >> >> My question is this: I'd like to set up a yum repository, available >> from the cobbler server, containing a few RPMs that aren't >> available from another repository (meaning, I'm not "mirroring" >> another repository.) One example would be that I want to have my >> own repository of the IBM TSM client RPMs, which one only gets by >> FTPing them from IBM's site: >> >> $ ls -1 *.rpm >> TIVsm-API64.i386.rpm >> TIVsm-API.i386.rpm >> TIVsm-BA.i386.rpm >> TIVsm-HSM.i386.rpm >> >> I'd like to have it so I could simply run "yum install tsm-client" >> on a system built from my cobbler server, and have the above RPMs >> and their dependencies (all included in rhel5 core) installed. >> This question might be best for a yum-specific list, but since >> cobbler has repository mirroring integrated, I'd give it a shot >> here. Is there an easy way to accomplish this? >> >> Thanks, >> -s- >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools >> > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From fj0588di at aa.jp.fujitsu.com Fri Feb 1 09:01:07 2008 From: fj0588di at aa.jp.fujitsu.com (S.Sakamoto) Date: Fri, 1 Feb 2008 18:01:07 +0900 Subject: [et-mgmt-tools] [PATCH] The return value of the argument error of option is wrong in virt-install. Message-ID: <200802011801.BBB09330.0GJE96K9@aa.jp.fujitsu.com> Hi, When I set a wrong value in option of virt-install,(ex. --cpuset=aaa, --mac=aaa) Error message appears and a command is finished, but the return value is 0. So, I make the patch which improved this problem so that a user is not confused. Thanks, Shigeki Sakamoto ========================================================== diff -r 9eccd3abccb6 virt-install --- a/virt-install Wed Jan 30 16:23:50 2008 -0500 +++ b/virt-install Thu Jan 31 16:07:44 2008 +0900 @@ -519,4 +519,5 @@ if __name__ == "__main__": sys.exit(e.code) except Exception, e: logging.exception(e) - + sys.exit(1) + From crobinso at redhat.com Fri Feb 1 14:59:54 2008 From: crobinso at redhat.com (Cole Robinson) Date: Fri, 01 Feb 2008 09:59:54 -0500 Subject: [et-mgmt-tools] [PATCH] The return value of the argument error of option is wrong in virt-install. In-Reply-To: <200802011801.BBB09330.0GJE96K9@aa.jp.fujitsu.com> References: <200802011801.BBB09330.0GJE96K9@aa.jp.fujitsu.com> Message-ID: <47A333EA.9010404@redhat.com> S.Sakamoto wrote: > Hi, > > When I set a wrong value in option of virt-install,(ex. --cpuset=aaa, --mac=aaa) > Error message appears and a command is finished, > but the return value is 0. > > So, I make the patch which improved this problem so that a user is not confused. > > > Thanks, > Shigeki Sakamoto > I've applied this: http://hg.et.redhat.com/virt/applications/virtinst--devel?cs=375d3f83844c Thanks! Cole From mdehaan at redhat.com Fri Feb 1 15:16:27 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 01 Feb 2008 10:16:27 -0500 Subject: [et-mgmt-tools] Creating a yum repo (via cobbler) from "scratch"? In-Reply-To: <47A2810A.2040905@gmail.com> References: <47A2810A.2040905@gmail.com> Message-ID: <47A337CB.100@redhat.com> Adam Rosenwald wrote: Thanks for the help Adam. A few quick comments: -- you don't actually have to call createrepo yourself, cobbler will do it for you. -- we don't currently have the ability to assign a list of repos to a distro, only a profile, however that might be a good idea for a future addition. -- there's a quick writeup on some of the repo management stuff here: http://cobbler.et.redhat.com/cobbler-repos.php --Michael From crobinso at redhat.com Fri Feb 1 23:19:24 2008 From: crobinso at redhat.com (Cole Robinson) Date: Fri, 01 Feb 2008 18:19:24 -0500 Subject: [et-mgmt-tools] [PATCH] virtinst: make hasFile not grab files Message-ID: <47A3A8FC.40703@redhat.com> The attached patch changes the implementation of hasFile from ImageFetcher classes to not actually acquire the file it is looking for, rather just test for its existence. http will use a HEAD request, local mounts will check the path, and ftp will query the size (which will work for files), and if that fails will try to cd to the location. I gave all the cases a test but I may be missing something, so if anyone has suggestions, please recommend :) This will be nice to have as it opens up some options, ex. quickly probing a tree for a paravirt kernel. Thanks, Cole # HG changeset patch # User "Cole Robinson " # Date 1201905006 18000 # Node ID 96dd51a7e06b2bdf0dbc9195d3047885f9f5eb2e # Parent 375d3f83844c8e374347d391fdc874c7b24fef1b Re-implement hasFile to not pull down any files, just test they exist. diff -r 375d3f83844c -r 96dd51a7e06b virtinst/DistroManager.py --- a/virtinst/DistroManager.py Fri Feb 01 09:56:47 2008 -0500 +++ b/virtinst/DistroManager.py Fri Feb 01 17:30:06 2008 -0500 @@ -32,7 +32,8 @@ from virtinst import _virtinst as _ from virtinst import _virtinst as _ from ImageFetcher import MountedImageFetcher -from ImageFetcher import URIImageFetcher +from ImageFetcher import FTPImageFetcher +from ImageFetcher import HTTPImageFetcher from ImageFetcher import DirectImageFetcher from OSDistro import FedoraDistro @@ -45,8 +46,10 @@ from OSDistro import MandrivaDistro from OSDistro import MandrivaDistro def _fetcherForURI(uri, scratchdir=None): - if uri.startswith("http://") or uri.startswith("ftp://"): - return URIImageFetcher(uri, scratchdir) + if uri.startswith("http://"): + return HTTPImageFetcher(uri, scratchdir) + elif uri.startswith("ftp://"): + return FTPImageFetcher(uri, scratchdir) elif uri.startswith("nfs://"): return MountedImageFetcher(uri, scratchdir) else: @@ -57,6 +60,7 @@ def _fetcherForURI(uri, scratchdir=None) def _storeForDistro(fetcher, baseuri, type, progresscb, distro=None, scratchdir=None): stores = [] + logging.debug("Attempting to detect distro:") if distro == "fedora" or distro is None: stores.append(FedoraDistro(baseuri, type, scratchdir)) if distro == "rhel" or distro is None: @@ -81,7 +85,7 @@ def _storeForDistro(fetcher, baseuri, ty raise ValueError, _("Could not find an installable distribution at '%s'" % baseuri) -# Method to fetch a krenel & initrd pair for a particular distro / HV type +# Method to fetch a kernel & initrd pair for a particular distro / HV type def acquireKernel(baseuri, progresscb, scratchdir="/var/tmp", type=None, distro=None): fetcher = _fetcherForURI(baseuri, scratchdir) diff -r 375d3f83844c -r 96dd51a7e06b virtinst/ImageFetcher.py --- a/virtinst/ImageFetcher.py Fri Feb 01 09:56:47 2008 -0500 +++ b/virtinst/ImageFetcher.py Fri Feb 01 17:30:06 2008 -0500 @@ -25,6 +25,9 @@ import subprocess import subprocess import urlgrabber.grabber as grabber import urlgrabber.progress as progress +import urllib2 +import urlparse +import ftplib import tempfile from virtinst import _virtinst as _ @@ -58,10 +61,10 @@ class ImageFetcher: def acquireFile(self, src, progresscb): raise "Must be implemented in subclass" - def hasFile(self, src, progresscb): + def hasFile(self, src): raise "Must be implemented in subclass" -# This is a fetcher capable of downloading from FTP / HTTP +# Base class for downloading from FTP / HTTP class URIImageFetcher(ImageFetcher): def prepareLocation(self, progresscb): @@ -72,7 +75,8 @@ class URIImageFetcher(ImageFetcher): return True except IOError, e: logging.debug("Opening URL %s failed." % (self.location,) + " " + str(e)) - raise ValueError(_("Opening URL %s failed.") % (self.location,)) + raise ValueError(_("Opening URL %s failed: %s") % \ + (self.location, e)) return False def acquireFile(self, filename, progresscb): @@ -85,7 +89,7 @@ class URIImageFetcher(ImageFetcher): progress_obj = progresscb, \ text = _("Retrieving file %s...") % base) except IOError, e: - raise ValueError, _("Invalid URL location given: %s %s") %\ + raise ValueError, _("Couldn't aquire file %s: %s") %\ ((self.location + "/" + filename), str(e)) tmpname = self.saveTemp(file, prefix=base + ".") logging.debug("Saved file to " + tmpname) @@ -94,15 +98,35 @@ class URIImageFetcher(ImageFetcher): if file: file.close() - def hasFile(self, filename, progresscb): - try: - tmpfile = self.acquireFile(filename, progresscb) - os.unlink(tmpfile) - return True +class HTTPImageFetcher(URIImageFetcher): + + def hasFile(self, filename): + try: + request = urllib2.Request(self.location + "/" + filename) + request.get_method = lambda: "HEAD" + http_file = urllib2.urlopen(request) except Exception, e: - logging.debug("Cannot find file %s" % filename) - return False - + logging.debug("HTTP hasFile: didn't find %s" % \ + (self.location + "/" + filename)) + return False + return True + +class FTPImageFetcher(URIImageFetcher): + + def hasFile(self, filename): + url = urlparse.urlparse(self.location + "/" + filename) + try: + ftp = ftplib.FTP(url[1]) + ftp.login() + try: + ftp.size(url[2]) # If a file + except ftplib.all_errors, e: + ftp.cwd(url[2]) # If a dir + except ftplib.all_errors, e: + logging.debug("FTP hasFile: couldn't access %s/%s" % \ + (url[1], url[2])) + return False + return True class LocalImageFetcher(ImageFetcher): @@ -132,14 +156,12 @@ class LocalImageFetcher(ImageFetcher): if file: file.close() - def hasFile(self, filename, progresscb): - try: - tmpfile = self.acquireFile(filename, progresscb) - if tmpfile is not None: - os.unlink(tmpfile) + def hasFile(self, filename): + if os.path.exists(os.path.abspath(self.srcdir + "/" + filename)): return True - except Exception, e: - logging.debug("Cannot find file %s" % filename) + else: + logging.debug("local hasFile: Couldn't find %s" % \ + (self.srcdir + "/" + filename)) return False # This is a fetcher capable of extracting files from a NFS server diff -r 375d3f83844c -r 96dd51a7e06b virtinst/OSDistro.py --- a/virtinst/OSDistro.py Fri Feb 01 09:56:47 2008 -0500 +++ b/virtinst/OSDistro.py Fri Feb 01 17:30:06 2008 -0500 @@ -26,7 +26,6 @@ import tempfile import tempfile from virtinst import _virtinst as _ -from ImageFetcher import ImageFetcher # An image store is a base class for retrieving either a bootable # ISO image, or a kernel+initrd pair for a particular OS distribution @@ -80,10 +79,10 @@ class RedHatDistro(Distro): # Fedora distro check class FedoraDistro(RedHatDistro): def isValidStore(self, fetcher, progresscb): - if fetcher.hasFile("fedora.css", progresscb): + if fetcher.hasFile("fedora.css"): logging.debug("Detected a Fedora distro") return True - if fetcher.hasFile("Fedora", progresscb): + if fetcher.hasFile("Fedora"): logging.debug("Detected a Fedora distro") return True return False @@ -91,13 +90,13 @@ class FedoraDistro(RedHatDistro): # Fedora distro check class RHELDistro(RedHatDistro): def isValidStore(self, fetcher, progresscb): - if fetcher.hasFile("Server", progresscb): + if fetcher.hasFile("Server"): logging.debug("Detected a RHEL 5 Server distro") return True - if fetcher.hasFile("Client", progresscb): + if fetcher.hasFile("Client"): logging.debug("Detected a RHEL 5 Client distro") return True - if fetcher.hasFile("RedHat", progresscb): + if fetcher.hasFile("RedHat"): logging.debug("Detected a RHEL 4 distro") return True return False @@ -105,7 +104,7 @@ class RHELDistro(RedHatDistro): # CentOS distro check class CentOSDistro(RedHatDistro): def isValidStore(self, fetcher, progresscb): - if fetcher.hasFile("CentOS", progresscb): + if fetcher.hasFile("CentOS"): logging.debug("Detected a CentOS distro") return True return False @@ -186,9 +185,9 @@ class SuseDistro(Distro): kernelrpm = dir + "/" + filename if kernelrpm is None: - raise _("Unable to determine kernel RPM path") + raise Exception(_("Unable to determine kernel RPM path")) if installinitrdrpm is None: - raise _("Unable to determine install-initrd RPM path") + raise Exception(_("Unable to determine install-initrd RPM path")) return (kernelrpm, installinitrdrpm) finally: filelistData.close() @@ -309,18 +308,9 @@ class SuseDistro(Distro): def isValidStore(self, fetcher, progresscb): # Suse distros always have a 'directory.yast' file in the top # level of install tree, which we use as the magic check - ignore = None - try: - try: - ignore = fetcher.acquireFile("directory.yast", progresscb) - logging.debug("Detected a Suse distro") - return True - except ValueError, e: - logging.debug("Doesn't look like a Suse distro " + str(e)) - pass - finally: - if ignore is not None: - os.unlink(ignore) + if fetcher.hasFile("directory.yast"): + logging.debug("Detected a Suse distro.") + return True return False @@ -333,7 +323,14 @@ class DebianDistro(Distro): file = None try: try: - file = fetcher.acquireFile("current/images/MANIFEST", progresscb) + file = None + if fetcher.hasFile("current/images/MANIFEST"): + file = fetcher.acquireFile("current/images/MANIFEST", + progresscb) + else: + logging.debug("Doesn't look like a Debian distro.") + return False + except ValueError, e: logging.debug("Doesn't look like a Debian distro " + str(e)) return False From ssklar at stanford.edu Sat Feb 2 00:13:40 2008 From: ssklar at stanford.edu (Sandor W. Sklar) Date: Fri, 1 Feb 2008 16:13:40 -0800 Subject: [et-mgmt-tools] Creating a yum repo (via cobbler) from "scratch"? In-Reply-To: <47A337CB.100@redhat.com> References: <47A2810A.2040905@gmail.com> <47A337CB.100@redhat.com> Message-ID: <3D305095-9D60-4074-9F83-D4A93293B846@stanford.edu> On Feb 1, 2008, at 7:16 AM, Michael DeHaan wrote: > Adam Rosenwald wrote: > > Thanks for the help Adam. Yes, again, thank you very much! > > > A few quick comments: > -- you don't actually have to call createrepo yourself, cobbler will > do it for you. OK, that is good to know. > > -- we don't currently have the ability to assign a list of repos to > a distro, only a profile, however that might be a good idea for a > future addition. > > -- there's a quick writeup on some of the repo management stuff > here: http://cobbler.et.redhat.com/cobbler-repos.php Thanks, I did read through that page (honest!) My new question is, is there any way around keeping two copies of each RPM? For example, I created my repo "base" in /var/repo/tsm-client/base, and then ran "cobbler repo add ..." and "cobbler reposync", which all worked perfectly. Now, I have ... # find /var -name TIVsm-API64.i386.rpm /var/repo/tsm-client/base/TIVsm-API64.i386.rpm /var/www/cobbler/repo_mirror/tsm-client/base/TIVsm-API64.i386.rpm ... two copies of each of the RPMs. I'd love to be able to delete the /var/repo/tsm-client directory, and if I have newer or changed RPMs to go in that repository, to be able to just drop them into /var/ www/cobbler/repo_mirror/tsm-client/base/ and then run something like "cobbler reposync" to have it automagically find the new files and do the right thing. Any chance that this is actually possible? Thanks, again ... I'm still amazed at how quickly I was able to get this far! -s- From ssklar at stanford.edu Sat Feb 2 01:35:00 2008 From: ssklar at stanford.edu (Sandor W. Sklar) Date: Fri, 1 Feb 2008 17:35:00 -0800 Subject: [et-mgmt-tools] Changing items in /tftpboot/pxelinux.cfg/{MACADDRESS} Message-ID: <248DB899-AE7B-4304-8215-8C18EBFD0C2B@stanford.edu> We've got a bunch of Sun X4200 M2 systems, and on these systems, the device physically labeled "NET0" actually comes up as eth2, when booted from any RH-like operating system. My problem is I can't determine a way of changing this value in the / tftpboot/pxelinux.cfg/MA:CA:DD:RE:SS file that gets generated by cobbler. I wind up with an entry like ... append ksdevice=eth0 lang= console=ttyS0,9600n8 text syslog=xxxxx.xxxx.xxx:25150 kssendmac initrd=/images/rhel5Server- x86_64/initrd.img ks=http://xxxxx.xxxx.xxx/cblr/kickstarts_sys/xxxxx.xxx.xxx/ks.cfg I don't want to change the "default" file in there, because for most of my systems, eth0 is the right device. I tried setting -- kopts="ksdevice=eth2" but that didn't make a difference in the pxelinux.cfg/... file. Am I missing something obvious? Thanks, -s- From ssklar at stanford.edu Sun Feb 3 05:58:30 2008 From: ssklar at stanford.edu (Sandor W. Sklar) Date: Sat, 2 Feb 2008 21:58:30 -0800 Subject: [et-mgmt-tools] Changing items in /tftpboot/pxelinux.cfg/{MACADDRESS} In-Reply-To: <248DB899-AE7B-4304-8215-8C18EBFD0C2B@stanford.edu> References: <248DB899-AE7B-4304-8215-8C18EBFD0C2B@stanford.edu> Message-ID: <20464F52-87A8-4164-B4B4-40DD57F7DB3C@stanford.edu> Sorry for responding to my own question, but I figured it out. Setting --kopts="ksdevice=eth2" does work, but ONLY if you also specify the MAC address when you add the system. -s- On Feb 1, 2008, at 5:35 PM, Sandor W. Sklar wrote: > > We've got a bunch of Sun X4200 M2 systems, and on these systems, the > device physically labeled "NET0" actually comes up as eth2, when > booted from any RH-like operating system. > > My problem is I can't determine a way of changing this value in the / > tftpboot/pxelinux.cfg/MA:CA:DD:RE:SS file that gets generated by > cobbler. I wind up with an entry like ... > > append ksdevice=eth0 lang= console=ttyS0,9600n8 text > syslog=xxxxx.xxxx.xxx:25150 kssendmac initrd=/images/rhel5Server- > x86_64/initrd.img ks=http://xxxxx.xxxx.xxx/cblr/kickstarts_sys/xxxxx.xxx.xxx/ks.cfg > > I don't want to change the "default" file in there, because for most > of my systems, eth0 is the right device. I tried setting -- > kopts="ksdevice=eth2" but that didn't make a difference in the > pxelinux.cfg/... file. > > Am I missing something obvious? Thanks, > > -s- > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From Hua.Zhang at Sun.COM Sun Feb 3 15:59:34 2008 From: Hua.Zhang at Sun.COM (Henry Zhang) Date: Sun, 03 Feb 2008 23:59:34 +0800 Subject: [et-mgmt-tools] Patch for serial console of virt-manager 0.4 for Solaris In-Reply-To: <20080201152647.742E173712@hormel.redhat.com> References: <20080201152647.742E173712@hormel.redhat.com> Message-ID: <47A5E4E6.2030301@sun.com> Hi there, I am not sure if I have provided this patch of virt-manager 0.4 for Solaris before, anyway let's list the patch here, and hope you all help to review. When I tried to run virt-manager 0.4 at Solaris, (we have not port 0.5.* to Solaris since there is some dependencies not ready.) we find the serial console can't run, so added the below patch, after that the serial console can run at Solaris, I took a quick test in Fedora, seems it also can run. If there is any problem, please let me know.. Thanks, Henry -------------- next part -------------- A non-text attachment was scrubbed... Name: serialcon.diff Type: text/x-patch Size: 1182 bytes Desc: not available URL: From berrange at redhat.com Sun Feb 3 17:42:25 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Sun, 3 Feb 2008 17:42:25 +0000 Subject: [et-mgmt-tools] Patch for serial console of virt-manager 0.4 for Solaris In-Reply-To: <47A5E4E6.2030301@sun.com> References: <20080201152647.742E173712@hormel.redhat.com> <47A5E4E6.2030301@sun.com> Message-ID: <20080203174225.GA9615@redhat.com> On Sun, Feb 03, 2008 at 11:59:34PM +0800, Henry Zhang wrote: > Hi there, > > I am not sure if I have provided this patch of virt-manager 0.4 for > Solaris before, anyway let's list the patch here, and hope you all help > to review. > > > When I tried to run virt-manager 0.4 at Solaris, (we have not port 0.5.* > to Solaris since there is some dependencies not ready.) we find the > serial console can't run, so added the below patch, after that the > serial console can run at Solaris, I took a quick test in Fedora, seems > it also can run. > > If there is any problem, please let me know.. It looks good - I'll apply it to the repo. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From Darrick.Waller at baesystems.com Mon Feb 4 13:44:39 2008 From: Darrick.Waller at baesystems.com (Waller, Darrick J (US SSA)) Date: Mon, 4 Feb 2008 07:44:39 -0600 Subject: [et-mgmt-tools] Changing items in/tftpboot/pxelinux.cfg/{MACADDRESS} References: <248DB899-AE7B-4304-8215-8C18EBFD0C2B@stanford.edu> <20464F52-87A8-4164-B4B4-40DD57F7DB3C@stanford.edu> Message-ID: <04799F26A23174449975FB1913E3A02F155D88@gldms20030.goldlnk.rootlnka.net> IIRC, you can also do ksdevice=link, and it will pick whichever network interface is live. -Darrick -----Original Message----- From: et-mgmt-tools-bounces at redhat.com [mailto:et-mgmt-tools-bounces at redhat.com] On Behalf Of Sandor W. Sklar Sent: Saturday, February 02, 2008 11:59 PM To: Fedora/Linux Management Tools Subject: Re: [et-mgmt-tools] Changing items in/tftpboot/pxelinux.cfg/{MACADDRESS} Sorry for responding to my own question, but I figured it out. Setting --kopts="ksdevice=eth2" does work, but ONLY if you also specify the MAC address when you add the system. -s- On Feb 1, 2008, at 5:35 PM, Sandor W. Sklar wrote: > > We've got a bunch of Sun X4200 M2 systems, and on these systems, the > device physically labeled "NET0" actually comes up as eth2, when > booted from any RH-like operating system. > > My problem is I can't determine a way of changing this value in the / > tftpboot/pxelinux.cfg/MA:CA:DD:RE:SS file that gets generated by > cobbler. I wind up with an entry like ... > > append ksdevice=eth0 lang= console=ttyS0,9600n8 text > syslog=xxxxx.xxxx.xxx:25150 kssendmac initrd=/images/rhel5Server- > x86_64/initrd.img ks=http://xxxxx.xxxx.xxx/cblr/kickstarts_sys/xxxxx.xxx.xxx/ks.cfg > > I don't want to change the "default" file in there, because for most > of my systems, eth0 is the right device. I tried setting -- > kopts="ksdevice=eth2" but that didn't make a difference in the > pxelinux.cfg/... file. > > Am I missing something obvious? Thanks, > > -s- > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools _______________________________________________ et-mgmt-tools mailing list et-mgmt-tools at redhat.com https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mdehaan at redhat.com Mon Feb 4 15:29:43 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 04 Feb 2008 10:29:43 -0500 Subject: [et-mgmt-tools] Changing items in /tftpboot/pxelinux.cfg/{MACADDRESS} In-Reply-To: <20464F52-87A8-4164-B4B4-40DD57F7DB3C@stanford.edu> References: <248DB899-AE7B-4304-8215-8C18EBFD0C2B@stanford.edu> <20464F52-87A8-4164-B4B4-40DD57F7DB3C@stanford.edu> Message-ID: <47A72F67.1090508@redhat.com> Sandor W. Sklar wrote: > Sorry for responding to my own question, but I figured it out. > > Setting --kopts="ksdevice=eth2" does work, but ONLY if you also > specify the MAC address when you add the system. > > -s- Right. Either the mac or the ip is needed so that syslinux can pick up the right settings file. You could also optionally add it to the profile and pick the profile from the boot menu (type menu while booting to activate the PXE menu). If you want to default to ksdevice=link across the board, you can change that value in /var/lib/cobbler/settings --Michael From ssklar at stanford.edu Mon Feb 4 16:11:10 2008 From: ssklar at stanford.edu (Sandor W. Sklar) Date: Mon, 4 Feb 2008 08:11:10 -0800 Subject: [et-mgmt-tools] Changing items in/tftpboot/pxelinux.cfg/{MACADDRESS} In-Reply-To: <04799F26A23174449975FB1913E3A02F155D88@gldms20030.goldlnk.rootlnka.net> References: <248DB899-AE7B-4304-8215-8C18EBFD0C2B@stanford.edu> <20464F52-87A8-4164-B4B4-40DD57F7DB3C@stanford.edu> <04799F26A23174449975FB1913E3A02F155D88@gldms20030.goldlnk.rootlnka.net> Message-ID: <8B540DE7-0D60-4935-96F0-55475DE605BF@stanford.edu> On Feb 4, 2008, at 5:44 AM, Waller, Darrick J (US SSA) wrote: > IIRC, you can also do ksdevice=link, and it will pick whichever > network > interface is live. Oh, that would be totally excellent. I'll try that out, thanks! -s- From rjones at redhat.com Mon Feb 4 17:39:57 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Mon, 04 Feb 2008 17:39:57 +0000 Subject: [et-mgmt-tools] Virt-p2v 0.9.1 - a physical-to-virtual migration live CD Message-ID: <47A74DED.8070803@redhat.com> I'm pleased to announce the latest version of virt-p2v, the open source physical-to-virtual migration live CD. http://et.redhat.com/~rjones/virt-p2v/ This is almost a complete rewrite, although largely based on the code in the old shell script. Amongst the new features are: - writes a configuration file for Xen/QEMU/KVM - auto-detects operating system(s) present on the machine - configurable so you can build customer-specific live CDs - auto-detects CPU & memory - support "ISO attachments" for quick development updates - allowed to specify an SSH username (not just 'root') - progress bar during copy - hopefully fixed the "stops at 4GB" bug reported by another user - lots and lots of bugfixes Documentation: http://et.redhat.com/~rjones/virt-p2v/virt-p2v.1.html Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3237 bytes Desc: S/MIME Cryptographic Signature URL: From mdehaan at redhat.com Mon Feb 4 19:11:20 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 04 Feb 2008 14:11:20 -0500 Subject: [et-mgmt-tools] Creating a yum repo (via cobbler) from "scratch"? In-Reply-To: <3D305095-9D60-4074-9F83-D4A93293B846@stanford.edu> References: <47A2810A.2040905@gmail.com> <47A337CB.100@redhat.com> <3D305095-9D60-4074-9F83-D4A93293B846@stanford.edu> Message-ID: <47A76358.1060707@redhat.com> Sandor W. Sklar wrote: > > On Feb 1, 2008, at 7:16 AM, Michael DeHaan wrote: > >> Adam Rosenwald wrote: >> >> Thanks for the help Adam. > > Yes, again, thank you very much! >> >> >> A few quick comments: >> -- you don't actually have to call createrepo yourself, cobbler will >> do it for you. > > OK, that is good to know. > >> >> -- we don't currently have the ability to assign a list of repos to a >> distro, only a profile, however that might be a good idea for a >> future addition. >> >> -- there's a quick writeup on some of the repo management stuff here: >> http://cobbler.et.redhat.com/cobbler-repos.php > > Thanks, I did read through that page (honest!) My new question is, is > there any way around keeping two copies of each RPM? For example, I > created my repo "base" in /var/repo/tsm-client/base, and then ran > "cobbler repo add ..." and "cobbler reposync", which all worked > perfectly. Cobbler currently doesn't have the code to represent a repo it is not mirroring. There's an RFE filed on that though. So you just want to throw stuff in /var/www/mystuff/reponame and run creatrepo there manually, and then mention that in your kickstart templates manually until that is in there. --Michael From jtatpbl at gmail.com Mon Feb 4 19:45:13 2008 From: jtatpbl at gmail.com (JimmyT _) Date: Mon, 4 Feb 2008 11:45:13 -0800 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! Message-ID: (moving this discussion thread from private email & irc chats to this list, per request) ------------------------------------------------------------------------------- if I, cd /tmp/xen wget http://mirrors.kernel.org/fedora/releases/8/Fedora/x86_64/os/images/xen/vmlinuz wget http://mirrors.kernel.org/fedora/releases/8/Fedora/x86_64/os/images/xen/initrd.img ls -al /tmp/xen cobbler distro add --name=f8 \ --kernel='/tmp/xen/vmlinuz' --initrd='/tmp/xen/initrd.img' \ --arch=x86_64 --breed=redhat cobbler profile add --name=f8pv --distro=f8 --virt-type=xenpv \ --virt-ram=512 --virt-cpus=1 --virt-bridge=eth0 --kickstart=/tmp/f8pv_ks.cfg cobbler sync ls -al /tmp/xen -rw-r--r-- 1 root root 6735017 2007-11-02 08:00 initrd.img -rw-r--r-- 1 root root 1961293 2007-11-02 08:00 vmlinuz koan --virt --virt-name=testvm --profile=f8pv --server=server.internal.net \ --virt-path=/dev/VG01/testvm \ --autonet --nogfx FAILS: - using kickstart: http://10.0.0.100/cblr/kickstarts/f8_min_pv/ks.cfg libvirtd (pid 2322) is running... downloading initrd initrd.img to /var/lib/xen/initrd.img url=http://server.internal.net/cobbler/images/fedora8/initrd.img - using kickstart: http://server.internal.net/cobbler/images/fedora8/initrd.img downloading kernel vmlinuz to /var/lib/xen/vmlinuz url=http://server.internal.net/cobbler/images/fedora8/vmlinuz - using kickstart: http://server.internal.net/cobbler/images/fedora8/vmlinuz libvir: Xen Daemon error : GET operation failed: libvir: Xen Daemon error : GET operation failed: libvir: Xen Daemon error : POST operation failed: (xend.err "Error creating domain: (2, 'Invalid kernel', 'xc_dom_find_loader: no loader found\\n')") Traceback (most recent call last): File "/usr/lib/python2.5/site-packages/koan/app.py", line 192, in main k.run() File "/usr/lib/python2.5/site-packages/koan/app.py", line 402, in run self.virt() File "/usr/lib/python2.5/site-packages/koan/app.py", line 670, in virt return self.net_install(after_download) File "/usr/lib/python2.5/site-packages/koan/app.py", line 624, in net_install after_download(self, profile_data) File "/usr/lib/python2.5/site-packages/koan/app.py", line 668, in after_download self.virt_net_install(profile_data) File "/usr/lib/python2.5/site-packages/koan/app.py", line 1243, in virt_net_install fullvirt = fullvirt File "/usr/lib/python2.5/site-packages/koan/xencreate.py", line 128, in start_install guest.start_install() File "/usr/lib/python2.5/site-packages/virtinst/Guest.py", line 813, in start_install return self._do_install(consolecb, meter) File "/usr/lib/python2.5/site-packages/virtinst/Guest.py", line 834, in _do_install self.domain = self.conn.createLinux(install_xml, 0) File "/usr/lib64/python2.5/site-packages/libvirt.py", line 585, in createLinux if ret is None:raise libvirtError('virDomainCreateLinux() failed', conn=self) libvirtError: virDomainCreateLinux() failed POST operation failed: (xend.err "Error creating domain: (2, 'Invalid kernel', 'xc_dom_find_loader: no loader found\\n')") Checking, ls -al /var/lib/xen/{vmlinuz*,initrd*} -rw-r--r-- 1 root root 609 2008-02-01 16:56 /var/lib/xen/initrd.img -rw-r--r-- 1 root root 609 2008-02-01 16:56 /var/lib/xen/vmlinuz Those sizes are clearly WRONG! What's "in there"? Oddly, cat /var/lib/xen/{vmlinuz,initrd.img) 400 Bad Request

Bad Request

Your browser sent a request that this server could not understand.
Reason: You're speaking plain HTTP to an SSL-enabled server port.
Instead use the HTTPS scheme to access this URL, please.

Hint: https://server.internal.net/

Apache/2.2.6 (Fedora) Server at server.internal.net Port 443
400 Bad Request

Bad Request

Your browser sent a request that this server could not understand.
Reason: You're speaking plain HTTP to an SSL-enabled server port.
Instead use the HTTPS scheme to access this URL, please.

Hint: https://server.internal.net/

Apache/2.2.6 (Fedora) Server at server.internal.net Port 443
This is happening, I believe because koan's DL'ing from: url=http://server.internal.net/cobbler/images/fedora8/initrd.img rather than, url=https://server.internal.net/cobbler/images/fedora8/initrd.img Checking, if I nav to: links http://server.internal.net/cobbler/images/fedora8/ I see, 400 Bad Request Bad Request Your browser sent a request that this server could not understand. Reason: You're speaking plain HTTP to an SSL-enabled server port. Instead use the HTTPS scheme to access this URL, please. Hint: https://server.internal.net/ --------------------------------------------------------------------------------------------------------------------------------------------------------------- Apache/2.2.6 (Fedora) Server at server.internal.net Port 443 which is what's in those "errant" koan DL's ... Whereas if I instead nav to: links https://server.internal.net/cobbler/images/fedora8/ i see as expected, Index of /cobbler/images/fedora8 Index of /cobbler/images/fedora8 [ICO] Name Last modified Size Description --------------------------------------------------------- [DIR] Parent Directory - [ ] initrd.img 01-Feb-2008 16:50 6.4M [ ] vmlinuz 01-Feb-2008 16:50 1.9M --------------------------------------------------------- Apache/2.2.6 (Fedora) Server at server.internal.net Port 443 Bottom line ... koan's NOT getting files via the required https://.... And, this, of course, would explain the "invalid kernel" messages I've been seeing. Now, the question is config change in koan/cobbler? apache? or a koan source problem? HTH, Jimmy ------------------------------------------------------------------------------- So to summarize, you have port 80 configured to be HTTPS and you have shut http:// off? To get things going you'll need to turn it back on in the Cobbler server. The address of the cobbler server and the port is configured in settings, but the URLs generated will be http://. It does not need to be port 80, but if it's not, you need to change the port value in /var/lib/cobbler/settings. This probably explains both your transfer error and the XMLRPC messages. It would be possible to allow cobbler to generate https:// URLs, though there is not much point in doing so -- Anaconda itself does not support authentication. --Michael ------------------------------------------------------------------------------- Michael, I decided to step away from it myself, and have someone else here setup a new/clean box. again, vanilla f8 + apache. They've pulled latest git heads of cobbler/koan. Now, @ their "koan ...", koan --virt --virt-name=test --server=http://server3.internal.net --profile=f8test \ --virt-path=/dev/VG03/test --autonet --nogfx Could not communicate with http://server3.internal.net:25151 checking @ server3 service iptables status iptables: Firewall is not running. service ip6tables status ip6tables: Firewall is not running. and, telnet server3.internal.net 25151 Trying 10.0.0.203... Connected to server3.internal.net. Escape character is '^]'. has something changed? ------------------------------------------------------------------------------- Is cobbler check clean? Is cobblerd running? ------------------------------------------------------------------------------- yes to both. cobbler check No setup problems found Manual review and editing of /var/lib/cobbler/settings is recommended to tailor cobbler to your particular configuration. Good luck. service cobblerd status cobblerd (pid 7761 7760 7759 7758) is running... ------------------------------------------------------------------------------- Can you actually telnet to that port from the remote box? FYI: Q&A should go to the public list. ------------------------------------------------------------------------------- yes. that's what the above, telnet server3.internal.net 25151 Trying 10.0.0.203... Connected to server3.internal.net. Escape character is '^]'. was. sorry, did not make it cleat that the telnet was from a remote. ------------------------------------------------------------------------------- From mdehaan at redhat.com Mon Feb 4 19:45:44 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 04 Feb 2008 14:45:44 -0500 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: References: Message-ID: <47A76B68.6000307@redhat.com> JimmyT _ wrote: > (moving this discussion thread from private email & irc chats to this > list, per request) > > > ------------------------------------------------------------------------------- > if I, > > cd /tmp/xen > wget http://mirrors.kernel.org/fedora/releases/8/Fedora/x86_64/os/images/xen/vmlinuz > wget http://mirrors.kernel.org/fedora/releases/8/Fedora/x86_64/os/images/xen/initrd.img > > ls -al /tmp/xen > > cobbler distro add --name=f8 \ > --kernel='/tmp/xen/vmlinuz' --initrd='/tmp/xen/initrd.img' \ > --arch=x86_64 --breed=redhat > > cobbler profile add --name=f8pv --distro=f8 --virt-type=xenpv \ > --virt-ram=512 --virt-cpus=1 --virt-bridge=eth0 --kickstart=/tmp/f8pv_ks.cfg > > cobbler sync > > ls -al /tmp/xen > -rw-r--r-- 1 root root 6735017 2007-11-02 08:00 initrd.img > -rw-r--r-- 1 root root 1961293 2007-11-02 08:00 vmlinuz > > koan --virt --virt-name=testvm --profile=f8pv --server=server.internal.net \ > --virt-path=/dev/VG01/testvm \ > --autonet --nogfx > > > FAILS: > > - using kickstart: http://10.0.0.100/cblr/kickstarts/f8_min_pv/ks.cfg > libvirtd (pid 2322) is running... > downloading initrd initrd.img to /var/lib/xen/initrd.img > url=http://server.internal.net/cobbler/images/fedora8/initrd.img > - using kickstart: > http://server.internal.net/cobbler/images/fedora8/initrd.img > downloading kernel vmlinuz to /var/lib/xen/vmlinuz > url=http://server.internal.net/cobbler/images/fedora8/vmlinuz > - using kickstart: http://server.internal.net/cobbler/images/fedora8/vmlinuz > libvir: Xen Daemon error : GET operation failed: > libvir: Xen Daemon error : GET operation failed: > libvir: Xen Daemon error : POST operation failed: (xend.err "Error > creating domain: (2, 'Invalid kernel', 'xc_dom_find_loader: no loader > found\\n')") > Traceback (most recent call last): > File "/usr/lib/python2.5/site-packages/koan/app.py", line 192, in main > k.run() > File "/usr/lib/python2.5/site-packages/koan/app.py", line 402, in run > self.virt() > File "/usr/lib/python2.5/site-packages/koan/app.py", line 670, in virt > return self.net_install(after_download) > File "/usr/lib/python2.5/site-packages/koan/app.py", line 624, in > net_install > after_download(self, profile_data) > File "/usr/lib/python2.5/site-packages/koan/app.py", line 668, in > after_download > self.virt_net_install(profile_data) > File "/usr/lib/python2.5/site-packages/koan/app.py", line 1243, > in virt_net_install > fullvirt = fullvirt > File "/usr/lib/python2.5/site-packages/koan/xencreate.py", line > 128, in start_install > guest.start_install() > File "/usr/lib/python2.5/site-packages/virtinst/Guest.py", line > 813, in start_install > return self._do_install(consolecb, meter) > File "/usr/lib/python2.5/site-packages/virtinst/Guest.py", line > 834, in _do_install > self.domain = self.conn.createLinux(install_xml, 0) > File "/usr/lib64/python2.5/site-packages/libvirt.py", line 585, > in createLinux > if ret is None:raise libvirtError('virDomainCreateLinux() > failed', conn=self) > libvirtError: virDomainCreateLinux() failed POST operation failed: > (xend.err "Error creating domain: (2, 'Invalid kernel', > 'xc_dom_find_loader: no loader found\\n')") > > > Checking, > > ls -al /var/lib/xen/{vmlinuz*,initrd*} > -rw-r--r-- 1 root root 609 2008-02-01 16:56 /var/lib/xen/initrd.img > -rw-r--r-- 1 root root 609 2008-02-01 16:56 /var/lib/xen/vmlinuz > > Those sizes are clearly WRONG! What's "in there"? > > > > This is happening, I believe because koan's DL'ing from: > > url=http://server.internal.net/cobbler/images/fedora8/initrd.img > This is the URLs it should use. > Checking, if I nav to: > > links http://server.internal.net/cobbler/images/fedora8/ > > I see, > > > 400 Bad Request > Bad > Request > > Your browser sent a request that this server could not > understand. > Reason: You're speaking plain HTTP to an SSL-enabled server > port. > Instead use the HTTPS scheme to access this URL, please. > > Hint: https://server.internal.net/ > This seems to indicate you have HTTPS running on port 80. This would be bad. This seems to be an Apache/site configuration issue which probably was also indiciate of the error you were seeing earlier -- namely that Apache Reverse proxying of your XMLRPC connection was not working, but seems to work for all other installs. > > Bottom line ... koan's NOT getting files via the required https://.... > > There is no reason why https:// should be required. ... snip ... If your Apache configuration is still http/https confused, koan is trying to contact the Apache proxy of the XMLRPC connection so that will still fail. I'm putting my money on the Apache config and port 80 not being plaintext. --Michael From jtatpbl at gmail.com Mon Feb 4 20:04:08 2008 From: jtatpbl at gmail.com (JimmyT _) Date: Mon, 4 Feb 2008 12:04:08 -0800 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <47A76B68.6000307@redhat.com> References: <47A76B68.6000307@redhat.com> Message-ID: The full configuration for those files that cobblerd serves up via apache should be provided in *it's* config file, /etc/httpd/conf.d/cobbler.conf. No? Or are you suggesting that we need to otherwise configure the stock/vanilla install of apache -- outside of the cobbler.conf? And, is that behind the new/latest: "Could not communicate with http://server3.internal.net:25151" issue as well? From mdehaan at redhat.com Mon Feb 4 20:10:41 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 04 Feb 2008 15:10:41 -0500 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: References: <47A76B68.6000307@redhat.com> Message-ID: <47A77141.3060001@redhat.com> > Or are you suggesting that we need to otherwise configure the > stock/vanilla install of apache -- outside of the cobbler.conf? > I'm thinking your main Apache config isn't actually stock, yes. > And, is that behind the new/latest: > > "Could not communicate with http://server3.internal.net:25151" > > issue as well? > You had it working before when it tried to download the kernel/initrd. What did you change? From jtatpbl at gmail.com Mon Feb 4 20:27:27 2008 From: jtatpbl at gmail.com (JimmyT _) Date: Mon, 4 Feb 2008 12:27:27 -0800 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <47A77141.3060001@redhat.com> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> Message-ID: I told you what we changed .... a new box, with a fresh download, and a vanilla install. Based on your reposnse, it's clear the perception is that "it's us". Fair enough. We'll stick to supported products. Thanks. From rainer at ultra-secure.de Mon Feb 4 21:01:55 2008 From: rainer at ultra-secure.de (Rainer Duffner) Date: Mon, 4 Feb 2008 22:01:55 +0100 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> Message-ID: <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> Am 04.02.2008 um 21:27 schrieb JimmyT _: > I told you what we changed .... a new box, with a fresh download, and > a vanilla install. > > Based on your reposnse, it's clear the perception is that "it's us". > > Fair enough. We'll stick to supported products. > > Do physical installs work? It works here, basically out-of-the-box. Sometimes, though, the RPC-XML interface can get "cobbled" and I've got to restart httpd+cobblerd. Haven't figured out why, though. But I don't do any kind of Xen-installations. And I use RHEL5 as base. Fedora is too much a moving target for us. I really wish I had something like cobbler for my FreeBSD-installations. There's a lot I don't like about Linux - but RHELs kickstart + cobbler works _very_ well. And I really get better "support" from Michael et.al. via IRC/ML than I get from other, wanna-be vendors via a paid contract. cheers, Rainer -- Rainer Duffner CISSP, LPI, MCSE rainer at ultra-secure.de From mdehaan at redhat.com Mon Feb 4 21:00:37 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 04 Feb 2008 16:00:37 -0500 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> Message-ID: <47A77CF5.100@redhat.com> Rainer Duffner wrote: > > Am 04.02.2008 um 21:27 schrieb JimmyT _: > >> I told you what we changed .... a new box, with a fresh download, and >> a vanilla install. >> >> Based on your reposnse, it's clear the perception is that "it's us". >> >> Fair enough. We'll stick to supported products. >> >> > > > Do physical installs work? > It works here, basically out-of-the-box. > > Sometimes, though, the RPC-XML interface can get "cobbled" and I've > got to restart httpd+cobblerd. > Haven't figured out why, though. Development cobbler from git (0.7.X, 0.8 candidate) has some logging updates that also include much better remote exception handling, so stay tuned for that, I suspect that should go away. I haven't seen any issues with cobblerd from the development build which I've been running in a /long/ time. I do recall some issues with cobblerd in previous releases. > > I really wish I had something like cobbler for my FreeBSD-installations. I'm not sure I've mentioned that to you yet or not (I know I have some other folks) -- Cobbler already has support in there to do SuSE/Debian type installs (they take an answer file, even if it's not technically a kickstart). If there's an easy enough way to adapt the code (mainly action_sync.py and item_distro.py) to grok the BSD differences, I'd be willing to take incorporate it. Honestly I'm not all that familiar with what it takes to network install BSDs -- patches definitely welcome though. > There's a lot I don't like about Linux - but RHELs kickstart + cobbler > works _very_ well. > And I really get better "support" from Michael et.al. via IRC/ML than > I get from other, wanna-be vendors via a paid contract. > Glad to hear that. Thanks! > > > cheers, > Rainer From rainer at ultra-secure.de Mon Feb 4 21:33:04 2008 From: rainer at ultra-secure.de (Rainer Duffner) Date: Mon, 4 Feb 2008 22:33:04 +0100 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <47A77CF5.100@redhat.com> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> <47A77CF5.100@redhat.com> Message-ID: <3918C3AE-3546-4CBA-B678-301D0BF3A195@ultra-secure.de> Am 04.02.2008 um 22:00 schrieb Michael DeHaan: > Rainer Duffner wrote: >> >> Am 04.02.2008 um 21:27 schrieb JimmyT _: >> >>> I told you what we changed .... a new box, with a fresh download, >>> and >>> a vanilla install. >>> >>> Based on your reposnse, it's clear the perception is that "it's us". >>> >>> Fair enough. We'll stick to supported products. >>> >>> >> >> >> Do physical installs work? >> It works here, basically out-of-the-box. >> >> Sometimes, though, the RPC-XML interface can get "cobbled" and >> I've got to restart httpd+cobblerd. >> Haven't figured out why, though. > > Development cobbler from git (0.7.X, 0.8 candidate) has some > logging updates that also include much better remote exception > handling, so stay tuned for that, I suspect that should go away. I > haven't seen any issues with cobblerd from the development build > which I've been running in a /long/ time. I do recall some issues > with cobblerd in previous releases. It's not very often. Maybe also was only in the vmware-image I tested this all with... > >> >> I really wish I had something like cobbler for my FreeBSD- >> installations. > > I'm not sure I've mentioned that to you yet or not (I know I have > some other folks) -- > > Cobbler already has support in there to do SuSE/Debian type > installs (they take an answer file, even if it's not technically a > kickstart). If there's an easy enough way to adapt > the code (mainly action_sync.py and item_distro.py) to grok the BSD > differences, I'd be willing to take incorporate it. Honestly I'm > not all that familiar with what it takes to > network install BSDs -- patches definitely welcome though. The problem is, that FreeBSD doesn't use pxelinux to boot - it has its own PXE-bootloader. Then, the equivalent of the ks.cfg file goes into the compressed msfroot filesystem image, which would be a major pain to adapt. It also likes to work via FTP only (all the example scripts have NFS commented out and setup FTP instead - as if everybody had tried and then settled for the obvious. I must look at that at some point - there's a howto about converting a FreeBSD-ISO for PXE-booting with pxelinux here: http:// phaq.phunsites.net/?s=pxe > >> There's a lot I don't like about Linux - but RHELs kickstart + >> cobbler works _very_ well. > >> And I really get better "support" from Michael et.al. via IRC/ML >> than I get from other, wanna-be vendors via a paid contract. >> > Glad to hear that. Thanks! >> >> Credit where credit is due ;-) cheers, Rainer -- Rainer Duffner CISSP, LPI, MCSE rainer at ultra-secure.de From mdehaan at redhat.com Mon Feb 4 21:31:08 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 04 Feb 2008 16:31:08 -0500 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <3918C3AE-3546-4CBA-B678-301D0BF3A195@ultra-secure.de> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> <47A77CF5.100@redhat.com> <3918C3AE-3546-4CBA-B678-301D0BF3A195@ultra-secure.de> Message-ID: <47A7841C.1060302@redhat.com> > > > The problem is, that FreeBSD doesn't use pxelinux to boot - it has its > own PXE-bootloader. > Then, the