From takatom at jp.fujitsu.com Fri May 2 08:39:41 2008 From: takatom at jp.fujitsu.com (Takahashi Tomohiro) Date: Fri, 2 May 2008 17:39:41 +0900 Subject: [et-mgmt-tools] [PATCH][virt-manager] Enable "Printscreen" toGuestOS from Virtual Machine console menu. References: <0a7601c8a6a6$2a3e5850$776f210a@fj98229700> Message-ID: <008e01c8ac30$1aeb8730$776f210a@fj98229700> Hi, Would you give me a comment on this virt-manager key-fixation patch? If not, Please commit it. Thanks Tomohiro Takahashi > Hi, > > Virtual Machine Console can not send > "Printscreen" to GuestOS from menu [Send key]-[Printscreen]. > > This patch fix this issue. > > And I checked that the others [Send key] has not a problem. > > Signed-off-by: Tomohiro Takahashi > > Thanks, > Tomohiro Takahashi > From berrange at redhat.com Fri May 2 13:36:57 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 2 May 2008 14:36:57 +0100 Subject: [et-mgmt-tools] [PATCH][virt-manager] Enable "Printscreen" toGuestOS from Virtual Machine console menu. In-Reply-To: <008e01c8ac30$1aeb8730$776f210a@fj98229700> References: <0a7601c8a6a6$2a3e5850$776f210a@fj98229700> <008e01c8ac30$1aeb8730$776f210a@fj98229700> Message-ID: <20080502133657.GE5681@redhat.com> On Fri, May 02, 2008 at 05:39:41PM +0900, Takahashi Tomohiro wrote: > Hi, > > Would you give me a comment on this virt-manager key-fixation patch? > If not, Please commit it. Sorry I missed this patch - it is comitted now. Regards, Daniel. -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From berrange at redhat.com Fri May 2 13:40:12 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 2 May 2008 14:40:12 +0100 Subject: [et-mgmt-tools] [RFE] virtual networks without DHCP In-Reply-To: <20080428075212.GA8144@amd.home.annexia.org> References: <48147852.9040605@redhat.com> <20080428075212.GA8144@amd.home.annexia.org> Message-ID: <20080502134012.GF5681@redhat.com> On Mon, Apr 28, 2008 at 08:52:12AM +0100, Richard W.M. Jones wrote: > On Sun, Apr 27, 2008 at 02:57:54PM +0200, Geert Jansen wrote: > > I think it would be nice if it were possible to create a virtual network > > with virt-manager without DHCP and DNS services. This is useful when you > > want to use a host-only network for testing a setup that includes a DHCP > > server. > > > > I am not sure whether this is a limitation in virt-manager or in libvirt. > > It's a libvirt thing. Replies redirected to libvir-list. > > My understanding is that if the network doesn't contain a > element in the network XML then no DHCP server will be created. > > http://libvirt.org/formatnetwork.html As Rich mentions, this is already supported in libvirt - simply omit the tag. This is not exposed in the virt-manager UI though - it always requires DHCP, so you'll need to use the command line to configure this. Dan -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From mdehaan at redhat.com Fri May 2 14:29:39 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 02 May 2008 10:29:39 -0400 Subject: [et-mgmt-tools] Installing without PXE, and without the live CD -- "cobbler buildiso" Message-ID: <481B2553.3030609@redhat.com> For a while Cobbler has had a solution based on koan and the live CD to enable network provisioning of bare metal in environments that either do not have DHCP, or do not have control over DHCP that is sufficient enough to set up a PXE environment. Naturally, if you can set up a PXE environment with Cobbler, it's very useful to have, and you'll want it. So, along came the live CD. The live CD was built using a recent Fedora, and it had the magic ability to install any distro -- /except/ you had to use Fedora to build the Live CD, and the build process was slow, and the live image was a bit bigger than it needed to be. However, due to driver constraints, it didn't always support the latest in storage technology -- that was a problem. We have finally implemented the low-tech solution, thanks to various folks in Red Hat GPS, and some hacking I've done to integrate that closer into cobbler. Using 0.9.X or later (it's checked in now into the git "devel" branch) you can do: cobbler buildiso [--iso=] [--tempdir=] [--profiles=] This will automatically generate an ISO that allows for menus just like Cobbler's PXE boot menus, for installing new systems. I also plan to allow a configurable default profile for those who want to mass deploy this image using remote management processors, etc. The menu will contain an entry for each bare metal bootable profile, with the current data set in Cobbler. Given that Cobbler now generates kickstarts in real time, changes to the kickstart templates can be made without reburning the CD! Again, the live CD is more dynamic, but this is much smaller, faster, and easier to build. The one thing the live CD still offers is simulation of the MAC address detection feature of PXE, but if you don't need to use cobbler system records to provision your lab/datacenter/etc, this will also get you there -- and is probably the perfect fit if the Live CD was not working out for you. Test release will be available soon, git is available now. --Michael From mdehaan at redhat.com Fri May 2 15:09:34 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 02 May 2008 11:09:34 -0400 Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? Message-ID: <481B2EAE.3050601@redhat.com> With John Eckersberg's recent patch (and likely future extensions), Cobbler will be acquiring very good support for bind, so Cobbler can manage DHCP and DNS at the same time using more scalable tools. Similarly, we can now manage DHCP without restarts (Pablo Iranzo Gomez's patch). So then, I'm wondering what are the reasons we would have for continuing to support dnsmasq? If you would be affected by removal of dnsmasq support, I'd like to hear from you, as to why you would like it to stay in place. This does not neccessarily eliminate the ability to have seperate options in Cobbler for (more modular) DHCP/DNS choices in the future. (Likely this is something we could use /etc/cobbler/modules.conf for). --Michael From hyclak at math.ohiou.edu Fri May 2 15:13:19 2008 From: hyclak at math.ohiou.edu (Matt Hyclak) Date: Fri, 2 May 2008 11:13:19 -0400 Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? In-Reply-To: <481B2EAE.3050601@redhat.com> References: <481B2EAE.3050601@redhat.com> Message-ID: <20080502151319.GE5991@math.ohiou.edu> On Fri, May 02, 2008 at 11:09:34AM -0400, Michael DeHaan enlightened us: > With John Eckersberg's recent patch (and likely future extensions), > Cobbler will be acquiring very good support for bind, so Cobbler can > manage DHCP and DNS at the same time using more scalable tools. > Similarly, we can now manage DHCP without restarts (Pablo Iranzo Gomez's > patch). So then, I'm wondering what are the reasons we would have for > continuing to support dnsmasq? > > If you would be affected by removal of dnsmasq support, I'd like to hear > from you, as to why you would like it to stay in place. This does not > neccessarily eliminate the ability to have seperate options in Cobbler > for (more modular) DHCP/DNS choices in the future. (Likely this is > something we could use /etc/cobbler/modules.conf for). > I use dnsmasq to handle the DNS for my internal server network and let cobbler manage it. Saves me headache. I have used bind before, and if the support is there I could probably make the switch back, but I'd recommend getting it in for lots of testing before removing dnsmasq support. Matt -- Matt Hyclak Department of Mathematics Department of Social Work Ohio University (740) 593-1263 From jeckersb at redhat.com Fri May 2 15:35:19 2008 From: jeckersb at redhat.com (John Eckersberg) Date: Fri, 02 May 2008 11:35:19 -0400 Subject: [et-mgmt-tools] [PATCH] BIND DNS support In-Reply-To: <48161299.8080507@redhat.com> References: <4815E61D.6070607@redhat.com> <80d7e4090804281104l193207cx6f024e72da8ac194@mail.gmail.com> <48161299.8080507@redhat.com> Message-ID: <481B34B7.3050204@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 John Eckersberg wrote: | Stephen John Smoogen wrote: | | On Mon, Apr 