equivalent of the ks.cfg file goes into the compressed > msfroot filesystem image, which would be a major pain to adapt. The seperately bootloader itself would not be a show stopper -- there is already some code there to supply elilo and alternate config files as opposed to syslinux. The ks.cfg problem, however would be annoying... however that doesn't sound that far off from what koan does with initrd's to get around DHCP timeouts. One potential solution would be to try to make a ks.cfg equivalent that includes the rest of the "kickstart" contents based on the result of a CGI script, so you only have to generate the initrd once. That's ugly though. The problem with generating all variances server side (which is why cobbler doesn't do it server side) for PXE setups (as opposed to reinstalls) is that you end up generating an initrd for every profile/system and that takes up a lot of space ... more importantly, it's just a slow thing to do. > It also likes to work via FTP only (all the example scripts have NFS > commented out and setup FTP instead - as if everybody had tried and > then settled for the obvious. > > I must look at that at some point - there's a howto about converting a > FreeBSD-ISO for PXE-booting with pxelinux here: > http://phaq.phunsites.net/?s=pxe Fun... From jwildebo at redhat.com Mon Feb 4 22:29:38 2008 From: jwildebo at redhat.com (Jan Wildeboer) Date: Mon, 04 Feb 2008 23:29:38 +0100 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> Message-ID: <1202164178.4093.49.camel@jhwx61.muc.redhat.com> On Mon, 2008-02-04 at 22:01 +0100, Rainer Duffner wrote: > Sometimes, though, the RPC-XML interface can get "cobbled" and I've > got to restart httpd+cobblerd. > Haven't figured out why, though. Sounds familiar. Please do a netstat -alp | grep 25 when the xmlrpc gets b0rked. In my case it *seems* that 25151 has been switched from python to dhcpd ?!? I am guessing that cobbler sync somehow mixes the ports sometimes, but I also didn't have time to investigate. Jan -- Jan H Wildeboer | Solution Architect, RHCE | Office: +49 (0)89 205071-207 Red Hat GmbH | Mobile: +49 (0)174 33 23 249 Otto-Hahn-Str.20 | Fax: +49 (0)89 205071-111 D-85609 Dornach/Munich | eMail: jan.wildeboer at redhat.com _____________________________________________________________________ GPG-Key-ID: 5DEBAFB0 GPG-Fingerprint: 6104 0F74 8513 F17E DFD5 E820 6F61 A078 5DEB AFB0 _____________________________________________________________________ Reg. Adresse: Red Hat GmbH, Hauptstaetter Strasse 58, 70178 Stuttgart Handelsregister: Amtsgericht Stuttgart HRB 153243 Geschaeftsfuehrer: Brendan Lane, Charlie Peters, Michael Cunningham, Werner Knoblich _____________________________________________________________________ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From mdehaan at redhat.com Mon Feb 4 22:24:00 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 04 Feb 2008 17:24:00 -0500 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <1202164178.4093.49.camel@jhwx61.muc.redhat.com> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> <1202164178.4093.49.camel@jhwx61.muc.redhat.com> Message-ID: <47A79080.3020401@redhat.com> Jan Wildeboer wrote: > On Mon, 2008-02-04 at 22:01 +0100, Rainer Duffner wrote: > > >> Sometimes, though, the RPC-XML interface can get "cobbled" and I've >> got to restart httpd+cobblerd. >> Haven't figured out why, though. >> > > Sounds familiar. > > Please do a netstat -alp | grep 25 when the xmlrpc gets b0rked. In my > case it *seems* that 25151 has been switched from python to dhcpd ?!? > > I am guessing that cobbler sync somehow mixes the ports sometimes, but I > also didn't have time to investigate. > That does not seem possible. I can believe it flipping out due to connection problems though (again, should be much happier in 0.7.X due to the new exception handling code). From clalance at redhat.com Mon Feb 4 22:51:58 2008 From: clalance at redhat.com (Chris Lalancette) Date: Mon, 04 Feb 2008 17:51:58 -0500 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <47A79080.3020401@redhat.com> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> <1202164178.4093.49.camel@jhwx61.muc.redhat.com> <47A79080.3020401@redhat.com> Message-ID: <47A7970E.7040303@redhat.com> Michael DeHaan wrote: >> >> Please do a netstat -alp | grep 25 when the xmlrpc gets b0rked. In my >> case it *seems* that 25151 has been switched from python to dhcpd ?!? >> >> I am guessing that cobbler sync somehow mixes the ports sometimes, but I >> also didn't have time to investigate. >> > > That does not seem possible. I can believe it flipping out due to > connection problems though (again, should be much happier in 0.7.X due > to the new exception handling code). (Disclaimer: I know nothing about cobbler internals in particular, this is just a general piece of information) Actually, it is possible, but it's not due to "mixed up ports". If cobbler has port 25151 open -> accept, and then it forks/execs dhcpd, then dhcpd can inherit that. I'm not sure if that is the case here, but that is one way it can happen. Chris Lalancette From jwildebo at redhat.com Mon Feb 4 23:06:53 2008 From: jwildebo at redhat.com (Jan Wildeboer) Date: Tue, 05 Feb 2008 00:06:53 +0100 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <47A79080.3020401@redhat.com> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> <1202164178.4093.49.camel@jhwx61.muc.redhat.com> <47A79080.3020401@redhat.com> Message-ID: <1202166414.4093.58.camel@jhwx61.muc.redhat.com> On Mon, 2008-02-04 at 17:24 -0500, Michael DeHaan wrote: > That does not seem possible. I can believe it flipping out due to > connection problems though (again, should be much happier in 0.7.X due > to the new exception handling code). I know it seems impossible. Let me sum up what I did, what I saw and how I corrected it. This is cobbler 0.6.5.1 After I had some problems with the web-ui (it seems impossible to delete distros, so I edited the distros file in /var/lib/cobbler manually) I decided to get some sleep. Next morning I logged in to cobbler and clicked on distros, after 30 seconds I got an error along the lines of "XMLRPC auth error, check your apache logs". I did netstat -alp | grep 25 and noticed something odd. Netstat claimed that port 25152 was owned by dhcpd. Hrmpf. Coffee. service cobblerd stop service dhcpd restart service cobblerd start netstat -alp | grep 25 All is fine again. Whatever. It *seems* to happen after I have clicked sync in the webui a few times. It is not really reproducable, but it happened a few times in the last week. HTH Jan -- Jan H Wildeboer | Solution Architect, RHCE | Office: +49 (0)89 205071-207 Red Hat GmbH | Mobile: +49 (0)174 33 23 249 Otto-Hahn-Str.20 | Fax: +49 (0)89 205071-111 D-85609 Dornach/Munich | eMail: jan.wildeboer at redhat.com _____________________________________________________________________ GPG-Key-ID: 5DEBAFB0 GPG-Fingerprint: 6104 0F74 8513 F17E DFD5 E820 6F61 A078 5DEB AFB0 _____________________________________________________________________ Reg. Adresse: Red Hat GmbH, Hauptstaetter Strasse 58, 70178 Stuttgart Handelsregister: Amtsgericht Stuttgart HRB 153243 Geschaeftsfuehrer: Brendan Lane, Charlie Peters, Michael Cunningham, Werner Knoblich _____________________________________________________________________ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From mdehaan at redhat.com Mon Feb 4 23:05:11 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 04 Feb 2008 18:05:11 -0500 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <1202166414.4093.58.camel@jhwx61.muc.redhat.com> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> <1202164178.4093.49.camel@jhwx61.muc.redhat.com> <47A79080.3020401@redhat.com> <1202166414.4093.58.camel@jhwx61.muc.redhat.com> Message-ID: <47A79A27.80104@redhat.com> > > I did netstat -alp | grep 25 and noticed something odd. Netstat claimed > that port 25152 was owned by dhcpd. > Not possible :) From jwildebo at redhat.com Mon Feb 4 23:22:33 2008 From: jwildebo at redhat.com (Jan Wildeboer) Date: Tue, 05 Feb 2008 00:22:33 +0100 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <47A79A27.80104@redhat.com> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> <1202164178.4093.49.camel@jhwx61.muc.redhat.com> <47A79080.3020401@redhat.com> <1202166414.4093.58.camel@jhwx61.muc.redhat.com> <47A79A27.80104@redhat.com> Message-ID: <1202167353.4093.60.camel@jhwx61.muc.redhat.com> On Mon, 2008-02-04 at 18:05 -0500, Michael DeHaan wrote: > > > > I did netstat -alp | grep 25 and noticed something odd. Netstat claimed > > that port 25152 was owned by dhcpd. > > > > Not possible :) And what is this? [root at vinf02 ~]# netstat -alp | grep 25 tcp 0 0 localhost.localdomain:25152 *:* LISTEN 28404/python tcp 0 0 *:25151 *:* LISTEN 28405/python tcp 0 0 localhost.localdomain:25152 localhost.localdomain:33122 TIME_WAIT - tcp 0 0 localhost.localdomain:33120 localhost.localdomain:25152 TIME_WAIT - tcp 0 0 localhost.localdomain:25152 localhost.localdomain:33126 TIME_WAIT - tcp 0 0 localhost.localdomain:25152 localhost.localdomain:33124 TIME_WAIT - tcp 0 0 vinf02.coe.muc.redhat.:http ::ffff:172.17.252.11:57593 TIME_WAIT - udp 0 0 *:25150 *:* 28402/python unix 2 [ ] DGRAM 3825 1144/rpc.statd [root at vinf02 ~]# service cobblerd restart Stopping cobbler daemon: [ OK ] Starting cobbler daemon: [ OK ] [root at vinf02 ~]# netstat -alp | grep 25 tcp 0 0 localhost.localdomain:25152 *:* LISTEN 29980/dhcpd tcp 0 0 *:25151 *:* LISTEN 30016/python tcp 0 0 localhost.localdomain:25152 localhost.localdomain:33122 TIME_WAIT - tcp 0 0 localhost.localdomain:33120 localhost.localdomain:25152 TIME_WAIT - tcp 0 0 localhost.localdomain:25152 localhost.localdomain:33126 TIME_WAIT - tcp 0 0 localhost.localdomain:25152 localhost.localdomain:33124 TIME_WAIT - tcp 0 0 vinf02.coe.muc.redhat.:http ::ffff:172.17.252.11:57593 TIME_WAIT - udp 0 0 *:25150 *:* 30013/python unix 2 [ ] DGRAM 3825 1144/rpc.statd -- Jan H Wildeboer | Solution Architect, RHCE | Office: +49 (0)89 205071-207 Red Hat GmbH | Mobile: +49 (0)174 33 23 249 Otto-Hahn-Str.20 | Fax: +49 (0)89 205071-111 D-85609 Dornach/Munich | eMail: jan.wildeboer at redhat.com _____________________________________________________________________ GPG-Key-ID: 5DEBAFB0 GPG-Fingerprint: 6104 0F74 8513 F17E DFD5 E820 6F61 A078 5DEB AFB0 _____________________________________________________________________ Reg. Adresse: Red Hat GmbH, Hauptstaetter Strasse 58, 70178 Stuttgart Handelsregister: Amtsgericht Stuttgart HRB 153243 Geschaeftsfuehrer: Brendan Lane, Charlie Peters, Michael Cunningham, Werner Knoblich _____________________________________________________________________ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From jwildebo at redhat.com Mon Feb 4 23:34:21 2008 From: jwildebo at redhat.com (Jan Wildeboer) Date: Tue, 05 Feb 2008 00:34:21 +0100 Subject: [et-mgmt-tools] re: "invalid kernel" -- found the problem! In-Reply-To: <1202167353.4093.60.camel@jhwx61.muc.redhat.com> References: <47A76B68.6000307@redhat.com> <47A77141.3060001@redhat.com> <0DE70409-12B7-48E4-AE62-180E5231EA30@ultra-secure.de> <1202164178.4093.49.camel@jhwx61.muc.redhat.com> <47A79080.3020401@redhat.com> <1202166414.4093.58.camel@jhwx61.muc.redhat.com> <47A79A27.80104@redhat.com> <1202167353.4093.60.camel@jhwx61.muc.redhat.com> Message-ID: <1202168061.4093.66.camel@jhwx61.muc.redhat.com> Some more details. I hammered cobbler sync, initiated by webui and console and could not reproduce. But a simple service cobbler restart did the trick. Now I have dhcpd listening on 25152 and a b0rk3d xmlrpc for cobbler. In the cobbler.log I see (not surprising): 2008-02-05 00:26:30,696 - cobbler.cobblerd - DEBUG - started 2008-02-05 00:26:30,865 - cobbler.cobblerd - INFO - XMLRPC running on 25151 2008-02-05 00:26:30,868 - cobbler.cobblerd - INFO - Exception occured: socket.error 2008-02-05 00:26:30,869 - cobbler.cobblerd - INFO - Exception value: (98, 'Address already in use') 2008-02-05 00:26:30,871 - cobbler.cobblerd - INFO - syslog running on 25150 2008-02-05 00:26:30,893 - cobbler.cobblerd - INFO - Exception Info: File "/usr/bin/cobblerd", line 66, in ? app.core(logger=logger) File "/usr/lib/python2.4/site-packages/cobbler/cobblerd.py", line 44, in core do_xmlrpc_tasks(bootapi, settings, xmlrpc_port, xmlrpc_port2, logger) File "/usr/lib/python2.4/site-packages/cobbler/cobblerd.py", line 55, in do_xmlrpc_tasks do_xmlrpc_rw(bootapi, settings, xmlrpc_port2, logger) File "/usr/lib/python2.4/site-packages/cobbler/cobblerd.py", line 114, in do_xmlrpc_rw server = remote.CobblerReadWriteXMLRPCServer(('127.0.0.1', port)) File "/usr/lib/python2.4/site-packages/cobbler/remote.py", line 805, in __init__ SimpleXMLRPCServer.SimpleXMLRPCServer.__init__(self,args) File "/usr/lib/python2.4/SimpleXMLRPCServer.py", line 473, in __init__ SocketServer.TCPServer.__init__(self, addr, requestHandler) File "/usr/lib/python2.4/SocketServer.py", line 330, in __init__ self.server_bind() File "/usr/lib/python2.4/SocketServer.py", line 341, in server_bind self.socket.bind(self.server_address) File "", line 1, in bind service dhcpd restart cleans up that part: [root at vinf02 ~]# service dhcpd restart Shutting down dhcpd: [ OK ] Starting dhcpd: [ OK ] [root at vinf02 ~]# netstat -alp | grep 25 tcp 0 0 *:25151 *:* LISTEN 30016/python udp 0 0 *:25150 *:* 30013/python But now still a missing 25152, so lets try with [root at vinf02 ~]# service cobblerd stop Stopping cobbler daemon: [ OK ] [root at vinf02 ~]# service cobblerd start Starting cobbler daemon: [ OK ] [root at vinf02 ~]# netstat -alp | grep 25 tcp 0 0 localhost.localdomain:25152 *:* LISTEN 30105/python tcp 0 0 *:25151 *:* LISTEN 30106/python udp 0 0 *:25150 *:* 30103/python unix 2 [ ] DGRAM 3825 1144/rpc.statd All fine again. Now for debugging that - not today, I am off to bed. Jan -- Jan H Wildeboer | Solution Architect, RHCE | Office: +49 (0)89 205071-207 Red Hat GmbH | Mobile: +49 (0)174 33 23 249 Otto-Hahn-Str.20 | Fax: +49 (0)89 205071-111 D-85609 Dornach/Munich | eMail: jan.wildeboer at redhat.com _____________________________________________________________________ GPG-Key-ID: 5DEBAFB0 GPG-Fingerprint: 6104 0F74 8513 F17E DFD5 E820 6F61 A078 5DEB AFB0 _____________________________________________________________________ Reg. Adresse: Red Hat GmbH, Hauptstaetter Strasse 58, 70178 Stuttgart Handelsregister: Amtsgericht Stuttgart HRB 153243 Geschaeftsfuehrer: Brendan Lane, Charlie Peters, Michael Cunningham, Werner Knoblich _____________________________________________________________________ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From sebastia at l00-bugdead-prods.de Tue Feb 5 13:18:01 2008 From: sebastia at l00-bugdead-prods.de (Sebastian Reitenbach) Date: Tue, 05 Feb 2008 14:18:01 +0100 Subject: [et-mgmt-tools] problem to connect to remote xend from virt-manager Message-ID: <20080205131802.68D7646C9A@smtp.l00-bugdead-prods.de> Hi, I am connecting from an opensuse 10.3 to a sles10sp1, using virt-manager. In the connection configuration I chose "Remote tunnel over SSH". On my first try, it did not find "nc" on the remote sles host. That was easy to fix, create a symlink from netcat to nc. But then, when I now try, it seems to try to use "nc" with an invalid option: virt-manager Password: nc: invalid option -- U Try `nc --help' for more information. libvir: Remote error : socket closed unexpectedly netcat-1.10 is installed on both hosts, on the client, libvirt-0.3.0-30.5, virt-manager-0.4.0-68 are installed, on the xen server, virt-manager-0.3.1-0.27, and libvirt-0.3.3-18.4 is installed. I've seen on OpenBSD, the netcat has option -U, which is there use a unix domain socket. Is there anything I can do about that? cheers Sebastian From agoodall at rm.com Tue Feb 5 15:49:31 2008 From: agoodall at rm.com (Adam Goodall) Date: Tue, 5 Feb 2008 15:49:31 -0000 Subject: [et-mgmt-tools] cobbler and multiple rhn repositories Message-ID: <565F003189BD8D4A971B2AEE3550ED7A1374B5@mail05.internal.rmplc.net> Hi I'm currently testing cobbler and so far it has done everything ive asked of it flawlessly. There is however one function I need some help with. Im trying to get it to pull down multiple rhn channels so that I can deploy/patch both rhel4 and rhel5 from the cobbler server. The box itself is RHEL 5 so it has no problem mirroring rhn://rhel-i386-server-5 I would also however like to mirror rhn://rhel-i386-es-4 , Whenever I try and run a reposync with this repo added, I of course get told that the system isn't entitled to download the rhel4 content. Using mrepo's tools and the instructions at: http://www.labmonkeys.org/systems_management:patch_management:installing _mrepo_on_rhel5 I have managed to get a separate systemid for the box which is allowed to pull down rhel4 content, and I can successfully create a local mirror of rhel-i386-es-4 in that fashion. I would however like to add this into my cobbler setup with the minimum of fuss. I could of course tell cobbler to create a repo from the local files but that means keeping 2 copies of the rpm's. Or I could just add the repository manually in the post of all my kickstart files. But both of these methods involve a bit more management as I still need to make sure that the rhel4 mirror is correctly synching - would be nice it it "just worked" Does anyone know of a way to get cobbler to pull down content from the rhel4 channel using the alternative systemid? Or of any other ingenious methods to achieve what I want? Thanks Adam ______________________________________________________________________ You might be interested in this... Over the next few months RM is hosting a number of free and exciting events... ...to ensure you don't miss out, visit: http://www.rm.com/busandconferences ______________________________________________________________________ P.S. Think Green - don't print this email unless you really need to. This message is confidential, so please treat it appropriately and for its intended purpose only. In particular, if it refers to any technical data, terms or prices not generally available or known, such items are "commercially sensitive information" within the terms of the Freedom of Information Act 2000 and related laws. As it would be prejudicial to RM's commercial interests if these were disclosed, please refrain from doing so. As Internet communications are not secure, please be aware that RM cannot accept responsibility for its contents. Any views or opinions presented are those of the author only and not of RM. If you are not the intended recipient of this e-mail, please accept our apologies and arrange for copies of it to be deleted. For your information, RM may intercept incoming and outgoing email communications. RM Education plc Registered Office: New Mill House, 183 Milton Park, Abingdon, Oxfordshire, OX14 4SE, England Registered Number: 1148594 -------------- next part -------------- An HTML attachment was scrubbed... URL: From marcmo at foxriver.com Tue Feb 5 16:21:53 2008 From: marcmo at foxriver.com (Marc Mondragon) Date: Tue, 5 Feb 2008 10:21:53 -0600 Subject: [et-mgmt-tools] RE: et-mgmt-tools Digest, Vol 18, Issue 7 In-Reply-To: <20080205155021.6176C7352D@hormel.redhat.com> References: <20080205155021.6176C7352D@hormel.redhat.com> Message-ID: FWIW, I have seen this as well ... usually after a weekend. In my mind I have been blaming logrotate. My fix is the same as Jan's, stop dhcpd, stop cobblerd, start cobblerd and finally start dhcpd and all is well. I have been running this on a RHEL 5 server if it matters. Marc Mondragon Fox River Partners LLC./CRT Trading LLC. 2100 Enterprise Avenue Geneva, IL 60134 marcmo at foxriver.com > Date: Tue, 05 Feb 2008 00:34:21 +0100 > From: Jan Wildeboer > Subject: Re: [et-mgmt-tools] re: "invalid kernel" -- found the > problem! > To: Fedora/Linux Management Tools > Message-ID: <1202168061.4093.66.camel at jhwx61.muc.redhat.com> > Content-Type: text/plain; charset="us-ascii" > > Some more details. > > I hammered cobbler sync, initiated by webui and console and could not > reproduce. > > But a simple service cobbler restart did the trick. Now I have dhcpd > listening on 25152 and a b0rk3d xmlrpc for cobbler. > > In the cobbler.log I see (not surprising): > > 2008-02-05 00:26:30,696 - cobbler.cobblerd - DEBUG - started > 2008-02-05 00:26:30,865 - cobbler.cobblerd - INFO - XMLRPC running on > 25151 > 2008-02-05 00:26:30,868 - cobbler.cobblerd - INFO - Exception occured: > socket.error > 2008-02-05 00:26:30,869 - cobbler.cobblerd - INFO - Exception value: > (98, 'Address already in use') > 2008-02-05 00:26:30,871 - cobbler.cobblerd - INFO - syslog running on > 25150 > 2008-02-05 00:26:30,893 - cobbler.cobblerd - INFO - Exception Info: > File "/usr/bin/cobblerd", line 66, in ? > app.core(logger=logger) > File "/usr/lib/python2.4/site-packages/cobbler/cobblerd.py", line 44, > in core > do_xmlrpc_tasks(bootapi, settings, xmlrpc_port, xmlrpc_port2, > logger) > File "/usr/lib/python2.4/site-packages/cobbler/cobblerd.py", line 55, > in do_xmlrpc_tasks > do_xmlrpc_rw(bootapi, settings, xmlrpc_port2, logger) > File "/usr/lib/python2.4/site-packages/cobbler/cobblerd.py", line > 114, in do_xmlrpc_rw > server = remote.CobblerReadWriteXMLRPCServer(('127.0.0.1', port)) > File "/usr/lib/python2.4/site-packages/cobbler/remote.py", line 805, > in __init__ > SimpleXMLRPCServer.SimpleXMLRPCServer.__init__(self,args) > File "/usr/lib/python2.4/SimpleXMLRPCServer.py", line 473, in > __init__ > SocketServer.TCPServer.__init__(self, addr, requestHandler) > File "/usr/lib/python2.4/SocketServer.py", line 330, in __init__ > self.server_bind() > File "/usr/lib/python2.4/SocketServer.py", line 341, in server_bind > self.socket.bind(self.server_address) > File "", line 1, in bind > > service dhcpd restart cleans up that part: > > [root at vinf02 ~]# service dhcpd restart > Shutting down dhcpd: [ OK ] > Starting dhcpd: [ OK ] > [root at vinf02 ~]# netstat -alp | grep 25 > tcp 0 0 *:25151 *:* > LISTEN 30016/python > udp 0 0 *:25150 *:* > 30013/python > > But now still a missing 25152, so lets try with > > [root at vinf02 ~]# service cobblerd stop > Stopping cobbler daemon: [ OK ] > [root at vinf02 ~]# service cobblerd start > Starting cobbler daemon: [ OK ] > [root at vinf02 ~]# netstat -alp | grep 25 > tcp 0 0 localhost.localdomain:25152 *:* > LISTEN 30105/python > tcp 0 0 *:25151 *:* > LISTEN 30106/python > udp 0 0 *:25150 *:* > 30103/python > unix 2 [ ] DGRAM 3825 > 1144/rpc.statd > > > All fine again. > > Now for debugging that - not today, I am off to bed. > > Jan > > -- > Jan H Wildeboer | > Solution Architect, RHCE | Office: +49 (0)89 205071-207 > Red Hat GmbH | Mobile: +49 (0)174 33 23 249 > Otto-Hahn-Str.20 | Fax: +49 (0)89 205071-111 > D-85609 Dornach/Munich | eMail: jan.wildeboer at redhat.com From mdehaan at redhat.com Tue Feb 5 17:27:47 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 05 Feb 2008 12:27:47 -0500 Subject: [et-mgmt-tools] RE: et-mgmt-tools Digest, Vol 18, Issue 7 In-Reply-To: References: <20080205155021.6176C7352D@hormel.redhat.com> Message-ID: <47A89C93.50703@redhat.com> Marc Mondragon wrote: > FWIW, I have seen this as well ... usually after a weekend. In my mind > I have been blaming logrotate. My fix is the same as Jan's, stop dhcpd, > stop cobblerd, start cobblerd and finally start dhcpd and all is well. > Older versions of cobbler didn't restart cobblerd on logrotate. What version are you running? From rjones at redhat.com Tue Feb 5 19:44:28 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 05 Feb 2008 19:44:28 +0000 Subject: [et-mgmt-tools] Re: virt-p2v In-Reply-To: <44cad7820802051139m432ff636qa1d2ea3d7cf54287@mail.gmail.com> References: <44cad7820802051139m432ff636qa1d2ea3d7cf54287@mail.gmail.com> Message-ID: <47A8BC9C.20104@redhat.com> Andrew Peabody wrote: > Hi Richard, > > I just wanted to let you know that virt-p2v is really awesome, I had > been looking tool to do just this. Sorry if contacting you directly is > not the correct way to do this, but I could only find you email address > in the readme file. > > It appears that currently it cannot properly handle machines running on > linux software raid. The disks appear only as /dev/* partitions, and > running virt-p2v on them fails during the snapshot operation, probably > because the partitions are type mdraid. However even when the partition > is change to 83, it still acts really weird. Looking through virt-p2v > it appears that it can only handle regular partitions and lvm. I was > just curious if you had thought of adding support for md devices, > otherwise I might try to do it myself. Any help is always welcome. Out of interest, which version are you using? Is your root directory RAIDed too? For general problems, the right mailing list to use is: http://www.redhat.com/mailman/listinfo/et-mgmt-tools Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3237 bytes Desc: S/MIME Cryptographic Signature URL: From apeabody at systemsbiology.org Tue Feb 5 19:58:35 2008 From: apeabody at systemsbiology.org (Andrew Peabody) Date: Tue, 5 Feb 2008 11:58:35 -0800 Subject: [et-mgmt-tools] Re: virt-p2v In-Reply-To: <47A8BC9C.20104@redhat.com> References: <44cad7820802051139m432ff636qa1d2ea3d7cf54287@mail.gmail.com> <47A8BC9C.20104@redhat.com> Message-ID: <44cad7820802051158t2eeab152m260649864cdad1f8@mail.gmail.com> On 2/5/08, Richard W.M. Jones wrote: > > > Any help is always welcome. > > Out of interest, which version are you using? > > Is your root directory RAIDed too? > > For general problems, the right mailing list to use is: > > http://www.redhat.com/mailman/listinfo/et-mgmt-tools > > Rich. > > -- > Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ > Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod > Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in > England and Wales under Company Registration No. 03798903 > > Hi Rich, Thanks for the pointer to the mailing list. I was using 0.9.1 with a RAIDed root directory. For example /dev/md0 /boot /dev/md1 swap /dev/md2 / However, currently the live CD does not autostart the raid devices. Thanks, Andrew -------------- next part -------------- An HTML attachment was scrubbed... URL: From rjones at redhat.com Tue Feb 5 19:59:45 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 05 Feb 2008 19:59:45 +0000 Subject: [et-mgmt-tools] Re: virt-p2v In-Reply-To: <44cad7820802051153k2f4cd6a4j8fe01fcba812e3a1@mail.gmail.com> References: <44cad7820802051139m432ff636qa1d2ea3d7cf54287@mail.gmail.com> <47A8BC9C.20104@redhat.com> <44cad7820802051153k2f4cd6a4j8fe01fcba812e3a1@mail.gmail.com> Message-ID: <47A8C031.5080104@redhat.com> Andrew Peabody wrote: > Hi Rich, > > Thanks for the pointer to the mailing list. > > I was using 0.9.1 with a RAIDed root directory. For example > > /dev/md0 /boot > /dev/md1 swap > /dev/md2 / > > However, currently the live CD does not autostart the raid devices. Yes, it only considers /dev/hd* and /dev/sd* to be hard disk devices. Patches welcome to fix this one ... I may get around to getting up a RAID array to test this, but probably not very soon. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3237 bytes Desc: S/MIME Cryptographic Signature URL: From berrange at redhat.com Tue Feb 5 20:01:59 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 5 Feb 2008 20:01:59 +0000 Subject: [et-mgmt-tools] Re: virt-p2v In-Reply-To: <47A8C031.5080104@redhat.com> References: <44cad7820802051139m432ff636qa1d2ea3d7cf54287@mail.gmail.com> <47A8BC9C.20104@redhat.com> <44cad7820802051153k2f4cd6a4j8fe01fcba812e3a1@mail.gmail.com> <47A8C031.5080104@redhat.com> Message-ID: <20080205200159.GK20888@redhat.com> On Tue, Feb 05, 2008 at 07:59:45PM +0000, Richard W.M. Jones wrote: > Andrew Peabody wrote: > >Hi Rich, > > > >Thanks for the pointer to the mailing list. > > > >I was using 0.9.1 with a RAIDed root directory. For example > > > >/dev/md0 /boot > >/dev/md1 swap > >/dev/md2 / > > > >However, currently the live CD does not autostart the raid devices. > > Yes, it only considers /dev/hd* and /dev/sd* to be hard disk devices. Can't you just ask HAL for the list of harddisks so you get everything including /dev/hda*, /dev/sd*, /dev/md*, and even the very wierd compaq stuff like /dev/cciss/* Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From rjones at redhat.com Tue Feb 5 20:04:16 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 05 Feb 2008 20:04:16 +0000 Subject: [et-mgmt-tools] Re: Hello! I have been testing the latest version of virt-p2v In-Reply-To: <2cc75dd80802051154v2d074a83v8656e0bea6f84167@mail.gmail.com> References: <2cc75dd80801251033m31fc0f52rf0e5df522edd186c@mail.gmail.com> <47A762C8.9080303@redhat.com> <2cc75dd80802051051u4b825d19y82c17ea4dcdeb022@mail.gmail.com> <47A8B234.30604@redhat.com> <2cc75dd80802051104t1e92e905o466344ec8e0516db@mail.gmail.com> <47A8B39C.4080309@redhat.com> <2cc75dd80802051113s46d455b1m427e55c8e26f48f1@mail.gmail.com> <47A8B7FD.4080102@redhat.com> <2cc75dd80802051134i6175a573x1330a951c2eaafcc@mail.gmail.com> <47A8BB24.4060208@redhat.com> <2cc75dd80802051154v2d074a83v8656e0bea6f84167@mail.gmail.com> Message-ID: <47A8C140.2090200@redhat.com> Lester M. wrote: [I'm CC-ing this to the et-mgmt-tools list since it's better if everyone can follow these discussions] > Looking over the configuration file... > > I see something strange in my config generated... Hmmm... > > > */usr/lib64/xen/bin/qemu-dm* > > Heres the problem of the above: This is a 32bit dom0 and the above path > / binary does not exist on the box. > > > However, this *does *exist in "/usr/lib/xen/bin/qemu-dm".... going to > test it with this in the config. .... Works! woo hoo! Ah right, that's actually a bug in the code around here (line 1384 in the current development version): let devices = let emulator = match state.hypervisor with | Some Xen -> [leaf "emulator" "/usr/lib64/xen/bin/qemu-dm"] (* XXX lib64? *) The fix for this is for virt-p2v to connect to the remote hypervisor (using libvirt/libvirtd) and issue a virConnectGetCapabilities call. In the capabilities XML there is the correct emulator path. However I haven't got around to this yet. > Interesting minor configuration error. Something to ask what > architecture (i386 or x86_64) at the begining of the p2v could fix that > based on a user choice, That seemed to fix it! :-D > > I have a few other machines I can test addendums to virt-p2v. Not > everyone uses 64 bit dom0's even though everyone should. :-) Please do! Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3237 bytes Desc: S/MIME Cryptographic Signature URL: From marcmo at foxriver.com Tue Feb 5 20:37:52 2008 From: marcmo at foxriver.com (Marc Mondragon) Date: Tue, 5 Feb 2008 14:37:52 -0600 Subject: [et-mgmt-tools] RE: et-mgmt-tools Digest, Vol 18, Issue 9 In-Reply-To: <20080205200457.D9B0A73639@hormel.redhat.com> References: <20080205200457.D9B0A73639@hormel.redhat.com> Message-ID: RPM: cobbler-0.6.5-3.el5.noarch Or cobbler -v 0.605 Marc Mondragon Fox River Partners LLC./CRT Trading LLC. 2100 Enterprise Avenue Geneva, IL 60134 marcmo at foxriver.com > Date: Tue, 05 Feb 2008 12:27:47 -0500 > From: Michael DeHaan > Subject: Re: [et-mgmt-tools] RE: et-mgmt-tools Digest, Vol 18, Issue 7 > To: Fedora/Linux Management Tools > Message-ID: <47A89C93.50703 at redhat.com> > Content-Type: text/plain; charset=us-ascii; format=flowed > > Marc Mondragon wrote: > > FWIW, I have seen this as well ... usually after a weekend. In my mind > > I have been blaming logrotate. My fix is the same as Jan's, stop dhcpd, > > stop cobblerd, start cobblerd and finally start dhcpd and all is well. > > > > Older versions of cobbler didn't restart cobblerd on logrotate. What > version are you running? > > > From crobinso at redhat.com Tue Feb 5 20:56:06 2008 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 05 Feb 2008 15:56:06 -0500 Subject: [et-mgmt-tools] [PATCH] virtinst: make hasFile not grab files In-Reply-To: <47A3A8FC.40703@redhat.com> References: <47A3A8FC.40703@redhat.com> Message-ID: <47A8CD66.50009@redhat.com> Cole Robinson wrote: > The attached patch changes the implementation of hasFile from > ImageFetcher classes to not actually acquire the file it > is looking for, rather just test for its existence. http > will use a HEAD request, local mounts will check the path, > and ftp will query the size (which will work for files), > and if that fails will try to cd to the location. > > I gave all the cases a test but I may be missing something, > so if anyone has suggestions, please recommend :) This will > be nice to have as it opens up some options, ex. quickly > probing a tree for a paravirt kernel. > I've committed this. http://hg.et.redhat.com/virt/applications/virtinst--devel?cs=96dd51a7e06b - Cole From mdehaan at redhat.com Wed Feb 6 20:33:22 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 06 Feb 2008 15:33:22 -0500 Subject: [et-mgmt-tools] RE: et-mgmt-tools Digest, Vol 18, Issue 9 In-Reply-To: References: <20080205200457.D9B0A73639@hormel.redhat.com> Message-ID: <47AA1992.7090100@redhat.com> Marc Mondragon wrote: > RPM: cobbler-0.6.5-3.el5.noarch > > Looks like the postrotate restart bits where added in December, so you'll pick those up in the coming release. --Michael From rdtennent at gmail.com Wed Feb 6 22:17:40 2008 From: rdtennent at gmail.com (Bob Tennent) Date: Wed, 06 Feb 2008 17:17:40 -0500 Subject: [et-mgmt-tools] user access to virt-manager guests Message-ID: <47aa3204.ysAVFzGAbkMaaAl0%rdtennent@gmail.com> Hi. Is it possible to allow non-root users to access a guest operating system? Once the guest is set up, it should be possible to choose to allow the guest to look after its own security and authentication. As far as I can see, this isn't now possible. Perhaps some /etc/sudoers incantation? The difficulty is that virt-manager is used both to set up *and* to use guests; those functions should be separated. Bob T. From berrange at redhat.com Wed Feb 6 22:26:55 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 6 Feb 2008 22:26:55 +0000 Subject: [et-mgmt-tools] user access to virt-manager guests In-Reply-To: <47aa3204.ysAVFzGAbkMaaAl0%rdtennent@gmail.com> References: <47aa3204.ysAVFzGAbkMaaAl0%rdtennent@gmail.com> Message-ID: <20080206222655.GJ9469@redhat.com> On Wed, Feb 06, 2008 at 05:17:40PM -0500, Bob Tennent wrote: > Hi. Is it possible to allow non-root users to access a guest operating > system? Once the guest is set up, it should be possible to choose to > allow the guest to look after its own security and authentication. As > far as I can see, this isn't now possible. Perhaps some /etc/sudoers > incantation? The difficulty is that virt-manager is used both > to set up *and* to use guests; those functions should be separated. Users can access the guest via SSH (or any network login service running inside the guest), or they can use a standalone viewer program like virt-viewer to access the graphical console without needing admin privileges. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From rdtennent at gmail.com Thu Feb 7 04:33:24 2008 From: rdtennent at gmail.com (Bob Tennent) Date: Wed, 06 Feb 2008 23:33:24 -0500 Subject: [et-mgmt-tools] user access to virt-manager guests Message-ID: <47aa8a14.YY8a1gASU6u1sWPu%rdtennent@gmail.com> >|> Hi. Is it possible to allow non-root users to access a guest operating >|> system? Once the guest is set up, it should be possible to choose to >|> allow the guest to look after its own security and authentication. As >|> far as I can see, this isn't now possible. Perhaps some /etc/sudoers >|> incantation? The difficulty is that virt-manager is used both >|> to set up *and* to use guests; those functions should be separated. >| >|Users can access the guest via SSH (or any network login service >|running inside the guest), or they can use a standalone viewer program >|like virt-viewer to access the graphical console without needing admin >|privileges. Thanks. virt-viewer isn't mentioned in "Virtualization in Fedora 8". Bob T. From fj0588di at aa.jp.fujitsu.com Thu Feb 7 05:52:40 2008 From: fj0588di at aa.jp.fujitsu.com (S.Sakamoto) Date: Thu, 7 Feb 2008 14:52:40 +0900 Subject: [et-mgmt-tools] [PATCH] The bridge of bonding and the tag VLAN is not displayed in virt-manager Message-ID: <200802071452.HGG65121.69J0EKG9@aa.jp.fujitsu.com> Hi, virt-manager can display only the bridge of the physics network device now when add a device and make a guest domain newly. A bridge of bonding and Tag VLAN is not displayed. So, I make the patch that a bridge of bonding and Tag VLAN came to be displayed. Do not getout from processing specific linux... but can use a bridge of bonding and the tag VLAN. Thanks, Shigeki Sakamoto -------------- next part -------------- A non-text attachment was scrubbed... Name: connection.patch Type: application/octet-stream Size: 9445 bytes Desc: not available URL: From tim.verhoeven.be at gmail.com Thu Feb 7 09:20:04 2008 From: tim.verhoeven.be at gmail.com (Tim Verhoeven) Date: Thu, 7 Feb 2008 10:20:04 +0100 Subject: [et-mgmt-tools] [PATCH] Cobbler - webui - hide random mac link when editing system Message-ID: <2a7fce340802070120g24028c0cu3cd90be4c031b011@mail.gmail.com> Hi, Attached is a small patch for cobbler that hides the Random MAC link when editing systems. This has no use as the needed JS code is also disabled in that case. Regards, Tim -------------- next part -------------- A non-text attachment was scrubbed... Name: cobbler-webui-randommac-hidelink.patch Type: text/x-patch Size: 444 bytes Desc: not available URL: From berrange at redhat.com Thu Feb 7 12:56:23 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 7 Feb 2008 12:56:23 +0000 Subject: [et-mgmt-tools] [PATCH] The bridge of bonding and the tag VLAN is not displayed in virt-manager In-Reply-To: <200802071452.HGG65121.69J0EKG9@aa.jp.fujitsu.com> References: <200802071452.HGG65121.69J0EKG9@aa.jp.fujitsu.com> Message-ID: <20080207125623.GB30230@redhat.com> On Thu, Feb 07, 2008 at 02:52:40PM +0900, S.Sakamoto wrote: > Hi, > > virt-manager can display only the bridge of the physics network device now > when add a device and make a guest domain newly. > A bridge of bonding and Tag VLAN is not displayed. > > So, I make the patch that a bridge of bonding and Tag VLAN came to be displayed. > Do not getout from processing specific linux... > but can use a bridge of bonding and the tag VLAN. I'd definitely welcome code to deal with VLANs - and we're already Linux specific in this code, so that's not a problem. We can't use the 'rhpl*' python libraries though - these are Red Hat specific libraries, so if we use them, virt-manager won't be able to run on Debian/SUSE/etc If you can adapt the code to remove use of 'import rhpl' then that'd be good. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From rjones at redhat.com Thu Feb 7 16:13:48 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 07 Feb 2008 16:13:48 +0000 Subject: [et-mgmt-tools] Re: Hello! I have been testing the latest version of virt-p2v In-Reply-To: <2cc75dd80802070800n3e2871d4tcee1e538065492c@mail.gmail.com> References: <2cc75dd80801251033m31fc0f52rf0e5df522edd186c@mail.gmail.com> <47A8B234.30604@redhat.com> <2cc75dd80802051104t1e92e905o466344ec8e0516db@mail.gmail.com> <47A8B39C.4080309@redhat.com> <2cc75dd80802051113s46d455b1m427e55c8e26f48f1@mail.gmail.com> <47A8B7FD.4080102@redhat.com> <2cc75dd80802051134i6175a573x1330a951c2eaafcc@mail.gmail.com> <47A8BB24.4060208@redhat.com> <2cc75dd80802051154v2d074a83v8656e0bea6f84167@mail.gmail.com> <47A8C140.2090200@redhat.com> <2cc75dd80802070800n3e2871d4tcee1e538065492c@mail.gmail.com> Message-ID: <47AB2E3C.2030803@redhat.com> Lester M. wrote: > Quick question, are fedora users able to do paravirt migrations with fedora > 8 currently? I noticed that for fedora 8 in particular there is no xenU > kernel package or update, so when a system is migrated it is as a full-virt. For paravirt-to-paravirt you can do an ordinary migration, like this: virsh migrate [--live] guest xen://otherhost/ But I'm guessing you mean P2V, and converting the physical machine into a paravirt guest (rather than as at the moment into a fullvirt guest), and the answer to this is _yes_, but not as far as I know in Fedora. For RHEL you can install the PV drivers after doing the P2V. The eventual plan for virt-p2v is to be able to do this automatically during the P2V operation. Hence I added the RHEL PV binaries to virt-p2v in preparation for this: http://hg.et.redhat.com/virt/applications/virt-p2v--devel?cs=bc808b3656fb Fedora policy prevents these drivers from being included in Fedora, because they are not part of the Linus tree. You'd have to compile them yourself and maintain them (or help get them upstream ...) BTW, installing these drivers doesn't make the guest completely paravirtual. The guest will still look like a fullvirt guest, particularly w.r.t. how it handles the CPU & memory. However they do improve the important things -- disk and network -- which means you get almost all of the performance benefit of paravirtualization. > Oh, another thing, a cool feature would be to have a select choice of which > type of virtualization to use (para or full.). Its a nice feature to have, > but with the current "state of xen" I guess the biggest hurdle which may add > "bloat" to the iso image would be to include xenU kernels. :-( See above - yes this in the eventual plan for virt-p2v. > I remember reading *somewhere* about fedora being behind with xen based > kernels and backporting changes, would that impact that idea somewhat? Yes. The latest Linux kernel supported by XenSource is something ancient, and Red Hat devotes some developer resources to forward-porting Xen to newer Linux kernels, but even so we are still a bit behind. > I have not tried a rhel 5 system yet (I will today) I am guessing by default > they will be a full virt migration right? See above. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3237 bytes Desc: S/MIME Cryptographic Signature URL: From tom at ng23.net Thu Feb 7 17:16:23 2008 From: tom at ng23.net (Tom Brown) Date: Thu, 07 Feb 2008 17:16:23 +0000 Subject: [et-mgmt-tools] Have 1 cobbler server know about boxes built by another Message-ID: <47AB3CE7.9050202@ng23.net> Hi In my situation i have a 'build' network where bare metal gets done - Once this is built it is moved into the correct network whether that be dev, qa or prod or whatever. Once in those networks they can be rebuild, using koan, by build machines in those networks, we dont allow machines to straddle networks hence having different build machines in each network. Is there a way to let other cobbler servers know that a box is build using a different cobbler server? ie bare metal server tells prod server about all its machines so that prod server can rebuild them? does that make sense? thanks From mdehaan at redhat.com Thu Feb 7 17:27:58 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 07 Feb 2008 12:27:58 -0500 Subject: [et-mgmt-tools] Have 1 cobbler server know about boxes built by another In-Reply-To: <47AB3CE7.9050202@ng23.net> References: <47AB3CE7.9050202@ng23.net> Message-ID: <47AB3F9E.4070003@redhat.com> Tom Brown wrote: > Hi > > In my situation i have a 'build' network where bare metal gets done - > Once this is built it is moved into the correct network whether that > be dev, qa or prod or whatever. Once in those networks they can be > rebuild, using koan, by build machines in those networks, we dont > allow machines to straddle networks hence having different build > machines in each network. Is there a way to let other cobbler servers > know that a box is build using a different cobbler server? > You can solve that problem in a different way, without having to keep track of things. Basically cobbler doesn't keep inventory of what you have installed so that is up to you. However I'm not sure that's a problem. > ie bare metal server tells prod server about all its machines so that > prod server can rebuild them? > > does that make sense? An approach many sites take is to create on master cobbler server, and replicate other Cobbler install servers off of it using --server-override and --dhcp-tag. You can have master profiles and then inherit subprofiles off of them like this: profile1 datacenter1-profile1 (--inherit from profile1) datacenter2-profile1 With the only difference is the prod versions point to different cobbler servers. As many folks are doing this, one of my goals is to get some documentation together on different ways to do multiple-cobbler server rollouts to make that easier, including instructions on how to most easily replicate config files and install trees. > > thanks > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From tom at ng23.net Thu Feb 7 17:40:37 2008 From: tom at ng23.net (Tom Brown) Date: Thu, 07 Feb 2008 17:40:37 +0000 Subject: [et-mgmt-tools] Have 1 cobbler server know about boxes built by another In-Reply-To: <47AB3F9E.4070003@redhat.com> References: <47AB3CE7.9050202@ng23.net> <47AB3F9E.4070003@redhat.com> Message-ID: <47AB4295.2070404@ng23.net> >> >> In my situation i have a 'build' network where bare metal gets done - >> Once this is built it is moved into the correct network whether that >> be dev, qa or prod or whatever. Once in those networks they can be >> rebuild, using koan, by build machines in those networks, we dont >> allow machines to straddle networks hence having different build >> machines in each network. Is there a way to let other cobbler servers >> know that a box is build using a different cobbler server? >> > You can solve that problem in a different way, without having to keep > track of things. Basically cobbler doesn't keep inventory of what > you have installed so that is up to you. However I'm not sure that's > a problem. > so how does the cobbler server know what to install when you reinstall using koan? I presumed that it remembers the name/mac or something so when it sees that again it knows what profile to install the machine back to? From mdehaan at redhat.com Thu Feb 7 18:19:51 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 07 Feb 2008 13:19:51 -0500 Subject: [et-mgmt-tools] Have 1 cobbler server know about boxes built by another In-Reply-To: <47AB4295.2070404@ng23.net> References: <47AB3CE7.9050202@ng23.net> <47AB3F9E.4070003@redhat.com> <47AB4295.2070404@ng23.net> Message-ID: <47AB4BC7.9030202@redhat.com> > > so how does the cobbler server know what to install when you reinstall > using koan? I presumed that it remembers the name/mac or something so > when it sees that again it knows what profile to install the machine > back to? Typically you tell koan what profile/system record you want to install, then koan fetches the data on that profile/system from the cobbler server you specify with --server. If you use --autodetect-system it will ask the cobbler server if a system has been configured with one of the matching MAC addresses, which is useful if you want to write a generic reinstall script. > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From tom at ng23.net Thu Feb 7 18:37:55 2008 From: tom at ng23.net (Tom Brown) Date: Thu, 07 Feb 2008 18:37:55 +0000 Subject: [et-mgmt-tools] Have 1 cobbler server know about boxes built by another In-Reply-To: <47AB4BC7.9030202@redhat.com> References: <47AB3CE7.9050202@ng23.net> <47AB3F9E.4070003@redhat.com> <47AB4295.2070404@ng23.net> <47AB4BC7.9030202@redhat.com> Message-ID: <47AB5003.8000604@ng23.net> > > Typically you tell koan what profile/system record you want to > install, then koan fetches the data on that profile/system from the > cobbler server you specify with --server. > > If you use --autodetect-system it will ask the cobbler server if a > system has been configured with one of the matching MAC addresses, > which is useful if you want to write a generic reinstall script. > yep - so i wonder where the mac is stored on the server? Surely if i 'share'' this info then the same build can be done from any server (providing servers are setup identically) ? From mdehaan at redhat.com Thu Feb 7 18:49:26 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 07 Feb 2008 13:49:26 -0500 Subject: [et-mgmt-tools] Have 1 cobbler server know about boxes built by another In-Reply-To: <47AB5003.8000604@ng23.net> References: <47AB3CE7.9050202@ng23.net> <47AB3F9E.4070003@redhat.com> <47AB4295.2070404@ng23.net> <47AB4BC7.9030202@redhat.com> <47AB5003.8000604@ng23.net> Message-ID: <47AB52B6.1090302@redhat.com> Tom Brown wrote: > >> >> Typically you tell koan what profile/system record you want to >> install, then koan fetches the data on that profile/system from the >> cobbler server you specify with --server. >> >> If you use --autodetect-system it will ask the cobbler server if a >> system has been configured with one of the matching MAC addresses, >> which is useful if you want to write a generic reinstall script. >> > > > yep - so i wonder where the mac is stored on the server? koan just checks the results of ifconfig. > Surely if i 'share'' this info then the same build can be done from > any server (providing servers are setup identically) ? yes. You can share the contents of /var/lib/cobbler but replace the server address in the settings file (/var/lib/cobbler/settings) -- if you leave the address the same things won't work. The alternative is to do the inherited profile approach with the --server-override flags and use different profiles based on location, though that grows a bit more complex. In the next few weeks, I'll see about putting up some suggested workflows for doing this on the Wiki and folks can contribute how their various setups are configured when they have more than one cobbler server. --Michael From fj0588di at aa.jp.fujitsu.com Fri Feb 8 08:31:42 2008 From: fj0588di at aa.jp.fujitsu.com (S.Sakamoto) Date: Fri, 8 Feb 2008 17:31:42 +0900 Subject: [et-mgmt-tools] [PATCH] The bridge of bonding and the tag VLAN isnot displayed in virt-manager In-Reply-To: <20080207125623.GB30230@redhat.com> References: <200802071452.HGG65121.69J0EKG9@aa.jp.fujitsu.com> <20080207125623.GB30230@redhat.com> Message-ID: <200802081731.DJC60956.EJG9K096@aa.jp.fujitsu.com> > > virt-manager can display only the bridge of the physics network device now > > when add a device and make a guest domain newly. > > A bridge of bonding and Tag VLAN is not displayed. > > > > So, I make the patch that a bridge of bonding and Tag VLAN came to be displayed. > > Do not getout from processing specific linux... > > but can use a bridge of bonding and the tag VLAN. > > I'd definitely welcome code to deal with VLANs - and we're already Linux > specific in this code, so that's not a problem. > > We can't use the 'rhpl*' python libraries though - these are Red Hat > specific libraries, so if we use them, virt-manager won't be able to run > on Debian/SUSE/etc > > If you can adapt the code to remove use of 'import rhpl' then that'd be > good. I remake the patch to remove use of 'import rhpl'. Regards, Shigeki Sakamoto. -------------- next part -------------- A non-text attachment was scrubbed... Name: connection.patch Type: application/octet-stream Size: 9321 bytes Desc: not available URL: From mdehaan at redhat.com Fri Feb 8 15:09:37 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 08 Feb 2008 10:09:37 -0500 Subject: [et-mgmt-tools] [PATCH] Cobbler - webui - hide random mac link when editing system In-Reply-To: <2a7fce340802070120g24028c0cu3cd90be4c031b011@mail.gmail.com> References: <2a7fce340802070120g24028c0cu3cd90be4c031b011@mail.gmail.com> Message-ID: <47AC70B1.8030403@redhat.com> Tim Verhoeven wrote: > Hi, > > Attached is a small patch for cobbler that hides the Random MAC link > when editing systems. This has no use as the needed JS code is also > disabled in that case. > > Regards, > Tim > Applied, Thanks! From rdtennent at gmail.com Fri Feb 8 15:32:01 2008 From: rdtennent at gmail.com (Bob Tennent) Date: Fri, 08 Feb 2008 10:32:01 -0500 Subject: [et-mgmt-tools] user access to virt-manager guests Message-ID: <47ac75f1.iLK8wgklSlVv5m0M%rdtennent@gmail.com> >|> Hi. Is it possible to allow non-root users to access a guest operating >|> system? Once the guest is set up, it should be possible to choose to >|> allow the guest to look after its own security and authentication. As >|> far as I can see, this isn't now possible. Perhaps some /etc/sudoers >|> incantation? The difficulty is that virt-manager is used both >|> to set up *and* to use guests; those functions should be separated. >| >|Users can access the guest via SSH (or any network login service >|running inside the guest), or they can use a standalone viewer program >|like virt-viewer to access the graphical console without needing admin >|privileges. I want to be able to "connect" and initiate a guest in a *script* (so it can be done automatically when the host boots), and then allow users (without admin privileges) to access the guest graphically. Is this a reasonable mode of operation? If so, how do I do it? If I call qemu-kvm directly the -nographic option doesn't seem to do what I expect. And I don't know how to use virt-viewer to view a guest initiated directly rather than by using virt-manager. If this kind of stuff is explained somewhere in documentation I haven't managed to find, please refer me to it and my apologies for using the mail-list inappropriately. It might be helpful if the list archives were searchable. Bob T. From berrange at redhat.com Fri Feb 8 16:02:05 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 8 Feb 2008 16:02:05 +0000 Subject: [et-mgmt-tools] user access to virt-manager guests In-Reply-To: <47ac75f1.iLK8wgklSlVv5m0M%rdtennent@gmail.com> References: <47ac75f1.iLK8wgklSlVv5m0M%rdtennent@gmail.com> Message-ID: <20080208160205.GA4921@redhat.com> On Fri, Feb 08, 2008 at 10:32:01AM -0500, Bob Tennent wrote: > >|> Hi. Is it possible to allow non-root users to access a guest operating > >|> system? Once the guest is set up, it should be possible to choose to > >|> allow the guest to look after its own security and authentication. As > >|> far as I can see, this isn't now possible. Perhaps some /etc/sudoers > >|> incantation? The difficulty is that virt-manager is used both > >|> to set up *and* to use guests; those functions should be separated. > >| > >|Users can access the guest via SSH (or any network login service > >|running inside the guest), or they can use a standalone viewer program > >|like virt-viewer to access the graphical console without needing admin > >|privileges. > > I want to be able to "connect" and initiate a guest in a *script* (so > it can be done automatically when the host boots), and then allow users > (without admin privileges) to access the guest graphically. Is this a > reasonable mode of operation? If so, how do I do it? If I call qemu-kvm > directly the -nographic option doesn't seem to do what I expect. And I > don't know how to use virt-viewer to view a guest initiated directly > rather than by using virt-manager. There is a manual page for virt-viewer which illustrates the examples on how to access the console. eg, one example is: virt-viewer --connect qemu:///system {NAME OF GUEST} If you want a guest to start automatically when a host boots, then you need to turn on the 'autostart' flag for that guest. This is described in the virsh manual page eg, virsh --connect qemu:///system autostart {NAME OF GUEST} > If this kind of stuff is explained somewhere in documentation I haven't > managed to find, please refer me to it and my apologies for using the > mail-list inappropriately. It might be helpful if the list archives were > searchable. The man pages for virsh, virt-viewer, and online help for virt-manager Finally the virt-manager Wiki site is available if anyone wants to add tips on use cases: http://virt-manager.org/wiki/ You can search the mailing list trivially using google. Just enter this in the google search box: site:redhat.com inurl:et-mgmt-tools {MORE SEARCH TERMS} Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From gavin at redhat.com Fri Feb 8 16:22:08 2008 From: gavin at redhat.com (Gavin Romig-Koch) Date: Fri, 08 Feb 2008 11:22:08 -0500 Subject: [et-mgmt-tools] PATCH: Get koan working on ia64 Message-ID: <47AC81B0.5070504@redhat.com> Please consider the following patch for koan. * Fri Feb 8 2008 Gavin Romig-Koch Get koan working on ia64 The following changes are needed to get koan working on ia64. Elilo expects the kernel and initrd to be on the EFI storage device, and grubby expects them to be within /boot/efi/EFI/redhat. Grubby's --bootloader-probe option only exists on x86 and x86_64, on other arches grubby errors out when given it, so grubby --bootloader-probe never returns "elilo". This patch changes koan to just let grubby figure out the appropriate bootloader to use (which it does correctly), and only after that checks to see if the boot loader is lilo, and if necessary runs lilo. -gavin... diff --git a/koan/app.py b/koan/app.py index 6d116de..3410b26 100755 --- a/koan/app.py +++ b/koan/app.py @@ -542,10 +542,15 @@ class Koan: # find the correct file download location if not self.is_virt: - download = "/boot" if self.live_cd: download = "/tmp/boot/boot" + elif os.path.exists("/boot/efi/EFI/redhat/elilo.conf"): + download = "/boot/efi/EFI/redhat" + + else: + download = "/boot" + else: # ensure we have a good virt type choice and know where # to download the kernel/initrd @@ -711,19 +716,7 @@ class Koan: k_args = k_args.replace("lang ","lang= ") - cmd = [ "/sbin/grubby", - "--bootloader-probe" ] - - which_loader = sub_process.Popen(cmd, stdout=sub_process.PIPE).communicate()[0] - - loader = "--grub" - if which_loader.find("elilo") != -1: - loader = "--elilo" - elif which_loader.find("lilo") != -1: - loader = "--lilo" - cmd = [ "/sbin/grubby", - loader, "--add-kernel", self.safe_load(profile_data,'kernel_local'), "--initrd", self.safe_load(profile_data,'initrd_local'), "--make-default", @@ -737,7 +730,14 @@ class Koan: cmd.append("--config-file=/tmp/boot/boot/grub/grub.conf") self.subprocess_call(cmd) - if loader == "--lilo": + + # if grubby --bootloader-probe returns lilo, + # apply lilo changes + cmd = [ "/sbin/grubby", "--bootloader-probe" ] + probe_process = sub_process.Popen(cmd, stdout=sub_process.PIPE) + which_loader = probe_process.communicate()[0] + if probe_process.returncode == 0 and \ + which_loader.find("lilo") != -1: print "- applying lilo changes" cmd = [ "/sbin/lilo" ] sub_process.Popen(cmd, stdout=sub_process.PIPE).communicate()[0] From mdehaan at redhat.com Fri Feb 8 16:24:59 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 08 Feb 2008 11:24:59 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.7.2 test release (0.8 candidate) Message-ID: <47AC825B.6010409@redhat.com> After a long time developing the next version of Cobbler, it's time to be finishing this version up. The motto should be "release early, release often" so we'll be doing this a lot more in the future. In about a week, I'll be releasing Cobbler 0.8, along with an updated koan. This will include three months worth of development on the devel branch as detailed here: *http://tinyurl.com/3y2el2 * So that that people have a change to test things first, I've uploaded SRC rpms for Cobbler 0.7.2 and koan 0.6.5 to: http://cobbler.et.redhat.com/download/testing/ I've been using these development versions for a long time, as well as many others, though additional testing before release is welcome. If all goes well, I'll push new versions out to the mirrors in about a week. As with the previous release, I'll support fixes to 0.8 as we work on the 0.9 development version. NOTE: We're going to do vmware in koan as a later update, the groundwork has already been laid in Cobbler so it will mostly be a change in koan. ALSO NOTE: If you're using the WebUI, setup instructions have changed, and you will need to run through setup again. These changes were needed to make things easier on new users as well as to accomodate future features. Instructions are at https://fedorahosted.org/cobbler/wiki/CobblerWebInterface Thanks, and as usual, let me know if you have any questions! --Michael From mdehaan at redhat.com Fri Feb 8 16:30:38 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 08 Feb 2008 11:30:38 -0500 Subject: [et-mgmt-tools] PATCH: Get koan working on ia64 In-Reply-To: <47AC81B0.5070504@redhat.com> References: <47AC81B0.5070504@redhat.com> Message-ID: <47AC83AE.2040107@redhat.com> Gavin Romig-Koch wrote: > Please consider the following patch for koan. Looks good ... I'm getting a "malformed patch" error, if you can send me this as an attachment (off-list is ok) I'll apply this. This can probably make the next koan release too. --Michael > > > * Fri Feb 8 2008 Gavin Romig-Koch > Get koan working on ia64 > > The following changes are needed to get koan working on ia64. Elilo > expects the kernel and initrd to be on the EFI storage device, and > grubby expects them to be within /boot/efi/EFI/redhat. > Grubby's --bootloader-probe option only exists on x86 and x86_64, on > other arches grubby errors out when given it, so grubby > --bootloader-probe > never returns "elilo". > This patch changes koan to just let grubby figure out the appropriate > bootloader to use (which it does correctly), and only after that checks > to see if the boot loader is lilo, and if necessary runs lilo. > > -gavin... > > > diff --git a/koan/app.py b/koan/app.py > index 6d116de..3410b26 100755 > --- a/koan/app.py > +++ b/koan/app.py > @@ -542,10 +542,15 @@ class Koan: > > # find the correct file download location if not > self.is_virt: > - download = "/boot" > if self.live_cd: > download = "/tmp/boot/boot" > > + elif os.path.exists("/boot/efi/EFI/redhat/elilo.conf"): > + download = "/boot/efi/EFI/redhat" > + > + else: > + download = "/boot" > + > else: > # ensure we have a good virt type choice and know where > # to download the kernel/initrd > @@ -711,19 +716,7 @@ class Koan: > > k_args = k_args.replace("lang ","lang= ") > > - cmd = [ "/sbin/grubby", - > "--bootloader-probe" ] > - > - which_loader = sub_process.Popen(cmd, > stdout=sub_process.PIPE).communicate()[0] > - - loader = "--grub" > - if which_loader.find("elilo") != -1: > - loader = "--elilo" > - elif which_loader.find("lilo") != -1: > - loader = "--lilo" > - > cmd = [ "/sbin/grubby", > - loader, > "--add-kernel", > self.safe_load(profile_data,'kernel_local'), > "--initrd", > self.safe_load(profile_data,'initrd_local'), > "--make-default", > @@ -737,7 +730,14 @@ class Koan: > cmd.append("--config-file=/tmp/boot/boot/grub/grub.conf") > self.subprocess_call(cmd) > > - if loader == "--lilo": > + > + # if grubby --bootloader-probe returns lilo, > + # apply lilo changes > + cmd = [ "/sbin/grubby", "--bootloader-probe" ] > + probe_process = sub_process.Popen(cmd, > stdout=sub_process.PIPE) > + which_loader = probe_process.communicate()[0] > + if probe_process.returncode == 0 and \ > + which_loader.find("lilo") != -1: > print "- applying lilo changes" > cmd = [ "/sbin/lilo" ] > sub_process.Popen(cmd, > stdout=sub_process.PIPE).communicate()[0] > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mdehaan at redhat.com Fri Feb 8 22:23:45 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 08 Feb 2008 17:23:45 -0500 Subject: [et-mgmt-tools] PATCH: Get koan working on ia64 In-Reply-To: <47AC81B0.5070504@redhat.com> References: <47AC81B0.5070504@redhat.com> Message-ID: <47ACD671.7090207@redhat.com> Gavin Romig-Koch wrote: > Please consider the following patch for koan. Applied. I did a quick test to make sure x86 is still happy, and it was. OT: I've also fixed a small bug where --list-profiles and --list-systems didn't work in the test versions I uploaded earlier today... this will be fixed when we do the koan 0.6.5 release. This patch will also be in 0.6.5 Thanks! > > > * Fri Feb 8 2008 Gavin Romig-Koch > Get koan working on ia64 > > The following changes are needed to get koan working on ia64. Elilo > expects the kernel and initrd to be on the EFI storage device, and > grubby expects them to be within /boot/efi/EFI/redhat. > Grubby's --bootloader-probe option only exists on x86 and x86_64, on > other arches grubby errors out when given it, so grubby > --bootloader-probe > never returns "elilo". > This patch changes koan to just let grubby figure out the appropriate > bootloader to use (which it does correctly), and only after that checks > to see if the boot loader is lilo, and if necessary runs lilo. > > -gavin... > > > diff --git a/koan/app.py b/koan/app.py > index 6d116de..3410b26 100755 > --- a/koan/app.py > +++ b/koan/app.py > @@ -542,10 +542,15 @@ class Koan: > > # find the correct file download location if not > self.is_virt: > - download = "/boot" > if self.live_cd: > download = "/tmp/boot/boot" > > + elif os.path.exists("/boot/efi/EFI/redhat/elilo.conf"): > + download = "/boot/efi/EFI/redhat" > + > + else: > + download = "/boot" > + > else: > # ensure we have a good virt type choice and know where > # to download the kernel/initrd > @@ -711,19 +716,7 @@ class Koan: > > k_args = k_args.replace("lang ","lang= ") > > - cmd = [ "/sbin/grubby", - > "--bootloader-probe" ] > - > - which_loader = sub_process.Popen(cmd, > stdout=sub_process.PIPE).communicate()[0] > - - loader = "--grub" > - if which_loader.find("elilo") != -1: > - loader = "--elilo" > - elif which_loader.find("lilo") != -1: > - loader = "--lilo" > - > cmd = [ "/sbin/grubby", > - loader, > "--add-kernel", > self.safe_load(profile_data,'kernel_local'), > "--initrd", > self.safe_load(profile_data,'initrd_local'), > "--make-default", > @@ -737,7 +730,14 @@ class Koan: > cmd.append("--config-file=/tmp/boot/boot/grub/grub.conf") > self.subprocess_call(cmd) > > - if loader == "--lilo": > + > + # if grubby --bootloader-probe returns lilo, > + # apply lilo changes > + cmd = [ "/sbin/grubby", "--bootloader-probe" ] > + probe_process = sub_process.Popen(cmd, > stdout=sub_process.PIPE) > + which_loader = probe_process.communicate()[0] > + if probe_process.returncode == 0 and \ > + which_loader.find("lilo") != -1: > print "- applying lilo changes" > cmd = [ "/sbin/lilo" ] > sub_process.Popen(cmd, > stdout=sub_process.PIPE).communicate()[0] > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From rdtennent at gmail.com Sat Feb 9 18:51:54 2008 From: rdtennent at gmail.com (Bob Tennent) Date: Sat, 09 Feb 2008 13:51:54 -0500 Subject: [et-mgmt-tools] sound configuration Message-ID: <47adf64a.ywWWI6mOJA/RbFg9%rdtennent@gmail.com> An earlier message http://www.redhat.com/archives/et-mgmt-tools/2007-June/msg00021.html says Audio devices are on the radar but there's no support yet. For the moment you'll have to manually modify the guest config. Could I see an example of a sound configuration in a guest.xml file? I can see in documentation examples what the command-line options look like but I don't know where the XML format is documented. Bob T. From tim.verhoeven.be at gmail.com Mon Feb 11 10:23:55 2008 From: tim.verhoeven.be at gmail.com (Tim Verhoeven) Date: Mon, 11 Feb 2008 11:23:55 +0100 Subject: [et-mgmt-tools] Cobbler - Issue with update_permissions function in action_reposync.py Message-ID: <2a7fce340802110223g231d787dk8fd0598766494e7f@mail.gmail.com> Hi, In the devel branc of Cobbler commit "a2d760b7369059731a6cfbe673e117a553511a20" introduced a new function "update_permissions" that is run during a sync. But there is a problem with this piece of that function : cmd2 = "chmod -R 640 %s" % repo_path sub_process.call(cmd2, shell=True) This basically means that all directories will loose there "x" bit and that also means that apache cannot enter anymore in the directories, this then results in that both a install and yum stop working. I would have send a patch myself to fix it but I'm not completely sure what problem the "update_permissions" function is suppose to solve. The SELinux part makes sense to me and can stay. But I'm guessing that for the UNIX style permissions part it is better to do something like this : chmod -R u+r,u+w,g+r,g-w,o-r,o-w I believe this accomplishes the same but without touching the "x" bit of directories. Regards, Tim -- Tim Verhoeven - tim.verhoeven.be at gmail.com - 0479 / 88 11 83 Hoping the problem magically goes away by ignoring it is the "microsoft approach to programming" and should never be allowed. (Linus Torvalds) From mdehaan at redhat.com Mon Feb 11 15:01:36 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 11 Feb 2008 10:01:36 -0500 Subject: [et-mgmt-tools] Cobbler - Issue with update_permissions function in action_reposync.py In-Reply-To: <2a7fce340802110223g231d787dk8fd0598766494e7f@mail.gmail.com> References: <2a7fce340802110223g231d787dk8fd0598766494e7f@mail.gmail.com> Message-ID: <47B06350.8050002@redhat.com> Tim Verhoeven wrote: > Hi, > > In the devel branc of Cobbler commit > "a2d760b7369059731a6cfbe673e117a553511a20" introduced a new function > "update_permissions" that is run during a sync. > > But there is a problem with this piece of that function : > > cmd2 = "chmod -R 640 %s" % repo_path > sub_process.call(cmd2, shell=True) > > This basically means that all directories will loose there "x" bit and > that also means that apache cannot enter anymore in the directories, > this then results in that both a install and yum stop working. > > I would have send a patch myself to fix it but I'm not completely sure > what problem the "update_permissions" function is suppose to solve. > The SELinux part makes sense to me and can stay. But I'm guessing that > for the UNIX style permissions part it is better to do something like > this : > > chmod -R u+r,u+w,g+r,g-w,o-r,o-w > > I believe this accomplishes the same but without touching the "x" bit > of directories. > > Regards, > Tim > > Nice feature :) s/640/750/ Actually it might as well be 755. I'll compare this versus the RPM permissions on that directory and straighten it out. The basic problem we were seeing was due the rsync originally doing "-a" to preserve permissions, which is less of an issue now that that parameter was removed. In other words, the code in question was being paranoid. --Michael From gavin at redhat.com Mon Feb 11 15:31:32 2008 From: gavin at redhat.com (Gavin Romig-Koch) Date: Mon, 11 Feb 2008 10:31:32 -0500 Subject: [et-mgmt-tools] PATCH: Get koan working on ia64 In-Reply-To: <47ACD671.7090207@redhat.com> References: <47AC81B0.5070504@redhat.com> <47ACD671.7090207@redhat.com> Message-ID: <47B06A54.8030603@redhat.com> Michael DeHaan wrote: > > Applied. I did a quick test to make sure x86 is still happy, and it > was. Thanks. -gavin.... From gavin at redhat.com Mon Feb 11 21:49:37 2008 From: gavin at redhat.com (Gavin Romig-Koch) Date: Mon, 11 Feb 2008 16:49:37 -0500 Subject: [et-mgmt-tools] [PATCH] Koan support for nfs install trees Message-ID: <47B0C2F1.2070905@redhat.com> Please consider adding the following patch to koan. This patch add support for kickstart's "nfs" command (along side the existing "url" command support) to koan. -gavin... -------------- next part -------------- A non-text attachment was scrubbed... Name: nfs-method.patch Type: text/x-patch Size: 7719 bytes Desc: not available URL: From mdehaan at redhat.com Mon Feb 11 22:26:43 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 11 Feb 2008 17:26:43 -0500 Subject: [et-mgmt-tools] [PATCH] Koan support for nfs install trees In-Reply-To: <47B0C2F1.2070905@redhat.com> References: <47B0C2F1.2070905@redhat.com> Message-ID: <47B0CBA3.8040507@redhat.com> Gavin Romig-Koch wrote: > > Please consider adding the following patch to koan. > > This patch add support for kickstart's "nfs" command (along side the > existing "url" command support) to koan. > > -gavin... > > > ------------------------------------------------------------------------ > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Nice. I was going to ask about that :). Cobbler's support for importing over NFS with "cobbler import ... --available-as=nfs://" will also be able to leverage this to make things more generic. As it stands it does some weird things with the kickstart template so there is an equivalent "url" line, so this will allow removing that code. Will test this out and integrate shortly. Thanks! --Michael From gavin at redhat.com Mon Feb 11 23:43:32 2008 From: gavin at redhat.com (Gavin Romig-Koch) Date: Mon, 11 Feb 2008 18:43:32 -0500 Subject: [et-mgmt-tools] PATCH: have --no-cobbler option handle file names Message-ID: <47B0DDA4.9080501@redhat.com> Please consider the attached patch for inclusion in koan This patch adds support for file names to the --no-cobbler option. -gavin... -------------- next part -------------- A non-text attachment was scrubbed... Name: relative-no-cobbler-option.patch Type: text/x-patch Size: 1617 bytes Desc: not available URL: From cranfill at imageworks.com Tue Feb 12 19:51:16 2008 From: cranfill at imageworks.com (Jennifer Cranfill) Date: Tue, 12 Feb 2008 11:51:16 -0800 Subject: [et-mgmt-tools] adding a new system with the cobbler api? Message-ID: <47B1F8B4.4060008@imageworks.com> I'm having trouble adding a new system using the cobbler python api with version 0.7.2 (I have not tried this with any other versions). Here is my python code: >>> from cobbler import api >>> cobbler_api = api.BootAPI() >>> new_sys = cobbler_api.new_system() >>> new_sys.name = 'testmachine' >>> new_sys.profile = 'cserver_centos51_x86' >>> cobbler_api.add_system(new_sys) True That works, but it doesn't show up in the cobbler command line: $ cobbler system list | grep test $ Am I missing something? Thanks! -- Jennifer Cranfill Systems Engineer Sony Pictures Imageworks cranfill at imageworks.com From mdehaan at redhat.com Tue Feb 12 20:06:20 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 12 Feb 2008 15:06:20 -0500 Subject: [et-mgmt-tools] adding a new system with the cobbler api? In-Reply-To: <47B1F8B4.4060008@imageworks.com> References: <47B1F8B4.4060008@imageworks.com> Message-ID: <47B1FC3C.6050803@redhat.com> Jennifer Cranfill wrote: > I'm having trouble adding a new system using the cobbler python api > with version 0.7.2 (I have not tried this with any other versions). > > Here is my python code: > > >>> from cobbler import api > >>> cobbler_api = api.BootAPI() > >>> new_sys = cobbler_api.new_system() > >>> new_sys.name = 'testmachine' > >>> new_sys.profile = 'cserver_centos51_x86' > >>> cobbler_api.add_system(new_sys) > True > > That works, but it doesn't show up in the cobbler command line: > $ cobbler system list | grep test > $ > > Am I missing something? > > Thanks! > > -- > Jennifer Cranfill > Systems Engineer > Sony Pictures Imageworks > cranfill at imageworks.com > Aha! You've uncovered a bug in one of the new API functions. Thanks for reporting this. In the old API, there was basically two types of adds ... one that adds things temporarily (which is what happens when the API is initialized, and another that happens when you save an object, that does nudge the serializer into saving state. Those two types are still there, but I added a higher level function to make needing to know about the collections interface less important. Looks like that function isn't doing it correctly though, and I need to fix that. The following will work for you now: cobbler_api.systems().add(new_sys,with_copy=True) And I'll fix the version in the 0.7.X branch to do the right thing when you call it. (Basically it just needs to pass some extra flags down to the systems().add call). --Michael From cranfill at imageworks.com Tue Feb 12 23:42:56 2008 From: cranfill at imageworks.com (Jennifer Cranfill) Date: Tue, 12 Feb 2008 15:42:56 -0800 Subject: [et-mgmt-tools] adding a new system with the cobbler api? In-Reply-To: <47B1FC3C.6050803@redhat.com> References: <47B1F8B4.4060008@imageworks.com> <47B1FC3C.6050803@redhat.com> Message-ID: <47B22F00.8000706@imageworks.com> Michael DeHaan wrote: > Aha! You've uncovered a bug in one of the new API functions. Thanks > for reporting this. > > In the old API, there was basically two types of adds ... one that adds > things temporarily (which is what happens when the API > is initialized, and another that happens when you save an object, that > does nudge the serializer into saving state. Those two types > are still there, but I added a higher level function to make needing to > know about the collections interface less important. Looks like that > function isn't doing it correctly though, and I need to fix that. > > The following will work for you now: > > cobbler_api.systems().add(new_sys,with_copy=True) Thanks Michael! That works great. Glad you added the higher level functions, they are very easy to use and understand. --Jennifer From fukuta.saori at jp.fujitsu.com Wed Feb 13 08:25:22 2008 From: fukuta.saori at jp.fujitsu.com (Saori Fukuta) Date: Wed, 13 Feb 2008 17:25:22 +0900 Subject: [et-mgmt-tools] TypeError in _vnc_auth_credential Message-ID: <20080213172500.12AD.FUKUTA.SAORI@jp.fujitsu.com> Hello, I am trying to connect to the guest domain which has the password setting by Graphic console of virt-manager. I expect that the 'Password' field is shown, but I cannot enter the password because there is no 'Password' field. So, the console remains 'connecting to console for guest'. It seems that the virt-manager could not retrieve the value of 'credList' at _vnc_auth_credential. Should I have any setting for using password ? I'm using Upstream virt-manager (changeset:682). Thanks, Saori Fukuta -------------- next part -------------- An embedded message was scrubbed... From: Saori Fukuta Subject: virt-manager.log Date: Wed, 13 Feb 2008 16:44:10 +0900 Size: 1578 URL: From atodorov at redhat.com Wed Feb 13 08:38:11 2008 From: atodorov at redhat.com (Alexander Todorov) Date: Wed, 13 Feb 2008 09:38:11 +0100 Subject: [et-mgmt-tools] OS type detection in virt-install/virt-manager Message-ID: <47B2AC73.5090607@redhat.com> Hello, I was looking at the recent code in OSDistro.py Can we switch to using the new .treeinfo file and the information there instead of the current if hasFile() mechanism? The .treeinfo provides with all the information needed to detect the family type (Fedora, RHEL, CentOS ???) and boot images location. Also improve the acquire$FILE functions with that information. In case this is not found then we may fall back to the current code. The reason I'm requesting this: in newer Fedora repos the packages dir is simply called "Packages" but not "Fedora" and wrt remixing tools it may be called with any name I think. OSDistro.py uses the presence of fedora.css in the repo to decide on a Fedora distribution. This file is not essential to the install process and may be missing in some repos. What do you think? Thanks, Alexander. From berrange at redhat.com Wed Feb 13 14:23:20 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 13 Feb 2008 14:23:20 +0000 Subject: [et-mgmt-tools] TypeError in _vnc_auth_credential In-Reply-To: <20080213172500.12AD.FUKUTA.SAORI@jp.fujitsu.com> References: <20080213172500.12AD.FUKUTA.SAORI@jp.fujitsu.com> Message-ID: <20080213142320.GA32737@redhat.com> On Wed, Feb 13, 2008 at 05:25:22PM +0900, Saori Fukuta wrote: > Hello, > > I am trying to connect to the guest domain which has the password setting > by Graphic console of virt-manager. I expect that the 'Password' field > is shown, but I cannot enter the password because there is no 'Password' > field. So, the console remains 'connecting to console for guest'. > > It seems that the virt-manager could not retrieve the value of 'credList' > at _vnc_auth_credential. > > Should I have any setting for using password ? I'm using Upstream > virt-manager (changeset:682). Can you tell me what version of GTK-VNC you are using ? Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From berrange at redhat.com Wed Feb 13 14:32:57 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 13 Feb 2008 14:32:57 +0000 Subject: [et-mgmt-tools] OS type detection in virt-install/virt-manager In-Reply-To: <47B2AC73.5090607@redhat.com> References: <47B2AC73.5090607@redhat.com> Message-ID: <20080213143257.GC32737@redhat.com> On Wed, Feb 13, 2008 at 09:38:11AM +0100, Alexander Todorov wrote: > Hello, > I was looking at the recent code in OSDistro.py > Can we switch to using the new .treeinfo file and the information there > instead of the current if hasFile() mechanism? > The .treeinfo provides with all the information needed to detect the > family type (Fedora, RHEL, CentOS ???) and boot images location. > Also improve the acquire$FILE functions with that information. What new .treeinfo file ? It sounds interesting, but this is the first I've heard of it... Is there anywhere that describes it ? I don't see a .treeinfo file on the Fedora download sites, eg http://download.fedora.redhat.com/pub/fedora/linux/development/x86_64/os/ http://download.fedora.redhat.com/pub/fedora/linux/releases/8/Fedora/x86_64/os/ > The reason I'm requesting this: in newer Fedora repos the packages dir > is simply called "Packages" but not "Fedora" and wrt remixing tools it > may be called with any name I think. OSDistro.py uses the presence of > fedora.css in the repo to decide on a Fedora distribution. This file > is not essential to the install process and may be missing in some repos. > > What do you think? If there's ways we can improve the OS detection, I'm open to any suggestions. Currently, we do the probing whether the user has manually selected a distro type or not. One optimization would be to only do the probing if the user has NOT explicitly selected a distro type. This would give users an easy way to force the install to continue even if the 'magic' files are missing. eg, if you select 'Fedora 8' in virt-manager it would just go straight to the kernel/initrd fetch stage. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From crobinso at redhat.com Wed Feb 13 14:53:44 2008 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 13 Feb 2008 09:53:44 -0500 Subject: [et-mgmt-tools] OS type detection in virt-install/virt-manager In-Reply-To: <20080213143257.GC32737@redhat.com> References: <47B2AC73.5090607@redhat.com> <20080213143257.GC32737@redhat.com> Message-ID: <47B30478.2070902@redhat.com> Daniel P. Berrange wrote: > On Wed, Feb 13, 2008 at 09:38:11AM +0100, Alexander Todorov wrote: >> Hello, >> I was looking at the recent code in OSDistro.py >> Can we switch to using the new .treeinfo file and the information there >> instead of the current if hasFile() mechanism? >> The .treeinfo provides with all the information needed to detect the >> family type (Fedora, RHEL, CentOS ???) and boot images location. >> Also improve the acquire$FILE functions with that information. > > What new .treeinfo file ? It sounds interesting, but this is the first > I've heard of it... Is there anywhere that describes it ? > > I don't see a .treeinfo file on the Fedora download sites, eg > > http://download.fedora.redhat.com/pub/fedora/linux/development/x86_64/os/ > > http://download.fedora.redhat.com/pub/fedora/linux/releases/8/Fedora/x86_64/os/ > >> The reason I'm requesting this: in newer Fedora repos the packages dir >> is simply called "Packages" but not "Fedora" and wrt remixing tools it >> may be called with any name I think. OSDistro.py uses the presence of >> fedora.css in the repo to decide on a Fedora distribution. This file >> is not essential to the install process and may be missing in some repos. >> >> What do you think? > > If there's ways we can improve the OS detection, I'm open to any suggestions. > > Currently, we do the probing whether the user has manually selected a > distro type or not. > > One optimization would be to only do the probing if the user has NOT > explicitly selected a distro type. This would give users an easy way > to force the install to continue even if the 'magic' files are missing. > > eg, if you select 'Fedora 8' in virt-manager it would just go straight > to the kernel/initrd fetch stage. > > Dan. I updated the hasFile function recently to not grab files and instead just test for their existence. This means we can change the distro detection to actually just be dependent on whether the install media is found or not. So the 'magic' file could be the file needed to actually run the install which should avoid any misses for goofy trees. - Cole From atodorov at redhat.com Wed Feb 13 15:43:45 2008 From: atodorov at redhat.com (Alexander Todorov) Date: Wed, 13 Feb 2008 16:43:45 +0100 Subject: [et-mgmt-tools] OS type detection in virt-install/virt-manager In-Reply-To: <20080213143257.GC32737@redhat.com> References: <47B2AC73.5090607@redhat.com> <20080213143257.GC32737@redhat.com> Message-ID: <47B31031.4050603@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Daniel P. Berrange wrote: > On Wed, Feb 13, 2008 at 09:38:11AM +0100, Alexander Todorov wrote: >> Hello, >> I was looking at the recent code in OSDistro.py >> Can we switch to using the new .treeinfo file and the information there >> instead of the current if hasFile() mechanism? >> The .treeinfo provides with all the information needed to detect the >> family type (Fedora, RHEL, CentOS ???) and boot images location. >> Also improve the acquire$FILE functions with that information. > > What new .treeinfo file ? It sounds interesting, but this is the first > I've heard of it... Is there anywhere that describes it ? > I'm afraid there's not. I've filed a BZ to request documentation. https://bugzilla.redhat.com/show_bug.cgi?id=432641 You probably didn't see the files because they start with a dot "." and are hidden files. Try looking closer under the os/ directory for already released versions. I've tried with Fedora 8 from my local mirror at http://fedora.mirrors.megalan.bg/linux/releases/8/Fedora/x86_64/os/ The files are supposed to provide a generic way to get some info for the tree incl. kernel and initrd locations. I was told that this is still work in progress. Greetings, Alexander. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Red Hat - http://enigmail.mozdev.org iD8DBQFHsxAvhmd3WOiFct4RCoxKAJ0fVAWrvJpT50Ua38hJt1h7c4Nr6wCfTZT9 3boMkiqOCik8FrS0CHJd8L8= =ahyt -----END PGP SIGNATURE----- From atodorov at redhat.com Wed Feb 13 16:05:48 2008 From: atodorov at redhat.com (Alexander Todorov) Date: Wed, 13 Feb 2008 17:05:48 +0100 Subject: [et-mgmt-tools] OS type detection in virt-install/virt-manager In-Reply-To: <47B31031.4050603@redhat.com> References: <47B2AC73.5090607@redhat.com> <20080213143257.GC32737@redhat.com> <47B31031.4050603@redhat.com> Message-ID: <47B3155C.2000203@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Alexander Todorov wrote: > > The files are supposed to provide a generic way to get some info for the tree > incl. kernel and initrd locations. I was told that this is still work in progress. > I forgot to say that this is a feature of anaconda. Greetings, Alexander. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Red Hat - http://enigmail.mozdev.org iD8DBQFHsxVThmd3WOiFct4RCjpJAJ9g3EEVrfXwQAcQTN+pOjAGo+aarACfX7Z9 xiRtj6CU4pKawADrDbJG4bA= =CZbm -----END PGP SIGNATURE----- From fukuta.saori at jp.fujitsu.com Thu Feb 14 00:12:18 2008 From: fukuta.saori at jp.fujitsu.com (Saori Fukuta) Date: Thu, 14 Feb 2008 09:12:18 +0900 Subject: [et-mgmt-tools] TypeError in _vnc_auth_credential In-Reply-To: <20080213142320.GA32737@redhat.com> References: <20080213172500.12AD.FUKUTA.SAORI@jp.fujitsu.com> <20080213142320.GA32737@redhat.com> Message-ID: <20080214091200.021B.FUKUTA.SAORI@jp.fujitsu.com> On Wed, 13 Feb 2008 14:23:20 +0000 "Daniel P. Berrange" wrote: > On Wed, Feb 13, 2008 at 05:25:22PM +0900, Saori Fukuta wrote: > > Hello, > > > > I am trying to connect to the guest domain which has the password setting > > by Graphic console of virt-manager. I expect that the 'Password' field > > is shown, but I cannot enter the password because there is no 'Password' > > field. So, the console remains 'connecting to console for guest'. > > > > It seems that the virt-manager could not retrieve the value of 'credList' > > at _vnc_auth_credential. > > > > Should I have any setting for using password ? I'm using Upstream > > virt-manager (changeset:682). > > Can you tell me what version of GTK-VNC you are using ? The version of GTK-VNC I was using was gtk-vnc-0.2.0-4.fc8. So I tried this again after updating the version to gtk-vnc-0.3.2-1.fc8 by using yum update. Then this problem does not occur. thanks so much ! Saori Fukuta From fukuta.saori at jp.fujitsu.com Thu Feb 14 00:58:22 2008 From: fukuta.saori at jp.fujitsu.com (Saori Fukuta) Date: Thu, 14 Feb 2008 09:58:22 +0900 Subject: [et-mgmt-tools] TypeError in _vnc_auth_credential In-Reply-To: <20080214091200.021B.FUKUTA.SAORI@jp.fujitsu.com> References: <20080213142320.GA32737@redhat.com> <20080214091200.021B.FUKUTA.SAORI@jp.fujitsu.com> Message-ID: <20080214095750.022A.FUKUTA.SAORI@jp.fujitsu.com> On Thu, 14 Feb 2008 09:12:18 +0900 Saori Fukuta wrote: > On Wed, 13 Feb 2008 14:23:20 +0000 "Daniel P. Berrange" wrote: > > Can you tell me what version of GTK-VNC you are using ? > > The version of GTK-VNC I was using was gtk-vnc-0.2.0-4.fc8. Let me correct the version. I was using Upstream GTK-VNC(Cset:134). gtk-vnc-0.2.0-4.fc8 works for virt-manager. > So I tried this again after updating the version to gtk-vnc-0.3.2-1.fc8 > by using yum update. Then this problem does not occur. I think something wrong with the following patch, maybe http://gtk-vnc.codemonkey.ws/hg/outgoing.hg/rev/a720c8172a85 So I will check with gtk-vnc code. Thanks, Saori Fukuta From berrange at redhat.com Thu Feb 14 03:43:05 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 14 Feb 2008 03:43:05 +0000 Subject: [et-mgmt-tools] [PATCH] The bridge of bonding and the tag VLAN isnot displayed in virt-manager In-Reply-To: <200802081731.DJC60956.EJG9K096@aa.jp.fujitsu.com> References: <200802071452.HGG65121.69J0EKG9@aa.jp.fujitsu.com> <20080207125623.GB30230@redhat.com> <200802081731.DJC60956.EJG9K096@aa.jp.fujitsu.com> Message-ID: <20080214034305.GA1039@redhat.com> On Fri, Feb 08, 2008 at 05:31:42PM +0900, S.Sakamoto wrote: > > > virt-manager can display only the bridge of the physics network device now > > > when add a device and make a guest domain newly. > > > A bridge of bonding and Tag VLAN is not displayed. > > > > > > So, I make the patch that a bridge of bonding and Tag VLAN came to be displayed. > > > Do not getout from processing specific linux... > > > but can use a bridge of bonding and the tag VLAN. > > > > I'd definitely welcome code to deal with VLANs - and we're already Linux > > specific in this code, so that's not a problem. > > > > We can't use the 'rhpl*' python libraries though - these are Red Hat > > specific libraries, so if we use them, virt-manager won't be able to run > > on Debian/SUSE/etc > > > > If you can adapt the code to remove use of 'import rhpl' then that'd be > > good. > > I remake the patch to remove use of 'import rhpl'. I've been testing this patch and hit a couple of network configs where it doesn't do the right thing. I'm experimenting with a variation on this patch & will post it for feedback tomorrow if it passes tests on my machines. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From berrange at redhat.com Thu Feb 14 03:45:25 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 14 Feb 2008 03:45:25 +0000 Subject: [et-mgmt-tools] OS type detection in virt-install/virt-manager In-Reply-To: <47B31031.4050603@redhat.com> References: <47B2AC73.5090607@redhat.com> <20080213143257.GC32737@redhat.com> <47B31031.4050603@redhat.com> Message-ID: <20080214034525.GB1039@redhat.com> On Wed, Feb 13, 2008 at 04:43:45PM +0100, Alexander Todorov wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA512 > > Daniel P. Berrange wrote: > > On Wed, Feb 13, 2008 at 09:38:11AM +0100, Alexander Todorov wrote: > >> Hello, > >> I was looking at the recent code in OSDistro.py > >> Can we switch to using the new .treeinfo file and the information there > >> instead of the current if hasFile() mechanism? > >> The .treeinfo provides with all the information needed to detect the > >> family type (Fedora, RHEL, CentOS ???) and boot images location. > >> Also improve the acquire$FILE functions with that information. > > > > What new .treeinfo file ? It sounds interesting, but this is the first > > I've heard of it... Is there anywhere that describes it ? > > > > I'm afraid there's not. I've filed a BZ to request documentation. > https://bugzilla.redhat.com/show_bug.cgi?id=432641 > > > You probably didn't see the files because they start with a dot "." and are > hidden files. Try looking closer under the os/ directory for already released > versions. I've tried with Fedora 8 from my local mirror at > http://fedora.mirrors.megalan.bg/linux/releases/8/Fedora/x86_64/os/ > > The files are supposed to provide a generic way to get some info for the tree > incl. kernel and initrd locations. I was told that this is still work in progress. It does look very promising - its definitely something we should try to make use of in virt-install as it ought to make us more robust wrt to future changes. If anyone's interested in supplying patches to make use of the .treeinfo files, please send them to the list, otherwise we'll consider this a TODO item for our development roadmap. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From berrange at redhat.com Thu Feb 14 03:50:08 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 14 Feb 2008 03:50:08 +0000 Subject: [et-mgmt-tools] OS type detection in virt-install/virt-manager In-Reply-To: <47B30478.2070902@redhat.com> References: <47B2AC73.5090607@redhat.com> <20080213143257.GC32737@redhat.com> <47B30478.2070902@redhat.com> Message-ID: <20080214035008.GC1039@redhat.com> On Wed, Feb 13, 2008 at 09:53:44AM -0500, Cole Robinson wrote: > Daniel P. Berrange wrote: > > On Wed, Feb 13, 2008 at 09:38:11AM +0100, Alexander Todorov wrote: > >> Hello, > >> I was looking at the recent code in OSDistro.py > >> Can we switch to using the new .treeinfo file and the information there > >> instead of the current if hasFile() mechanism? > >> The .treeinfo provides with all the information needed to detect the > >> family type (Fedora, RHEL, CentOS ???) and boot images location. > >> Also improve the acquire$FILE functions with that information. > > > > What new .treeinfo file ? It sounds interesting, but this is the first > > I've heard of it... Is there anywhere that describes it ? > > > > I don't see a .treeinfo file on the Fedora download sites, eg > > > > http://download.fedora.redhat.com/pub/fedora/linux/development/x86_64/os/ > > > > http://download.fedora.redhat.com/pub/fedora/linux/releases/8/Fedora/x86_64/os/ > > > >> The reason I'm requesting this: in newer Fedora repos the packages dir > >> is simply called "Packages" but not "Fedora" and wrt remixing tools it > >> may be called with any name I think. OSDistro.py uses the presence of > >> fedora.css in the repo to decide on a Fedora distribution. This file > >> is not essential to the install process and may be missing in some repos. > >> > >> What do you think? > > > > If there's ways we can improve the OS detection, I'm open to any suggestions. > > > > Currently, we do the probing whether the user has manually selected a > > distro type or not. > > > > One optimization would be to only do the probing if the user has NOT > > explicitly selected a distro type. This would give users an easy way > > to force the install to continue even if the 'magic' files are missing. > > > > eg, if you select 'Fedora 8' in virt-manager it would just go straight > > to the kernel/initrd fetch stage. > > > > Dan. > > I updated the hasFile function recently to not grab files and instead just > test for their existence. This means we can change the distro detection > to actually just be dependent on whether the install media is found or > not. So the 'magic' file could be the file needed to actually run the > install which should avoid any misses for goofy trees. Yes, my original goal was that the file 'magic' would just serve as hinting for detecting distros / sanity checking. In the end I made it a mandatory check before doing anything - in retrospect this was a mistake because it made the system fragile wrt to change, and also adds download overhead. The 'hasFile' changes eliminate the overhead, but we still need to address the reliability issue - the treeinfo would help here for Fedora at least. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From tom at ng23.net Thu Feb 14 09:38:34 2008 From: tom at ng23.net (Tom Brown) Date: Thu, 14 Feb 2008 09:38:34 +0000 Subject: [et-mgmt-tools] Moving Cobbler Server In-Reply-To: <475D69A4.3080509@redhat.com> References: <475D5CA0.3040901@ng23.net> <475D61AB.8030805@redhat.com> <475D69A4.3080509@redhat.com> Message-ID: <47B40C1A.9030108@ng23.net> Michael DeHaan wrote: > > Here it is... > https://hosted.fedoraproject.org/projects/cobbler/wiki/MovingYourInstallToAnotherBox > > > I added a little bit more than was in my original email. > i tried this and hit a snag - changing the 'server' field in the setting file and then restarting cobblerd and also running the cobbler sync produces no errors however when i check on the generated kickstarts etc it still contains the old location in the URL's - i would have thought these would update dynamically. Any ideas? thanks From atodorov at redhat.com Thu Feb 14 10:30:22 2008 From: atodorov at redhat.com (Alexander Todorov) Date: Thu, 14 Feb 2008 11:30:22 +0100 Subject: [et-mgmt-tools] [PATCH] virtinst use .treeinfo for OS type detection files path In-Reply-To: <20080214034525.GB1039@redhat.com> References: <47B2AC73.5090607@redhat.com> <20080213143257.GC32737@redhat.com> <47B31031.4050603@redhat.com> <20080214034525.GB1039@redhat.com> Message-ID: <47B4183E.50508@redhat.com> Daniel P. Berrange wrote: > It does look very promising - its definitely something we should try to > make use of in virt-install as it ought to make us more robust wrt to > future changes. If anyone's interested in supplying patches to make use of > the .treeinfo files, please send them to the list, otherwise we'll consider > this a TODO item for our development roadmap. > Here's a completely UNTESTED patch. The .treeinfo file is available in Fedora and will be available for RHEL 5.2 and later and probably CentOS 5.2 as well. Fall backs to old code are still in place. Greetings, Alexander. -------------- next part -------------- A non-text attachment was scrubbed... Name: use_treeinfo.patch Type: text/x-patch Size: 5049 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 252 bytes Desc: OpenPGP digital signature URL: From agoodall at rm.com Thu Feb 14 10:33:38 2008 From: agoodall at rm.com (Adam Goodall) Date: Thu, 14 Feb 2008 10:33:38 -0000 Subject: [et-mgmt-tools] Moving Cobbler Server References: <475D5CA0.3040901@ng23.net> <475D61AB.8030805@redhat.com><475D69A4.3080509@redhat.com> <47B40C1A.9030108@ng23.net> Message-ID: <565F003189BD8D4A971B2AEE3550ED7A1D39BF@mail05.internal.rmplc.net> >-----Original Message----- >From: et-mgmt-tools-bounces at redhat.com [mailto:et-mgmt-tools->bounces at redhat.com] On Behalf Of Tom Brown >Sent: 14 February 2008 09:39 >To: Fedora/Linux Management Tools >Subject: Re: [et-mgmt-tools] Moving Cobbler Server > >i tried this and hit a snag - changing the 'server' field in the setting >file and then restarting cobblerd and also running the cobbler sync >produces no errors however when i check on the generated kickstarts etc >it still contains the old location in the URL's - i would have thought >these would update dynamically. > >Any ideas? I did this recently and had no problems with changing the server IP. I presume you do have @@server@@ In your kickstarts/snippets instead of a hardcoded address. If not Its time to sed all those files and swap it for that variable. With that in place you only need to update it in the settings file for it to propagate with a cobbler sync. ______________________________________________________________________ You might be interested in this... Over the next few months RM is hosting a number of free and exciting events... ...to ensure you don't miss out, visit: http://www.rm.com/busandconferences ______________________________________________________________________ P.S. Think Green - don't print this email unless you really need to. This message is confidential, so please treat it appropriately and for its intended purpose only. In particular, if it refers to any technical data, terms or prices not generally available or known, such items are "commercially sensitive information" within the terms of the Freedom of Information Act 2000 and related laws. As it would be prejudicial to RM's commercial interests if these were disclosed, please refrain from doing so. As Internet communications are not secure, please be aware that RM cannot accept responsibility for its contents. Any views or opinions presented are those of the author only and not of RM. If you are not the intended recipient of this e-mail, please accept our apologies and arrange for copies of it to be deleted. For your information, RM may intercept incoming and outgoing email communications. RM Education plc Registered Office: New Mill House, 183 Milton Park, Abingdon, Oxfordshire, OX14 4SE, England Registered Number: 1148594 From tom at ng23.net Thu Feb 14 11:13:19 2008 From: tom at ng23.net (Tom Brown) Date: Thu, 14 Feb 2008 11:13:19 +0000 Subject: [et-mgmt-tools] Moving Cobbler Server In-Reply-To: <565F003189BD8D4A971B2AEE3550ED7A1D39BF@mail05.internal.rmplc.net> References: <475D5CA0.3040901@ng23.net> <475D61AB.8030805@redhat.com><475D69A4.3080509@redhat.com> <47B40C1A.9030108@ng23.net> <565F003189BD8D4A971B2AEE3550ED7A1D39BF@mail05.internal.rmplc.net> Message-ID: <47B4224F.5000602@ng23.net> > > I did this recently and had no problems with changing the server IP. I > presume you do have > @@server@@ > In your kickstarts/snippets instead of a hardcoded address. If not Its > time to sed all those files and swap it for that variable. With that in > place you only need to update it in the settings file for it to > propagate with a cobbler sync. > indeed - thankyou, that was my issue From lippold at gmail.com Thu Feb 14 20:40:41 2008 From: lippold at gmail.com (Aaron Lippold) Date: Thu, 14 Feb 2008 15:40:41 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.7.2 test release (0.8 candidate) In-Reply-To: <47AC825B.6010409@redhat.com> References: <47AC825B.6010409@redhat.com> Message-ID: <39d2723b0802141240x2a29fe5ewf22bb6160d22d679@mail.gmail.com> Hi, I can't seem to find these cobbler 0.7.2 in the testing EPEL repo. Have they been uploaded? Jucain is a Cobbler newbie and is setting a cobbler server up on a Dell Xeon 1U box to help with some provisioning. If someone could send us the current links, I would appreciate it. Thanks, Aaron http://cobbler.et.redhat.com On Fri, Feb 8, 2008 at 11:24 AM, Michael DeHaan wrote: > After a long time developing the next version of Cobbler, it's time to > be finishing this version up. The motto should > be "release early, release often" so we'll be doing this a lot more in > the future. > > In about a week, I'll be releasing Cobbler 0.8, along with an updated > koan. This will include three months worth of > development on the devel branch as detailed here: > > *http://tinyurl.com/3y2el2 > * > So that that people have a change to test things first, I've uploaded > SRC rpms for Cobbler 0.7.2 and koan 0.6.5 to: > > http://cobbler.et.redhat.com/download/testing/ > > I've been using these development versions for a long time, as well as > many others, though additional testing before release is welcome. > If all goes well, I'll push new versions out to the mirrors in about a week. > > As with the previous release, I'll support fixes to 0.8 as we work on > the 0.9 development version. > > NOTE: We're going to do vmware in koan as a later update, the > groundwork has already been laid in Cobbler so it will mostly be a > change in koan. > > ALSO NOTE: If you're using the WebUI, setup instructions have changed, > and you will need to run through setup again. These changes > were needed to make things easier on new users as well as to accomodate > future features. Instructions are at > https://fedorahosted.org/cobbler/wiki/CobblerWebInterface > > Thanks, and as usual, let me know if you have any questions! > > --Michael > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From mdehaan at redhat.com Thu Feb 14 20:38:24 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 14 Feb 2008 15:38:24 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.7.2 test release (0.8 candidate) In-Reply-To: <39d2723b0802141240x2a29fe5ewf22bb6160d22d679@mail.gmail.com> References: <47AC825B.6010409@redhat.com> <39d2723b0802141240x2a29fe5ewf22bb6160d22d679@mail.gmail.com> Message-ID: <47B4A6C0.5060501@redhat.com> Aaron Lippold wrote: > Hi, > > I can't seem to find these cobbler 0.7.2 in the testing EPEL repo. > Have they been uploaded? > Jucain is a Cobbler newbie and is setting a cobbler server up on a > Dell Xeon 1U box to help with some provisioning. If someone could send > us the current links, I would appreciate it. > No problem. The email mentioned http://cobbler.et.redhat.com/download/testing/ Though I'll be releasing updated versions of those (with fixes from everyone's testing) to Fedora/EPEL tomorrow as 0.8/stable. --Michael From lippold at gmail.com Thu Feb 14 20:59:55 2008 From: lippold at gmail.com (Aaron Lippold) Date: Thu, 14 Feb 2008 15:59:55 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.7.2 test release (0.8 candidate) In-Reply-To: <47B4A6C0.5060501@redhat.com> References: <47AC825B.6010409@redhat.com> <39d2723b0802141240x2a29fe5ewf22bb6160d22d679@mail.gmail.com> <47B4A6C0.5060501@redhat.com> Message-ID: <39d2723b0802141259n3f74b11br36629d2d0959f91b@mail.gmail.com> Ah, Very cool. When you build the el5 rpm just send me and Jucain a link and we can let them rip :). Thanks, Aaron On Thu, Feb 14, 2008 at 3:38 PM, Michael DeHaan wrote: > Aaron Lippold wrote: > > Hi, > > > > I can't seem to find these cobbler 0.7.2 in the testing EPEL repo. > > Have they been uploaded? > > > Jucain is a Cobbler newbie and is setting a cobbler server up on a > > Dell Xeon 1U box to help with some provisioning. If someone could send > > us the current links, I would appreciate it. > > > > No problem. The email mentioned > > http://cobbler.et.redhat.com/download/testing/ > > Though I'll be releasing updated versions of those (with fixes from > everyone's testing) to Fedora/EPEL tomorrow as 0.8/stable. > > > > --Michael > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From lippold at gmail.com Thu Feb 14 21:00:57 2008 From: lippold at gmail.com (Aaron Lippold) Date: Thu, 14 Feb 2008 16:00:57 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.7.2 test release (0.8 candidate) In-Reply-To: <47B4A6C0.5060501@redhat.com> References: <47AC825B.6010409@redhat.com> <39d2723b0802141240x2a29fe5ewf22bb6160d22d679@mail.gmail.com> <47B4A6C0.5060501@redhat.com> Message-ID: <39d2723b0802141300h67ec01e2g95e55f6ba23b76c@mail.gmail.com> Hi, Well yes, but I was just letting you know that the el5 rpms weren't in the EPEL repo. I thought you had said you were going to put them there. I guess the process is that EL5 is built from the src rpms? Thanks, Aaron On Thu, Feb 14, 2008 at 3:38 PM, Michael DeHaan wrote: > Aaron Lippold wrote: > > Hi, > > > > I can't seem to find these cobbler 0.7.2 in the testing EPEL repo. > > Have they been uploaded? > > > Jucain is a Cobbler newbie and is setting a cobbler server up on a > > Dell Xeon 1U box to help with some provisioning. If someone could send > > us the current links, I would appreciate it. > > > > No problem. The email mentioned > > http://cobbler.et.redhat.com/download/testing/ > > Though I'll be releasing updated versions of those (with fixes from > everyone's testing) to Fedora/EPEL tomorrow as 0.8/stable. > > > > --Michael > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From mdehaan at redhat.com Thu Feb 14 20:53:11 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 14 Feb 2008 15:53:11 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.7.2 test release (0.8 candidate) In-Reply-To: <39d2723b0802141300h67ec01e2g95e55f6ba23b76c@mail.gmail.com> References: <47AC825B.6010409@redhat.com> <39d2723b0802141240x2a29fe5ewf22bb6160d22d679@mail.gmail.com> <47B4A6C0.5060501@redhat.com> <39d2723b0802141300h67ec01e2g95e55f6ba23b76c@mail.gmail.com> Message-ID: <47B4AA37.4040500@redhat.com> Aaron Lippold wrote: > Hi, > > Well yes, but I was just letting you know that the el5 rpms weren't in > the EPEL repo. I thought you had said you were going to put them > there. I guess the process is that EL5 is built from the src rpms? > > Thanks, > > Aaron > > On Thu, Feb 14, 2008 at 3:38 PM, Michael DeHaan wrote: > >> Aaron Lippold wrote: >> > Hi, >> > >> > I can't seem to find these cobbler 0.7.2 in the testing EPEL repo. >> > Have they been uploaded? >> >> > Jucain is a Cobbler newbie and is setting a cobbler server up on a >> > Dell Xeon 1U box to help with some provisioning. If someone could send >> > us the current links, I would appreciate it. >> > >> >> No problem. The email mentioned >> >> http://cobbler.et.redhat.com/download/testing/ >> >> Though I'll be releasing updated versions of those (with fixes from >> everyone's testing) to Fedora/EPEL tomorrow as 0.8/stable. >> >> >> >> --Michael >> >> >> >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools >> >> > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > EPEL is built from the tarballs, technically, though since I am not in control of what goes to EPEL stable vs testing (as is anyone), I choose to not push "testing" content to EPEL, so that people can use EPEL testing to get the latest -- even if EPEL has not yet rolled testing over to stable. Previously this happened roughly quarterly -- it's more often now IIRC -- but there's still an interest in the ability to supply a newer "stable" than EPEL "stable" allows. Make sense? :) From mdehaan at redhat.com Thu Feb 14 20:54:30 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 14 Feb 2008 15:54:30 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.7.2 test release (0.8 candidate) In-Reply-To: <39d2723b0802141259n3f74b11br36629d2d0959f91b@mail.gmail.com> References: <47AC825B.6010409@redhat.com> <39d2723b0802141240x2a29fe5ewf22bb6160d22d679@mail.gmail.com> <47B4A6C0.5060501@redhat.com> <39d2723b0802141259n3f74b11br36629d2d0959f91b@mail.gmail.com> Message-ID: <47B4AA86.8000101@redhat.com> Aaron Lippold wrote: > Ah, > > Very cool. > > When you build the el5 rpm just send me and Jucain a link and we can > let them rip :). > > Expect an announcemenent to the list tomorrow. From lippold at gmail.com Thu Feb 14 21:59:07 2008 From: lippold at gmail.com (Aaron Lippold) Date: Thu, 14 Feb 2008 16:59:07 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.7.2 test release (0.8 candidate) In-Reply-To: <47B4AA86.8000101@redhat.com> References: <47AC825B.6010409@redhat.com> <39d2723b0802141240x2a29fe5ewf22bb6160d22d679@mail.gmail.com> <47B4A6C0.5060501@redhat.com> <39d2723b0802141259n3f74b11br36629d2d0959f91b@mail.gmail.com> <47B4AA86.8000101@redhat.com> Message-ID: <39d2723b0802141359kab4c46ao2c1a5090dbceca08@mail.gmail.com> thanks On Thu, Feb 14, 2008 at 3:54 PM, Michael DeHaan wrote: > Aaron Lippold wrote: > > Ah, > > > > Very cool. > > > > When you build the el5 rpm just send me and Jucain a link and we can > > let them rip :). > > > > > Expect an announcemenent to the list tomorrow. > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From hbrock at redhat.com Thu Feb 14 22:04:56 2008 From: hbrock at redhat.com (Hugh O. Brock) Date: Thu, 14 Feb 2008 17:04:56 -0500 Subject: [et-mgmt-tools] Announcing oVirt Message-ID: <20080214220456.GJ15422@redhat.com> Announcing oVirt ================ It is my pleasure to announce oVirt, the next step in open virtual machine management. oVirt is: * A small OS image that runs libvirt and hosts virtual machines * A Web-based virtual machine management console oVirt goals: * Empower virtual machine owners without giving up control of hardware * Automate virtual machine clustering, load balancing, and SLA maintenance * Simplify management of large numbers of machines * Work across platforms and architectures oVirt uses: * A kerberos/LDAP server for authentication and authorization (oVirt ships with FreeIPA) * DNS/DHCP services on the local LAN -- or provides them for oVirt hosts over a private network if desired * Libvirt for virtual machine management, storage management, and secure remote communication * collectd for stats gathering and monitoring * Rails for rapid, flexible development oVirt mailing list: http://www.redhat.com/mailman/listinfo/ovirt-devel oVirt IRC: irc.freenode.net/#ovirt oVirt website: http://ovirt.org We encourage anyone interested to download the source (git clone git://git.et.redhat.com/ovirt) or the prebuilt appliance (http://ovirt.org/download). Let us know what you think! Enjoy, --Hugh Brock hbrock at redhat.com From markmc at redhat.com Fri Feb 15 10:26:38 2008 From: markmc at redhat.com (Mark McLoughlin) Date: Fri, 15 Feb 2008 10:26:38 +0000 Subject: [et-mgmt-tools] [PATCH] Fix virt-install disks typo Message-ID: # HG changeset patch # User Mark McLoughlin # Date 1203068244 0 # Node ID fd6a7fa855d11ca890f3c60f669c1830b9874f25 # Parent 07091304b75962460fd01c419a358e1a30fecb65 Fix virt-install disks typo Fixes this traceback: Traceback (most recent call last): File "./virt-install", line 498, in main() File "./virt-install", line 462, in main dom = guest.start_install(conscb,progresscb) File "/home/markmc/projects/virt/virtinst--devel/virtinst/Guest.py", line 813, in start_install return self._do_install(consolecb, meter) File "/home/markmc/projects/virt/virtinst--devel/virtinst/Guest.py", line 829, in _do_install self._create_devices(meter) File "/home/markmc/projects/virt/virtinst--devel/virtinst/Guest.py", line 725, in _create_devices disk.setup(progresscb) AttributeError: 'str' object has no attribute 'setup' Looks to have been introduced by cset 351:07091304b759 Signed-off-by: Mark McLoughlin diff -r 07091304b759 -r fd6a7fa855d1 virt-install --- a/virt-install Thu Feb 14 12:47:16 2008 -0500 +++ b/virt-install Fri Feb 15 09:37:24 2008 +0000 @@ -91,7 +91,7 @@ def get_disk(disk, size, sparse, guest, disk = None continue - guest.disks.append(disk) + guest.disks.append(d) break def get_disks(disk, size, sparse, nodisks, guest, hvm, conn): From hbrock at redhat.com Fri Feb 15 12:55:04 2008 From: hbrock at redhat.com (Hugh O. Brock) Date: Fri, 15 Feb 2008 07:55:04 -0500 Subject: [et-mgmt-tools] Re: et-mgnt] Announcing oVirt In-Reply-To: References: <20080214220456.GJ15422@redhat.com> Message-ID: <20080215125504.GO15422@redhat.com> On Thu, Feb 14, 2008 at 05:54:44PM -0500, R P Herrold wrote: > On Thu, 14 Feb 2008, Hugh O. Brock wrote: > >> Announcing oVirt > > license: GPL? the website seems not to say readily Yes, GPL2. It is in the source, we do need to add it to the website though. Actually it might be in the FAQ too... --Hugh From berrange at redhat.com Fri Feb 15 14:22:37 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 15 Feb 2008 14:22:37 +0000 Subject: [et-mgmt-tools] Re: et-mgnt] Announcing oVirt In-Reply-To: <20080215125504.GO15422@redhat.com> References: <20080214220456.GJ15422@redhat.com> <20080215125504.GO15422@redhat.com> Message-ID: <20080215142237.GA15860@redhat.com> On Fri, Feb 15, 2008 at 07:55:04AM -0500, Hugh O. Brock wrote: > On Thu, Feb 14, 2008 at 05:54:44PM -0500, R P Herrold wrote: > > On Thu, 14 Feb 2008, Hugh O. Brock wrote: > > > >> Announcing oVirt > > > > license: GPL? the website seems not to say readily > > Yes, GPL2. It is in the source, we do need to add it to the website > though. Actually it might be in the FAQ too... Yes, I clarified the FAQ entry yesterday to explicitly say GPLv2, and Chris added the GPLv2 header to all source files Regards, Dan -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From crobinso at redhat.com Fri Feb 15 15:33:12 2008 From: crobinso at redhat.com (Cole Robinson) Date: Fri, 15 Feb 2008 10:33:12 -0500 Subject: [et-mgmt-tools] [PATCH] Fix virt-install disks typo In-Reply-To: References: Message-ID: <47B5B0B8.8060903@redhat.com> Mark McLoughlin wrote: > # HG changeset patch > # User Mark McLoughlin > # Date 1203068244 0 > # Node ID fd6a7fa855d11ca890f3c60f669c1830b9874f25 > # Parent 07091304b75962460fd01c419a358e1a30fecb65 > Fix virt-install disks typo > > Fixes this traceback: > > Traceback (most recent call last): > File "./virt-install", line 498, in > main() > File "./virt-install", line 462, in main > dom = guest.start_install(conscb,progresscb) > File "/home/markmc/projects/virt/virtinst--devel/virtinst/Guest.py", line 813, in start_install > return self._do_install(consolecb, meter) > File "/home/markmc/projects/virt/virtinst--devel/virtinst/Guest.py", line 829, in _do_install > self._create_devices(meter) > File "/home/markmc/projects/virt/virtinst--devel/virtinst/Guest.py", line 725, in _create_devices > disk.setup(progresscb) > AttributeError: 'str' object has no attribute 'setup' > > Looks to have been introduced by cset 351:07091304b759 Argh, yes it was. Spent all my time testing the virt-clone pieces and forgot to test virt-install :\ Applied now: http://hg.et.redhat.com/virt/applications/virtinst--devel?cs=fd6a7fa855d1 Thanks, Cole From berrange at redhat.com Fri Feb 15 18:25:49 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 15 Feb 2008 18:25:49 +0000 Subject: [et-mgmt-tools] [PATCH] The bridge of bonding and the tag VLAN isnot displayed in virt-manager In-Reply-To: <200802081731.DJC60956.EJG9K096@aa.jp.fujitsu.com> References: <200802071452.HGG65121.69J0EKG9@aa.jp.fujitsu.com> <20080207125623.GB30230@redhat.com> <200802081731.DJC60956.EJG9K096@aa.jp.fujitsu.com> Message-ID: <20080215182549.GA24578@redhat.com> On Fri, Feb 08, 2008 at 05:31:42PM +0900, S.Sakamoto wrote: > > > virt-manager can display only the bridge of the physics network device now > > > when add a device and make a guest domain newly. > > > A bridge of bonding and Tag VLAN is not displayed. > > > > > > So, I make the patch that a bridge of bonding and Tag VLAN came to be displayed. > > > Do not getout from processing specific linux... > > > but can use a bridge of bonding and the tag VLAN. > > > > I'd definitely welcome code to deal with VLANs - and we're already Linux > > specific in this code, so that's not a problem. > > > > We can't use the 'rhpl*' python libraries though - these are Red Hat > > specific libraries, so if we use them, virt-manager won't be able to run > > on Debian/SUSE/etc > > > > If you can adapt the code to remove use of 'import rhpl' then that'd be > > good. > > I remake the patch to remove use of 'import rhpl'. After much testing on some horrible different networking configs, here is an updated patch which I think should work correctly. - Instead of using /etc/modprobe.conf, it reads /sys/class/net/bonding_masters to get an accurate list of bonding devices - It doesn't use the HAL linux.sysfs_path patch when looking for VLANs because that is /sys/devices/pci0000:00/0000:00:1e.0/0000:01:00.0/net/eth0 but the VLANs are only visible under /sys/class/net/eth0.* - It filters out any physical devices that are part of a bonding master so they don't appear in the list twice I've tested with: - eth0 as bridge, and peth0 as physical device - br0 as bridge and eth0 as physical device - bond0 with eth0 as slave, and brbond0 as bridge - eth0.5 vlan with breth05 as bridge Which I think covers the use cases your patch was trying to address. The algorithm it uses is thus: - Find all bonding master devices - Add them to netdev list - Mark shared if in a bridge - Find all physical devices from HAL - Skip if a slave in a bonding device - Check if renamed from ethN -> pethN - Add them to netdev list - Mark shared if in a bridge - For all VLANs associated with physical device - Add them to netdev list - Mark shared if in a bridge Let me know if this patch shows the correct lists for you - if not the /root/.virt-manager/virt-manager.log file should contain useful debug output diff -r 1892867ca5c7 src/virtManager/connection.py --- a/src/virtManager/connection.py Thu Jan 31 11:39:10 2008 -0500 +++ b/src/virtManager/connection.py Fri Feb 15 13:18:56 2008 -0500 @@ -22,6 +22,7 @@ import libvirt import libvirt import logging import os, sys +import glob import traceback from time import time import logging @@ -29,6 +30,7 @@ import dbus import dbus import threading import gtk +import string from virtManager.domain import vmmDomain from virtManager.network import vmmNetwork @@ -134,9 +136,7 @@ class vmmConnection(gobject.GObject): self.record = [] self.hostinfo = None - self.detect_network_devices() - - def detect_network_devices(self): + # Probe for network devices try: # Get a connection to the SYSTEM bus self.bus = dbus.SystemBus() @@ -145,12 +145,21 @@ class vmmConnection(gobject.GObject): self.hal_iface = dbus.Interface(hal_object, 'org.freedesktop.Hal.Manager') # Track device add/removes so we can detect newly inserted CD media - self.hal_iface.connect_to_signal("DeviceAdded", self._device_added) - self.hal_iface.connect_to_signal("DeviceRemoved", self._device_removed) - - # Find info about all current present media + self.hal_iface.connect_to_signal("DeviceAdded", self._net_phys_device_added) + self.hal_iface.connect_to_signal("DeviceRemoved", self._net_phys_device_removed) + + # find all bonding master devices and register them + # XXX bonding stuff is linux specific + bondMasters = self._net_get_bonding_masters() + for bond in bondMasters: + sysfspath = "/sys/class/net/" + bond + mac = self._net_get_mac_address(bond, sysfspath) + self._net_device_added(bond, mac, sysfspath) + + # Find info about all current present physical net devices + # This is OS portable... for path in self.hal_iface.FindDeviceByCapability("net"): - self._device_added(path) + self._net_phys_device_added(path) except: (type, value, stacktrace) = sys.exc_info () logging.error("Unable to connect to HAL to list network devices: '%s'" + \ @@ -159,56 +168,70 @@ class vmmConnection(gobject.GObject): self.bus = None self.hal_iface = None - def _device_added(self, path): + def _net_phys_device_added(self, path): + logging.debug("Got physical device %s" % path) obj = self.bus.get_object("org.freedesktop.Hal", path) if obj.QueryCapability("net"): name = obj.GetPropertyString("net.interface") + # XXX ...but this is Linux specific again - patches welcomed + #sysfspath = obj.GetPropertyString("linux.sysfs_path") + # XXX hal gives back paths to /sys/devices/pci0000:00/0000:00:1e.0/0000:01:00.0/net/eth0 + # which doesnt' work so well - we want this: + sysfspath = "/sys/class/net/" + name + + # If running a device in bridged mode, there's a reasonable + # chance that the actual ethernet device has been renamed to + # something else. ethN -> pethN + psysfspath = sysfspath[0:len(sysfspath)-len(name)] + "p" + name + if os.path.exists(psysfspath): + logging.debug("Device %s named to p%s" % (name, name)) + name = "p" + name + sysfspath = psysfspath + + # Ignore devices that are slaves of a bond + if self._net_is_bonding_slave(name, sysfspath): + logging.debug("Skipping device %s in bonding slave" % name) + return + mac = obj.GetPropertyString("net.address") - # Now magic to determine if the device is part of a bridge - shared = False - bridge = None - try: - # XXX Linux specific - needs porting for other OS - patches - # welcomed... - sysfspath = obj.GetPropertyString("linux.sysfs_path") - - # If running a device in bridged mode, there's a reasonable - # chance that the actual ethernet device has been renamed to - # something else. ethN -> pethN - psysfspath = sysfspath[0:len(sysfspath)-len(name)] + "p" + name - if os.path.exists(psysfspath): - name = "p" + name - sysfspath = psysfspath - - brportpath = os.path.join(sysfspath, "brport") - - if os.path.exists(brportpath): - shared = True - brlinkpath = os.path.join(brportpath, "bridge") - dest = os.readlink(brlinkpath) - (head,tail) = os.path.split(dest) - bridge = tail - except: - (type, value, stacktrace) = sys.exc_info () - logging.error("Unable to determine if device is shared:" + - str(type) + " " + str(value) + "\n" + \ - traceback.format_exc (stacktrace)) - - if self.netdevs.has_key(path): - currDev = self.netdevs[path] - if currDev.get_info() == (name, mac, shared, bridge): - return - del self.netdevs[path] - dev = vmmNetDevice(self.config, self, name, mac, shared, bridge) - self.netdevs[path] = dev - self.emit("netdev-added", dev.get_name()) - - def _device_removed(self, path): - if self.netdevs.has_key(path): - dev = self.netdevs[path] + # Add the main NIC + self._net_device_added(name, mac, sysfspath) + + # Add any associated VLANs + logging.debug("Checking for VLANs on %s" % sysfspath) + for vlanpath in glob.glob(sysfspath + ".*"): + if os.path.exists(vlanpath): + logging.debug("Process VLAN %s" % vlanpath) + vlanmac = self._net_get_mac_address(name, vlanpath) + (ignore,vlanname) = os.path.split(vlanpath) + self._net_device_added(vlanname, vlanmac, vlanpath) + + def _net_device_added(self, name, mac, sysfspath): + # Race conditions mean we can occassionally see device twice + if self.netdevs.has_key(name): + return + + bridge = self._net_get_bridge_owner(name, sysfspath) + shared = False + if bridge is not None: + shared = True + + logging.debug("Adding net device %s %s %s bridge %s" % (name, mac, sysfspath, str(bridge))) + + dev = vmmNetDevice(self.config, self, name, mac, shared, bridge) + self.netdevs[name] = dev + self.emit("netdev-added", dev.get_name()) + + def _net_phys_device_removed(self, path): + obj = self.bus.get_object("org.freedesktop.Hal", path) + if obj.QueryCapability("net"): + name = obj.GetPropertyString("net.interface") + + if self.netdevs.has_key(name): + dev = self.netdevs[name] self.emit("netdev-removed", dev.get_name()) - del self.netdevs[path] + del self.netdevs[name] def is_read_only(self): return self.readOnly @@ -545,13 +568,6 @@ class vmmConnection(gobject.GObject): newInactiveNetNames = self.vmm.listDefinedNetworks() except: logging.warn("Unable to list inactive networks") - - # check of net devices - newPaths = [] - if self.hal_iface: - newPaths = self.hal_iface.FindDeviceByCapability("net") - for newPath in newPaths: - self._device_added(newPath) for name in newActiveNetNames: net = self.vmm.networkLookupByName(name) @@ -847,5 +863,47 @@ class vmmConnection(gobject.GObject): else: return _("Unknown") + def _net_get_bridge_owner(self, name, sysfspath): + # Now magic to determine if the device is part of a bridge + brportpath = os.path.join(sysfspath, "brport") + try: + if os.path.exists(brportpath): + brlinkpath = os.path.join(brportpath, "bridge") + dest = os.readlink(brlinkpath) + (ignore,bridge) = os.path.split(dest) + return bridge + except: + (type, value, stacktrace) = sys.exc_info () + logging.error("Unable to determine if device is shared:" + + str(type) + " " + str(value) + "\n" + \ + traceback.format_exc (stacktrace)) + + return None + + def _net_get_mac_address(self, name, sysfspath): + mac = None + addrpath = sysfspath + "/address" + if os.path.exists(addrpath): + df = open(addrpath, 'r') + mac = df.readline() + df.close() + return mac.strip(" \n\t") + + def _net_get_bonding_masters(self): + masters = [] + f = open("/sys/class/net/bonding_masters") + while True: + rline = f.readline() + if not rline: break + if rline == "\x00": continue + masters.append(rline.strip(" \n\t")) + return masters + + def _net_is_bonding_slave(self, name, sysfspath): + masterpath = sysfspath + "/master" + if os.path.exists(masterpath): + return True + return False + gobject.type_register(vmmConnection) Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From mdehaan at redhat.com Fri Feb 15 20:27:42 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 15 Feb 2008 15:27:42 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.8, Koan 0.8 Message-ID: <47B5F5BE.3060707@redhat.com> Hi folks, Cobbler 0.8 and Koan 0.8 are now built and released to the mirrors. You can also find src RPMs at http://cobbler.et.redhat.com/download The executive summary of features -- reworked command line now provides help, can be extended with modules -- improved API -- improved logging -- lots of bugfixes -- lots of miscellaneous options and feature requests -- easier to install Web UI, WebUI/XMLRPC has pluggable auth (more work to be done here, esp WRT kerberos) -- recursive syncs mean editing an object with a cobbler command means you don't have to run "sync" -- ability to do recursive deletes, able to move and rename objects that have child objects, etc -- koan can be used without a cobbler server if you want to reinstall systems (though it's best with one!) Enjoy... 0.8.X will now be kept as a stable release while new development will happen on the 0.9 devel branch. As usual, new feature requests, ideas, and patches are always welcome. Thanks again to everyone who helped make this release! --Michael From mdehaan at redhat.com Fri Feb 15 20:32:37 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 15 Feb 2008 15:32:37 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.8, Koan 0.8 In-Reply-To: <47B5F5BE.3060707@redhat.com> References: <47B5F5BE.3060707@redhat.com> Message-ID: <47B5F6E5.6000901@redhat.com> Michael DeHaan wrote: > Hi folks, > > Cobbler 0.8 and Koan 0.8 are now built and released to the mirrors. > > You can also find src RPMs at http://cobbler.et.redhat.com/download > > The executive summary of features > -- reworked command line now provides help, can be extended with > modules > -- improved API > -- improved logging > -- lots of bugfixes > -- lots of miscellaneous options and feature requests > -- easier to install Web UI, WebUI/XMLRPC has pluggable auth (more > work to be done here, esp WRT kerberos) > -- recursive syncs mean editing an object with a cobbler command > means you don't have to run "sync" > -- ability to do recursive deletes, able to move and rename objects > that have child objects, etc -- koan can be used without a cobbler > server if you want to reinstall systems (though it's best with one!) > > Enjoy... 0.8.X will now be kept as a stable release while new > development will happen on the 0.9 devel branch. > > As usual, new feature requests, ideas, and patches are always welcome. > Thanks again to everyone who helped make this release! > > --Michael > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools One quick (but important) note... If you are using the Web user interface and were running a previous version of Cobbler, please go through the setup instructions here. They have changed since the 0.6.5 release. https://fedorahosted.org/cobbler/wiki/CobblerWebInterface --Michael From jeremy at rosengren.org Sat Feb 16 06:24:51 2008 From: jeremy at rosengren.org (Jeremy Rosengren) Date: Sat, 16 Feb 2008 00:24:51 -0600 Subject: [et-mgmt-tools] [PATCH] Fix typo in action_reposync.py self.settings call and change http_port to "str" Message-ID: This is my first patch generated with git, so hopefully I've done this correctly. I installed Cobbler 0.8 today and noticed that one of my repos which is defined as an existing yum repository (http://) generated an error. This patch fixes the typo which caused the error and changed the default http_port setting to be a string instead of an integer - hopefully that doesn't break something else. Thanks, -- jeremy --- cobbler/action_reposync.py | 2 +- cobbler/settings.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/cobbler/action_reposync.py b/cobbler/action_reposync.py index 06340aa..6e62489 100644 --- a/cobbler/action_reposync.py +++ b/cobbler/action_reposync.py @@ -244,7 +244,7 @@ class RepoSync: config_file.write("%s=%s\n" % (x, repo.yumopts[x])) else: line = "baseurl=%s\n" % repo.mirror - http_server = self.setting.server + ":" + self.settings.http_port + http_server = self.settings.server + ":" + self.settings.http_port line = line.replace("@@server@@",http_server) config_file.write(line) config_file.write("enabled=1\n") diff --git a/cobbler/settings.py b/cobbler/settings.py index 86a24af..8cd7032 100644 --- a/cobbler/settings.py +++ b/cobbler/settings.py @@ -36,7 +36,7 @@ DEFAULTS = { "dnsmasq_bin" : "/usr/sbin/dnsmasq", "dnsmasq_conf" : "/etc/dnsmasq.conf", "httpd_bin" : "/usr/sbin/httpd", - "http_port" : 80, + "http_port" : "80", "kerberos_realm" : "example.org", "kernel_options" : { "lang" : " ", -- 1.5.4 From Hua.Zhang at Sun.COM Sat Feb 16 09:06:50 2008 From: Hua.Zhang at Sun.COM (Henry Zhang) Date: Sat, 16 Feb 2008 17:06:50 +0800 Subject: [et-mgmt-tools] [Patch] Make Host windows About to show about information In-Reply-To: <20071021160016.7DC81732FE@hormel.redhat.com> References: <20071021160016.7DC81732FE@hormel.redhat.com> Message-ID: <47B6A7AA.4080107@sun.com> Hi there, Please help to review this little patch, since when I run virt-manager 0.4.0 at Solaris, I find About in Host will not work, seems it's trying to show help infor when select About, but in fact there is no help infor. for it, and I think it should link to about information, so I add the attached patch... Thanks, Henry -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-manager-05-host-about.diff Type: text/x-patch Size: 2303 bytes Desc: not available URL: From Hua.Zhang at Sun.COM Sat Feb 16 10:24:30 2008 From: Hua.Zhang at Sun.COM (Henry Zhang) Date: Sat, 16 Feb 2008 18:24:30 +0800 Subject: [et-mgmt-tools] [Patch] Build virt-manager at Solaris Message-ID: <47B6B9DE.30702@sun.com> Hi there, When I tried to build virt-manager 0.4.0 at Solaris, failed, because several parameters can't pass at Solaris, I have to remove them, please review, if it's possible to remove them from your source codes... Thanks, Henry -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-manager-01-build-flags.diff Type: text/x-patch Size: 600 bytes Desc: not available URL: From berrange at redhat.com Sat Feb 16 17:52:16 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Sat, 16 Feb 2008 17:52:16 +0000 Subject: [et-mgmt-tools] [Patch] Make Host windows About to show about information In-Reply-To: <47B6A7AA.4080107@sun.com> References: <20071021160016.7DC81732FE@hormel.redhat.com> <47B6A7AA.4080107@sun.com> Message-ID: <20080216175216.GD24669@redhat.com> On Sat, Feb 16, 2008 at 05:06:50PM +0800, Henry Zhang wrote: > Hi there, > > Please help to review this little patch, since when I run virt-manager > 0.4.0 at Solaris, I find About in Host will not work, seems it's trying > to show help infor when select About, but in fact there is no help > infor. for it, and I think it should link to about information, so I add > the attached patch... THanks for the patch - i've applied it. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From berrange at redhat.com Sat Feb 16 18:36:19 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Sat, 16 Feb 2008 18:36:19 +0000 Subject: [et-mgmt-tools] Re: [PATCH] virtinst use .treeinfo for OS type detection files path In-Reply-To: <47B4183E.50508@redhat.com> References: <47B2AC73.5090607@redhat.com> <20080213143257.GC32737@redhat.com> <47B31031.4050603@redhat.com> <20080214034525.GB1039@redhat.com> <47B4183E.50508@redhat.com> Message-ID: <20080216183619.GE24669@redhat.com> On Thu, Feb 14, 2008 at 11:30:22AM +0100, Alexander Todorov wrote: > Daniel P. Berrange wrote: > > It does look very promising - its definitely something we should try to > > make use of in virt-install as it ought to make us more robust wrt to > > future changes. If anyone's interested in supplying patches to make use of > > the .treeinfo files, please send them to the list, otherwise we'll consider > > this a TODO item for our development roadmap. > > > Here's a completely UNTESTED patch. > The .treeinfo file is available in Fedora and will be available for RHEL 5.2 and > later and probably CentOS 5.2 as well. > Fall backs to old code are still in place. I tested it against Fedora and it works great, so I've committed this. BTW, please just use spaces and not tabs for patches - python complained about inconsistent indentation due to the tabs. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From smooge at gmail.com Sun Feb 17 00:35:37 2008 From: smooge at gmail.com (Stephen John Smoogen) Date: Sat, 16 Feb 2008 17:35:37 -0700 Subject: [et-mgmt-tools] Cobbler reposync error with 0.8 Message-ID: <80d7e4090802161635g16beff5axbe3300d962229e78@mail.gmail.com> System is RHEL-5 with cobbler-0.8 updates. /var/www/cobbler/repo_mirror/rhel-i386-server-cluster-5/getPackage/system-config-cluster-1.0.50-1.3.noarch.rpm already exists and appears to be complete - creating: /var/www/cobbler/repo_mirror/.origin/rhel-i386-server-cluster-5.repo Traceback (most recent call last): File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 54, in main return BootCLI().run(sys.argv) File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 43, in run return self.loader.run(args) File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line 69, in run return fn.run() File "/usr/lib/python2.4/site-packages/cobbler/modules/cli_misc.py", line 227, in run return self.api.reposync(self.options.only) File "/usr/lib/python2.4/site-packages/cobbler/api.py", line 332, in reposync return reposync.run(name) File "/usr/lib/python2.4/site-packages/cobbler/action_reposync.py", line 74, in run self.do_reposync(repo) File "/usr/lib/python2.4/site-packages/cobbler/action_reposync.py", line 180, in do_reposync temp_file = self.create_local_file(repo, temp_path, output=False) File "/usr/lib/python2.4/site-packages/cobbler/action_reposync.py", line 247, in create_local_file http_server = self.setting.server + ":" + self.settings.http_port AttributeError: RepoSync instance has no attribute 'setting' Doing this with any repository (using --only=fedora-8-updates for instance) gives the same error. My guess is that there is a 0.6.5->0.8 config change missing. This led to another error [root at snorbtz ~]# cobbler reposync -h usage: cobbler reposync [ARGS] options: -h, --help show this help message and exit --only=ONLY update only this repository name Traceback (most recent call last): File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 54, in main return BootCLI().run(sys.argv) File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 43, in run return self.loader.run(args) File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line 66, in run loaded_ok = fn.parse_args(args) File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line 190, in parse_args (self.options, self.args) = p.parse_args(args) File "/usr/lib/python2.4/optparse.py", line 1275, in parse_args stop = self._process_args(largs, rargs, values) File "/usr/lib/python2.4/optparse.py", line 1319, in _process_args self._process_short_opts(rargs, values) File "/usr/lib/python2.4/optparse.py", line 1426, in _process_short_opts option.process(opt, value, values, self) File "/usr/lib/python2.4/optparse.py", line 707, in process return self.take_action( File "/usr/lib/python2.4/optparse.py", line 729, in take_action parser.exit() File "/usr/lib/python2.4/optparse.py", line 1449, in exit sys.exit(status) SystemExit: 0 -- Stephen J Smoogen. -- CSIRT/Linux System Administrator How far that little candle throws his beams! So shines a good deed in a naughty world. = Shakespeare. "The Merchant of Venice" From jeremy at rosengren.org Sun Feb 17 18:02:31 2008 From: jeremy at rosengren.org (Jeremy Rosengren) Date: Sun, 17 Feb 2008 12:02:31 -0600 Subject: [et-mgmt-tools] Cobbler reposync error with 0.8 In-Reply-To: <80d7e4090802161635g16beff5axbe3300d962229e78@mail.gmail.com> References: <80d7e4090802161635g16beff5axbe3300d962229e78@mail.gmail.com> Message-ID: <50970541-2AFB-45C3-8C9F-E74A9A5D2A49@rosengren.org> The patch I submitted the other day addresses the first problem - the last line indicates the error: self.setting.server should be self.settings.server. Once I fixed that typo, there was another error because the "http_port" default setting was an 'int' and not a 'str', so I fixed that as well by making http_port a string in settings.py. If any other parts of the code rely on http_port being an integer, it should be trivial to fix the http_server line. -- j On Feb 16, 2008, at 6:35 PM, Stephen John Smoogen wrote: > System is RHEL-5 with cobbler-0.8 updates. > > /var/www/cobbler/repo_mirror/rhel-i386-server-cluster-5/getPackage/ > system-config-cluster-1.0.50-1.3.noarch.rpm > already exists and appears to be complete > - creating: /var/www/cobbler/repo_mirror/.origin/rhel-i386-server- > cluster-5.repo > Traceback (most recent call last): > File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line > 54, in main > return BootCLI().run(sys.argv) > File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line > 43, in run > return self.loader.run(args) > File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line > 69, in run > return fn.run() > File "/usr/lib/python2.4/site-packages/cobbler/modules/cli_misc.py", > line 227, in run > return self.api.reposync(self.options.only) > File "/usr/lib/python2.4/site-packages/cobbler/api.py", line 332, > in reposync > return reposync.run(name) > File "/usr/lib/python2.4/site-packages/cobbler/action_reposync.py", > line 74, in run > self.do_reposync(repo) > File "/usr/lib/python2.4/site-packages/cobbler/action_reposync.py", > line 180, in do_reposync > temp_file = self.create_local_file(repo, temp_path, output=False) > File "/usr/lib/python2.4/site-packages/cobbler/action_reposync.py", > line 247, in create_local_file > http_server = self.setting.server + ":" + self.settings.http_port > AttributeError: RepoSync instance has no attribute 'setting' > > Doing this with any repository (using --only=fedora-8-updates for > instance) gives the same error. My guess is that there is a 0.6.5->0.8 > config change missing. > > This led to another error > > [root at snorbtz ~]# cobbler reposync -h > usage: cobbler reposync [ARGS] > > options: > -h, --help show this help message and exit > --only=ONLY update only this repository name > Traceback (most recent call last): > File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line > 54, in main > return BootCLI().run(sys.argv) > File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line > 43, in run > return self.loader.run(args) > File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line > 66, in run > loaded_ok = fn.parse_args(args) > File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line > 190, in parse_args > (self.options, self.args) = p.parse_args(args) > File "/usr/lib/python2.4/optparse.py", line 1275, in parse_args > stop = self._process_args(largs, rargs, values) > File "/usr/lib/python2.4/optparse.py", line 1319, in _process_args > self._process_short_opts(rargs, values) > File "/usr/lib/python2.4/optparse.py", line 1426, in > _process_short_opts > option.process(opt, value, values, self) > File "/usr/lib/python2.4/optparse.py", line 707, in process > return self.take_action( > File "/usr/lib/python2.4/optparse.py", line 729, in take_action > parser.exit() > File "/usr/lib/python2.4/optparse.py", line 1449, in exit > sys.exit(status) > SystemExit: 0 > > > > -- > Stephen J Smoogen. -- CSIRT/Linux System Administrator > How far that little candle throws his beams! So shines a good deed > in a naughty world. = Shakespeare. "The Merchant of Venice" > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From berrange at redhat.com Sun Feb 17 21:49:32 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Sun, 17 Feb 2008 21:49:32 +0000 Subject: [et-mgmt-tools] FYI: bugfix to virt-image to allow non-raw images Message-ID: <20080217214932.GB10935@redhat.com> The virt-image tool is explicitly rejecting any disk images which are not in the raw or iso format. Some patches I did to the Fedora live CD tools allow creation of appliances & virt-image XML using any disk format supported by qemu-img: http://www.redhat.com/archives/fedora-livecd-list/2008-February/msg00085.html http://www.redhat.com/archives/fedora-livecd-list/2008-February/msg00090.html If I loosen the virt-image disk format checks to allow qcow, qcow2, and vmdk it all works fine, so I've comitted this change http://hg.et.redhat.com/virt/applications/virtinst--devel?cs=3b2bc0e2b060 Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From fj0588di at aa.jp.fujitsu.com Mon Feb 18 08:27:22 2008 From: fj0588di at aa.jp.fujitsu.com (S.Sakamoto) Date: Mon, 18 Feb 2008 17:27:22 +0900 Subject: [et-mgmt-tools] [PATCH] The bridge of bonding and the tag VLANisnot displayed in virt-manager In-Reply-To: <20080215182549.GA24578@redhat.com> References: <200802071452.HGG65121.69J0EKG9@aa.jp.fujitsu.com> <20080207125623.GB30230@redhat.com> <200802081731.DJC60956.EJG9K096@aa.jp.fujitsu.com> <20080215182549.GA24578@redhat.com> Message-ID: <200802181727.HFI21810.J9KEG690@aa.jp.fujitsu.com> > After much testing on some horrible different networking configs, here is > an updated patch which I think should work correctly. > > - Instead of using /etc/modprobe.conf, it reads /sys/class/net/bonding_masters > to get an accurate list of bonding devices > - It doesn't use the HAL linux.sysfs_path patch when looking for VLANs > because that is /sys/devices/pci0000:00/0000:00:1e.0/0000:01:00.0/net/eth0 > but the VLANs are only visible under /sys/class/net/eth0.