28, 2008 at 8:58 AM, John Eckersberg | wrote: | |> -----BEGIN PGP SIGNED MESSAGE----- | |> Hash: SHA1 | |> | |> Working well but probably needs a bit more cleanup... | |> | | | | Well a 0 length patch usually works well. What are you wanting to | | patch it against? | | | | | |> Patches to my patch are welcome :) | |> | |> - -- | |> John Eckersberg | |> Production Operations | |> Red Hat, Inc. | |> | |> This is an updated patch: http://jeckersb.fedorapeople.org/cobbler/0001-BIND-dns-support.patch Which does two things: 1. default entry in the named.conf is now nobody.example.com instead of hostmaster.redhat.com. I don't want email for everybody using the default template :) 2. If manage_{forward/reverse}_zones are not set in the config file, dns maps for all hosts will be generated (before it would generate none). Maybe taking a regex would be nice for the future but for now this will work. - -- John Eckersberg Production Operations Red Hat, Inc. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFIGzS3O5VSO/Qqm+gRAlKMAKCFNRKCOlGIYT/7sH47R042HDjIdACeJV6W NbPOR9ruywq49JkoPtmkQdM= =/JOI -----END PGP SIGNATURE----- From mdehaan at redhat.com Fri May 2 15:54:40 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 02 May 2008 11:54:40 -0400 Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? In-Reply-To: <20080502151319.GE5991@math.ohiou.edu> References: <481B2EAE.3050601@redhat.com> <20080502151319.GE5991@math.ohiou.edu> Message-ID: <481B3940.6090103@redhat.com> Matt Hyclak wrote: > On Fri, May 02, 2008 at 11:09:34AM -0400, Michael DeHaan enlightened us: > >> With John Eckersberg's recent patch (and likely future extensions), >> Cobbler will be acquiring very good support for bind, so Cobbler can >> manage DHCP and DNS at the same time using more scalable tools. >> Similarly, we can now manage DHCP without restarts (Pablo Iranzo Gomez's >> patch). So then, I'm wondering what are the reasons we would have for >> continuing to support dnsmasq? >> >> If you would be affected by removal of dnsmasq support, I'd like to hear >> from you, as to why you would like it to stay in place. This does not >> neccessarily eliminate the ability to have seperate options in Cobbler >> for (more modular) DHCP/DNS choices in the future. (Likely this is >> something we could use /etc/cobbler/modules.conf for). >> >> > > I use dnsmasq to handle the DNS for my internal server network and let > cobbler manage it. Saves me headache. > > I have used bind before, and if the support is there I could probably make > the switch back, but I'd recommend getting it in for lots of testing before > removing dnsmasq support. > > Matt > > One idea that came up on IRC and that I've been thinking about is to use the cobbler/modules.conf system to select the DNS types, and move the DHCP/DNS code into the cobbler modules directory. This would allow cleaning up a lot of the underlying infrastructure and still keeping this around. Still, more feedback welcome... as I'm curious as to how widely this is being used (and also how high dnsmasq is scaling) Thanks! --Michael From mdehaan at redhat.com Fri May 2 15:59:34 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 02 May 2008 11:59:34 -0400 Subject: [et-mgmt-tools] [PATCH] BIND DNS support In-Reply-To: <481B34B7.3050204@redhat.com> References: <4815E61D.6070607@redhat.com> <80d7e4090804281104l193207cx6f024e72da8ac194@mail.gmail.com> <48161299.8080507@redhat.com> <481B34B7.3050204@redhat.com> Message-ID: <481B3A66.90202@redhat.com> John Eckersberg wrote: Applied and very cool! I'm going to be working on integrating this into /etc/cobbler/modules.conf so don't be too alarmed if the code moves around slightly. Thanks! --Michael > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > John Eckersberg wrote: > | Stephen John Smoogen wrote: > | | On Mon, Apr 28, 2008 at 8:58 AM, John Eckersberg > > | wrote: > | |> -----BEGIN PGP SIGNED MESSAGE----- > | |> Hash: SHA1 > | |> > | |> Working well but probably needs a bit more cleanup... > | |> > | | > | | Well a 0 length patch