* > - It filters out any physical devices that are part of a bonding master > so they don't appear in the list twice > > I've tested with: > > - eth0 as bridge, and peth0 as physical device > - br0 as bridge and eth0 as physical device > - bond0 with eth0 as slave, and brbond0 as bridge > - eth0.5 vlan with breth05 as bridge > > Which I think covers the use cases your patch was trying to address. > > The algorithm it uses is thus: > > - Find all bonding master devices > - Add them to netdev list > - Mark shared if in a bridge > > - Find all physical devices from HAL > - Skip if a slave in a bonding device > - Check if renamed from ethN -> pethN > - Add them to netdev list > - Mark shared if in a bridge > > - For all VLANs associated with physical device > - Add them to netdev list > - Mark shared if in a bridge > > Let me know if this patch shows the correct lists for you - if not the > /root/.virt-manager/virt-manager.log file should contain useful debug > output Thank you for improving a patch. I tested your patch. I think that do not have any problem basically. But, there are two points that should have revised. - A tag on bond is not considered. - cannot acquire correct bond name from bonding_master. I made the patch that revised for the above points. In this revision, the problem for me is solved. Thanks, Shigeki Sakamoto. -------------- next part -------------- A non-text attachment was scrubbed... Name: connection.patch Type: application/octet-stream Size: 9865 bytes Desc: not available URL: From berrange at redhat.com Mon Feb 18 15:02:41 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 18 Feb 2008 15:02:41 +0000 Subject: [et-mgmt-tools] [PATCH] The bridge of bonding and the tag VLANisnot displayed in virt-manager In-Reply-To: <200802181727.HFI21810.J9KEG690@aa.jp.fujitsu.com> References: <200802071452.HGG65121.69J0EKG9@aa.jp.fujitsu.com> <20080207125623.GB30230@redhat.com> <200802081731.DJC60956.EJG9K096@aa.jp.fujitsu.com> <20080215182549.GA24578@redhat.com> <200802181727.HFI21810.J9KEG690@aa.jp.fujitsu.com> Message-ID: <20080218150241.GB30725@redhat.com> On Mon, Feb 18, 2008 at 05:27:22PM +0900, S.Sakamoto wrote: > > The algorithm it uses is thus: > > > > - Find all bonding master devices > > - Add them to netdev list > > - Mark shared if in a bridge > > > > - Find all physical devices from HAL > > - Skip if a slave in a bonding device > > - Check if renamed from ethN -> pethN > > - Add them to netdev list > > - Mark shared if in a bridge > > > > - For all VLANs associated with physical device > > - Add them to netdev list > > - Mark shared if in a bridge > > > > Let me know if this patch shows the correct lists for you - if not the > > /root/.virt-manager/virt-manager.log file should contain useful debug > > output > > Thank you for improving a patch. > I tested your patch. > I think that do not have any problem basically. > > But, there are two points that should have revised. > - A tag on bond is not considered. > - cannot acquire correct bond name from bonding_master. > > I made the patch that revised for the above points. > In this revision, the problem for me is solved. Excellant - this works for me too, so I've committed to the repo http://hg.et.redhat.com/virt/applications/virt-manager--devel?cs=6a155cfe437c Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From atodorov at redhat.com Mon Feb 18 15:32:46 2008 From: atodorov at redhat.com (Alexander Todorov) Date: Mon, 18 Feb 2008 16:32:46 +0100 Subject: [et-mgmt-tools] Re: [PATCH] virtinst use .treeinfo for OS type detection files path In-Reply-To: <20080216183619.GE24669@redhat.com> References: <47B2AC73.5090607@redhat.com> <20080213143257.GC32737@redhat.com> <47B31031.4050603@redhat.com> <20080214034525.GB1039@redhat.com> <47B4183E.50508@redhat.com> <20080216183619.GE24669@redhat.com> Message-ID: <47B9A51E.9070405@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Daniel P. Berrange wrote: > I tested it against Fedora and it works great, so I've committed this. > Glad to hear it worked. > BTW, please just use spaces and not tabs for patches - python complained > about inconsistent indentation due to the tabs. > Errr. I suppose that's what "Tab to spaces" is for. However it seems to not work every time. Thanks for fixing the indents. Greetings, Alexander. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Red Hat - http://enigmail.mozdev.org iD8DBQFHuaUbhmd3WOiFct4RCueqAJ4lO60q1LHcLK0yTwyokTYVuGZP0QCgsqq+ 4dUKmLpYG0r++Yt4DmOoUtM= =YN0d -----END PGP SIGNATURE----- From mdehaan at redhat.com Mon Feb 18 15:44:42 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 18 Feb 2008 10:44:42 -0500 Subject: [et-mgmt-tools] Cobbler reposync error with 0.8 In-Reply-To: <50970541-2AFB-45C3-8C9F-E74A9A5D2A49@rosengren.org> References: <80d7e4090802161635g16beff5axbe3300d962229e78@mail.gmail.com> <50970541-2AFB-45C3-8C9F-E74A9A5D2A49@rosengren.org> Message-ID: <47B9A7EA.4050408@redhat.com> Jeremy Rosengren wrote: > The patch I submitted the other day addresses the first problem - the > last line indicates the error: self.setting.server should be > self.settings.server. > Jeremy, very much appreciated. I'll give it a few days to see if any other problems come up and then we can roll them into a 0.8.1. --Michael > Once I fixed that typo, there was another error because the > "http_port" default setting was an 'int' and not a 'str', so I fixed > that as well by making http_port a string in settings.py. If any other > parts of the code rely on http_port being an integer, it should be > trivial to fix the http_server line. > > -- j > > On Feb 16, 2008, at 6:35 PM, Stephen John Smoogen wrote: > >> System is RHEL-5 with cobbler-0.8 updates. >> >> /var/www/cobbler/repo_mirror/rhel-i386-server-cluster-5/getPackage/system-config-cluster-1.0.50-1.3.noarch.rpm >> >> already exists and appears to be complete >> - creating: >> /var/www/cobbler/repo_mirror/.origin/rhel-i386-server-cluster-5.repo >> Traceback (most recent call last): >> File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 54, >> in main >> return BootCLI().run(sys.argv) >> File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 43, >> in run >> return self.loader.run(args) >> File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line 69, >> in run >> return fn.run() >> File "/usr/lib/python2.4/site-packages/cobbler/modules/cli_misc.py", >> line 227, in run >> return self.api.reposync(self.options.only) >> File "/usr/lib/python2.4/site-packages/cobbler/api.py", line 332, in >> reposync >> return reposync.run(name) >> File "/usr/lib/python2.4/site-packages/cobbler/action_reposync.py", >> line 74, in run >> self.do_reposync(repo) >> File "/usr/lib/python2.4/site-packages/cobbler/action_reposync.py", >> line 180, in do_reposync >> temp_file = self.create_local_file(repo, temp_path, output=False) >> File "/usr/lib/python2.4/site-packages/cobbler/action_reposync.py", >> line 247, in create_local_file >> http_server = self.setting.server + ":" + self.settings.http_port >> AttributeError: RepoSync instance has no attribute 'setting' >> >> Doing this with any repository (using --only=fedora-8-updates for >> instance) gives the same error. My guess is that there is a 0.6.5->0.8 >> config change missing. >> >> This led to another error >> >> [root at snorbtz ~]# cobbler reposync -h >> usage: cobbler reposync [ARGS] >> >> options: >> -h, --help show this help message and exit >> --only=ONLY update only this repository name >> Traceback (most recent call last): >> File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 54, >> in main >> return BootCLI().run(sys.argv) >> File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 43, >> in run >> return self.loader.run(args) >> File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line 66, >> in run >> loaded_ok = fn.parse_args(args) >> File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line >> 190, in parse_args >> (self.options, self.args) = p.parse_args(args) >> File "/usr/lib/python2.4/optparse.py", line 1275, in parse_args >> stop = self._process_args(largs, rargs, values) >> File "/usr/lib/python2.4/optparse.py", line 1319, in _process_args >> self._process_short_opts(rargs, values) >> File "/usr/lib/python2.4/optparse.py", line 1426, in _process_short_opts >> option.process(opt, value, values, self) >> File "/usr/lib/python2.4/optparse.py", line 707, in process >> return self.take_action( >> File "/usr/lib/python2.4/optparse.py", line 729, in take_action >> parser.exit() >> File "/usr/lib/python2.4/optparse.py", line 1449, in exit >> sys.exit(status) >> SystemExit: 0 >> >> >> >> -- >> Stephen J Smoogen. -- CSIRT/Linux System Administrator >> How far that little candle throws his beams! So shines a good deed >> in a naughty world. = Shakespeare. "The Merchant of Venice" >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mdehaan at redhat.com Mon Feb 18 15:50:13 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 18 Feb 2008 10:50:13 -0500 Subject: [et-mgmt-tools] [PATCH] Fix typo in action_reposync.py self.settings call and change http_port to "str" In-Reply-To: References: Message-ID: <47B9A935.7040404@redhat.com> Jeremy Rosengren wrote: > This is my first patch generated with git, so hopefully I've done this > correctly. > > I installed Cobbler 0.8 today and noticed that one of my repos which > is defined as an existing yum repository (http://) generated an error. > This patch fixes the typo which caused the error and changed the > default http_port setting to be a string instead of an integer - > hopefully that doesn't break something else. > > Thanks, > > -- jeremy > This is now applied in git (both branches). Thanks! From Sean.Embry at nisd.net Mon Feb 18 23:07:23 2008 From: Sean.Embry at nisd.net (Sean Embry) Date: Mon, 18 Feb 2008 17:07:23 -0600 Subject: [et-mgmt-tools] Problems with Xen PXE installs Message-ID: Pardon if this was covered before... I looked through the last 2 months archive, and found only "Invalid Kernel", and that appeared to be all distros, not just xen. Did a bare metal install to what would become the cobbler server via DVD of Fedora 8 i686, (32 bit) non-xen install. Set up 8 IPs in /var/lib/cobbler for DHCP in a subnet without DHCP, imported images for Fedora 8 (32 and 64 bit), and RH 5.2 (32 and 64). All non Xen installs work fine, but no Xen kernels will. From the PXE boot prompt, selecting "menu", then any Xen kernel gives a "Invalid kernel" error. I don't see anything out of place in the error logs, and I'm pretty much stuck on dead center. I guess I could just stay with non-xen and add it later via "Add/Remove Software", but that seems to be a pain in the neck I'd rather do without... I've not tried xen installs using ftp or http yet, that's for the morning... Please let me know what other info I can provide to trouble shoot this problem. Thanks in advance, Sean Embry Systems/Database Administrator NISD.NET From steve.woodcock at gmail.com Mon Feb 18 23:31:17 2008 From: steve.woodcock at gmail.com (Steve Woodcock) Date: Mon, 18 Feb 2008 23:31:17 +0000 Subject: [et-mgmt-tools] [PATCH] Some manpage cleanups In-Reply-To: References: Message-ID: Hi A few trivial manpage cleanups for you Cheers, Steve -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Some-manpage-cleanups.patch Type: application/octet-stream Size: 8302 bytes Desc: not available URL: From berrange at redhat.com Tue Feb 19 03:42:21 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 19 Feb 2008 03:42:21 +0000 Subject: [et-mgmt-tools] [Patch] Build virt-manager at Solaris In-Reply-To: <47B6B9DE.30702@sun.com> References: <47B6B9DE.30702@sun.com> Message-ID: <20080219034221.GA13958@redhat.com> On Sat, Feb 16, 2008 at 06:24:30PM +0800, Henry Zhang wrote: > Hi there, > > When I tried to build virt-manager 0.4.0 at Solaris, failed, because > several parameters can't pass at Solaris, I have to remove them, please > review, if it's possible to remove them from your source codes... I've looked at this quickly - one of the Makefile's you're changing no longer exists - the po/Makefile. The other one will need to have an automake conditional defined, so we can do pass different flags for Solaris vs Linux. eg in configure.in need to call with some param based on whether it is solaris or not AM_CONDITIONAL And then in the makefile if $(OS) = "SunOS" ... else .... endif I've been meaning to try this, but not had a chance yet.. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From capel at stone-it.com Tue Feb 19 07:49:44 2008 From: capel at stone-it.com (Jasper Capel) Date: Tue, 19 Feb 2008 08:49:44 +0100 Subject: [et-mgmt-tools] Problems with Xen PXE installs In-Reply-To: References: Message-ID: <47BA8A18.5010000@stone-it.com> I'm always using a non-xen kernel for my PXE-boots, I think the provided xen-kernel was meant to be used for DomU-installs. Why don't you just modify your kickstart file to include the required Xen-software? You only have to add xen and kernel-xen to the packages section. For more information on how to use kickstart files, see: http://www.centos.org/docs/5/html/5.1/Installation_Guide/ch-kickstart2.html Jasper Sean Embry wrote: > Pardon if this was covered before... I looked through the last > 2 months archive, and found only "Invalid Kernel", and that > appeared to be all distros, not just xen. > > Did a bare metal install to what would become > the cobbler server via DVD of Fedora 8 i686, (32 bit) > non-xen install. Set up 8 IPs in /var/lib/cobbler for DHCP > in a subnet without DHCP, imported images for > Fedora 8 (32 and 64 bit), and RH 5.2 (32 and 64). > > All non Xen installs work fine, but no Xen kernels will. From > the PXE boot prompt, selecting "menu", then any Xen > kernel gives a "Invalid kernel" error. I don't see anything > out of place in the error logs, and I'm pretty much stuck > on dead center. I guess I could just stay with non-xen > and add it later via "Add/Remove Software", but that > seems to be a pain in the neck I'd rather do without... > > I've not tried xen installs using ftp or http yet, that's for the > morning... > > Please let me know what other info I can provide to > trouble shoot this problem. > > Thanks in advance, > > Sean Embry > Systems/Database Administrator > NISD.NET > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -- Kind regards, Jasper Capel email : capel at stone-it.com web : http://www.stone-it.com/ .~. /V\ Stone-IT tel : +31 (0)62 86 50 854 // \\ LINUX FOR e-BUSINESS fax : +31 (0)30 22 87 997 /( )\ ^^-^^ Groenekanseweg 246, 3737 AL Groenekan, Nederland From mdehaan at redhat.com Tue Feb 19 14:23:08 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 19 Feb 2008 09:23:08 -0500 Subject: [et-mgmt-tools] Problems with Xen PXE installs In-Reply-To: <47BA8A18.5010000@stone-it.com> References: <47BA8A18.5010000@stone-it.com> Message-ID: <47BAE64C.8000405@redhat.com> Jasper Capel wrote: > I'm always using a non-xen kernel for my PXE-boots, I think the > provided xen-kernel was meant to be used for DomU-installs. Why don't > you just modify your kickstart file to include the required > Xen-software? You only have to add xen and kernel-xen to the packages > section. Absolutely correct. You don't want to pick any profile with "xen" in the name for bare metal installs, you instead add the packages to your kickstart. Of course when doing the guest OS's you do need the ones with "xen" in the name. Ultimately having an "exclude-from-menu" feature in Cobbler (on the distro) might be worthwhile, as this can be detected during import. This would cause any cobbler profiles attached to the distro to not show up in the menus because they weren't PXE compatible. (Note that this will only ever apply to Xen paravirt kernels). Make sense? --Michael > > Jasper > > Sean Embry wrote: >> Pardon if this was covered before... I looked through the last >> 2 months archive, and found only "Invalid Kernel", and that >> appeared to be all distros, not just xen. >> >> Did a bare metal install to what would become >> the cobbler server via DVD of Fedora 8 i686, (32 bit) >> non-xen install. Set up 8 IPs in /var/lib/cobbler for DHCP >> in a subnet without DHCP, imported images for >> Fedora 8 (32 and 64 bit), and RH 5.2 (32 and 64). >> >> All non Xen installs work fine, but no Xen kernels will. From >> the PXE boot prompt, selecting "menu", then any Xen >> kernel gives a "Invalid kernel" error. I don't see anything >> out of place in the error logs, and I'm pretty much stuck >> on dead center. I guess I could just stay with non-xen >> and add it later via "Add/Remove Software", but that >> seems to be a pain in the neck I'd rather do without... >> >> I've not tried xen installs using ftp or http yet, that's for the >> morning... >> >> Please let me know what other info I can provide to >> trouble shoot this problem. >> >> Thanks in advance, >> >> Sean Embry >> Systems/Database Administrator >> NISD.NET >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools >> > > From jeremy at rosengren.org Tue Feb 19 14:52:21 2008 From: jeremy at rosengren.org (Jeremy Rosengren) Date: Tue, 19 Feb 2008 08:52:21 -0600 Subject: [et-mgmt-tools] [PATCH] Fix traceback on commands run with -h Message-ID: <47BAED25.7040604@rosengren.org> This patch should fix the traceback that occurs when you run a command with -h (cobbler reposync -h). Looking back through the history of cobbler.py, it looked like a simple matter of the conditional expressions becoming reversed in a recent commit. Thanks, -- jeremy --- cobbler/cobbler.py | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/cobbler/cobbler.py b/cobbler/cobbler.py index fb9b273..0f201f7 100755 --- a/cobbler/cobbler.py +++ b/cobbler/cobbler.py @@ -56,9 +56,9 @@ def main(): print str(exc)[1:-1] # remove framing air quotes except Exception, exc2: if str(type(exc2)).find("CX") == -1: - traceback.print_exc() - else: print str(exc2)[1:-1] # remove framing air quotes + else: + traceback.print_exc() return 1 return 1 -- 1.5.2.1 From chris.sarginson at ukfast.net Tue Feb 19 15:12:07 2008 From: chris.sarginson at ukfast.net (Chris Sarginson - Technical Support) Date: Tue, 19 Feb 2008 15:12:07 +0000 Subject: [et-mgmt-tools] Cobbler rename Message-ID: <47BAF1C7.8030303@ukfast.net> Hi Guys, I've just used the cobbler (distro) rename feature for the first time, and have picked up a couple of things about it that could be handled better IMVHO Although it renames *most* files/links it leaves the symlink in /var/www/cobbler/links/ and the folder /var/www/cobbler/ks_mirror/ with the original name - I appreciate that this is more of a "vanity" request than a functionality one, having (what are essentially) randomly named folders following the rename is not a great solution. This has all come about as I have had need to update the installer used for Centos 5 and RHEL 5 to use Update 1 due to the inclusion of additional network drivers that are included in this, and rather than mess around recreating all my profiles for RHEL/Centos I decided to move the Centos/RHEL 5.0 and then put the 5.1 import in its place. Have I been clear about what I meant here? Is there any chance of this behaviour being adjusted? I don't currently *get* python, so if anybody could point me in the direction of some good resources for learning they would be gratefully received. -- Kind regards Chris Sarginson Technical Support UKFast.Net Ltd (t) 0870 111 8866 (f) 0870 458 4545 "The UK's Best Hosting Provider" ISPA Awards 2007, 2006 and 2005 Dedicated Servers - Managed Hosting - Domain Names- http://www.ukfast.net UKFast.Net Ltd, City Tower, Piccadilly Plaza, Manchester, M1 4BT Registered in England. Number 384 5616 From mdehaan at redhat.com Tue Feb 19 15:44:07 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 19 Feb 2008 10:44:07 -0500 Subject: [et-mgmt-tools] [PATCH] Fix traceback on commands run with -h In-Reply-To: <47BAED25.7040604@rosengren.org> References: <47BAED25.7040604@rosengren.org> Message-ID: <47BAF947.6010602@redhat.com> Jeremy Rosengren wrote: > This patch should fix the traceback that occurs when you run a command > with -h (cobbler reposync -h). Looking back through the history of > cobbler.py, it looked like a simple matter of the conditional > expressions becoming reversed in a recent commit. cobbler reposync -h works for me as is, from latest git, and in the released version. What version of Cobbler are you using (rpm -q cobbler)? The code in question below is already correct, because it prints the traceback only for non-cobbler ("CX") exceptions. --Michael > > Thanks, > > -- jeremy > > --- > cobbler/cobbler.py | 4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/cobbler/cobbler.py b/cobbler/cobbler.py > index fb9b273..0f201f7 100755 > --- a/cobbler/cobbler.py > +++ b/cobbler/cobbler.py > @@ -56,9 +56,9 @@ def main(): > print str(exc)[1:-1] # remove framing air quotes > except Exception, exc2: > if str(type(exc2)).find("CX") == -1: > - traceback.print_exc() > - else: > print str(exc2)[1:-1] # remove framing air quotes > + else: > + traceback.print_exc() > return 1 > return 1 > > -- > 1.5.2.1 > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mdehaan at redhat.com Tue Feb 19 15:53:49 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 19 Feb 2008 10:53:49 -0500 Subject: [et-mgmt-tools] Cobbler rename In-Reply-To: <47BAF1C7.8030303@ukfast.net> References: <47BAF1C7.8030303@ukfast.net> Message-ID: <47BAFB8D.6060503@redhat.com> Chris Sarginson - Technical Support wrote: > Hi Guys, > > I've just used the cobbler (distro) rename feature for the first time, > and have picked up a couple of things about it that could be handled > better IMVHO > > Although it renames *most* files/links it leaves the symlink in > /var/www/cobbler/links/ and the folder /var/www/cobbler/ks_mirror/ > with the original name - I appreciate that this is more of a "vanity" > request than a functionality one, having (what are essentially) > randomly named folders following the rename is not a great solution. Thanks for the feedback. The reason cobbler doesn't rename the source directories is twofold. (A) Renaming the folders could break other users who are referencing the kickstart tree directly through Apache (outside of Cobbler). (B) Also in some cases, cobbler will not own the mirrored location, for instance, the tree may be stored on NFS. However, it's /probably/ safe to assume if the directory starts with /var/www/cobbler (or more specifically, self.settings.webroot), it's changeable. We would of course want to ensure that the directory we were moving it to was not taken. > > This has all come about as I have had need to update the installer > used for Centos 5 and RHEL 5 to use Update 1 due to the inclusion of > additional network drivers that are included in this, and rather than > mess around recreating all my profiles for RHEL/Centos I decided to > move the Centos/RHEL 5.0 and then put the 5.1 import in its place. > > Have I been clear about what I meant here? Is there any chance of > this behaviour being adjusted? I don't currently *get* python, so if > anybody could point me in the direction of some good resources for > learning they would be gratefully received. Yes, it's changeable. I think it's reasonable to assume if the directory is in ks_mirror we can do renames. As for Python, there's some introductory info here... http://www.python.org/doc/ http://www.diveintopython.org/ Reading Cobbler's source is also a pretty good start. I'll file a RFE ticket in Trac (https://fedorahosted.org/cobbler) this so we can keep track of the idea. --Michael From jeremy at rosengren.org Tue Feb 19 16:12:56 2008 From: jeremy at rosengren.org (Jeremy Rosengren) Date: Tue, 19 Feb 2008 10:12:56 -0600 Subject: [et-mgmt-tools] [PATCH] Fix traceback on commands run with -h In-Reply-To: <47BAF947.6010602@redhat.com> References: <47BAED25.7040604@rosengren.org> <47BAF947.6010602@redhat.com> Message-ID: <47BB0008.7030505@rosengren.org> I compiled from latest git and am seeing what I've copied below. Stephen Smoogen also reported the problem using the latest published RPM version (which is also where I first saw it): https://www.redhat.com/archives/et-mgmt-tools/2008-February/msg00112.html From what I saw in cobbler.py from latest git, the conditional in main() was giving the traceback all the time. The "else" case of print str(exc2)[1:-1] had flipped sides of the conditional statement from a couple of revisions back. -- jeremy [root at raid ~]# cobbler reposync -h usage: cobbler reposync [ARGS] options: -h, --help show this help message and exit --only=ONLY update only this repository name Traceback (most recent call last): File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 54, in main return BootCLI().run(sys.argv) File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 43, in run return self.loader.run(args) File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line 66, in run loaded_ok = fn.parse_args(args) File "/usr/lib/python2.4/site-packages/cobbler/commands.py", line 190, in parse_args (self.options, self.args) = p.parse_args(args) File "/usr/lib/python2.4/optparse.py", line 1275, in parse_args stop = self._process_args(largs, rargs, values) File "/usr/lib/python2.4/optparse.py", line 1319, in _process_args self._process_short_opts(rargs, values) File "/usr/lib/python2.4/optparse.py", line 1426, in _process_short_opts option.process(opt, value, values, self) File "/usr/lib/python2.4/optparse.py", line 707, in process return self.take_action( File "/usr/lib/python2.4/optparse.py", line 729, in take_action parser.exit() File "/usr/lib/python2.4/optparse.py", line 1449, in exit sys.exit(status) SystemExit: 0 Michael DeHaan wrote: > Jeremy Rosengren wrote: >> This patch should fix the traceback that occurs when you run a >> command with -h (cobbler reposync -h). Looking back through the >> history of cobbler.py, it looked like a simple matter of the >> conditional expressions becoming reversed in a recent commit. > > cobbler reposync -h works for me as is, from latest git, and in the > released version. > > What version of Cobbler are you using (rpm -q cobbler)? > The code in question below is already correct, because it prints the > traceback only for non-cobbler ("CX") exceptions. > > --Michael > > >> >> Thanks, >> >> -- jeremy >> >> --- >> cobbler/cobbler.py | 4 ++-- >> 1 files changed, 2 insertions(+), 2 deletions(-) >> >> diff --git a/cobbler/cobbler.py b/cobbler/cobbler.py >> index fb9b273..0f201f7 100755 >> --- a/cobbler/cobbler.py >> +++ b/cobbler/cobbler.py >> @@ -56,9 +56,9 @@ def main(): >> print str(exc)[1:-1] # remove framing air quotes >> except Exception, exc2: >> if str(type(exc2)).find("CX") == -1: >> - traceback.print_exc() >> - else: >> print str(exc2)[1:-1] # remove framing air quotes >> + else: >> + traceback.print_exc() >> return 1 >> return 1 >> >> -- >> 1.5.2.1 >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mdehaan at redhat.com Tue Feb 19 16:29:55 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 19 Feb 2008 11:29:55 -0500 Subject: [et-mgmt-tools] [PATCH] Fix traceback on commands run with -h In-Reply-To: <47BB0008.7030505@rosengren.org> References: <47BAED25.7040604@rosengren.org> <47BAF947.6010602@redhat.com> <47BB0008.7030505@rosengren.org> Message-ID: <47BB0403.5000505@redhat.com> Jeremy Rosengren wrote: > I compiled from latest git and am seeing what I've copied below. > Stephen Smoogen also reported the problem using the latest published > RPM version (which is also where I first saw it): > https://www.redhat.com/archives/et-mgmt-tools/2008-February/msg00112.html > > From what I saw in cobbler.py from latest git, the conditional in > main() was giving the traceback all the time. The "else" case of > print str(exc2)[1:-1] had flipped sides of the conditional statement > from a couple of revisions back. > I don't agree with the patch, but it looks like your particular version of optparse is throwing system.exit in a way that cobbler doesn't like and we need to catch that /also/. I'll look into it. Thanks. From jeremy at rosengren.org Tue Feb 19 16:41:59 2008 From: jeremy at rosengren.org (Jeremy Rosengren) Date: Tue, 19 Feb 2008 10:41:59 -0600 Subject: [et-mgmt-tools] [PATCH] Fix traceback on commands run with -h In-Reply-To: <47BB0403.5000505@redhat.com> References: <47BAED25.7040604@rosengren.org> <47BAF947.6010602@redhat.com> <47BB0008.7030505@rosengren.org> <47BB0403.5000505@redhat.com> Message-ID: <47BB06D7.6050900@rosengren.org> Michael DeHaan wrote: > I don't agree with the patch, but it looks like your particular > version of optparse is throwing system.exit in a way that cobbler > doesn't like and we need to catch that /also/. > > I'll look into it. Thanks. > > I'm not claiming that what I did was agreeable :) I've copied my python version information, should that help - I'm running on updated CentOS 5.1. Thanks! -- j [root at raid ~]# rpm -qi python Name : python Relocations: (not relocatable) Version : 2.4.3 Vendor: CentOS Release : 19.el5 Build Date: Wed 14 Mar 2007 05:59:57 PM CDT Install Date: Tue 24 Apr 2007 04:23:26 PM CDT Build Host: builder4.centos.org Group : Development/Languages Source RPM: python-2.4.3-19.el5.src.rpm Size : 21602087 License: PSF - see LICENSE Signature : DSA/SHA1, Tue 03 Apr 2007 07:27:09 PM CDT, Key ID a8a447dce8562897 URL : http://www.python.org/ Summary : An interpreted, interactive, object-oriented programming language. Description : Python is an interpreted, interactive, object-oriented programming language often compared to Tcl, Perl, Scheme or Java. Python includes modules, classes, exceptions, very high level dynamic data types and dynamic typing. Python supports interfaces to many system calls and libraries, as well as to various windowing systems (X11, Motif, Tk, Mac and MFC). Programmers can write new built-in modules for Python in C or C++. Python can be used as an extension language for applications that need a programmable interface. This package contains most of the standard Python modules, as well as modules for interfacing to the Tix widget set for Tk and RPM. Note that documentation for Python is provided in the python-docs package. From Sean.Embry at nisd.net Tue Feb 19 17:15:58 2008 From: Sean.Embry at nisd.net (Sean Embry) Date: Tue, 19 Feb 2008 11:15:58 -0600 Subject: [et-mgmt-tools] Problems with Xen PXE installs Message-ID: I don't know if top or bottom posting is correct, so I'll bottom post for now: >>> mdehaan at redhat.com 02/19/08 8:23 AM >>> Jasper Capel wrote: > I'm always using a non-xen kernel for my PXE-boots, I think the > provided xen-kernel was meant to be used for DomU-installs. Why don't > you just modify your kickstart file to include the required > Xen-software? You only have to add xen and kernel-xen to the packages > section. >Absolutely correct. You don't want to pick any profile with "xen" in >the name for bare metal installs, you instead add the packages to your >kickstart. Of course when doing the guest OS's you do need the ones >with "xen" in the name. >Ultimately having an "exclude-from-menu" feature in Cobbler (on the >distro) might be worthwhile, as this can be detected during import. >This would cause >any cobbler profiles attached to the distro to not show up in the menus >because they weren't PXE compatible. (Note that this will only ever apply >to Xen paravirt kernels). >Make sense? >--Michael > > Jasper > > Sean Embry wrote: >> Pardon if this was covered before... I looked through the last >> 2 months archive, and found only "Invalid Kernel", and that >> appeared to be all distros, not just xen. >> >> Did a bare metal install to what would become >> the cobbler server via DVD of Fedora 8 i686, (32 bit) >> non-xen install. Set up 8 IPs in /var/lib/cobbler for DHCP >> in a subnet without DHCP, imported images for >> Fedora 8 (32 and 64 bit), and RH 5.2 (32 and 64). >> >> All non Xen installs work fine, but no Xen kernels will. From >> the PXE boot prompt, selecting "menu", then any Xen >> kernel gives a "Invalid kernel" error. I don't see anything >> out of place in the error logs, and I'm pretty much stuck >> on dead center. I guess I could just stay with non-xen >> and add it later via "Add/Remove Software", but that >> seems to be a pain in the neck I'd rather do without... >> >> I've not tried xen installs using ftp or http yet, that's for the >> morning... >> >> Please let me know what other info I can provide to >> trouble shoot this problem. >> >> Thanks in advance, >> >> Sean Embry >> Systems/Database Administrator >> NISD.NET >> snip sigs Jasper, thank you. I did just that and it worked out fine. Michael, Thank you. I've got that working now. WOOT! Sean From mdehaan at redhat.com Tue Feb 19 17:34:44 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 19 Feb 2008 12:34:44 -0500 Subject: [et-mgmt-tools] Problems with Xen PXE installs In-Reply-To: References: Message-ID: <47BB1334.9010500@redhat.com> Sean Embry wrote: > I don't know if top or bottom posting is correct, so I'll bottom post > for now: > > Bottom/inline posting with trimming the parts you don't need to quote :) > snip sigs > > Jasper, thank you. I did just that and it worked out fine. > > Michael, > > Thank you. I've got that working now. WOOT! > > Sean > Good deal. From jeremy at rosengren.org Tue Feb 19 17:59:21 2008 From: jeremy at rosengren.org (Jeremy Rosengren) Date: Tue, 19 Feb 2008 11:59:21 -0600 Subject: [et-mgmt-tools] [PATCH] Fix traceback on commands run with -h In-Reply-To: <47BB06D7.6050900@rosengren.org> References: <47BAED25.7040604@rosengren.org> <47BAF947.6010602@redhat.com> <47BB0008.7030505@rosengren.org> <47BB0403.5000505@redhat.com> <47BB06D7.6050900@rosengren.org> Message-ID: <47BB18F9.1070307@rosengren.org> Jeremy Rosengren wrote: > Michael DeHaan wrote: >> I don't agree with the patch, but it looks like your particular >> version of optparse is throwing system.exit in a way that cobbler >> doesn't like and we need to catch that /also/. >> >> I'll look into it. Thanks. >> I looked through the Python documentation and it looks like the SystemExit exception needs to be caught, so I added that to the other exception handling in cobbler.py I don't claim to know for sure that this is the best way to do it, just that it solves the problem better than my last attempt. Thanks, -- j --- cobbler/cobbler.py | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/cobbler/cobbler.py b/cobbler/cobbler.py index fb9b273..9f95f1e 100755 --- a/cobbler/cobbler.py +++ b/cobbler/cobbler.py @@ -52,7 +52,7 @@ def main(): try: # FIXME: redo locking code? return BootCLI().run(sys.argv) - except CX, exc: + except (CX, SystemExit), exc: print str(exc)[1:-1] # remove framing air quotes except Exception, exc2: if str(type(exc2)).find("CX") == -1: -- 1.5.2.1 From mdehaan at redhat.com Tue Feb 19 18:47:45 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 19 Feb 2008 13:47:45 -0500 Subject: [et-mgmt-tools] [PATCH] Fix traceback on commands run with -h In-Reply-To: <47BB18F9.1070307@rosengren.org> References: <47BAED25.7040604@rosengren.org> <47BAF947.6010602@redhat.com> <47BB0008.7030505@rosengren.org> <47BB0403.5000505@redhat.com> <47BB06D7.6050900@rosengren.org> <47BB18F9.1070307@rosengren.org> Message-ID: <47BB2451.1080707@redhat.com> Jeremy Rosengren wrote: > Jeremy Rosengren wrote: >> Michael DeHaan wrote: >>> I don't agree with the patch, but it looks like your particular >>> version of optparse is throwing system.exit in a way that cobbler >>> doesn't like and we need to catch that /also/. >>> >>> I'll look into it. Thanks. >>> > I looked through the Python documentation and it looks like the > SystemExit exception needs to be caught, so I added that to the other > exception handling in cobbler.py > > I don't claim to know for sure that this is the best way to do it, > just that it solves the problem better than my last attempt. > > Thanks, > > -- j > > --- > cobbler/cobbler.py | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/cobbler/cobbler.py b/cobbler/cobbler.py > index fb9b273..9f95f1e 100755 > --- a/cobbler/cobbler.py > +++ b/cobbler/cobbler.py > @@ -52,7 +52,7 @@ def main(): > try: > # FIXME: redo locking code? > return BootCLI().run(sys.argv) > - except CX, exc: > + except (CX, SystemExit), exc: > print str(exc)[1:-1] # remove framing air quotes > except Exception, exc2: > if str(type(exc2)).find("CX") == -1: Will apply -- (not sure if the double exception catching on one line is legal syntax -- but something like it anyway). Thanks. From berrange at redhat.com Tue Feb 19 18:50:39 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 19 Feb 2008 18:50:39 +0000 Subject: [et-mgmt-tools] [PATCH] Fix traceback on commands run with -h In-Reply-To: <47BB2451.1080707@redhat.com> References: <47BAED25.7040604@rosengren.org> <47BAF947.6010602@redhat.com> <47BB0008.7030505@rosengren.org> <47BB0403.5000505@redhat.com> <47BB06D7.6050900@rosengren.org> <47BB18F9.1070307@rosengren.org> <47BB2451.1080707@redhat.com> Message-ID: <20080219185039.GL17472@redhat.com> On Tue, Feb 19, 2008 at 01:47:45PM -0500, Michael DeHaan wrote: > Jeremy Rosengren wrote: > >Jeremy Rosengren wrote: > >>Michael DeHaan wrote: > >>>I don't agree with the patch, but it looks like your particular > >>>version of optparse is throwing system.exit in a way that cobbler > >>>doesn't like and we need to catch that /also/. > >>> > >>>I'll look into it. Thanks. > >>> > >I looked through the Python documentation and it looks like the > >SystemExit exception needs to be caught, so I added that to the other > >exception handling in cobbler.py > > > >I don't claim to know for sure that this is the best way to do it, > >just that it solves the problem better than my last attempt. > > > >Thanks, > > > >-- j > > > >--- > >cobbler/cobbler.py | 2 +- > >1 files changed, 1 insertions(+), 1 deletions(-) > > > >diff --git a/cobbler/cobbler.py b/cobbler/cobbler.py > >index fb9b273..9f95f1e 100755 > >--- a/cobbler/cobbler.py > >+++ b/cobbler/cobbler.py > >@@ -52,7 +52,7 @@ def main(): > > try: > > # FIXME: redo locking code? > > return BootCLI().run(sys.argv) > >- except CX, exc: > >+ except (CX, SystemExit), exc: > > print str(exc)[1:-1] # remove framing air quotes > > except Exception, exc2: > > if str(type(exc2)).find("CX") == -1: > > Will apply -- (not sure if the double exception catching on one line is > legal syntax -- but something like it anyway). It is only legal in python >= 2.5, so you'll need to fix that if you support 2.4 or earlier for cobbler Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From mdehaan at redhat.com Tue Feb 19 18:52:27 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 19 Feb 2008 13:52:27 -0500 Subject: [et-mgmt-tools] [PATCH] Fix traceback on commands run with -h In-Reply-To: <20080219185039.GL17472@redhat.com> References: <47BAED25.7040604@rosengren.org> <47BAF947.6010602@redhat.com> <47BB0008.7030505@rosengren.org> <47BB0403.5000505@redhat.com> <47BB06D7.6050900@rosengren.org> <47BB18F9.1070307@rosengren.org> <47BB2451.1080707@redhat.com> <20080219185039.GL17472@redhat.com> Message-ID: <47BB256B.3010309@redhat.com> > > It is only legal in python >= 2.5, so you'll need to fix that if you > support 2.4 or earlier for cobbler > > Dan. > Yep, I settled on just: except SystemExit: pass just because on SystemExit optparse has already printed what it wants to print, and we don't use SystemExit elsewhere. This has been your daily python lesson on et-mgmt-tools at redhat.com :) --Michael From fukuta.saori at jp.fujitsu.com Wed Feb 20 08:02:56 2008 From: fukuta.saori at jp.fujitsu.com (Saori Fukuta) Date: Wed, 20 Feb 2008 17:02:56 +0900 Subject: [et-mgmt-tools] [PATCH] readOnly flags for virConnectOpenAuth Message-ID: <20080220170236.0FBC.FUKUTA.SAORI@jp.fujitsu.com> Hello, I cannot connect to localhost Xen with non-root user. The virt-manager says the following message, Unable to open connection to hypervisor URI 'xen:///': virConnectOpenAuth() failed Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/connection.py", line 414, in _open_thread None], flags) File "/usr/lib/python2.5/site-packages/libvirt.py", line 94, in openAuth if ret is None:raise libvirtError('virConnectOpenAuth() failed') libvirtError: virConnectOpenAuth() failed I guess this problem occurs from Cset:680 http://hg.et.redhat.com/virt/applications/virt-manager--devel?cs=1892867ca5c7 Before that Cset, connection was able to succeed since the virt-manager tried to connect with libvirt.openReadOnly after failed to libvirt.open, even if 'self.readOnly' was None. After that Cset, 'flags' for libvirt.openAuth depend on 'self.readOnly' though 'self.readOnly' is never set. I'm not sure where the readOnly flags should be set, but how about this fix for readOnly flags ? --- diff -r 07ff9bffe54d src/virtManager/connection.py --- a/src/virtManager/connection.py Mon Feb 18 10:02:21 2008 -0500 +++ b/src/virtManager/connection.py Wed Feb 20 17:00:41 2008 +0900 @@ -403,8 +403,9 @@ class vmmConnection(gobject.GObject): logging.debug("Background thread is running") try: flags = 0 - if self.readOnly: + if os.getuid() != 0: flags = libvirt.VIR_CONNECT_RO + self.readOnly = True self.vmm = libvirt.openAuth(self.uri, [[libvirt.VIR_CRED_AUTHNAME, Regards, Saori Fukuta From berrange at redhat.com Wed Feb 20 13:05:22 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 20 Feb 2008 13:05:22 +0000 Subject: [et-mgmt-tools] [PATCH] readOnly flags for virConnectOpenAuth In-Reply-To: <20080220170236.0FBC.FUKUTA.SAORI@jp.fujitsu.com> References: <20080220170236.0FBC.FUKUTA.SAORI@jp.fujitsu.com> Message-ID: <20080220130522.GB31194@redhat.com> On Wed, Feb 20, 2008 at 05:02:56PM +0900, Saori Fukuta wrote: > Hello, > > I cannot connect to localhost Xen with non-root user. The virt-manager > says the following message, > Unable to open connection to hypervisor URI 'xen:///': > virConnectOpenAuth() failed > Traceback (most recent call last): > File "/usr/share/virt-manager/virtManager/connection.py", line 414, in _open_thread > None], flags) > File "/usr/lib/python2.5/site-packages/libvirt.py", line 94, in openAuth > if ret is None:raise libvirtError('virConnectOpenAuth() failed') > libvirtError: virConnectOpenAuth() failed > > I guess this problem occurs from Cset:680 > http://hg.et.redhat.com/virt/applications/virt-manager--devel?cs=1892867ca5c7 > > Before that Cset, connection was able to succeed since the virt-manager > tried to connect with libvirt.openReadOnly after failed to libvirt.open, > even if 'self.readOnly' was None. > After that Cset, 'flags' for libvirt.openAuth depend on 'self.readOnly' > though 'self.readOnly' is never set. This is intentional - virt-manager will authenticate with libvirt when attempting to connect, so it gets a full read-write connection. In Fedora 8 or later (or any distro with PolicyKit available) this should 'just work' with you being prompted for password. > I'm not sure where the readOnly flags should be set, but how about > this fix for readOnly flags ? Yes & no - we must only try a read-only connection if we are connecting to a local hypervisor, and don't have PolicyKit available. This change makes all connections read only, even remote ones, which is too mcuh. > --- > diff -r 07ff9bffe54d src/virtManager/connection.py > --- a/src/virtManager/connection.py Mon Feb 18 10:02:21 2008 -0500 > +++ b/src/virtManager/connection.py Wed Feb 20 17:00:41 2008 +0900 > @@ -403,8 +403,9 @@ class vmmConnection(gobject.GObject): > logging.debug("Background thread is running") > try: > flags = 0 > - if self.readOnly: > + if os.getuid() != 0: > flags = libvirt.VIR_CONNECT_RO > + self.readOnly = True > > self.vmm = libvirt.openAuth(self.uri, > [[libvirt.VIR_CRED_AUTHNAME, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From atodorov at redhat.com Wed Feb 20 13:29:57 2008 From: atodorov at redhat.com (Alexander Todorov) Date: Wed, 20 Feb 2008 14:29:57 +0100 Subject: [et-mgmt-tools] Re-usage of existing disks/virt-install Message-ID: <47BC2B55.1040408@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Hello, I've been using a shell script to help me with some testing which installs the same Xen guest over and over again. Now I've noticed that in python-virtinst-0.300.2-3.fc8 there's new functionality. virt-install will check if the file exists and ask to overwrite, then it will ask if you want to use the same file if it's being used by another guest (in my case inactive). That's breaking the cli by making it interactive. This has impact on old scripts which do not require further interaction. Can we consider a --yes or a similar option to deal with this problem? Any opinions are welcome. Thanks, Alexander. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Red Hat - http://enigmail.mozdev.org iD8DBQFHvCtUhmd3WOiFct4RCldHAJ9hsZ/GWyBPfPwT7huHE+ttyKNmHACeLebg qZjSViPcXWjnc8vEyAX0d54= =fMBR -----END PGP SIGNATURE----- From crobinso at redhat.com Wed Feb 20 15:02:27 2008 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 20 Feb 2008 10:02:27 -0500 Subject: [et-mgmt-tools] Re-usage of existing disks/virt-install In-Reply-To: <47BC2B55.1040408@redhat.com> References: <47BC2B55.1040408@redhat.com> Message-ID: <47BC4103.2080103@redhat.com> Alexander Todorov wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA512 > > Hello, > I've been using a shell script to help me with some testing which installs the > same Xen guest over and over again. Now I've noticed that in > python-virtinst-0.300.2-3.fc8 there's new functionality. virt-install will check > if the file exists and ask to overwrite, then it will ask if you want to use the > same file if it's being used by another guest (in my case inactive). > > That's breaking the cli by making it interactive. This has impact on old scripts > which do not require further interaction. > > Can we consider a --yes or a similar option to deal with this problem? > Any opinions are welcome. > > Thanks, > Alexander. Complete agreement here, I actually cooked up something similar to this a while ago and posted it to the list, but never committed it. There should probably be an option to answer yes where applicable, but also an option to outright fail on any prompt. Something like --noprompt={fail,yes}, defaults to fail. Unfortunately this doesn't look very intuitive but my brain isn't thinking of anything cleaner at the moment. - Cole From berrange at redhat.com Wed Feb 20 15:10:09 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 20 Feb 2008 15:10:09 +0000 Subject: [et-mgmt-tools] Re-usage of existing disks/virt-install In-Reply-To: <47BC4103.2080103@redhat.com> References: <47BC2B55.1040408@redhat.com> <47BC4103.2080103@redhat.com> Message-ID: <20080220151009.GA14209@redhat.com> On Wed, Feb 20, 2008 at 10:02:27AM -0500, Cole Robinson wrote: > Alexander Todorov wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > > Hash: SHA512 > > > > Hello, > > I've been using a shell script to help me with some testing which installs the > > same Xen guest over and over again. Now I've noticed that in > > python-virtinst-0.300.2-3.fc8 there's new functionality. virt-install will check > > if the file exists and ask to overwrite, then it will ask if you want to use the > > same file if it's being used by another guest (in my case inactive). > > > > That's breaking the cli by making it interactive. This has impact on old scripts > > which do not require further interaction. > > > > Can we consider a --yes or a similar option to deal with this problem? > > Any opinions are welcome. > > > > Thanks, > > Alexander. > > Complete agreement here, I actually cooked up something similar to this a while > ago and posted it to the list, but never committed it. There should probably > be an option to answer yes where applicable, but also an option to outright fail > on any prompt. > > Something like --noprompt={fail,yes}, defaults to fail. Unfortunately this doesn't > look very intuitive but my brain isn't thinking of anything cleaner at the moment. Personally I think the check for a file existing on disk is an utter waste of time & just serves to seriously annoy & should be removed. Fine checking for it being in use in an existing guest though. The -f, or --force is the common nomenculture used to rm/cp/mv to stop it prompting Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From tom at ng23.net Wed Feb 20 17:17:00 2008 From: tom at ng23.net (Tom Brown) Date: Wed, 20 Feb 2008 17:17:00 +0000 Subject: [et-mgmt-tools] cobbler/koan virt install Message-ID: <47BC608C.3030908@ng23.net> Hi Is there anywhere tht shows a simple how to for virt installs at all? All i mean is an example of a full command line that i could run on a box to install a guest on that box. thanks From jwildebo at redhat.com Wed Feb 20 17:22:14 2008 From: jwildebo at redhat.com (Jan Wildeboer) Date: Wed, 20 Feb 2008 18:22:14 +0100 Subject: [et-mgmt-tools] cobbler/koan virt install In-Reply-To: <47BC608C.3030908@ng23.net> References: <47BC608C.3030908@ng23.net> Message-ID: <1203528134.10578.4.camel@jhwx61.muc.redhat.com> On Wed, 2008-02-20 at 17:17 +0000, Tom Brown wrote: > Is there anywhere tht shows a simple how to for virt installs at all? > All i mean is an example of a full command line that i could run on a > box to install a guest on that box. koan --server=myCobblerBox --virt --system=VirtSystemName or koan --server=myCobblerBox --virt --profile=MyVirtProfile HTH ;-) Jan -- Jan H Wildeboer | Solution Architect, RHCE | Office: +49 (0)89 205071-207 Red Hat GmbH | Mobile: +49 (0)174 33 23 249 Otto-Hahn-Str.20 | Fax: +49 (0)89 205071-111 D-85609 Dornach/Munich | eMail: jan.wildeboer at redhat.com _____________________________________________________________________ GPG-Key-ID: 5DEBAFB0 GPG-Fingerprint: 6104 0F74 8513 F17E DFD5 E820 6F61 A078 5DEB AFB0 _____________________________________________________________________ Reg. Adresse: Red Hat GmbH, Hauptstaetter Strasse 58, 70178 Stuttgart Handelsregister: Amtsgericht Stuttgart HRB 153243 Geschaeftsfuehrer: Brendan Lane, Charlie Peters, Michael Cunningham, Werner Knoblich _____________________________________________________________________ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From dlutter at redhat.com Wed Feb 20 18:37:38 2008 From: dlutter at redhat.com (David Lutterkort) Date: Wed, 20 Feb 2008 10:37:38 -0800 Subject: [et-mgmt-tools] FYI: bugfix to virt-image to allow non-raw images In-Reply-To: <20080217214932.GB10935@redhat.com> References: <20080217214932.GB10935@redhat.com> Message-ID: <1203532658.30187.19.camel@localhost.localdomain> On Sun, 2008-02-17 at 21:49 +0000, Daniel P. Berrange wrote: > If I loosen the virt-image disk format checks to allow qcow, qcow2, > and vmdk > it all works fine, so I've comitted this change > > http://hg.et.redhat.com/virt/applications/virtinst--devel?cs=3b2bc0e2b060 ACK David From tom at ng23.net Wed Feb 20 18:52:01 2008 From: tom at ng23.net (Tom Brown) Date: Wed, 20 Feb 2008 18:52:01 +0000 Subject: [et-mgmt-tools] cobbler/koan virt install In-Reply-To: <1203528134.10578.4.camel@jhwx61.muc.redhat.com> References: <47BC608C.3030908@ng23.net> <1203528134.10578.4.camel@jhwx61.muc.redhat.com> Message-ID: <47BC76D1.3010705@ng23.net> > On Wed, 2008-02-20 at 17:17 +0000, Tom Brown wrote: > > >> Is there anywhere tht shows a simple how to for virt installs at all? >> All i mean is an example of a full command line that i could run on a >> box to install a guest on that box. >> > > koan --server=myCobblerBox --virt --system=VirtSystemName > > or > > koan --server=myCobblerBox --virt --profile=MyVirtProfile > > ah! ok so as i understand it you add a system as per normal and then reference this system name to koan to install with those credentials - makes more sense now as i was wondering how to specify IP info etc. If this is the case where does the mac of the virt system come from or is that not important ? thanks From mdehaan at redhat.com Wed Feb 20 18:56:29 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 20 Feb 2008 13:56:29 -0500 Subject: [et-mgmt-tools] cobbler/koan virt install In-Reply-To: <47BC76D1.3010705@ng23.net> References: <47BC608C.3030908@ng23.net> <1203528134.10578.4.camel@jhwx61.muc.redhat.com> <47BC76D1.3010705@ng23.net> Message-ID: <47BC77DD.3080604@redhat.com> Tom Brown wrote: > >> On Wed, 2008-02-20 at 17:17 +0000, Tom Brown wrote: >> >> >>> Is there anywhere tht shows a simple how to for virt installs at >>> all? All i mean is an example of a full command line that i could >>> run on a box to install a guest on that box. >>> >> >> koan --server=myCobblerBox --virt --system=VirtSystemName >> >> or >> >> koan --server=myCobblerBox --virt --profile=MyVirtProfile >> >> > > > ah! ok so as i understand it you add a system as per normal and then > reference this system name to koan to install with those credentials - > makes more sense now as i was wondering how to specify IP info etc. If > this is the case where does the mac of the virt system come from or is > that not important ? "as i was wondering how to specify IP info etc" See the manpage for options on "cobbler system add", or alternatively, set up the web user interface. I prefer the command line. The MAC if not specified is randomly generated in the range reserved for Xen MAC addresses, and it's only important if you want to set up DHCP reservations (fixing an IP to the MAC) or, transitively, fixing a hostname to the IP. Cobbler's DHCP management features can do both for you (the latter requires dnsmasq not DHCPd) -- or you could manage those yourself. If you are ok with random mac's, IP's, and so on, then you probably don't care. > > thanks > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From tom at ng23.net Wed Feb 20 19:00:28 2008 From: tom at ng23.net (Tom Brown) Date: Wed, 20 Feb 2008 19:00:28 +0000 Subject: [et-mgmt-tools] cobbler/koan virt install In-Reply-To: <47BC77DD.3080604@redhat.com> References: <47BC608C.3030908@ng23.net> <1203528134.10578.4.camel@jhwx61.muc.redhat.com> <47BC76D1.3010705@ng23.net> <47BC77DD.3080604@redhat.com> Message-ID: <47BC78CC.5070602@ng23.net> > > "as i was wondering how to specify IP info etc" > > See the manpage for options on "cobbler system add", or alternatively, > set up the web user interface. I prefer the command line. > > The MAC if not specified is randomly generated in the range reserved > for Xen MAC addresses, and it's only important if you want to set up > DHCP reservations (fixing an IP to the MAC) or, transitively, fixing a > hostname to the IP. Cobbler's DHCP management features can do both > for you (the latter requires dnsmasq not DHCPd) -- or you could manage > those yourself. If you are ok with random mac's, IP's, and so on, then > you probably don't care. > yes thanks - i have been using this fine for a few months now with 'real' systems so it was assigning IP's to a virt system that was slightly puzzling me. So i take it that i can add a system and just omit the reference to the mac address and then reference that system to koan to get the virt machine installed. From mdehaan at redhat.com Wed Feb 20 19:05:45 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 20 Feb 2008 14:05:45 -0500 Subject: [et-mgmt-tools] cobbler/koan virt install In-Reply-To: <47BC78CC.5070602@ng23.net> References: <47BC608C.3030908@ng23.net> <1203528134.10578.4.camel@jhwx61.muc.redhat.com> <47BC76D1.3010705@ng23.net> <47BC77DD.3080604@redhat.com> <47BC78CC.5070602@ng23.net> Message-ID: <47BC7A09.301@redhat.com> Tom Brown wrote: > >> >> "as i was wondering how to specify IP info etc" >> >> See the manpage for options on "cobbler system add", or >> alternatively, set up the web user interface. I prefer the command >> line. >> >> The MAC if not specified is randomly generated in the range reserved >> for Xen MAC addresses, and it's only important if you want to set up >> DHCP reservations (fixing an IP to the MAC) or, transitively, fixing >> a hostname to the IP. Cobbler's DHCP management features can do >> both for you (the latter requires dnsmasq not DHCPd) -- or you could >> manage those yourself. If you are ok with random mac's, IP's, and so >> on, then >> you probably don't care. > > yes thanks - i have been using this fine for a few months now with > 'real' systems so it was assigning IP's to a virt system that was > slightly puzzling me. So i take it that i can add a system and just > omit the reference to the mac address and then reference that system > to koan to get the virt machine installed. Totally. If you want to name the virtual system "foosball" you can, or you can just decide to provision based on profiles. If you do that, the best invocation is: koan --virt --profile=profile-xen --server=cobblerBox --virt-name=foosball That way you get a better name than the one cobbler will generate for you, and it will make more sense when you look at it in virsh or virt-manager later. The reason to use virt provisioning based on the system would be if you wanted DHCP control, or if you were doing per-system templating. In most cases, I just use --profile for virt. --Michael From tom at ng23.net Wed Feb 20 19:13:42 2008 From: tom at ng23.net (Tom Brown) Date: Wed, 20 Feb 2008 19:13:42 +0000 Subject: [et-mgmt-tools] cobbler/koan virt install In-Reply-To: <47BC78CC.5070602@ng23.net> References: <47BC608C.3030908@ng23.net> <1203528134.10578.4.camel@jhwx61.muc.redhat.com> <47BC76D1.3010705@ng23.net> <47BC77DD.3080604@redhat.com> <47BC78CC.5070602@ng23.net> Message-ID: <47BC7BE6.6000305@ng23.net> > > yes thanks - i have been using this fine for a few months now with > 'real' systems so it was assigning IP's to a virt system that was > slightly puzzling me. So i take it that i can add a system and just > omit the reference to the mac address and then reference that system > to koan to get the virt machine installed. > but thinking out loud here i think i am missing a step as i'd need to assign a mac as otherwise dhcp will not know to give out the assigned IP to the virt box. Again i am only talking about virt machines here, i can build and rebuild real machines without any issue. From tom at ng23.net Wed Feb 20 19:14:56 2008 From: tom at ng23.net (Tom Brown) Date: Wed, 20 Feb 2008 19:14:56 +0000 Subject: [et-mgmt-tools] cobbler/koan virt install In-Reply-To: <47BC7A09.301@redhat.com> References: <47BC608C.3030908@ng23.net> <1203528134.10578.4.camel@jhwx61.muc.redhat.com> <47BC76D1.3010705@ng23.net> <47BC77DD.3080604@redhat.com> <47BC78CC.5070602@ng23.net> <47BC7A09.301@redhat.com> Message-ID: <47BC7C30.2090203@ng23.net> > Totally. If you want to name the virtual system "foosball" you can, > or you can just decide to provision based on profiles. If you do > that, the best invocation is: > > koan --virt --profile=profile-xen --server=cobblerBox > --virt-name=foosball > > That way you get a better name than the one cobbler will generate for > you, and it will make more sense when you look at it in virsh or > virt-manager later. > > The reason to use virt provisioning based on the system would be if > you wanted DHCP control, or if you were doing per-system templating. > In most cases, I just use --profile for virt. do you mind sharing what a virt profile looks like ? From mdehaan at redhat.com Wed Feb 20 19:17:01 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 20 Feb 2008 14:17:01 -0500 Subject: [et-mgmt-tools] cobbler/koan virt install In-Reply-To: <47BC7BE6.6000305@ng23.net> References: <47BC608C.3030908@ng23.net> <1203528134.10578.4.camel@jhwx61.muc.redhat.com> <47BC76D1.3010705@ng23.net> <47BC77DD.3080604@redhat.com> <47BC78CC.5070602@ng23.net> <47BC7BE6.6000305@ng23.net> Message-ID: <47BC7CAD.4030208@redhat.com> Tom Brown wrote: >> >> yes thanks - i have been using this fine for a few months now with >> 'real' systems so it was assigning IP's to a virt system that was >> slightly puzzling me. So i take it that i can add a system and just >> omit the reference to the mac address and then reference that system >> to koan to get the virt machine installed. >> > > but thinking out loud here i think i am missing a step as i'd need to > assign a mac as otherwise dhcp will not know to give out the assigned > IP to the virt box. > > Again i am only talking about virt machines here, i can build and > rebuild real machines without any issue. > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools cobbler system add --name=foosball --mac=AA:BB:CC:DD:EE:FF [--ip=192.168.1.42] [etc] [etc] koan --virt --system=foosball --server=cobbler.example.org That'll do it. Koan will create the VM guest named "foosball" with the MAC as set in cobbler. --Michael From mdehaan at redhat.com Wed Feb 20 19:18:39 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 20 Feb 2008 14:18:39 -0500 Subject: [et-mgmt-tools] cobbler/koan virt install In-Reply-To: <47BC7C30.2090203@ng23.net> References: <47BC608C.3030908@ng23.net> <1203528134.10578.4.camel@jhwx61.muc.redhat.com> <47BC76D1.3010705@ng23.net> <47BC77DD.3080604@redhat.com> <47BC78CC.5070602@ng23.net> <47BC7A09.301@redhat.com> <47BC7C30.2090203@ng23.net> Message-ID: <47BC7D0F.2040602@redhat.com> Tom Brown wrote: > >> Totally. If you want to name the virtual system "foosball" you can, >> or you can just decide to provision based on profiles. If you do >> that, the best invocation is: >> >> koan --virt --profile=profile-xen --server=cobblerBox >> --virt-name=foosball >> >> That way you get a better name than the one cobbler will generate for >> you, and it will make more sense when you look at it in virsh or >> virt-manager later. >> >> The reason to use virt provisioning based on the system would be if >> you wanted DHCP control, or if you were doing per-system >> templating. In most cases, I just use --profile for virt. > > > do you mind sharing what a virt profile looks like ? > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Just like a regular one. See the manpage for additional parameters starting with --virt that are only used by virtual machines. For instance, you can set the amount of RAM the guest will get, as well as the amount of disk space -- which is determined by the actual hardware instead for bare metal. Incidentally, you may be interested in joining #cobbler on irc.freenode.net. It's somewhat easier than email for quick questions. --Michael From mdehaan at redhat.com Wed Feb 20 20:06:57 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 20 Feb 2008 15:06:57 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler 0.8.1 Message-ID: <47BC8861.10508@redhat.com> This is a quick bugfix release to fix to earlier reported problems. -- reposync error with http_port -- extra SystemExit traceback printing on some installs There are also some minor manpage corrections. This has been pushed to Fedora/EPEL and is also available as src rpms from http://cobbler.et.redhat.com/download --Michael From fj0588di at aa.jp.fujitsu.com Thu Feb 21 05:46:30 2008 From: fj0588di at aa.jp.fujitsu.com (S.Sakamoto) Date: Thu, 21 Feb 2008 14:46:30 +0900 Subject: [et-mgmt-tools] [PATCH] Add bonding_masters existence check In-Reply-To: <20080218150241.GB30725@redhat.com> References: <20080207125623.GB30230@redhat.com> <200802081731.DJC60956.EJG9K096@aa.jp.fujitsu.com> <20080215182549.GA24578@redhat.com> <200802181727.HFI21810.J9KEG690@aa.jp.fujitsu.com> <20080218150241.GB30725@redhat.com> Message-ID: <200802211446.CHF34808.69K0JEG9@aa.jp.fujitsu.com> Hi, Current virt-manager reads bonding infomation without checking "bonding_masters" existence. This causes an error in reading bridge information on bonding part and NIC list is shown as NULL. I make a patch to solve this problem. Regards, Shigeki Sakamoto. -------------- next part -------------- A non-text attachment was scrubbed... Name: connection.patch Type: application/octet-stream Size: 2191 bytes Desc: not available URL: From berrange at redhat.com Thu Feb 21 12:38:12 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 21 Feb 2008 12:38:12 +0000 Subject: [et-mgmt-tools] [PATCH] Add bonding_masters existence check In-Reply-To: <200802211446.CHF34808.69K0JEG9@aa.jp.fujitsu.com> References: <20080207125623.GB30230@redhat.com> <200802081731.DJC60956.EJG9K096@aa.jp.fujitsu.com> <20080215182549.GA24578@redhat.com> <200802181727.HFI21810.J9KEG690@aa.jp.fujitsu.com> <20080218150241.GB30725@redhat.com> <200802211446.CHF34808.69K0JEG9@aa.jp.fujitsu.com> Message-ID: <20080221123812.GA30015@redhat.com> On Thu, Feb 21, 2008 at 02:46:30PM +0900, S.Sakamoto wrote: > Hi, > > Current virt-manager reads bonding infomation > without checking "bonding_masters" existence. > This causes an error in reading bridge information > on bonding part and NIC list is shown as NULL. Thanks, I have applied your patch. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From tom at ng23.net Thu Feb 21 16:09:37 2008 From: tom at ng23.net (Tom Brown) Date: Thu, 21 Feb 2008 16:09:37 +0000 Subject: [et-mgmt-tools] virt-manager - no x windows Message-ID: <47BDA241.3090805@ng23.net> Hi Is it possible to run virtmanager in purely command line mode without using X at all? I cant seem to find a man page for it etc and i am trying it without X use virt-manager or reconnect with virsh console virtbox [root at xxxxxx] # virt-manager Traceback (most recent call last): File "/usr/share/virt-manager/virt-manager.py", line 59, in ? import gtk File "/usr/lib/python2.4/site-packages/gtk-2.0/gtk/__init__.py", line 76, in ? _init() File "/usr/lib/python2.4/site-packages/gtk-2.0/gtk/__init__.py", line 64, in _init _gtk.init_check() RuntimeError: could not open display thanks From chris.sarginson at ukfast.net Thu Feb 21 16:14:34 2008 From: chris.sarginson at ukfast.net (Chris Sarginson - Technical Support) Date: Thu, 21 Feb 2008 16:14:34 +0000 Subject: [et-mgmt-tools] virt-manager - no x windows In-Reply-To: <47BDA241.3090805@ng23.net> References: <47BDA241.3090805@ng23.net> Message-ID: <47BDA36A.8060305@ukfast.net> AFAIK virt-manager is the gui interface. virsh is the command line. virsh help should see you on your way. Kind regards Chris Sarginson Technical Support UKFast.Net Ltd (t) 0870 111 8866 (f) 0870 458 4545 "The UK's Best Hosting Provider" ISPA Awards 2007, 2006 and 2005 Dedicated Servers - Managed Hosting - Domain Names- http://www.ukfast.net UKFast.Net Ltd, City Tower, Piccadilly Plaza, Manchester, M1 4BT Registered in England. Number 384 5616 Tom Brown wrote: > Hi > > Is it possible to run virtmanager in purely command line mode without > using X at all? > > I cant seem to find a man page for it etc and i am trying it without X > > use virt-manager or reconnect with virsh console virtbox > > [root at xxxxxx] # virt-manager > Traceback (most recent call last): > File "/usr/share/virt-manager/virt-manager.py", line 59, in ? > import gtk > File "/usr/lib/python2.4/site-packages/gtk-2.0/gtk/__init__.py", line > 76, in ? > _init() > File "/usr/lib/python2.4/site-packages/gtk-2.0/gtk/__init__.py", line > 64, in _init > _gtk.init_check() > RuntimeError: could not open display > > thanks > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mnielsen at redhat.com Thu Feb 21 16:19:07 2008 From: mnielsen at redhat.com (Mark Nielsen) Date: Thu, 21 Feb 2008 11:19:07 -0500 Subject: [et-mgmt-tools] virt-manager - no x windows In-Reply-To: <47BDA36A.8060305@ukfast.net> References: <47BDA241.3090805@ng23.net> <47BDA36A.8060305@ukfast.net> Message-ID: <1203610747.5924.0.camel@localhost.localdomain> virt-install -h On Thu, 2008-02-21 at 16:14 +0000, Chris Sarginson - Technical Support wrote: > AFAIK virt-manager is the gui interface. > > virsh is the command line. > > virsh help should see you on your way. > > Kind regards > > Chris Sarginson > Technical Support > UKFast.Net Ltd > > (t) 0870 111 8866 > (f) 0870 458 4545 > > "The UK's Best Hosting Provider" ISPA Awards 2007, 2006 and 2005 > > Dedicated Servers - Managed Hosting - Domain Names- http://www.ukfast.net > > UKFast.Net Ltd, City Tower, Piccadilly Plaza, Manchester, M1 4BT > Registered in England. Number 384 5616 > > Tom Brown wrote: > > Hi > > > > Is it possible to run virtmanager in purely command line mode without > > using X at all? > > > > I cant seem to find a man page for it etc and i am trying it without X > > > > use virt-manager or reconnect with virsh console virtbox > > > > [root at xxxxxx] # virt-manager > > Traceback (most recent call last): > > File "/usr/share/virt-manager/virt-manager.py", line 59, in ? > > import gtk > > File "/usr/lib/python2.4/site-packages/gtk-2.0/gtk/__init__.py", line > > 76, in ? > > _init() > > File "/usr/lib/python2.4/site-packages/gtk-2.0/gtk/__init__.py", line > > 64, in _init > > _gtk.init_check() > > RuntimeError: could not open display > > > > thanks > > > > _______________________________________________ > > et-mgmt-tools mailing list > > et-mgmt-tools at redhat.com > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From chris.sarginson at ukfast.net Thu Feb 21 16:23:27 2008 From: chris.sarginson at ukfast.net (Chris Sarginson - Technical Support) Date: Thu, 21 Feb 2008 16:23:27 +0000 Subject: [et-mgmt-tools] virt-manager - no x windows In-Reply-To: <1203610747.5924.0.camel@localhost.localdomain> References: <47BDA241.3090805@ng23.net> <47BDA36A.8060305@ukfast.net> <1203610747.5924.0.camel@localhost.localdomain> Message-ID: <47BDA57F.2090400@ukfast.net> Mark Nielsen wrote: > virt-install -h Thanks mark, completely forgot about virt-install :-/ Chris From mdehaan at redhat.com Thu Feb 21 16:39:27 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 21 Feb 2008 11:39:27 -0500 Subject: [et-mgmt-tools] virt-manager - no x windows In-Reply-To: <1203610747.5924.0.camel@localhost.localdomain> References: <47BDA241.3090805@ng23.net> <47BDA36A.8060305@ukfast.net> <1203610747.5924.0.camel@localhost.localdomain> Message-ID: <47BDA93F.70500@redhat.com> Mark Nielsen wrote: > virt-install -h > see also Cobbler/koan if you are doing repeated installs :) From berrange at redhat.com Thu Feb 21 16:42:00 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 21 Feb 2008 16:42:00 +0000 Subject: [et-mgmt-tools] virt-manager - no x windows In-Reply-To: <47BDA93F.70500@redhat.com> References: <47BDA241.3090805@ng23.net> <47BDA36A.8060305@ukfast.net> <1203610747.5924.0.camel@localhost.localdomain> <47BDA93F.70500@redhat.com> Message-ID: <20080221164200.GM15883@redhat.com> On Thu, Feb 21, 2008 at 11:39:27AM -0500, Michael DeHaan wrote: > Mark Nielsen wrote: > >virt-install -h > > > > see also Cobbler/koan if you are doing repeated installs :) Yep, for console based management the suite of tools is: - virt-install - virt-clone - virt-image - koan (via cobbler) - virsh And as a standlone graphical viewer - virt-viewer Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From tomg at mcclatchyinteractive.com Thu Feb 21 19:26:28 2008 From: tomg at mcclatchyinteractive.com (Tom Georgoulias) Date: Thu, 21 Feb 2008 14:26:28 -0500 Subject: [et-mgmt-tools] rhn/local repo name mismatch error preventing cobblerd from starting In-Reply-To: <47BC8861.10508@redhat.com> References: <47BC8861.10508@redhat.com> Message-ID: <47BDD064.4030208@mcclatchyinteractive.com> I upgraded from cobbler 0.6.3-2 to 0.8.1-1 this afternoon and got the following error during the cobblerd restart: [root at radm011p ~]# rpm -Uvh cobbler-0.8.1-1.noarch.rpm Preparing... ########################################### [100%] 1:cobbler warning: /var/lib/cobbler/settings created as /var/lib/cobbler/settings.rpmnew ########################################### [100%] Since mirror is RHN rhn://rhel-i386-server-5, the repo must also be named rhel-i386-server-5 Stopping cobbler daemon: [ OK ] Starting cobbler daemon: Traceback (most recent call last): File "/usr/bin/cobblerd", line 32, in ? api = bootapi.BootAPI() File "/usr/lib/python2.4/site-packages/cobbler/api.py", line 70, in __init__ self.deserialize() File "/usr/lib/python2.4/site-packages/cobbler/api.py", line 364, in deserialize return self._config.deserialize() File "/usr/lib/python2.4/site-packages/cobbler/config.py", line 169, in deserialize if not serializer.deserialize(x,topological=True): File "/usr/lib/python2.4/site-packages/cobbler/serializer.py", line 92, in deserialize rc = storage_module.deserialize(obj,topological) File "/usr/lib/python2.4/site-packages/cobbler/modules/serializer_yaml.py", line 110, in deserialize obj.from_datastruct(datastruct) File "/usr/lib/python2.4/site-packages/cobbler/collection.py", line 99, in from_datastruct self.add(item) File "/usr/lib/python2.4/site-packages/cobbler/collection.py", line 171, in add if ref is None or not ref.is_valid(): File "/usr/lib/python2.4/site-packages/cobbler/item_repo.py", line 158, in is_valid raise CX(_("Since mirror is RHN %(m1)s, the repo must also be named %(m2)s") % args) cobbler.cexceptions.CX: 'Since mirror is RHN rhn://rhel-i386-server-5, the repo must also be named rhel-i386-server-5' [FAILED] error: %post(cobbler-0.8.1-1.noarch) scriptlet failed, exit status 1 If I rename the repo in this file to match the rhn mirror name: [root at radm011p cobbler]# cat /var/lib/cobbler/repos --- - arch: '' createrepo_flags: -c cache depth: 2 keep_updated: True mirror: 'rhn://rhel-i386-server-5' name: rhel5-i386-updates parent: '' rpm_list: '' I get this when I start cobbler: [root at radm011p cobbler]# service cobblerd start Starting cobbler daemon: warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates warning: repository not found: rhel5-i386-updates [ OK ] Is this the best way to solve this problem (updating all references to the previous repo name), or is there a smarter way to fix this? Thanks, Tom From mdehaan at redhat.com Thu Feb 21 19:32:06 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 21 Feb 2008 14:32:06 -0500 Subject: [et-mgmt-tools] rhn/local repo name mismatch error preventing cobblerd from starting In-Reply-To: <47BDD064.4030208@mcclatchyinteractive.com> References: <47BC8861.10508@redhat.com> <47BDD064.4030208@mcclatchyinteractive.com> Message-ID: <47BDD1B6.8040909@redhat.com> Tom Georgoulias wrote: > I upgraded from cobbler 0.6.3-2 to 0.8.1-1 this afternoon and got the > following error during the cobblerd restart: > > cobbler.cexceptions.CX: 'Since mirror is RHN rhn://rhel-i386-server-5, > the repo must also be named rhel-i386-server-5' ... snip ... Yes, this was applied in a recent version of cobbler to simplify the way RHN repos are mirrored, it's unfortunate that the upgrade wasn't seamless. "cobbler repo rename" is definitely the way to go. Cobbler probably should not be restarting cobblerd on upgrade. You may want to try --noscripts. > [FAILED] > error: %post(cobbler-0.8.1-1.noarch) scriptlet failed, exit status 1 > > If I rename the repo in this file to match the rhn mirror name: > > [root at radm011p cobbler]# cat /var/lib/cobbler/repos > --- > - > arch: '' > createrepo_flags: -c cache > depth: 2 > keep_updated: True > mirror: 'rhn://rhel-i386-server-5' > name: rhel5-i386-updates > parent: '' > rpm_list: '' > > > I get this when I start cobbler: > > [root at radm011p cobbler]# service cobblerd start > Starting cobbler daemon: warning: repository not found: > rhel5-i386-updates Right, you probably need to re-run "cobbler reposync" now because the directories have changed. Or you may want to remove your repo entries and resync them from the begining. I don't forsee the repo management code for those types of repos changing much in the future, so hopefully this should be the only compatibility problems you face. --Michael From tomg at mcclatchyinteractive.com Thu Feb 21 19:54:58 2008 From: tomg at mcclatchyinteractive.com (Tom Georgoulias) Date: Thu, 21 Feb 2008 14:54:58 -0500 Subject: [et-mgmt-tools] rhn/local repo name mismatch error preventing cobblerd from starting In-Reply-To: <47BDD1B6.8040909@redhat.com> References: <47BC8861.10508@redhat.com> <47BDD064.4030208@mcclatchyinteractive.com> <47BDD1B6.8040909@redhat.com> Message-ID: <47BDD712.500@mcclatchyinteractive.com> Michael DeHaan wrote: > Right, you probably need to re-run "cobbler reposync" now because the > directories have changed. Or you may want to remove your repo > entries and resync them from the begining. Thanks for the help. I just changed all of the references to the old repo name in /var/lib/cobbler/repos and /var/lib/cobbler/profile, the ran sync and reposync. Everything seems OK now. I've been happy with my 0.6.3 release and never had any problems with it, but thought that I should upgrade now to get more in line with what you are doing and to minimize these kinds of gotchas in the future. Tom From tiagocruz at forumgdh.net Thu Feb 21 20:46:15 2008 From: tiagocruz at forumgdh.net (Tiago Cruz) Date: Thu, 21 Feb 2008 17:46:15 -0300 Subject: [et-mgmt-tools] virt-manager on RH AS 5.1 Message-ID: <1203626775.12787.12.camel@tuxkiller.ig.com.br> Hello, I'm running "Red Hat Enterprise Linux Server release 5.1 (Tikanga)" on my server with: [root at xen-7 ~]# rpm -qa | grep virt libvirt-0.2.3-9.el5 libvirt-python-0.2.3-9.el5 virt-manager-0.4.0-3.el5 python-virtinst-0.103.0-3.el5_1.1 # egrep -v '(^#|^$)' /etc/libvirt/libvirtd.conf listen_tls = 1 listen_tcp = 1 tls_port = "16514" tcp_port = "16509" unix_sock_group = "libvirtd" unix_sock_ro_perms = "0777" unix_sock_rw_perms = "0770" auth_unix_ro = "none" auth_unix_rw = "none" # ps -ef | grep libvirt root 31142 1 0 17:29 ? 00:00:00 libvirt_qemud --system --daemon nobody 31167 31142 0 17:29 ? 00:00:00 dnsmasq --keep-in-foreground --strict-order --bind-interfaces --pid-file --conf-file --listen-address 192.168.122.1 --except-interface lo --dhcp-leasefile=/var/lib/libvirt/dhcp-default.leases --dhcp-range 192.168.122.2,192.168.122.254 But I can't see any port open on netstat: # netstat -nlp | grep 16509 # (nothing) So, on my desktop-client (Ubuntu hardy), the virt-manager (0.5.3) says: "Unable to open a connection to the libvirt management daemon. Verify that: - The 'libvirtd' daemon has been started" What can I do to fix this? Thanks -- Tiago Cruz http://everlinux.com Linux User #282636 From fukuta.saori at jp.fujitsu.com Fri Feb 22 05:19:17 2008 From: fukuta.saori at jp.fujitsu.com (Saori Fukuta) Date: Fri, 22 Feb 2008 14:19:17 +0900 Subject: [et-mgmt-tools] [PATCH] readOnly flags for virConnectOpenAuth In-Reply-To: <20080220130522.GB31194@redhat.com> References: <20080220170236.0FBC.FUKUTA.SAORI@jp.fujitsu.com> <20080220130522.GB31194@redhat.com> Message-ID: <20080222141833.F0A6.FUKUTA.SAORI@jp.fujitsu.com> On Wed, 20 Feb 2008 13:05:22 +0000 "Daniel P. Berrange" wrote: > On Wed, Feb 20, 2008 at 05:02:56PM +0900, Saori Fukuta wrote: > > I guess this problem occurs from Cset:680 > > http://hg.et.redhat.com/virt/applications/virt-manager--devel?cs=1892867ca5c7 > > This is intentional - virt-manager will authenticate with libvirt when > attempting to connect, so it gets a full read-write connection. In Fedora 8 > or later (or any distro with PolicyKit available) this should 'just work' > with you being prompted for password. > > > I'm not sure where the readOnly flags should be set, but how about > > this fix for readOnly flags ? > > Yes & no - we must only try a read-only connection if we are connecting > to a local hypervisor, and don't have PolicyKit available. This change > makes all connections read only, even remote ones, which is too mcuh. Okey, I understand the intention and I re-work to set the read-only flag under the following conditions, - destination is a local hypervisor, (i.e. including "///" in URI) - user is a non-root user, (i.e. uid is not "0") - PolicyKit is not available with libvirt, (i.e. "/usr/share/PolicyKit/policy/libvirtd.policy" has not been created by libvirt) Could you check the attached patch ? Thanks, Saori Fukuta -------------- next part -------------- A non-text attachment was scrubbed... Name: set_readOnly.patch Type: application/octet-stream Size: 1150 bytes Desc: not available URL: From jensampol at yahoo.de Fri Feb 22 15:08:01 2008 From: jensampol at yahoo.de (Jens Ahrens) Date: Fri, 22 Feb 2008 16:08:01 +0100 (CET) Subject: [et-mgmt-tools] cobbler 0.8 - --virt-bridge0= option unknown? Message-ID: <522434.64060.qm@web25501.mail.ukl.yahoo.com> Hi everybody, i'm trying to install a system with 2 NICs with cobbler (cobbler-0.8.0-1.fc7) running on FC7. When adding a profile or system using options like e.g. --virt-bridge0= --ip0= cobbler complains saying: Usage: cobbler system add [ARGS] cobbler: error: no such option: --virt-bridge0 here is an example line: cobbler system add --name=testsys --profile=testprofile --virt-path="/dev/xendisks/testdisk" --virt-bridge0=xenbr0 --virt-bridge1=xenbr1 --ip0=192.168.1.x ip1=192.168.2.x --hostname=testsystem.privnet.de --netboot-enabled=Y In the man pages of cobbler these options exist, so why doesn't cobler want to accept them? Thanks for any help, Jens __________________________________ Ihre erste Baustelle? Wissenswertes f?r Bastler und Hobby Handwerker. www.yahoo.de/clever From mdehaan at redhat.com Fri Feb 22 16:18:10 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 22 Feb 2008 11:18:10 -0500 Subject: [et-mgmt-tools] cobbler 0.8 - --virt-bridge0= option unknown? In-Reply-To: <522434.64060.qm@web25501.mail.ukl.yahoo.com> References: <522434.64060.qm@web25501.mail.ukl.yahoo.com> Message-ID: <47BEF5C2.9050106@redhat.com> Jens Ahrens wrote: > Hi everybody, > > i'm trying to install a system with 2 NICs with > cobbler (cobbler-0.8.0-1.fc7) running on FC7. > > When adding a profile or system using options like > e.g. > --virt-bridge0= > --ip0= > > cobbler complains saying: > Usage: cobbler system add [ARGS] > > cobbler: error: no such option: --virt-bridge0 > Easy one. Some users reported that the way multiple interfaces were specified and edited was confusing, so in order to simplify the way multiple interfaces work to users, we decided to do this... # for the first interface... cobbler system add --name=foo --virt-bridge=br1 --ip=192.168.1.55 # etc To edit additional interfaces you specify which interface you are editing with --interface. Using this option changes the value of the interface being edited. Note that you can only add/edit one interface at a time. cobbler system edit --name=foo --interface=1 --ip=192.168.1.56 This has been updated in the manpage. If I've missed any online docs, let me know, and I'll take care of them. As before, if you need to delete additional interfaces after you add one too many, you'll need to use the WebUI for that. The WebUI for editing multiple interfaces works as before. --Michael From tiagocruz at forumgdh.net Fri Feb 22 16:50:02 2008 From: tiagocruz at forumgdh.net (Tiago Cruz) Date: Fri, 22 Feb 2008 13:50:02 -0300 Subject: [et-mgmt-tools] virt-manager on RH AS 5.1 In-Reply-To: <1203626775.12787.12.camel@tuxkiller.ig.com.br> References: <1203626775.12787.12.camel@tuxkiller.ig.com.br> Message-ID: <1203699002.12787.35.camel@tuxkiller.ig.com.br> Someone here use RH AS and virt-manager remote? On Thu, 2008-02-21 at 17:46 -0300, Tiago Cruz wrote: > Hello, > > I'm running "Red Hat Enterprise Linux Server release 5.1 (Tikanga)" on > my server with: > > [root at xen-7 ~]# rpm -qa | grep virt > libvirt-0.2.3-9.el5 > libvirt-python-0.2.3-9.el5 > virt-manager-0.4.0-3.el5 > python-virtinst-0.103.0-3.el5_1.1 > > # egrep -v '(^#|^$)' /etc/libvirt/libvirtd.conf > listen_tls = 1 > listen_tcp = 1 > tls_port = "16514" > tcp_port = "16509" > unix_sock_group = "libvirtd" > unix_sock_ro_perms = "0777" > unix_sock_rw_perms = "0770" > auth_unix_ro = "none" > auth_unix_rw = "none" > > # ps -ef | grep libvirt > root 31142 1 0 17:29 ? 00:00:00 libvirt_qemud --system > --daemon > nobody 31167 31142 0 17:29 ? 00:00:00 dnsmasq > --keep-in-foreground --strict-order --bind-interfaces --pid-file > --conf-file --listen-address 192.168.122.1 --except-interface lo > --dhcp-leasefile=/var/lib/libvirt/dhcp-default.leases --dhcp-range > 192.168.122.2,192.168.122.254 > > > But I can't see any port open on netstat: > > # netstat -nlp | grep 16509 > # (nothing) > > So, on my desktop-client (Ubuntu hardy), the virt-manager (0.5.3) says: > > "Unable to open a connection to the libvirt management daemon. > > Verify that: > - The 'libvirtd' daemon has been started" > > What can I do to fix this? > > Thanks > -- Tiago Cruz http://everlinux.com Linux User #282636 > From tiagocruz at forumgdh.net Fri Feb 22 18:35:33 2008 From: tiagocruz at forumgdh.net (Tiago Cruz) Date: Fri, 22 Feb 2008 15:35:33 -0300 Subject: [et-mgmt-tools] libvirt_qemud Message-ID: <1203705333.12787.42.camel@tuxkiller.ig.com.br> Hello, libvirt_qemud its the same of libvirtd of documentation? >From http://libvirt.org/remote.html: -- libvirtd configuration file Libvirtd (the remote daemon) is configured from a file called /etc/libvirt/libvirtd.conf, or specified on the command line using -f filename or --config filename. --- >From $ libvirt_qemud --help Usage: libvirt_qemud [options] Options: -v | --verbose Verbose messages. -d | --daemon Run as a daemon & write PID file. -s | --system Run as system daemon. -t | --timeout Exit after timeout period. -p | --pid-file Change name of PID file. Notes: For '--system' option you must be running this daemon as root. The '--timeout' applies only when the daemon is not servicing clients. Default paths: Sockets (in system mode): /var/run/libvirt/qemud-sock /var/run/libvirt/qemud-sock-ro Sockets (not in system mode): $HOME/.libvirt/qemud-sock (in Unix abstract namespace) PID file (unless overridden by --pid-file): /var/run/libvirt_qemud.pid Thanks From tomackell at duke-energy.com Fri Feb 22 19:04:21 2008 From: tomackell at duke-energy.com (Mackell, Thomas O) Date: Fri, 22 Feb 2008 14:04:21 -0500 Subject: [et-mgmt-tools] Connection Failure Message-ID: <1B3160038A79B54DB674532195C68A582DD8E3@IMCLTEXCP15.nam.ent.duke-energy.com> I am very new to the cobbler setup. I just installed, Imported a distribution and did the rest of the setup. No major problems at this point. Set up the wiu interface, and I am receiving an error on almost all of the selections I try from there. The doc selection works fine. When I select any of the menu functions, I receive the following: MOD_PYTHON ERROR ProcessId: 27185 Interpreter: 'linuxtst' ServerName: 'linuxtst' DocumentRoot: '/var/www/html' URI: '/cobbler/web/' Location: None Directory: '/var/www/cobbler/web/' Filename: '/var/www/cobbler/web/' PathInfo: '' Phase: 'PythonHandler' Handler: 'index' Traceback (most recent call last): File "/usr/lib/python2.5/site-packages/mod_python/importer.py", line 1537, in HandlerDispatch default=default_handler, arg=req, silent=hlist.silent) File "/usr/lib/python2.5/site-packages/mod_python/importer.py", line 1229, in _process_target result = _execute_target(config, req, object, arg) File "/usr/lib/python2.5/site-packages/mod_python/importer.py", line 1128, in _execute_target result = object(arg) File "/var/www/cobbler/web/index.py", line 97, in handler content = func( **form ) File "/usr/lib/python2.5/site-packages/cobbler/webui/CobblerWeb.py", line 141, in distro_list if not self.__xmlrpc_setup(): File "/usr/lib/python2.5/site-packages/cobbler/webui/CobblerWeb.py", line 70, in __xmlrpc_setup raise e error: (111, 'Connection refused') MODULE CACHE DETAILS Accessed: Fri Feb 22 13:50:26 2008 Generation: 1 _mp_43be29589f5d909275df7d4c30cbcdd3 { FileName: '/var/www/cobbler/web/index.py' Instance: 1 Generation: 1 Modified: Thu Jan 31 14:46:14 2008 Imported: Fri Feb 22 13:48:49 2008 } >From the cobbler.log file I see: 2008-02-22 13:38:04,497 - api - sync 2008-02-22 13:48:35,178 - api - sync 2008-02-22 13:49:56,127 - remote - cobblerd started 2008-02-22 13:49:56,159 - remote - syslog running on 25150 2008-02-22 13:49:56,151 - remote - XMLRPC running on 25151 2008-02-22 13:50:11,431 - remote - login attempt; user(tomacke) 2008-02-22 13:50:11,435 - api - authenticate; ['tomacke', True] 2008-02-22 13:50:11,440 - remote - login succeeded; user(tomacke) I am running cobbler-0.8.0-1.fc8 Under Fedora Core 8 I must have missed something, At this point I don't see what it is. Thanks in advance, Tom Mackell -------------- next part -------------- An HTML attachment was scrubbed... URL: From mdehaan at redhat.com Fri Feb 22 20:08:53 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 22 Feb 2008 15:08:53 -0500 Subject: [et-mgmt-tools] Connection Failure In-Reply-To: <1B3160038A79B54DB674532195C68A582DD8E3@IMCLTEXCP15.nam.ent.duke-energy.com> References: <1B3160038A79B54DB674532195C68A582DD8E3@IMCLTEXCP15.nam.ent.duke-energy.com> Message-ID: <47BF2BD5.1010105@redhat.com> Mackell, Thomas O wrote: > > I am very new to the cobbler setup. > > I just installed, Imported a distribution and did the rest of the setup. > > No major problems at this point. > > Set up the wiu interface, and I am receiving an error on almost all of > the selections I try from there. > > The doc selection works fine. > Sounds like the web user interface is unable to connect to cobblerd, even though your log does indicate it is started. For starters, make sure you've followed the instructions here: https://fedorahosted.org/cobbler/wiki/CobblerWebInterface and that "cobbler check" does not report any problems. > When I select any of the menu functions, I receive the following: > > MOD_PYTHON ERROR > > ProcessId: 27185 > Interpreter: 'linuxtst' > > ServerName: 'linuxtst' > DocumentRoot: '/var/www/html' > > URI: '/cobbler/web/' > Location: None > Directory: '/var/www/cobbler/web/' > Filename: '/var/www/cobbler/web/' > PathInfo: '' > > Phase: 'PythonHandler' > Handler: 'index' > > Traceback (most recent call last): > > File "/usr/lib/python2.5/site-packages/mod_python/importer.py", line 1537, in HandlerDispatch > default=default_handler, arg=req, silent=hlist.silent) > > File "/usr/lib/python2.5/site-packages/mod_python/importer.py", line 1229, in _process_target > result = _execute_target(config, req, object, arg) > > File "/usr/lib/python2.5/site-packages/mod_python/importer.py", line 1128, in _execute_target > result = object(arg) > > File "/var/www/cobbler/web/index.py", line 97, in handler > content = func( **form ) > > File "/usr/lib/python2.5/site-packages/cobbler/webui/CobblerWeb.py", line 141, in distro_list > if not self.__xmlrpc_setup(): > > File "/usr/lib/python2.5/site-packages/cobbler/webui/CobblerWeb.py", line 70, in __xmlrpc_setup > raise e > > error: (111, 'Connection refused') > > > MODULE CACHE DETAILS > > Accessed: Fri Feb 22 13:50:26 2008 > Generation: 1 > > _mp_43be29589f5d909275df7d4c30cbcdd3 { > FileName: '/var/www/cobbler/web/index.py' > Instance: 1 > Generation: 1 > Modified: Thu Jan 31 14:46:14 2008 > Imported: Fri Feb 22 13:48:49 2008 > } > > From the cobbler.log file I see: > 2008-02-22 13:38:04,497 - api - sync > 2008-02-22 13:48:35,178 - api - sync > 2008-02-22 13:49:56,127 - remote - cobblerd started > 2008-02-22 13:49:56,159 - remote - syslog running on 25150 > 2008-02-22 13:49:56,151 - remote - XMLRPC running on 25151 > 2008-02-22 13:50:11,431 - remote - login attempt; user(tomacke) > 2008-02-22 13:50:11,435 - api - authenticate; ['tomacke', True] > 2008-02-22 13:50:11,440 - remote - login succeeded; user(tomacke) > > I am running cobbler-0.8.0-1.fc8 > Under Fedora Core 8 > > I must have missed something, At this point I don?t see what it is. > > > Thanks in advance, > > Tom Mackell > > ------------------------------------------------------------------------ > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From berrange at redhat.com Fri Feb 22 21:11:11 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 22 Feb 2008 21:11:11 +0000 Subject: [et-mgmt-tools] libvirt_qemud In-Reply-To: <1203705333.12787.42.camel@tuxkiller.ig.com.br> References: <1203705333.12787.42.camel@tuxkiller.ig.com.br> Message-ID: <20080222211111.GA17153@redhat.com> On Fri, Feb 22, 2008 at 03:35:33PM -0300, Tiago Cruz wrote: > Hello, > > libvirt_qemud its the same of libvirtd of documentation? No, they are different things. libvirt_qemud was an early prototype dedicated to providing the virtual networking (virbr0 bridge) and a QEMU/KVM driver. libvirtd provides a general purpose daemon doing these, and providing full remote management from off-host via SSL x509, and now SASL / Kerberos. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From mdehaan at redhat.com Fri Feb 22 21:45:00 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 22 Feb 2008 16:45:00 -0500 Subject: [et-mgmt-tools] Connection Failure In-Reply-To: <47BF2BD5.1010105@redhat.com> References: <1B3160038A79B54DB674532195C68A582DD8E3@IMCLTEXCP15.nam.ent.duke-energy.com> <47BF2BD5.1010105@redhat.com> Message-ID: <47BF425C.30400@redhat.com> Michael DeHaan wrote: > Mackell, Thomas O wrote: >> >> I am very new to the cobbler setup. >> >> I just installed, Imported a distribution and did the rest of the setup. >> >> No major problems at this point. >> >> Set up the wiu interface, and I am receiving an error on almost all >> of the selections I try from there. >> >> The doc selection works fine. >> > > Sounds like the web user interface is unable to connect to cobblerd, > even though your log does indicate it is started. > > For starters, make sure you've followed the instructions here: > https://fedorahosted.org/cobbler/wiki/CobblerWebInterface and > that "cobbler check" does not report any problems. > > >> When I select any of the menu functions, I receive the following: >> >> MOD_PYTHON ERROR >> >> ProcessId: 27185 >> Interpreter: 'linuxtst' >> >> ServerName: 'linuxtst' >> DocumentRoot: '/var/www/html' >> >> URI: '/cobbler/web/' >> Location: None >> Directory: '/var/www/cobbler/web/' >> Filename: '/var/www/cobbler/web/' >> PathInfo: '' >> >> Phase: 'PythonHandler' >> Handler: 'index' >> >> Traceback (most recent call last): >> >> File "/usr/lib/python2.5/site-packages/mod_python/importer.py", >> line 1537, in HandlerDispatch >> default=default_handler, arg=req, silent=hlist.silent) >> >> File "/usr/lib/python2.5/site-packages/mod_python/importer.py", >> line 1229, in _process_target >> result = _execute_target(config, req, object, arg) >> >> File "/usr/lib/python2.5/site-packages/mod_python/importer.py", >> line 1128, in _execute_target >> result = object(arg) >> >> File "/var/www/cobbler/web/index.py", line 97, in handler >> content = func( **form ) >> >> File >> "/usr/lib/python2.5/site-packages/cobbler/webui/CobblerWeb.py", line >> 141, in distro_list >> if not self.__xmlrpc_setup(): >> >> File >> "/usr/lib/python2.5/site-packages/cobbler/webui/CobblerWeb.py", line >> 70, in __xmlrpc_setup >> raise e >> >> error: (111, 'Connection refused') >> >> >> MODULE CACHE DETAILS >> >> Accessed: Fri Feb 22 13:50:26 2008 >> Generation: 1 >> >> _mp_43be29589f5d909275df7d4c30cbcdd3 { >> FileName: '/var/www/cobbler/web/index.py' >> Instance: 1 >> Generation: 1 >> Modified: Thu Jan 31 14:46:14 2008 >> Imported: Fri Feb 22 13:48:49 2008 >> } >> >> From the cobbler.log file I see: >> 2008-02-22 13:38:04,497 - api - sync >> 2008-02-22 13:48:35,178 - api - sync >> 2008-02-22 13:49:56,127 - remote - cobblerd started >> 2008-02-22 13:49:56,159 - remote - syslog running on 25150 >> 2008-02-22 13:49:56,151 - remote - XMLRPC running on 25151 >> 2008-02-22 13:50:11,431 - remote - login attempt; user(tomacke) >> 2008-02-22 13:50:11,435 - api - authenticate; ['tomacke', True] >> 2008-02-22 13:50:11,440 - remote - login succeeded; user(tomacke) >> >> I am running cobbler-0.8.0-1.fc8 >> Under Fedora Core 8 >> >> I must have missed something, At this point I don?t see what it is. >> >> >> Thanks in advance, >> >> Tom Mackell Also of WebUI note... Someone recently reported an error submitting the profile webform with Python 2.5 and mod_python. I've got a workaround posted in git. If you don't want to apply that, you can still use the command line to assign repos to profiles. If you are using the WebUI and also using cobbler's repo management features, and are seeing cobbler failing to save your repo choices, be aware of that, and we'll include that in the next bugfix release. Long term I am pondering TurboGears (the switch would be reasonably easy) but the existing implementation has the virtue of already working pretty well on older platforms. If someone wants to take a crack at a TurboGears port that could be interesting. (Namely I think latching on the authentication providers could streamline some things). --Michael From mdehaan at redhat.com Fri Feb 22 21:52:33 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 22 Feb 2008 16:52:33 -0500 Subject: [et-mgmt-tools] Connection Failure In-Reply-To: <47BF425C.30400@redhat.com> References: <1B3160038A79B54DB674532195C68A582DD8E3@IMCLTEXCP15.nam.ent.duke-energy.com> <47BF2BD5.1010105@redhat.com> <47BF425C.30400@redhat.com> Message-ID: <47BF4421.20503@redhat.com> Removing extra text so people can read this ... > > Also of WebUI note... > > Someone recently reported an error submitting the profile webform with > Python 2.5 and mod_python. > > I've got a workaround posted in git. If you don't want to apply > that, you can still use the command line > to assign repos to profiles. > > If you are using the WebUI and also using cobbler's repo management > features, and are seeing cobbler > failing to save your repo choices, be aware of that, and we'll include > that in the next bugfix release. > > Long term I am pondering TurboGears (the switch would be reasonably > easy) but the existing implementation > has the virtue of already working pretty well on older platforms. If > someone wants to take a crack > at a TurboGears port that could be interesting. (Namely I think > latching on the authentication providers > could streamline some things). > > --Michael > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From jensampol at yahoo.de Sat Feb 23 16:49:23 2008 From: jensampol at yahoo.de (Jens Ahrens) Date: Sat, 23 Feb 2008 17:49:23 +0100 (CET) Subject: [et-mgmt-tools] cobbler 0.8 - --virt-bridge0= option unknown? In-Reply-To: <47BEF5C2.9050106@redhat.com> Message-ID: <309386.16769.qm@web25502.mail.ukl.yahoo.com> --- Michael DeHaan schrieb: > Jens Ahrens wrote: > > Hi everybody, > > > > i'm trying to install a system with 2 NICs with > > cobbler (cobbler-0.8.0-1.fc7) running on FC7. > > > > When adding a profile or system using options like > > e.g. > > --virt-bridge0= > > --ip0= > > > > cobbler complains saying: > > Usage: cobbler system add [ARGS] > > > > cobbler: error: no such option: --virt-bridge0 > > > > Easy one. Some users reported that the way > multiple interfaces were > specified and edited was confusing, so > in order to simplify the way multiple interfaces > work to users, we > decided to do this... > > # for the first interface... > > cobbler system add --name=foo --virt-bridge=br1 > --ip=192.168.1.55 # etc > > To edit additional interfaces you specify which > interface you are > editing with --interface. Using this option > changes the value of the interface being edited. > Note that you can > only add/edit one interface at a time. > > cobbler system edit --name=foo --interface=1 > --ip=192.168.1.56 > > This has been updated in the manpage. If I've > missed any online docs, > let me know, and I'll take care of them. > > As before, if you need to delete additional > interfaces after you add one > too many, you'll need to use the WebUI for that. > > The WebUI for editing multiple interfaces works as > before. > > --Michael > > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > Thanks, that seems to work. By the way - the man page of cobbler says that if one wanted to import a mirror from a dvd image he should use: cobbler import --path=/mnt/dvd --name=anyname but that doesn't work: cobbler: error: no such option: --path Like this it works: cobbler import --mirror=/mnt/dvd --name=anyname Cheers and thanks again, Jens Lesen Sie Ihre E-Mails jetzt einfach von unterwegs. www.yahoo.de/go From berrange at redhat.com Sat Feb 23 20:33:22 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Sat, 23 Feb 2008 20:33:22 +0000 Subject: [et-mgmt-tools] virt-manager allows fully virt install from kernel/initrd Message-ID: <20080223203322.GA30832@redhat.com> Traditionally paravirt Xen installs only supported NFS/FTP/HTTP based network installs due to the requirement to use a kernel/initrd, while fullvirt Xen installs only spported ISO/CDROM due to requirement to boot from CDROM. With the introduction of KVM though, it became possible to install fullyvirt guests off a NFS/FTP/HTTP tree directly in same manner as paravirt. Xen also recently gained this capability for fullvirt guests. Thus I have just pushed changes to virt-manager to allow fullyvirt install to be performed using NFS/FTP/HTTP sources, in addition to existing option of doing ISO/CDROM installs. http://hg.et.redhat.com/virt/applications/virt-manager--devel?cs=ea6903353938 Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From berrange at redhat.com Sat Feb 23 20:40:44 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Sat, 23 Feb 2008 20:40:44 +0000 Subject: [et-mgmt-tools] [PATCH] readOnly flags for virConnectOpenAuth In-Reply-To: <20080222141833.F0A6.FUKUTA.SAORI@jp.fujitsu.com> References: <20080220170236.0FBC.FUKUTA.SAORI@jp.fujitsu.com> <20080220130522.GB31194@redhat.com> <20080222141833.F0A6.FUKUTA.SAORI@jp.fujitsu.com> Message-ID: <20080223204044.GB30832@redhat.com> On Fri, Feb 22, 2008 at 02:19:17PM +0900, Saori Fukuta wrote: > On Wed, 20 Feb 2008 13:05:22 +0000 "Daniel P. Berrange" wrote: > > On Wed, Feb 20, 2008 at 05:02:56PM +0900, Saori Fukuta wrote: > > > I guess this problem occurs from Cset:680 > > > http://hg.et.redhat.com/virt/applications/virt-manager--devel?cs=1892867ca5c7 > > > > This is intentional - virt-manager will authenticate with libvirt when > > attempting to connect, so it gets a full read-write connection. In Fedora 8 > > or later (or any distro with PolicyKit available) this should 'just work' > > with you being prompted for password. > > > > > I'm not sure where the readOnly flags should be set, but how about > > > this fix for readOnly flags ? > > > > Yes & no - we must only try a read-only connection if we are connecting > > to a local hypervisor, and don't have PolicyKit available. This change > > makes all connections read only, even remote ones, which is too mcuh. > > Okey, I understand the intention and I re-work to set the read-only > flag under the following conditions, > - destination is a local hypervisor, > (i.e. including "///" in URI) This also matches 'xen:///', so I changed it to use the 'is_remote' method to check. > - user is a non-root user, > (i.e. uid is not "0") > - PolicyKit is not available with libvirt, > (i.e. "/usr/share/PolicyKit/policy/libvirtd.policy" has not > been created by libvirt) > > Could you check the attached patch ? I applied it with a minor change: http://hg.et.redhat.com/virt/applications/virt-manager--devel?cs=d1c6390bbea9 Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From berrange at redhat.com Sun Feb 24 17:05:32 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Sun, 24 Feb 2008 17:05:32 +0000 Subject: [et-mgmt-tools] Patched virt-install to use capabilities XML Message-ID: <20080224170532.GD5500@redhat.com> FYI, I've just patched virt-install to make use of the libvirt capabilities XML instead of poking around in /proc & /sys & /dev to determine hypervisor capabilities. This removes some of the hardcoded assumptions about HV support and is a step towards being able to run virt-install remotely. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From tim.verhoeven.be at gmail.com Mon Feb 25 10:53:19 2008 From: tim.verhoeven.be at gmail.com (Tim Verhoeven) Date: Mon, 25 Feb 2008 11:53:19 +0100 Subject: [et-mgmt-tools] Cobbler & Func presentation at Fosdem Message-ID: <2a7fce340802250253m720f2613xeb5f4f00ff2936d8@mail.gmail.com> Hi, Just for your information. I've done a talk about Virtualization in the CentOS developers room at Fosdem (last weekend). But I've also talked and demo'ed Cobbler and Func during my talk. The slides from that talk are online on the CentOS wiki at http://wiki.centos.org/Events/Fosdem2008. Regards, Tim -- Tim Verhoeven - tim.verhoeven.be at gmail.com - 0479 / 88 11 83 Hoping the problem magically goes away by ignoring it is the "microsoft approach to programming" and should never be allowed. (Linus Torvalds) From tom at ng23.net Mon Feb 25 12:48:16 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 12:48:16 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble Message-ID: <47C2B910.1020904@ng23.net> Hi I am still having issues trying to get xen based VM's running, here is what i have done so far. Have cobbler build a x86_64 machine with the xen kernel etc 2.6.18-53.el5xen Create a xen based profile so that i can install a xen system cobbler profile add --name=RHEL4-u5-xen-i386 --distro=RHEL4-u5-xen-i386 --kickstart=/etc/cobbler/kickstart_RHEL4-u5-xen-i386.ks Then create a system in cobbler to use this profile cobbler system add --name=virtbox --hostname=virtbox --ip=192.168.28.247 --subnet=255.255.255.0 --gateway=192.168.12.2 --profile=RHEL4-u5-xen-i386 Then on the machine that i built using cobbler use koan to do the following koan --virt --server=192.168.12.161 --system=virtbox This appears to work fine - using kickstart from cobbler: http://192.168.12.161/cblr/kickstarts_sys/virtbox/ks.cfg - no virt-type specified, auto-selecting xenpv libvirt_qemud (pid 2345) is running... downloading initrd initrd.img to /var/lib/xen/initrd.img url=http://192.168.12.161/cobbler/images/RHEL4-u5-xen-i386/initrd.img - using kickstart from cobbler: http://192.168.12.161/cobbler/images/RHEL4-u5-xen-i386/initrd.img downloading kernel vmlinuz to /var/lib/xen/vmlinuz url=http://192.168.12.161/cobbler/images/RHEL4-u5-xen-i386/vmlinuz - using kickstart from cobbler: http://192.168.12.161/cobbler/images/RHEL4-u5-xen-i386/vmlinuz Kernel arguments: ks=http://192.168.12.161/cblr/kickstarts_sys/virtbox/ks.cfg ksdevice=eth0 lang= kssendmac syslog=192.168.12.161:25150 text use virt-manager or reconnect with virsh console virtbox as i can see that the virt box is created but it only ever shows as 'blocked' # virsh Welcome to virsh, the virtualization interactive terminal. Type: 'help' for help with commands 'quit' to quit virsh # list Id Name State ---------------------------------- 0 Domain-0 running 4 virtbox blocked Any ideas here to what i am doing wrong? thanks From tim.verhoeven.be at gmail.com Mon Feb 25 13:05:54 2008 From: tim.verhoeven.be at gmail.com (Tim Verhoeven) Date: Mon, 25 Feb 2008 14:05:54 +0100 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2B910.1020904@ng23.net> References: <47C2B910.1020904@ng23.net> Message-ID: <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> On Mon, Feb 25, 2008 at 1:48 PM, Tom Brown wrote: > > I am still having issues trying to get xen based VM's running, here is > what i have done so far. > ...snip... > > Then on the machine that i built using cobbler use koan to do the following > > koan --virt --server=192.168.12.161 --system=virtbox > > This appears to work fine > > - using kickstart from cobbler: > http://192.168.12.161/cblr/kickstarts_sys/virtbox/ks.cfg > - no virt-type specified, auto-selecting xenpv > libvirt_qemud (pid 2345) is running... > downloading initrd initrd.img to /var/lib/xen/initrd.img > url=http://192.168.12.161/cobbler/images/RHEL4-u5-xen-i386/initrd.img > - using kickstart from cobbler: > http://192.168.12.161/cobbler/images/RHEL4-u5-xen-i386/initrd.img > downloading kernel vmlinuz to /var/lib/xen/vmlinuz > url=http://192.168.12.161/cobbler/images/RHEL4-u5-xen-i386/vmlinuz > - using kickstart from cobbler: > http://192.168.12.161/cobbler/images/RHEL4-u5-xen-i386/vmlinuz > Kernel arguments: > ks=http://192.168.12.161/cblr/kickstarts_sys/virtbox/ks.cfg > ksdevice=eth0 lang= kssendmac syslog=192.168.12.161:25150 text > use virt-manager or reconnect with virsh console virtbox > > as i can see that the virt box is created but it only ever shows as > 'blocked' > > # virsh > Welcome to virsh, the virtualization interactive terminal. > > Type: 'help' for help with commands > 'quit' to quit > > virsh # list > Id Name State > ---------------------------------- > 0 Domain-0 running > 4 virtbox blocked > > Any ideas here to what i am doing wrong? It looks like you are not doing anything wrong. Block in Xen terms mean that at the moment you looked the VM is not using CPU but is waiting I/O to happen before it can continue. So nothing to worry about. Did you actually looked at the console of the virtual machine if the installation is running or not ? I think it probably will be performing a install. Regards, Tim -- Tim Verhoeven - tim.verhoeven.be at gmail.com - 0479 / 88 11 83 Hoping the problem magically goes away by ignoring it is the "microsoft approach to programming" and should never be allowed. (Linus Torvalds) From tom at ng23.net Mon Feb 25 13:25:42 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 13:25:42 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> Message-ID: <47C2C1D6.3070706@ng23.net> > It looks like you are not doing anything wrong. Block in Xen terms > mean that at the moment you looked the VM is not using CPU but is > waiting I/O to happen before it can continue. So nothing to worry > about. > > Did you actually looked at the console of the virtual machine if the > installation is running or not ? I think it probably will be > performing a install. > > thanks for the tip - using virsh to connect to the console it appears to be stuck here -- snip -- Using IPI No-Shortcut mode XENBUS: Device with no driver: device/vbd/51712 XENBUS: Device with no driver: device/vif/0 md: Autodetecting RAID arrays. md: autorun ... md: ... autorun DONE. RAMDISK: ext2 filesystem found at block 0 RAMDISK: Loading 8192KiB [1 disk] into ram disk... done. VFS: Mounted root (ext2 filesystem). From berrange at redhat.com Mon Feb 25 13:29:19 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 25 Feb 2008 13:29:19 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2C1D6.3070706@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> Message-ID: <20080225132919.GD2614@redhat.com> On Mon, Feb 25, 2008 at 01:25:42PM +0000, Tom Brown wrote: > > >It looks like you are not doing anything wrong. Block in Xen terms > >mean that at the moment you looked the VM is not using CPU but is > >waiting I/O to happen before it can continue. So nothing to worry > >about. > > > >Did you actually looked at the console of the virtual machine if the > >installation is running or not ? I think it probably will be > >performing a install. > > > > > > thanks for the tip - using virsh to connect to the console it appears to > be stuck here It most likely /isn't/ stuck - boot up messages will be on the graphical console instead of the text console. You can use virt-viewer to access the graphical console Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From tom at ng23.net Mon Feb 25 13:45:33 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 13:45:33 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <20080225132919.GD2614@redhat.com> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> Message-ID: <47C2C67D.6030402@ng23.net> > > It most likely /isn't/ stuck - boot up messages will be on the graphical > console instead of the text console. You can use virt-viewer to access > the graphical console > > > Is there an rpm for virt-viewer available for RHEL 5 ? I am seeing them for Fedora but not RHEL - or a src rpm or something similar? thanks From mnielsen at redhat.com Mon Feb 25 13:50:59 2008 From: mnielsen at redhat.com (Mark Nielsen) Date: Mon, 25 Feb 2008 08:50:59 -0500 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2C67D.6030402@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> Message-ID: <1203947459.3346.4.camel@localhost.localdomain> On Mon, 2008-02-25 at 13:45 +0000, Tom Brown wrote: > > > > It most likely /isn't/ stuck - boot up messages will be on the graphical > > console instead of the text console. You can use virt-viewer to access > > the graphical console > > > > > > > > Is there an rpm for virt-viewer available for RHEL 5 ? I am seeing them > for Fedora but not RHEL - or a src rpm or something similar? > you can also use virt-manager and get a graphical console from that Mark > thanks > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From tim.verhoeven.be at gmail.com Mon Feb 25 13:50:53 2008 From: tim.verhoeven.be at gmail.com (Tim Verhoeven) Date: Mon, 25 Feb 2008 14:50:53 +0100 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2C67D.6030402@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> Message-ID: <2a7fce340802250550s998370fr4b9d5ab51c0f81a9@mail.gmail.com> On Mon, Feb 25, 2008 at 2:45 PM, Tom Brown wrote: > > > > > It most likely /isn't/ stuck - boot up messages will be on the graphical > > console instead of the text console. You can use virt-viewer to access > > the graphical console > > > > > > > > Is there an rpm for virt-viewer available for RHEL 5 ? I am seeing them > for Fedora but not RHEL - or a src rpm or something similar? I don't think virt-viewer will work on RHEL5, but just start the virt-manager and look at the console in there. Regards, Tim -- Tim Verhoeven - tim.verhoeven.be at gmail.com - 0479 / 88 11 83 Hoping the problem magically goes away by ignoring it is the "microsoft approach to programming" and should never be allowed. (Linus Torvalds) From berrange at redhat.com Mon Feb 25 13:51:59 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 25 Feb 2008 13:51:59 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2C67D.6030402@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> Message-ID: <20080225135159.GF2614@redhat.com> On Mon, Feb 25, 2008 at 01:45:33PM +0000, Tom Brown wrote: > > > > >It most likely /isn't/ stuck - boot up messages will be on the graphical > >console instead of the text console. You can use virt-viewer to access > >the graphical console > > > > > > > > Is there an rpm for virt-viewer available for RHEL 5 ? I am seeing them > for Fedora but not RHEL - or a src rpm or something similar? Its not yet available in RHEL-5 - just use vncviewer or virt-manager there. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From tom at ng23.net Mon Feb 25 13:55:37 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 13:55:37 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <20080225135159.GF2614@redhat.com> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> Message-ID: <47C2C8D9.6020608@ng23.net> > Its not yet available in RHEL-5 - just use vncviewer or virt-manager > there. > > > I cant - i dont have X avilable to me - So i cant look at the console if i dont have X ? From tim.verhoeven.be at gmail.com Mon Feb 25 14:05:23 2008 From: tim.verhoeven.be at gmail.com (Tim Verhoeven) Date: Mon, 25 Feb 2008 15:05:23 +0100 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2C8D9.6020608@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> Message-ID: <2a7fce340802250605x43672517m686f6b7333373e08@mail.gmail.com> On Mon, Feb 25, 2008 at 2:55 PM, Tom Brown wrote: > > > Its not yet available in RHEL-5 - just use vncviewer or virt-manager > > there. > > > > > > > > I cant - i dont have X avilable to me - So i cant look at the console if > i dont have X ? Just use X forwarding. Login with ssh -X and you should be able to see the virt-manager on your local console. Regards, Tim -- Tim Verhoeven - tim.verhoeven.be at gmail.com - 0479 / 88 11 83 Hoping the problem magically goes away by ignoring it is the "microsoft approach to programming" and should never be allowed. (Linus Torvalds) From capel at stone-it.com Mon Feb 25 14:06:21 2008 From: capel at stone-it.com (Jasper Capel) Date: Mon, 25 Feb 2008 15:06:21 +0100 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2C8D9.6020608@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> Message-ID: <47C2CB5D.1020303@stone-it.com> Tom Brown wrote: > > I cant - i dont have X avilable to me - So i cant look at the console > if i dont have X ? > You can, if you tunnel your X over ssh from a machine on which you're logged in graphically. You can do this by passing the -X or -Y parameter to the ssh-client. Make sure xauth is installed on the server machine. -- Kind regards, Jasper Capel email : capel at stone-it.com web : http://www.stone-it.com/ .~. /V\ Stone-IT tel : +31 (0)62 86 50 854 // \\ LINUX FOR e-BUSINESS fax : +31 (0)30 22 87 997 /( )\ ^^-^^ Groenekanseweg 246, 3737 AL Groenekan, Nederland From mdehaan at redhat.com Mon Feb 25 15:47:59 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 25 Feb 2008 10:47:59 -0500 Subject: [et-mgmt-tools] cobbler 0.8 - --virt-bridge0= option unknown? In-Reply-To: <309386.16769.qm@web25502.mail.ukl.yahoo.com> References: <309386.16769.qm@web25502.mail.ukl.yahoo.com> Message-ID: <47C2E32F.8030500@redhat.com> > > By the way - the man page of cobbler says that if one > wanted to import a mirror from a dvd image he should > use: > cobbler import --path=/mnt/dvd --name=anyname > > but that doesn't work: > cobbler: error: no such option: --path > > Like this it works: > cobbler import --mirror=/mnt/dvd --name=anyname > Yep. I'll fix the manpage for the next release. Thanks! --Michael From mdehaan at redhat.com Mon Feb 25 15:59:16 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 25 Feb 2008 10:59:16 -0500 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2C8D9.6020608@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> Message-ID: <47C2E5D4.9040505@redhat.com> Tom Brown wrote: > >> Its not yet available in RHEL-5 - just use vncviewer or virt-manager >> there. >> >> >> > > I cant - i dont have X avilable to me - So i cant look at the console > if i dont have X ? > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Alternatively, pass --nogfx to koan when you're installing and you'll be able to use virsh instead of VNC... --Michael From mdehaan at redhat.com Mon Feb 25 16:01:43 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 25 Feb 2008 11:01:43 -0500 Subject: [et-mgmt-tools] virt-manager allows fully virt install from kernel/initrd In-Reply-To: <20080223203322.GA30832@redhat.com> References: <20080223203322.GA30832@redhat.com> Message-ID: <47C2E667.1020904@redhat.com> Daniel P. Berrange wrote: > Traditionally paravirt Xen installs only supported NFS/FTP/HTTP based network > installs due to the requirement to use a kernel/initrd, while fullvirt Xen > installs only spported ISO/CDROM due to requirement to boot from CDROM. > > With the introduction of KVM though, it became possible to install fullyvirt > guests off a NFS/FTP/HTTP tree directly in same manner as paravirt. Xen also > recently gained this capability for fullvirt guests. > > Thus I have just pushed changes to virt-manager to allow fullyvirt install > to be performed using NFS/FTP/HTTP sources, in addition to existing option > of doing ISO/CDROM installs. > > http://hg.et.redhat.com/virt/applications/virt-manager--devel?cs=ea6903353938 > > Dan. > Dan, Cool. I take it this is the features-in-upstream announcement? When is this going to be available for the various distros (so I can possibly obsolete the Xen PXE install method for Linux-FV)? --Michael From berrange at redhat.com Mon Feb 25 16:07:51 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 25 Feb 2008 16:07:51 +0000 Subject: [et-mgmt-tools] virt-manager allows fully virt install from kernel/initrd In-Reply-To: <47C2E667.1020904@redhat.com> References: <20080223203322.GA30832@redhat.com> <47C2E667.1020904@redhat.com> Message-ID: <20080225160751.GI2614@redhat.com> On Mon, Feb 25, 2008 at 11:01:43AM -0500, Michael DeHaan wrote: > Daniel P. Berrange wrote: > >Traditionally paravirt Xen installs only supported NFS/FTP/HTTP based > >network > >installs due to the requirement to use a kernel/initrd, while fullvirt Xen > >installs only spported ISO/CDROM due to requirement to boot from CDROM. > > > >With the introduction of KVM though, it became possible to install > >fullyvirt > >guests off a NFS/FTP/HTTP tree directly in same manner as paravirt. Xen > >also > >recently gained this capability for fullvirt guests. > > > >Thus I have just pushed changes to virt-manager to allow fullyvirt install > >to be performed using NFS/FTP/HTTP sources, in addition to existing option > >of doing ISO/CDROM installs. > > > >http://hg.et.redhat.com/virt/applications/virt-manager--devel?cs=ea6903353938 > > Cool. I take it this is the features-in-upstream announcement? > > When is this going to be available for the various distros (so I can > possibly obsolete the Xen PXE install method for Linux-FV)? It requires corresponding support in libvirt (next pending release) and Xen (next pending release). This will all be included from Fedora 9, via the forthcoming libvirt release, and the Xen bits are already backported. http://fedoraproject.org/wiki/Features/XenFullvirtKernelBoot Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From mdehaan at redhat.com Mon Feb 25 16:12:59 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 25 Feb 2008 11:12:59 -0500 Subject: [et-mgmt-tools] Cobbler & Func presentation at Fosdem In-Reply-To: <2a7fce340802250253m720f2613xeb5f4f00ff2936d8@mail.gmail.com> References: <2a7fce340802250253m720f2613xeb5f4f00ff2936d8@mail.gmail.com> Message-ID: <47C2E90B.1020700@redhat.com> Tim Verhoeven wrote: > Hi, > > Just for your information. I've done a talk about Virtualization in > the CentOS developers room at Fosdem (last weekend). But I've also > talked and demo'ed Cobbler and Func during my talk. The slides from > that talk are online on the CentOS wiki at > http://wiki.centos.org/Events/Fosdem2008. > > Regards, > Tim > > Very nice! Thanks for demoing! --Michael From tom at ng23.net Mon Feb 25 16:24:44 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 16:24:44 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2E5D4.9040505@redhat.com> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> Message-ID: <47C2EBCC.10508@ng23.net> > > Alternatively, pass --nogfx to koan when you're installing and you'll > be able to use virsh instead of VNC... > thanks for all the hints - i would like to not use X at all, locally or remote, if possible so i have used the --nogfx option and it works, sort of. i seem to have some font issue - as the console appears like below which is pretty unreadable - any hints? thanks W?????? ?? R?? H?? E????????? L???? +--------------------+ C???????? TCP/IP +---------------------+ ? ? ? P????? ????? ??? IP ????????????? ??? ???? ???????. E??? ? ? ???? ?????? ?? ??????? ?? ?? IP ??????? ?? ??????-??????? ? ? ???????? (??? ???????, 1.2.3.4). ? ? ? ? [*] U?? ??????? IP ????????????? (BOOTP/DHCP) ? ? ? ? IP ???????: ________________ ? ? N??????: ________________ ? ? D?????? ??????? (IP): ________________ ? ? P?????? ??????????: ________________ ? ? ? ? +----+ +------+ ? ? ? OK ? ? B??? ? ? ? +----+ +------+ ? ? ? ? ? +-------------------------------------------------------------+ / ??????? ???????? ? ??????? ? ???? ?????? From 1.cobbler at msquared.id.au Mon Feb 25 16:30:39 2008 From: 1.cobbler at msquared.id.au (Msquared) Date: Tue, 26 Feb 2008 01:30:39 +0900 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2C8D9.6020608@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> Message-ID: <20080225163038.GN4042@sliderule.msquared.net.au> On Mon, Feb 25, 2008 at 01:55:37PM +0000, Tom Brown wrote: > I cant - i dont have X avilable to me - So i cant look at the console if > i dont have X ? If you have X on another machine, you can do something like this: vncviewer -via virtualhost localhost`ssh root at virtualhost virsh vncdisplay virtualguest` Where: * virtualhost is the physical machine * virtualguest is the Xen guest you are trying to connect to If you don't have X (ie: your desktop workstation is Windows), then you can do the same thing, but it's a little more fiddly: 1. SSH into the host as root and run "virsh vncdisplay virtualguest" to find out what display virtualguest is using (eg: ":2") 2. Add 5900 to the number from step 1 (eg: 5902). 3. SSH into the host again, but forward any local port to the port number from step 2. (eg: forward 5905 to 5902) [Note that PuTTY can forward ports for you] 4. VNC to the forwarded port on the local host (eg: 5905). Regards, Msquared... From mdehaan at redhat.com Mon Feb 25 16:36:38 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 25 Feb 2008 11:36:38 -0500 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2EBCC.10508@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> Message-ID: <47C2EE96.5070702@redhat.com> Tom Brown wrote: > >> >> Alternatively, pass --nogfx to koan when you're installing and you'll >> be able to use virsh instead of VNC... >> > > thanks for all the hints - i would like to not use X at all, locally > or remote, if possible so i have used the --nogfx option and it works, > sort of. > > i seem to have some font issue - as the console appears like below > which is pretty unreadable - > > any hints? > > thanks It's always worked fine for me through gnome-terminal. What are you using? From mnielsen at redhat.com Mon Feb 25 16:43:49 2008 From: mnielsen at redhat.com (Mark Nielsen) Date: Mon, 25 Feb 2008 11:43:49 -0500 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <20080225163038.GN4042@sliderule.msquared.net.au> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <20080225163038.GN4042@sliderule.msquared.net.au> Message-ID: <1203957829.3346.12.camel@localhost.localdomain> On Tue, 2008-02-26 at 01:30 +0900, Msquared wrote: > On Mon, Feb 25, 2008 at 01:55:37PM +0000, Tom Brown wrote: > > > I cant - i dont have X avilable to me - So i cant look at the console if > > i dont have X ? > > If you have X on another machine, you can do something like this: > > vncviewer -via virtualhost localhost`ssh root at virtualhost virsh vncdisplay virtualguest` > > Where: > > * virtualhost is the physical machine > * virtualguest is the Xen guest you are trying to connect to > > If you don't have X (ie: your desktop workstation is Windows), then you > can do the same thing, but it's a little more fiddly: > I like using Xming. http://sourceforge.net/projects/xming it is a very small X server for windows, and doesn't even require admin privileges to install. Then you run putty with X forwarding enabled. > 1. SSH into the host as root and run "virsh vncdisplay virtualguest" > to find out what display virtualguest is using (eg: ":2") > > 2. Add 5900 to the number from step 1 (eg: 5902). > > 3. SSH into the host again, but forward any local port to the port > number from step 2. (eg: forward 5905 to 5902) [Note that PuTTY > can forward ports for you] > > 4. VNC to the forwarded port on the local host (eg: 5905). > > Regards, Msquared... > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From tom at ng23.net Mon Feb 25 16:48:21 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 16:48:21 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <1203957829.3346.12.camel@localhost.localdomain> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <20080225163038.GN4042@sliderule.msquared.net.au> <1203957829.3346.12.camel@localhost.localdomain> Message-ID: <47C2F155.2040507@ng23.net> Mark Nielsen wrote: > On Tue, 2008-02-26 at 01:30 +0900, Msquared wrote: > >> On Mon, Feb 25, 2008 at 01:55:37PM +0000, Tom Brown wrote: >> >> >>> I cant - i dont have X avilable to me - So i cant look at the console if >>> i dont have X ? >>> >> If you have X on another machine, you can do something like this: >> >> vncviewer -via virtualhost localhost`ssh root at virtualhost virsh vncdisplay virtualguest` >> >> Where: >> >> * virtualhost is the physical machine >> * virtualguest is the Xen guest you are trying to connect to >> >> If you don't have X (ie: your desktop workstation is Windows), then you >> can do the same thing, but it's a little more fiddly: >> >> > > I like using Xming. > http://sourceforge.net/projects/xming > it is a very small X server for windows, and doesn't even require admin > privileges to install. Then you run putty with X forwarding enabled. > i am on leopard so will look to see about finding a solution for that From tom at ng23.net Mon Feb 25 16:50:49 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 16:50:49 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2EE96.5070702@redhat.com> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> Message-ID: <47C2F1E9.7020906@ng23.net> > > It's always worked fine for me through gnome-terminal. > > What are you using? > an ssh session From mdehaan at redhat.com Mon Feb 25 18:01:26 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 25 Feb 2008 13:01:26 -0500 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C2F1E9.7020906@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> Message-ID: <47C30276.2080604@redhat.com> Tom Brown wrote: > >> >> It's always worked fine for me through gnome-terminal. >> >> What are you using? >> > > an ssh session > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools I meant for a terminal :) You said you were using Leopard, so I'll assume you're using the stock terminal and we can blame that. Otherwise possibly try iTerm. From tom at ng23.net Mon Feb 25 19:02:12 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 19:02:12 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C30276.2080604@redhat.com> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> <47C30276.2080604@redhat.com> Message-ID: <47C310B4.8080700@ng23.net> > > I meant for a terminal :) > > You said you were using Leopard, so I'll assume you're using the stock > terminal and > we can blame that. Otherwise possibly try iTerm. iTerm handles things much better i can see things perfectly now BUT - why is there always a but!! At random points during the install the virt machine crashes with output like the below, it does not happen at the same time everytime but the installer never gets past 50% - Any clues as to what i can do to avoid this crash? thanks Red Hat Enterprise Linux ES (C) 2004 Red Hat, Inc. +------------------+ Package Installation +------------------+ | | | Name : libtermcap-2.0.8-39-i386 | | Size : 16k | | Summary: A basic system library for accessing the | | termcap database. | | | | Status: Installing... | | | | 100% | | | | Packages Bytes Time | | Total : 310 731M 0:00:38 | | Completed: 35 266M 0:00:14 | | Remaining: 275 465M 0:00:24 | | | | 36% Oops: 0002 [#1] | | SMP +------------------------------------------------------------+ Modules linked in: dm_snapshot dm_mirror dm_zero dm_mod ext3 jbd msdos raid6 raid5 xor raid1 raid0 xenblk xennet sr_mod sd_mod scsi_mod cdrom loop nfs nfs_acl lockd sunrpc vfat fat cramfslements | selects | next screen CPU: 0 EIP: 0061:[] Not tainted VLI EFLAGS: 00010206 (2.6.9-55.ELxenU) EIP is at copy_page_range+0x485/0x520 eax: da9bd000 ebx: 1a95d045 ecx: c13537c0 edx: 80000000 esi: 6ec08000 edi: 0001a95d ebp: 00020800 esp: dd8aaec0 ds: 007b es: 007b ss: 0068 Process anaconda (pid: 385, threadinfo=dd8aa000 task=dde5d220) Stack: 1a95e067 00000000 6ec08000 da9bd000 dd8b4258 dfebe258 00000001 0b705000 09600000 d0686000 dfe7d000 dd8a9344 dbf5238c dfa62a14 dd8a9300 c0119ea8 dd8a9300 dfa5b040 dbf5238c 00002fb5 00000000 dbf523a8 dbf523b0 dbf52398 Call Trace: [] copy_mm+0x2d9/0x396 [] copy_process+0x6b5/0xb0b [] do_fork+0x8a/0x16b [] sigprocmask+0xb6/0xd0 [] sys_rt_sigprocmask+0x97/0x14c [] sys_fork+0x16/0x1a [] syscall_call+0x7/0xb Code: d8 25 bf fd ff ff 89 c3 89 f2 83 e3 df f0 ff 41 04 8b 74 24 40 ff 86 80 00 00 00 f6 41 10 01 74 06 ff 86 84 00 00 00 8b 74 24 08 <89> 56 04 89 1e f0 ff 41 08 81 44 24 20 00 10 00 00 8b 44 24 1c <0>Fatal exception: panic in 5 seconds Kernel panic - not syncing: Fatal exception From mnielsen at redhat.com Mon Feb 25 19:10:28 2008 From: mnielsen at redhat.com (Mark Nielsen) Date: Mon, 25 Feb 2008 14:10:28 -0500 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C310B4.8080700@ng23.net> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> <47C30276.2080604@redhat.com> <47C310B4.8080700@ng23.net> Message-ID: <1203966628.3346.14.camel@localhost.localdomain> On Mon, 2008-02-25 at 19:02 +0000, Tom Brown wrote: > > > > > Red Hat Enterprise Linux ES (C) 2004 Red Hat, > Inc. > > > +------------------+ Package Installation > +------------------+ > | > | > | Name : libtermcap-2.0.8-39-i386 Is this a 64 bit dom0 you are installing to? > > | > | Size : 16k > | > | Summary: A basic system library for accessing the > | > | termcap database. > | > | > | > | Status: Installing... > | > | > | > | 100% > | > | > | > | Packages Bytes Time > | > | Total : 310 731M 0:00:38 > | > | Completed: 35 266M 0:00:14 > | > | Remaining: 275 465M 0:00:24 > | > | > | > | 36% > Oops: 0002 [#1] > | | > SMP > +------------------------------------------------------------+ > Modules linked in: dm_snapshot dm_mirror dm_zero dm_mod ext3 jbd msdos > raid6 raid5 xor raid1 raid0 xenblk xennet sr_mod sd_mod scsi_mod cdrom > loop nfs nfs_acl lockd sunrpc vfat fat cramfslements | > selects | next screen > CPU: 0 > EIP: 0061:[] Not tainted VLI > EFLAGS: 00010206 (2.6.9-55.ELxenU) > EIP is at copy_page_range+0x485/0x520 > eax: da9bd000 ebx: 1a95d045 ecx: c13537c0 edx: 80000000 > esi: 6ec08000 edi: 0001a95d ebp: 00020800 esp: dd8aaec0 > ds: 007b es: 007b ss: 0068 > Process anaconda (pid: 385, threadinfo=dd8aa000 task=dde5d220) > Stack: 1a95e067 00000000 6ec08000 da9bd000 dd8b4258 dfebe258 00000001 > 0b705000 > 09600000 d0686000 dfe7d000 dd8a9344 dbf5238c dfa62a14 dd8a9300 > c0119ea8 > dd8a9300 dfa5b040 dbf5238c 00002fb5 00000000 dbf523a8 dbf523b0 > dbf52398 > Call Trace: > [] copy_mm+0x2d9/0x396 > [] copy_process+0x6b5/0xb0b > [] do_fork+0x8a/0x16b > [] sigprocmask+0xb6/0xd0 > [] sys_rt_sigprocmask+0x97/0x14c > [] sys_fork+0x16/0x1a > [] syscall_call+0x7/0xb > Code: d8 25 bf fd ff ff 89 c3 89 f2 83 e3 df f0 ff 41 04 8b 74 24 40 ff > 86 80 00 00 00 f6 41 10 01 74 06 ff 86 84 00 00 00 8b 74 24 08 <89> 56 > 04 89 1e f0 ff 41 08 81 44 24 20 00 10 00 00 8b 44 24 1c > <0>Fatal exception: panic in 5 seconds > Kernel panic - not syncing: Fatal exception > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From tom at ng23.net Mon Feb 25 19:16:53 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 19:16:53 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <1203966628.3346.14.camel@localhost.localdomain> References: <47C2B910.1020904@ng23.net> <2a7fce340802250505s59add146idf79a88cbe9397fa@mail.gmail.com> <47C2C1D6.3070706@ng23.net> <20080225132919.GD2614@redhat.com> <47C2C67D.6030402@ng23.net> <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> <47C30276.2080604@redhat.com> <47C310B4.8080700@ng23.net> <1203966628.3346.14.camel@localhost.localdomain> Message-ID: <47C31425.9010200@ng23.net> >> Red Hat Enterprise Linux ES (C) 2004 Red Hat, >> Inc. >> >> >> +------------------+ Package Installation >> +------------------+ >> | >> | >> | Name : libtermcap-2.0.8-39-i386 >> > > > Is this a 64 bit dom0 you are installing to? > > > yeap - RHEL 5.1 x86_64 is the host and the guest is RHEL4 u5 i386 From berrange at redhat.com Mon Feb 25 19:19:07 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 25 Feb 2008 19:19:07 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C31425.9010200@ng23.net> References: <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> <47C30276.2080604@redhat.com> <47C310B4.8080700@ng23.net> <1203966628.3346.14.camel@localhost.localdomain> <47C31425.9010200@ng23.net> Message-ID: <20080225191907.GK2614@redhat.com> On Mon, Feb 25, 2008 at 07:16:53PM +0000, Tom Brown wrote: > > >>Red Hat Enterprise Linux ES (C) 2004 Red Hat, > >>Inc. > >> > >> > >> +------------------+ Package Installation > >>+------------------+ > >> | > >>| > >> | Name : libtermcap-2.0.8-39-i386 > >> > > > > > >Is this a 64 bit dom0 you are installing to? > > > > yeap - RHEL 5.1 x86_64 is the host and the guest is RHEL4 u5 i386 That is not a supported combination for RHEL-5.1. Running a 32-bit DomU on a 64-bit Dom0 is known to randomly crash. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=| From mnielsen at redhat.com Mon Feb 25 19:21:27 2008 From: mnielsen at redhat.com (Mark Nielsen) Date: Mon, 25 Feb 2008 14:21:27 -0500 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <20080225191907.GK2614@redhat.com> References: <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> <47C30276.2080604@redhat.com> <47C310B4.8080700@ng23.net> <1203966628.3346.14.camel@localhost.localdomain> <47C31425.9010200@ng23.net> <20080225191907.GK2614@redhat.com> Message-ID: <1203967287.3346.16.camel@localhost.localdomain> On Mon, 2008-02-25 at 19:19 +0000, Daniel P. Berrange wrote: > On Mon, Feb 25, 2008 at 07:16:53PM +0000, Tom Brown wrote: > > > > >>Red Hat Enterprise Linux ES (C) 2004 Red Hat, > > >>Inc. > > >> > > >> > > >> +------------------+ Package Installation > > >>+------------------+ > > >> | > > >>| > > >> | Name : libtermcap-2.0.8-39-i386 > > >> > > > > > > > > >Is this a 64 bit dom0 you are installing to? > > > > > > > yeap - RHEL 5.1 x86_64 is the host and the guest is RHEL4 u5 i386 > > That is not a supported combination for RHEL-5.1. Running a 32-bit DomU > on a 64-bit Dom0 is known to randomly crash. > Dan. http://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/Virtualization-en-US/ch-virt-hw-support.html From tom at ng23.net Mon Feb 25 21:06:18 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 21:06:18 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <1203967287.3346.16.camel@localhost.localdomain> References: <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> <47C30276.2080604@redhat.com> <47C310B4.8080700@ng23.net> <1203966628.3346.14.camel@localhost.localdomain> <47C31425.9010200@ng23.net> <20080225191907.GK2614@redhat.com> <1203967287.3346.16.camel@localhost.localdomain> Message-ID: <47C32DCA.7010601@ng23.net> > http://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/Virtualization-en-US/ch-virt-hw-support.html > > thanks for that - changing the dom0 to 32bit and the install completes which is great! only issue now is that during the reboot after installation the domU gets destroyed, so i cant actually use them yet! [2008-02-25 19:44:50 xend 3109] INFO (XendDomain:370) Domain virtbox (1) unpaused. [2008-02-25 19:50:00 xend.XendDomainInfo 3109] INFO (XendDomainInfo:947) Domain has shutdown: name=virtbox id=1 reason=reboot. [2008-02-25 19:50:00 xend.XendDomainInfo 3109] DEBUG (XendDomainInfo:1557) XendDomainInfo.destroy: domid=1 [2008-02-25 19:50:00 xend.XendDomainInfo 3109] DEBUG (XendDomainInfo:1566) XendDomainInfo.destroyDomain(1) this is in the domU config - generated with koan on_poweroff = "destroy" on_reboot = "restart" on_crash = "restart" any ideas why this happens? From mdehaan at redhat.com Mon Feb 25 21:10:14 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 25 Feb 2008 16:10:14 -0500 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C32DCA.7010601@ng23.net> References: <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> <47C30276.2080604@redhat.com> <47C310B4.8080700@ng23.net> <1203966628.3346.14.camel@localhost.localdomain> <47C31425.9010200@ng23.net> <20080225191907.GK2614@redhat.com> <1203967287.3346.16.camel@localhost.localdomain> <47C32DCA.7010601@ng23.net> Message-ID: <47C32EB6.8000706@redhat.com> > > this is in the domU config - generated with koan > > on_poweroff = "destroy" > on_reboot = "restart" > on_crash = "restart" > > any ideas why this happens? > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Destroy just means "yank power and make the box go away" in Xen terms. Yes, it's badly named :) Don't worry, you can get it back! Do a "virsh status --all" and you should see your guest and be able to start it with virsh. From tom at ng23.net Mon Feb 25 21:14:22 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 21:14:22 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C32EB6.8000706@redhat.com> References: <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> <47C30276.2080604@redhat.com> <47C310B4.8080700@ng23.net> <1203966628.3346.14.camel@localhost.localdomain> <47C31425.9010200@ng23.net> <20080225191907.GK2614@redhat.com> <1203967287.3346.16.camel@localhost.localdomain> <47C32DCA.7010601@ng23.net> <47C32EB6.8000706@redhat.com> Message-ID: <47C32FAE.2030003@ng23.net> Michael DeHaan wrote: > > Destroy just means "yank power and make the box go away" in Xen > terms. Yes, it's badly named :) > > Don't worry, you can get it back! > > Do a "virsh status --all" and you should see your guest and be able to > start it with virsh. > # virsh status --all error: unknown command: 'status' this is on RHEL 5.1 From mdehaan at redhat.com Mon Feb 25 21:19:54 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 25 Feb 2008 16:19:54 -0500 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C32FAE.2030003@ng23.net> References: <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> <47C30276.2080604@redhat.com> <47C310B4.8080700@ng23.net> <1203966628.3346.14.camel@localhost.localdomain> <47C31425.9010200@ng23.net> <20080225191907.GK2614@redhat.com> <1203967287.3346.16.camel@localhost.localdomain> <47C32DCA.7010601@ng23.net> <47C32EB6.8000706@redhat.com> <47C32FAE.2030003@ng23.net> Message-ID: <47C330FA.5020001@redhat.com> Tom Brown wrote: > Michael DeHaan wrote: >> >> Destroy just means "yank power and make the box go away" in Xen >> terms. Yes, it's badly named :) >> >> Don't worry, you can get it back! >> >> Do a "virsh status --all" and you should see your guest and be able >> to start it with virsh. >> > > # virsh status --all > error: unknown command: 'status' > > this is on RHEL 5.1 > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools err, "virsh list --all" http://fedoraproject.org/wiki/Docs/Fedora8VirtQuickStart#head-d507982a8694217b2721b86797a04e127e13c6f6 From tom at ng23.net Mon Feb 25 21:50:44 2008 From: tom at ng23.net (Tom Brown) Date: Mon, 25 Feb 2008 21:50:44 +0000 Subject: [et-mgmt-tools] cobbler-0.6.5-3.el4 - xen virt trouble In-Reply-To: <47C330FA.5020001@redhat.com> References: <20080225135159.GF2614@redhat.com> <47C2C8D9.6020608@ng23.net> <47C2E5D4.9040505@redhat.com> <47C2EBCC.10508@ng23.net> <47C2EE96.5070702@redhat.com> <47C2F1E9.7020906@ng23.net> <47C30276.2080604@redhat.com> <47C310B4.8080700@ng23.net> <1203966628.3346.14.camel@localhost.localdomain> <47C31425.9010200@ng23.net> <20080225191907.GK2614@redhat.com> <1203967287.3346.16.camel@localhost.localdomain> <47C32DCA.7010601@ng23.net> <47C32EB6.8000706@redhat.com> <47C32FAE.2030003@ng23.net> <47C330FA.5020001@redhat.com> Message-ID: <47C33834.5020803@ng23.net> > > err, "virsh list --all" > > http://fedoraproject.org/wiki/Docs/Fedora8VirtQuickStart#head-d507982a8694217b2721b86797a04e127e13c6f6 > > awsome thanks - working, finally!! From duffy at redhat.com Mon Feb 25 21:56:20 2008 From: duffy at redhat.com (=?ISO-8859-1?Q?M=E1ir=EDn_Duffy?=) Date: Mon, 25 Feb 2008 16:56:20 -0500 Subject: [et-mgmt-tools] Software Content Management (Introducing Pulp) Message-ID: <47C33984.3020104@redhat.com> Hi folks, As some of you who have participated in them already know, over the past couple of years or so Red Hat has been conducting some studies on how folks manage their systems using the Red Hat Network and Satellite products. We've learned a lot about the processes many of you have established for managing your systems and the strengths and weaknesses of the RHN products in supporting those processes. In addition to this, it is also clear that the free and open source management tools available for Fedora, RHEL, and CentOS (as well as other *nixes) don't sufficiently cover some of the areas of need that the current Satellite product addresses. Over the past few weeks some Red Hat folks and Fedora community members have been working on a free and open source project that will not only attempt to fill one of the gaps in free & open source systems management tools, but also to take some of the things we've learned from talking with Satellite and RHN customers and improve upon how we could address one area of systems management. This project is called 'Pulp', and its scope is centered around the management of software content. From the Pulp Fedora project website [1]: "Pulp is an application for managing the software installed on your systems. Suppose you want to control what machines on your network get what software updates, to establish testing/stage repositories, to mirror 3rd party content, to create your own repositories, or to add new content to existing repositories. Pulp will provide an easy web, web-services, and command line interface for managing all of this." REPOSITORY CREATION AND MIRRORING MANAGEMENT To start, we were thinking that Pulp could be a way of improving upon the custom channel management capabilities of Satellite, using yum repositories instead of RHN channels. Last month Michael DeHaan hosted a discussion introducing Pulp at FUDCon in Raleigh [2]. What we have taken away from the participants of that session is that they would like to see less emphasis on system <=> content mapping, and would like tools that focus on mirroring contents from many different 'upstream' sources and organizing them neatly in one place. It could be kind of like mirror manager, except instead of managing the mirroring of a particular set of content across many sites, it would manage the mirroring of MANY different sets of content at ONE particular site. On top of this it could greatly simplify the creation and management of yum repositories from this mirrored content as well as from other local content sources. (This today has some annoying manual process involved.) CONTENT INVENTORY, ACCESS CONTROL, AND DELIVERY We have also thought about Pulp as a way of managing which content gets to which systems and maintaining an inventory of which content is which systems. For example, maybe using Pulp to get a list of which systems are allowed to connect to which repositories, and maybe on a more granular level, using Pulp to store black or whitelists of packages that the system is allowed to access. Or maybe using it to create a system whereby using some logical/policy statements you can create virtual yum repositories that compose content from many sources in a particular way and then contrl access to those. The group at FUDCon seemed to care less about content access control and delivery, seeming to prefer letting their configuration management systems (eg cfengine) handle content access and delivery to systems and having Pulp stop at providing yum repos for these configuration management tools to access. I do think, from talking with several different types of Satellite and RHN users, that some folks may still be interested in content access control, but at this point it seems that repository creation and mirroring management is one area that both groups of people would find great value in. DISCUSSION Many of the folks subscribed to these lists are seasoned Linux system engineers, system administrators, and/or release engineers for software content, so we would love to hear some of your thoughts on what problems areas you'd like to see addressed by free and open source management tools like Pulp. If you have any thoughts on the following topics or others that are related but maybe not mentioned here, please let's discuss them here and see if we figure out the best way to make Pulp useful for you!: - Do you host internal mirrors of external content? What kind of content? How many mirrors? Do you have mirrors available for multiple geographic locations within your organization? - How many different 'upstream' sources of content need to be made available for systems at your organization? Hardware drivers from hardware vendors? Operating systems from OS vendors or from FOSS repos? Non-FOSS proprietary applications from application vendors? In-house application/software development teams? - How often do you pull down content ('sync' maybe could be a term) from these different upstream content sources? - How do you organize all of the software content that is delivered to your systems right now? What are the strengths you've found to your approach today? What are the weaknesses you'd like to address? - How much customization/general 'mucking' do you do with the content you pull down from various sources? Are you more interested in simply making all the content available or do you have requirements for modifying/customizing it as well? - If you do customize the content, to what extent do you need to do this? Branding? Localization? Etc.? - How strict are your policies for which systems have access to which kind of content? Is access completely open, is access constrained by which system owners have purchased licenses/entitlements to which content? Is access constrained by security concerns? Is access constrained by stability concerns (e.g., production systems must never be able to have development level content deployed to them?) - What kind of requirements do you have for producing data about which systems had which content installed when, if any? - How many different environments do you manage content for? Do you manage content for development / qa / production environments? - How do you prefer to deploy content to systems? Do you prefer to have a software management tool to do that or do you prefer to tie this into a configuration management tool? - At what level of granularity do you perform software-management related tasks on your systems? For example, do you find yourself most often: - automatically selecting and deploying content to many systems at once in a uniform fashion - automatically selecting and deploying content to smaller groupings of systems with carefully defined templates - manually selecting and deploying content to many systems at once - manually selecting and deploying content to individual systems one-by-one What level of importance does each of these abilities have to you? SHAMELESS PLUG Pulp is an open project, stop by the mailing list (cc'ed :) ) to say hi! Feedback, bug reports, ideas, and patches are always welcome. :) Thanks, ~m and the Pulp Team :) [1] https://fedorahosted.org/pulp [2] Notes available here: https://fedorahosted.org/pulp/wiki/FudConOhEightNotes From adrian.revill at shazamteam.com Tue Feb 26 09:49:51 2008 From: adrian.revill at shazamteam.com (Adrian Revill) Date: Tue, 26 Feb 2008 09:49:51 +0000 Subject: [et-mgmt-tools] Error during cobbler reposync Message-ID: <47C3E0BF.5060303@shazamteam.com> Hi, I am just starting with cobbler and i have setup a mirror of rhn repository rhn://rhel-x86_64-server-5 and i am often getting the error Traceback (most recent call last): File "/usr/bin/reposync", line 196, in ? main() File "/usr/bin/reposync", line 190, in main path = repo.getPackage(pkg) File "/usr/lib/python2.4/site-packages/yum/yumRepo.py", line 447, in getPackage cache=cache File "/usr/lib/yum-plugins/rhnplugin.py", line 209, in _getFile start, end, copy_local, checkfunc, text, reget, cache) File "/usr/lib/yum-plugins/rhnplugin.py", line 298, in _noExceptionWrappingGet return result UnboundLocalError: local variable 'result' referenced before assignment cobbler reposync failed If i re-run cobbler reposync it continues without error. Any ideas? regards adrian From mdehaan at redhat.com Tue Feb 26 15:57:17 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 26 Feb 2008 10:57:17 -0500 Subject: [et-mgmt-tools] Error during cobbler reposync In-Reply-To: <47C3E0BF.5060303@shazamteam.com> References: <47C3E0BF.5060303@shazamteam.com> Message-ID: <47C436DD.8040404@redhat.com> Adrian Revill wrote: > Hi, > I am just starting with cobbler and i have setup a mirror of rhn > repository rhn://rhel-x86_64-server-5 and i am often getting the error The yum_rhn_plugin stuff was kind of tempramental in older versions of yum, though folks have found it to be working fine in later versions. Do you have the most recent yum and yum-utils? (at least 1.04 or later for yum-utils?) I haven't seen this exact error before, so versions of those packages (if already updated) might be useful ... and I can get this info to someone that can address this. --Michael > > Traceback (most recent call last): > File "/usr/bin/reposync", line 196, in ? > main() > File "/usr/bin/reposync", line 190, in main > path = repo.getPackage(pkg) > File "/usr/lib/python2.4/site-packages/yum/yumRepo.py", line 447, in > getPackage > cache=cache > File "/usr/lib/yum-plugins/rhnplugin.py", line 209, in _getFile > start, end, copy_local, checkfunc, text, reget, cache) > File "/usr/lib/yum-plugins/rhnplugin.py", line 298, in > _noExceptionWrappingGet > return result > UnboundLocalError: local variable 'result' referenced before assignment > cobbler reposync failed > > If i re-run cobbler reposync it continues without error. Any ideas? > > regards > > adrian > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mblack at platform.com Tue Feb 26 22:12:45 2008 From: mblack at platform.com (Mark Black) Date: Tue, 26 Feb 2008 17:12:45 -0500 Subject: [et-mgmt-tools] Handling of non-linux partitions by Vert-P2V Message-ID: <0B6E8E0E306CD044B4CE0552D5DC9D6C01DD0C1A@catoexm06.noam.corp.platform.com> Hi Does virt-p2v handle non-linux partitions like ntfs? If so does it treat it as a raw partition and compress and sent the entire partition, or does it mount it and extract out the files? Mark -------------- next part -------------- An HTML attachment was scrubbed... URL: From adrian.revill at shazamteam.com Wed Feb 27 10:28:30 2008 From: adrian.revill at shazamteam.com (Adrian Revill) Date: Wed, 27 Feb 2008 10:28:30 +0000 Subject: [et-mgmt-tools] Error during cobbler reposync Message-ID: <47C53B4E.4000804@shazamteam.com> Hi Michael, The yum packages are; Name : yum Arch : noarch Version: 3.0.1 Release: 5.el5 Name : yum-utils Arch : noarch Version: 1.0.4 Release: 3.el5 I am running on rhel5u1 regards Adrian Michael DeHaan wrote: The yum_rhn_plugin stuff was kind of tempramental in older versions of yum, though folks have found it to be working fine in later versions. Do you have the most recent yum and yum-utils? (at least 1.04 or later for yum-utils?) I haven't seen this exact error before, so versions of those packages (if already updated) might be useful ... and I can get this info to someone that can address this. From rjones at redhat.com Wed Feb 27 10:56:05 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Wed, 27 Feb 2008 10:56:05 +0000 Subject: [et-mgmt-tools] Handling of non-linux partitions by Vert-P2V In-Reply-To: <0B6E8E0E306CD044B4CE0552D5DC9D6C01DD0C1A@catoexm06.noam.corp.platform.com> References: <0B6E8E0E306CD044B4CE0552D5DC9D6C01DD0C1A@catoexm06.noam.corp.platform.com> Message-ID: <20080227105605.GA32458@amd.home.annexia.org> On Tue, Feb 26, 2008 at 05:12:45PM -0500, Mark Black wrote: > Does virt-p2v handle non-linux partitions like ntfs? If so does it > treat it as a raw partition and compress and sent the entire > partition, or does it mount it and extract out the files? At the moment it just treats them as opaque partitions. If they are on a disk which you have selected for migration, then they are migrated along with the rest. The plan is to look into them more deeply and rewrite key files / install PV drivers as necessary, but Windows / NTFS isn't high on my list of priorities. Patches are welcome of course. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://et.redhat.com/~rjones/virt-top From adrian.revill at shazamteam.com Wed Feb 27 14:49:32 2008 From: adrian.revill at shazamteam.com (Adrian Revill) Date: Wed, 27 Feb 2008 14:49:32 +0000 Subject: [et-mgmt-tools] virt-path modification sugestion Message-ID: <47C5787C.9060004@shazamteam.com> I was trying to to do an install onto an existing logical volume, VolVirt/el5test. I have not found any documentation on the use of the logical volumes, so I set virt-path to VolVirt/el5test But i got an error saying "The volume group [VolVirt/el5test] does not exist." I have looked at the koan code app.py, calc_virt_path2. It first checks to see if the volume group exists. It runs /usr/sbin/vgs -o vg_name which correctly lists the volume groups. VG VolGroup00 VolVirt But It then searches the list of volume groups looking for VolVirt/el5test, which of course fails, it should be spliting the location string and only looking for the volume group name. Reading on, It then checks the size of the volume group by running /usr/sbin/vgs --noheadings -o vg_free --units g VolVirt/el5test Which also fails as it is expecting just the volume group name "VolVirt" After this it starts to check the logical volume. It correctly runs lvs -o lv_name VolVirt/el5test LV el5test It now defines the logical volume name as Wed_Feb_27_14_19_24_2008-disk0 (The date is inserted as i have not defined a system name.) Now the logical volume is created with the command /usr/sbin/lvcreate -L 5G -n Wed_Feb_27_14_19_24_2008-disk0 VolVirt/el5test Which fails. ----- So from this i can see that when using logical volumes the volume names are defined by setting the virt-path to the volume group name and the logical volume name will be sytstemName-disk0 If the calc_virt_path2 function was modified slightly, to first split the location on / and using the first part as the volume group name, and then if the second part is not null set the logical volume name to the second part, else use the default name. It would be possible to also explicitly specify the logical volume in the virt-path using the form volumeGroup/logicalVolume I would submit a patch to do it, but im not a python programmer. regards Adrian Revill From tomg at mcclatchyinteractive.com Wed Feb 27 15:06:35 2008 From: tomg at mcclatchyinteractive.com (Tom Georgoulias) Date: Wed, 27 Feb 2008 10:06:35 -0500 Subject: [et-mgmt-tools] mirroring 32 bit and 64 bit RHN channels on a single cobbler server Message-ID: <47C57C7B.9000408@mcclatchyinteractive.com> I'm currently taking advantage of a RHN mirror for the rhel-i386-server-5 channel to deploy systems that are fully updated with errata upon first boot. The cobbler server is 32 bit and subscribed to the rhel-i386-server-5 channel. Is there any way I can mirror the 64 bit channel (rhel-x86_64-server-5) onto my cobbler server so that my 64 bit servers get the same treatment? I have entitlements for all of my servers at RHN so I'm not trying to work around any agreements or restrictions.... Thanks, Tom From mdehaan at redhat.com Wed Feb 27 15:09:22 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 27 Feb 2008 10:09:22 -0500 Subject: [et-mgmt-tools] virt-path modification sugestion In-Reply-To: <47C5787C.9060004@shazamteam.com> References: <47C5787C.9060004@shazamteam.com> Message-ID: <47C57D22.4020908@redhat.com> > > ----- > > So from this i can see that when using logical volumes the volume > names are defined by setting the virt-path to the volume group name > and the logical volume name will be sytstemName-disk0 > > If the calc_virt_path2 function was modified slightly, to first split > the location on / and using the first part as the volume group name, > and then if the second part is not null set the logical volume name to > the second part, else use the default name. > It would be possible to also explicitly specify the logical volume in > the virt-path using the form volumeGroup/logicalVolume > > I would submit a patch to do it, but im not a python programmer. Python's easy :) Anyhow, if --virt-path is an LVM volume, the idea with systemName-diskN nomenclature is that you can specify that a virtual machine gets multiple disks cobbler profile edit --name=foo --virt-size=5,6,7 In this example, it would get 3 disks -- 5GB, 6GB, and 7GB The disks would be named as follows: virtName-disk0 virtName-disk1 virtName-disk2 Essentially the thought is that allowing too much variance in naming here is going to cause some confusion, and using the virt name and the disk numbers ensure things don't stomp over each other. If you want to specify a specific LVM partition, you can still pre-create it and then specify it with it's full "/dev" path. Make sense? I believe there are a several people using the LVM/virt stuff as above so I'm not sure I'd want to change it and possibly break them. --Michael From mdehaan at redhat.com Wed Feb 27 15:26:17 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 27 Feb 2008 10:26:17 -0500 Subject: [et-mgmt-tools] mirroring 32 bit and 64 bit RHN channels on a single cobbler server In-Reply-To: <47C57C7B.9000408@mcclatchyinteractive.com> References: <47C57C7B.9000408@mcclatchyinteractive.com> Message-ID: <47C58119.8010307@redhat.com> Tom Georgoulias wrote: > I'm currently taking advantage of a RHN mirror for the > rhel-i386-server-5 channel to deploy systems that are fully updated > with errata upon first boot. The cobbler server is 32 bit and > subscribed to the rhel-i386-server-5 channel. Is there any way I can > mirror the 64 bit channel (rhel-x86_64-server-5) onto my cobbler > server so that my 64 bit servers get the same treatment? I've been told that apparently only works for the arch you're subscribed to. From tomg at mcclatchyinteractive.com Wed Feb 27 15:29:32 2008 From: tomg at mcclatchyinteractive.com (Tom Georgoulias) Date: Wed, 27 Feb 2008 10:29:32 -0500 Subject: [et-mgmt-tools] mirroring 32 bit and 64 bit RHN channels on a single cobbler server In-Reply-To: <47C58119.8010307@redhat.com> References: <47C57C7B.9000408@mcclatchyinteractive.com> <47C58119.8010307@redhat.com> Message-ID: <47C581DC.3030209@mcclatchyinteractive.com> Michael DeHaan wrote: > Tom Georgoulias wrote: >> Is there any way I can >> mirror the 64 bit channel (rhel-x86_64-server-5) onto my cobbler >> server so that my 64 bit servers get the same treatment? > > I've been told that apparently only works for the arch you're subscribed > to. Yup, that's absolutely true. Can't find a way to get around that, either. Oh well, thought I'd ask anyway. Tom From adrian.revill at shazamteam.com Wed Feb 27 15:33:05 2008 From: adrian.revill at shazamteam.com (Adrian Revill) Date: Wed, 27 Feb 2008 15:33:05 +0000 Subject: [et-mgmt-tools] virt-path modification sugestion Message-ID: <47C582B1.2060504@shazamteam.com> Ah yes very good point, of course i can use the /dev/volGroup/logicalVol Dohh blindingly obvious. Thx for the quick reply, BTW cobbler is damm excellent, im very impressed. So far ;) Michael DeHaan wrote: Python's easy :) Anyhow, if --virt-path is an LVM volume, the idea with systemName-diskN nomenclature is that you can specify that a virtual machine gets multiple disks cobbler profile edit --name=foo --virt-size=5,6,7 In this example, it would get 3 disks -- 5GB, 6GB, and 7GB The disks would be named as follows: virtName-disk0 virtName-disk1 virtName-disk2 Essentially the thought is that allowing too much variance in naming here is going to cause some confusion, and using the virt name and the disk numbers ensure things don't stomp over each other. If you want to specify a specific LVM partition, you can still pre-create it and then specify it with it's full "/dev" path. Make sense? I believe there are a several people using the LVM/virt stuff as above so I'm not sure I'd want to change it and possibly break them. From mdehaan at redhat.com Wed Feb 27 16:15:53 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 27 Feb 2008 11:15:53 -0500 Subject: [et-mgmt-tools] Error during cobbler reposync In-Reply-To: <47C53B4E.4000804@shazamteam.com> References: <47C53B4E.4000804@shazamteam.com> Message-ID: <47C58CB9.9030607@redhat.com> Adrian Revill wrote: > Hi Michael, > > The yum packages are; > Name : yum > Arch : noarch > Version: 3.0.1 > Release: 5.el5 > > Name : yum-utils > Arch : noarch > Version: 1.0.4 > Release: 3.el5 Per #yum, you want to have the following: yum-3.2.8-7.el5 and yum-utils-1.1.10-6.el5 From mdehaan at redhat.com Wed Feb 27 16:32:07 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 27 Feb 2008 11:32:07 -0500 Subject: [et-mgmt-tools] Error during cobbler reposync In-Reply-To: <47C58CB9.9030607@redhat.com> References: <47C53B4E.4000804@shazamteam.com> <47C58CB9.9030607@redhat.com> Message-ID: <47C59087.8010308@redhat.com> Michael DeHaan wrote: > Adrian Revill wrote: >> Hi Michael, >> >> The yum packages are; >> Name : yum >> Arch : noarch >> Version: 3.0.1 >> Release: 5.el5 >> >> Name : yum-utils >> Arch : noarch >> Version: 1.0.4 >> Release: 3.el5 > > Per #yum, you want to have the following: > > yum-3.2.8-7.el5 and yum-utils-1.1.10-6.el5 > Apparently fixed in RHEL 5.2 https://bugzilla.redhat.com/show_bug.cgi?id=378911 From steve.pearson at prodeasystems.com Wed Feb 27 17:57:01 2008 From: steve.pearson at prodeasystems.com (Steve Pearson) Date: Wed, 27 Feb 2008 11:57:01 -0600 Subject: [et-mgmt-tools] Cobbler-PXE-Kickstart Message-ID: <56DA5079467D1C48B857C6FE59D5D4FCF3E5AE@prodeaserve.prodea.local> Hi, I have some very basic questions on this relationship. I have read through the Cobbler man page and some of the other available information. I plan to setup a RHEL5 Cobbler server to do a PXE based network installation of several servers. Here are my questions: 1) What network protocol is used to do the actual network OS installation? This detail seems to be missing from the docs but from what I have read it appears to be http...I am wanting to know how the PXE client accesses the files on the cobbler server. 2) I am going to Import a loop mounted DVD iso for my RHEL5 distro (so I do not have to expand the install tree). I plan to use Kickstart Configurator to generate unique kickstarts...I may use ks templating later. In the ks file, what do I put for "Installation Method" (CD, NFS, HTTP, Hard-drive)? If I choose HTTP, what path do I use for the "HTTP Directory"? It is supposed to be the location of the "Server" directory from the distro but since I used iso's I do not have one. Will Cobbler explode the iso and create an Install Tree in /var/www/cobbler? If so I can specify that dir... Sorry, I know I could do some experimenting but do not have the setup for this yet and am in planning stage. Understanding some of these basics will help me out greatly. Thanks! Steve Pearson System Engineering Prodea Systems 214.278.1805 www.prodeasystems.com This message is confidential to Prodea Systems, Inc unless otherwise indicated or apparent from its nature. This message is directed to the intended recipient only, who may be readily determined by the sender of this message and its contents. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient:(a)any dissemination or copying of this message is strictly prohibited; and(b)immediately notify the sender by return message and destroy any copies of this message in any form(electronic, paper or otherwise) that you have.The delivery of this message and its information is neither intended to be nor constitutes a disclosure or waiver of any trade secrets, intellectual property, attorney work product, or attorney-client communications. The authority of the individual sending this message to legally bind Prodea Systems is neither apparent nor implied,and must be independently verified. Steve Pearson System Engineering Prodea Systems 214.278.1805 www.prodeasystems.com This message is confidential to Prodea Systems, Inc unless otherwise indicated or apparent from its nature. This message is directed to the intended recipient only, who may be readily determined by the sender of this message and its contents. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient:(a)any dissemination or copying of this message is strictly prohibited; and(b)immediately notify the sender by return message and destroy any copies of this message in any form(electronic, paper or otherwise) that you have.The delivery of this message and its information is neither intended to be nor constitutes a disclosure or waiver of any trade secrets, intellectual property, attorney work product, or attorney-client communications. The authority of the individual sending this message to legally bind Prodea Systems is neither apparent nor implied,and must be independently verified. -------------- next part -------------- An HTML attachment was scrubbed... URL: From mdehaan at redhat.com Wed Feb 27 18:02:22 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 27 Feb 2008 13:02:22 -0500 Subject: [et-mgmt-tools] Cobbler-PXE-Kickstart In-Reply-To: <56DA5079467D1C48B857C6FE59D5D4FCF3E5AE@prodeaserve.prodea.local> References: <56DA5079467D1C48B857C6FE59D5D4FCF3E5AE@prodeaserve.prodea.local> Message-ID: <47C5A5AE.4060101@redhat.com> Steve Pearson wrote: > > Hi, > > I have some very basic questions on this relationship. I have read > through the Cobbler man page and some of the other available > information. I plan to setup a RHEL5 Cobbler server to do a PXE based > network installation of several servers. Here are my questions: > > 1) What network protocol is used to do the actual network OS > installation? This detail seems to be missing from the docs but from > what I have read it appears to be http?I am wanting to know how the > PXE client accesses the files on the cobbler server. > First there's DHCP PXE uses TFTP Kickstarts are served over http:// if using Cobbler templating, but also can be ftp:// or nfs:// The install by default, if using Cobbler import is http://, which most people use, but can also be ftp:// or nfs:// > 2) I am going to Import a loop mounted DVD iso for my RHEL5 distro (so > I do not have to expand the install tree). I plan to use Kickstart > Configurator to generate unique kickstarts?I may use ks templating > later. In the ks file, what do I put for ?Installation Method? (CD, > NFS, HTTP, Hard-drive)? If I choose HTTP, what path do I use for the > ?HTTP Directory?? It is supposed to be the location of the ?Server? > directory from the distro but since I used iso?s I do not have one. > Will Cobbler explode the iso and create an Install Tree in > /var/www/cobbler? If so I can specify that dir? > The easiest way to start out is to just import the mounted DVD cobbler import --name=foo --mirror=/mnt/dvdlocation That will create cobbler distros and profiles, as well as assign a base kickstart (which lives in /etc/cobbler) If you want to use templating you can, or you can use an existing kickstart you have available in your web space cobbler profile edit --name=foo --kickstart=http://bootserver.example.org/foo.ks > Sorry, I know I could do some experimenting but do not have the setup > for this yet and am in planning stage. Understanding some of these > basics will help me out greatly. > No problem... You may also be interested in #cobbler on irc.freenode.net which is also good for quick questions like this. > Thanks! > > *Steve Pearson* > > System Engineering > > Prodea Systems > > 214.278.1805 > www.prodeasystems.com > > > > This message is confidential to Prodea Systems, Inc unless otherwise indicated > or apparent from its nature. This message is directed to the intended recipient > only, who may be readily determined by the sender of this message and its > contents. If the reader of this message is not the intended recipient, or an > employee or agent responsible for delivering this message to the intended > recipient:(a)any dissemination or copying of this message is strictly > prohibited; and(b)immediately notify the sender by return message and destroy > any copies of this message in any form(electronic, paper or otherwise) that you > have.The delivery of this message and its information is neither intended to be > nor constitutes a disclosure or waiver of any trade secrets, intellectual > property, attorney work product, or attorney-client communications. The > authority of the individual sending this message to legally bind Prodea Systems > is neither apparent nor implied,and must be independently verified. > > *Steve Pearson* > > System Engineering > > Prodea Systems > > 214.278.1805 > www.prodeasystems.com > > This message is confidential to Prodea Systems, Inc unless otherwise indicated > or apparent from its nature. This message is directed to the intended recipient > only, who may be readily determined by the sender of this message and its > contents. If the reader of this message is not the intended recipient, or an > employee or agent responsible for delivering this message to the intended > recipient:(a)any dissemination or copying of this message is strictly > prohibited; and(b)immediately notify the sender by return message and destroy > any copies of this message in any form(electronic, paper or otherwise) that you > have.The delivery of this message and its information is neither intended to be > nor constitutes a disclosure or waiver of any trade secrets, intellectual > property, attorney work product, or attorney-client communications. The > authority of the individual sending this message to legally bind Prodea Systems > is neither apparent nor implied,and must be independently verified. > ------------------------------------------------------------------------ > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mdehaan at redhat.com Wed Feb 27 21:08:40 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 27 Feb 2008 16:08:40 -0500 Subject: [et-mgmt-tools] Signups for FudCON Boston (it's free!) Message-ID: <47C5D158.9000302@redhat.com> You'll need a Fedora account to edit the page (this may be moved to barcamp.org later?), though... https://fedoraproject.org/wiki/FUDCon/FUDConF10 https://admin.fedoraproject.org/accounts The list of talks/subjects for FudCON is still rolling in, but both Cobbler and Func will be there. This FudCON is also at the same time (and the day after also) as Red Hat Summit, so that may be something you are interested in also. http://www.redhat.com/promo/summit/ ... Summit last year was a lot of fun and had some great talks and was also a great chance for developers and users alike to get together, meet, and talk shop. There should be a good round of talks on subjects of interest to et-mgmt-tools subscribers at both, so if you are interested, check them out! --Michael From mdehaan at redhat.com Wed Feb 27 22:26:17 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 27 Feb 2008 17:26:17 -0500 Subject: [et-mgmt-tools] Signups for FudCON Boston (it's free!) In-Reply-To: <47C5D158.9000302@redhat.com> References: <47C5D158.9000302@redhat.com> Message-ID: <47C5E389.8000409@redhat.com> Michael DeHaan wrote: > You'll need a Fedora account to edit the page (this may be moved to > barcamp.org later?), though... > > https://fedoraproject.org/wiki/FUDCon/FUDConF10 > https://admin.fedoraproject.org/accounts > > The list of talks/subjects for FudCON is still rolling in, but both > Cobbler and Func will be there. > > This FudCON is also at the same time (and the day after also) as Red > Hat Summit, so that may be > something you are interested in also. > http://www.redhat.com/promo/summit/ ... > > Summit last year was a lot of fun and had some great talks and was > also a great chance for developers and users > alike to get together, meet, and talk shop. > There should be a good round of talks on subjects of interest to > et-mgmt-tools subscribers at both, so > if you are interested, check them out! > > --Michael > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools One quick addition... (suggested on another list) If you don't have a Fedora account and want to go, email me off list and I can add you to the Wiki. That can save you some trouble. (Though Fedora accounts are great too!) --Michael From adrian.revill at shazamteam.com Thu Feb 28 09:07:19 2008 From: adrian.revill at shazamteam.com (Adrian Revill) Date: Thu, 28 Feb 2008 09:07:19 +0000 Subject: [et-mgmt-tools] Error during cobbler reposync Message-ID: <47C679C7.3010000@shazamteam.com> Hi Michael, Do you know if these rpms (srpms) are available from rhn? or should I wait for rhel5.2? Michael DeHaan wrote: Per #yum, you want to have the following: yum-3.2.8-7.el5 and yum-utils-1.1.10-6.el5 From Pablo.Iranzo at redhat.com Thu Feb 28 09:11:52 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Thu, 28 Feb 2008 10:11:52 +0100 Subject: [et-mgmt-tools] Error during cobbler reposync In-Reply-To: <47C679C7.3010000@shazamteam.com> References: <47C679C7.3010000@shazamteam.com> Message-ID: <1204189912.11610.21.camel@iranzo.users.redhat.com> I had to compile them from sources and upgrade yum, yum-utils, yum-updatesd and other packages. Regards Pablo El jue, 28-02-2008 a las 09:07 +0000, Adrian Revill escribi?: > Hi Michael, > > Do you know if these rpms (srpms) are available from rhn? or should I > wait for rhel5.2? > > Michael DeHaan wrote: > > Per #yum, you want to have the following: > > yum-3.2.8-7.el5 and yum-utils-1.1.10-6.el5 > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools -- Pablo Iranzo G?mez (Pablo.Iranzo at redhat.com) RHCE/Global Profesional Services Consultant Spain Phone: +34 645 01 01 49 (CET/CEST) GnuPG KeyID: 0xFAD3CF0D --- Direcci?n Comercial: C/Jose Bardasano Baos, 9, Edif. Gorbea 3, planta 3?D, 28016 Madrid, Spain Direcci?n Registrada: Red Hat S.L., C/ Velazquez 63, Madrid 28001, Spain Inscrita en el Reg. Mercantil de Madrid ? C.I.F. B82657941 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Esta parte del mensaje est? firmada digitalmente URL: From swissslinky at gmail.com Thu Feb 28 16:12:41 2008 From: swissslinky at gmail.com (Dan Dengate) Date: Thu, 28 Feb 2008 17:12:41 +0100 Subject: [et-mgmt-tools] Signups for FudCON Boston (it's free!) In-Reply-To: <47C5E389.8000409@redhat.com> References: <47C5D158.9000302@redhat.com> <47C5E389.8000409@redhat.com> Message-ID: why is America so far away :( On 27/02/2008, Michael DeHaan wrote: > > Michael DeHaan wrote: > > You'll need a Fedora account to edit the page (this may be moved to > > barcamp.org later?), though... > > > > https://fedoraproject.org/wiki/FUDCon/FUDConF10 > > https://admin.fedoraproject.org/accounts > > > > The list of talks/subjects for FudCON is still rolling in, but both > > Cobbler and Func will be there. > > > > This FudCON is also at the same time (and the day after also) as Red > > Hat Summit, so that may be > > something you are interested in also. > > http://www.redhat.com/promo/summit/ ... > > > > Summit last year was a lot of fun and had some great talks and was > > also a great chance for developers and users > > alike to get together, meet, and talk shop. > > There should be a good round of talks on subjects of interest to > > et-mgmt-tools subscribers at both, so > > if you are interested, check them out! > > > > --Michael > > > > > > _______________________________________________ > > et-mgmt-tools mailing list > > et-mgmt-tools at redhat.com > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > > One quick addition... (suggested on another list) > > If you don't have a Fedora account and want to go, email me off list and > I can add you to the Wiki. > > That can save you some trouble. (Though Fedora accounts are great too!) > > > --Michael > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -- MT Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: From mdehaan at redhat.com Thu Feb 28 23:08:08 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 28 Feb 2008 18:08:08 -0500 Subject: [et-mgmt-tools] Working on VMware support for cobbler/koan.. Message-ID: <47C73ED8.7030802@redhat.com> I'm working on adding VMware support for koan. Obviously it's not Free as in Speech, but it's widely deployed, and I think that would be a nice addition to the idea for further abstracting out provisioning infrastructure. Most likely this will work by allowing cobbler to keep track of CD ISO's in the "distro" objects. So we'll be able to do things like: cobbler distro edit --name=RHEL5 --rescuecd=/path/*.iso cobbler profile edit --name=RHEL5-examplefoo --virt-type=vmware # on the remote system koan --server=cobbler.example.com --virt --profile=examplefoo Once we do this (and are able to represent ISO file tracking), we can also support Xen Fullvirt without the need for PXE infrastructure (which means potentially being able to install non-Linux distros also as well as support appliance-like stuff, though we may need a few more options there as that rolls on). Comments/ideas welcome... In other news, vmware server is really annoying to setup compared to Xen/KVM :) --Michael From sakaia at jp.fujitsu.com Fri Feb 29 06:04:15 2008 From: sakaia at jp.fujitsu.com (Atsushi SAKAI) Date: Fri, 29 Feb 2008 15:04:15 +0900 Subject: [et-mgmt-tools] [PATCH] fix typos for virtinst Message-ID: <200802290604.m1T64GFU029480@fjmscan502.ms.jp.fujitsu.com> Hi, It fix typos. Before applying, ImageFetcher.py patch effects po file. Please care. man/en/virt-install.1 | 2 +- virtinst/ImageFetcher.py | 2 +- virtinst/OSDistro.py | 2 +- virtinst/util.py | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) Thanks Atsushi SAKAI -------------- next part -------------- A non-text attachment was scrubbed... Name: fix_typo_virtinst0229.patch Type: application/octet-stream Size: 2510 bytes Desc: not available URL: From bkearney at redhat.com Fri Feb 29 13:49:25 2008 From: bkearney at redhat.com (Bryan Kearney) Date: Fri, 29 Feb 2008 08:49:25 -0500 Subject: [et-mgmt-tools] Working on VMware support for cobbler/koan.. In-Reply-To: <47C73ED8.7030802@redhat.com> References: <47C73ED8.7030802@redhat.com> Message-ID: <47C80D65.4020806@redhat.com> What do you think the possibility of of being able to have one kickstart file feed vmware and xen images? It would be great to be able to make the choice of vmwware/xen at koan time and not cobbler time. -- bk Michael DeHaan wrote: > I'm working on adding VMware support for koan. Obviously it's not Free > as in Speech, but it's widely deployed, and I think that > would be a nice addition to the idea for further abstracting out > provisioning infrastructure. > Most likely this will work by allowing cobbler to keep track of CD ISO's > in the "distro" objects. > > So we'll be able to do things like: > cobbler distro edit --name=RHEL5 --rescuecd=/path/*.iso > cobbler profile edit --name=RHEL5-examplefoo --virt-type=vmware > # on the remote system > koan --server=cobbler.example.com --virt --profile=examplefoo > > Once we do this (and are able to represent ISO file tracking), we can > also support Xen Fullvirt without the need > for PXE infrastructure (which means potentially being able to install > non-Linux distros also as well as support > appliance-like stuff, though we may need a few more options there as > that rolls on). > > Comments/ideas welcome... > > In other news, vmware server is really annoying to setup compared to > Xen/KVM :) > > --Michael > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mdehaan at redhat.com Fri Feb 29 15:01:04 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 29 Feb 2008 10:01:04 -0500 Subject: [et-mgmt-tools] Working on VMware support for cobbler/koan.. In-Reply-To: <47C80D65.4020806@redhat.com> References: <47C73ED8.7030802@redhat.com> <47C80D65.4020806@redhat.com> Message-ID: <47C81E30.9040703@redhat.com> Bryan Kearney wrote: > What do you think the possibility of of being able to have one > kickstart file feed vmware and xen images? It would be great to be > able to make the choice of vmwware/xen at koan time and not cobbler time. > > -- bk Koan already takes a --virt-type for an override for the guest type, so that's the plan. Right now, the same kickstarts are usable for metal, Xen pv/fv, and qemu/KVM so I wouldn't expect a change. --Michael > > > Michael DeHaan wrote: >> I'm working on adding VMware support for koan. Obviously it's not >> Free as in Speech, but it's widely deployed, and I think that >> would be a nice addition to the idea for further abstracting out >> provisioning infrastructure. Most likely this will work by allowing >> cobbler to keep track of CD ISO's in the "distro" objects. >> >> So we'll be able to do things like: >> cobbler distro edit --name=RHEL5 --rescuecd=/path/*.iso >> cobbler profile edit --name=RHEL5-examplefoo --virt-type=vmware >> # on the remote system >> koan --server=cobbler.example.com --virt --profile=examplefoo >> >> Once we do this (and are able to represent ISO file tracking), we can >> also support Xen Fullvirt without the need >> for PXE infrastructure (which means potentially being able to install >> non-Linux distros also as well as support >> appliance-like stuff, though we may need a few more options there as >> that rolls on). >> >> Comments/ideas welcome... >> >> In other news, vmware server is really annoying to setup compared to >> Xen/KVM :) >> >> --Michael >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From mdehaan at redhat.com Fri Feb 29 15:25:22 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 29 Feb 2008 10:25:22 -0500 Subject: [et-mgmt-tools] Working on VMware support for cobbler/koan.. In-Reply-To: <47C73ED8.7030802@redhat.com> References: <47C73ED8.7030802@redhat.com> Message-ID: <47C823E2.7010901@redhat.com> > > > Once we do this (and are able to represent ISO file tracking), we can > also support Xen Fullvirt without the need > for PXE infrastructure (which means potentially being able to install > non-Linux distros also as well as support > appliance-like stuff, though we may need a few more options there as > that rolls on). Rethinking this, the above ISO approach might not allow passing ks=foo without rebuilding ISOs, in which case we may still need to do PXE for both of those after all to keep them fully automated. Stay tuned :) From mdehaan at redhat.com Fri Feb 29 17:27:58 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 29 Feb 2008 12:27:58 -0500 Subject: [et-mgmt-tools] The "Who Uses Cobbler" page Message-ID: <47C8409E.3090309@redhat.com> I have an opt-in page here of people/projects that are using Cobbler on the Wiki: https://fedorahosted.org/cobbler/wiki/WhoUsesCobbler The list of of course a lot bigger than this, so if you would like to add your company or project, please do. If you do not have a Fedora account (which you need to edit the Wiki), you can email me (off-list) and I can add you. Thanks! --Michael From mdehaan at redhat.com Fri Feb 29 23:08:41 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 29 Feb 2008 18:08:41 -0500 Subject: [et-mgmt-tools] Working on VMware support for cobbler/koan.. In-Reply-To: <47C823E2.7010901@redhat.com> References: <47C73ED8.7030802@redhat.com> <47C823E2.7010901@redhat.com> Message-ID: <47C89079.90206@redhat.com> Michael DeHaan wrote: > >> >> >> Once we do this (and are able to represent ISO file tracking), we can >> also support Xen Fullvirt without the need >> for PXE infrastructure (which means potentially being able to install >> non-Linux distros also as well as support >> appliance-like stuff, though we may need a few more options there as >> that rolls on). > > Rethinking this, the above ISO approach might not allow passing ks=foo > without rebuilding ISOs, in which case we may still need to do > PXE for both of those after all to keep them fully automated. Stay > tuned :) > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools It's looking that way. So Cobbler will manage the PXE tree and koan can feed the vmx file info on ram/disk requirements, which it will then create the "shell" VM which can then PXE via cobbler. So you will probably want to do it by system records in Cobbler... More info later as I hack on this further... --Michael