usually works well. What are you wanting to > | | patch it against? > | | > | | > | |> Patches to my patch are welcome :) > | |> > | |> - -- > | |> John Eckersberg > | |> Production Operations > | |> Red Hat, Inc. > | |> > | |> > > This is an updated patch: > > http://jeckersb.fedorapeople.org/cobbler/0001-BIND-dns-support.patch > > Which does two things: > > 1. default entry in the named.conf is now nobody.example.com instead of > hostmaster.redhat.com. I don't want email for everybody using the > default template :) > > 2. If manage_{forward/reverse}_zones are not set in the config file, > dns maps for all hosts will be generated (before it would generate > none). Maybe taking a regex would be nice for the future but for now > this will work. > > - -- > John Eckersberg > Production Operations > Red Hat, Inc. > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.7 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iD8DBQFIGzS3O5VSO/Qqm+gRAlKMAKCFNRKCOlGIYT/7sH47R042HDjIdACeJV6W > NbPOR9ruywq49JkoPtmkQdM= > =/JOI > -----END PGP SIGNATURE----- > > _______________________________________________ > 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 May 2 16:00:46 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 02 May 2008 12:00:46 -0400 Subject: [et-mgmt-tools] Cobbler patch OMAPI v2 ;) In-Reply-To: <1209537427.8766.47.camel@iranzo.users.redhat.com> References: <4817D8FB.8090404@redhat.com> <1209537427.8766.47.camel@iranzo.users.redhat.com> Message-ID: <481B3AAE.6080606@redhat.com> Pablo Iranzo G?mez wrote: > Morning! > > El mar, 29-04-2008 a las 22:27 -0400, Michael DeHaan escribi?: > Applied to devel branch. As with John's patch, I'm going to be working on moving this around into the 'modules' directory, though this is very good to see. Thanks! --Michael >> Should be applied Thursday or so barring any other comments, and that >> will make this easier for people to test. I also need to do a test >> release for 0.9 soon (let's call this 0.9.1 ... there will be more test >> releases with some other features likely being added prior to the 1.0). >> > > Count with me for the beta-testing :) > > > >> Quick question -- Should omapi in settings be off by default? It only >> works with modifications to the dhcp.template to enable it -- though we >> /could/ enable that by default in the template and solve the problem. >> > > For me, as cobbler is rewriting dhcp info, should be enabled by > default. This could need another check for the code I submitted > yesterday to check if "manage dhcp is enabled for isc". > > For example, in trigger, the code is: > > if manage_dhcp_mode == "isc": > if not omapi: > if not omapi_port: > rc = os.system("/sbin/service dhcpd restart") > > So, if manage mode is "ISC", and omapi is not enabled nor omapi_port, > then it does a restart. > > I don't think that this could hurt anyone if already using Cobbler for > managing DHCP, if not, system will just do the other tasks, but nothing > about DHCP > > >> Comments from folks who use Cobbler for DHCP management? >> >> The main goal of this is to keep DHCP running throughout sync operations >> -- but also to allow for less reasons to ever need to do "sync". Since >> kickstart generation is now dynamic in 0.9/1.0, the only real reason to >> run sync is to rebuild some of the tftpboot tree, the rest of the >> "partial" data gets built each time you make a change to the associated >> objects (and all descendents automatically update). >> > > > Well, regarding this, right now I had to put the new stanza for > cleaning leases just in case of machines removal. The best thing will be > to put the code in "cobbler system add" "cobbler system remove" and > "cobbler system rename" in order to just make this calls when needed, > remmoving the need for the "leases-cleanup code", as the systems will > get created on DHCP or removed dinamically. > > >> We can probably even make that less important as time goes on, by >> knowing basic things like if we add a profile, we can also quickly >> rebuild the pxemenus (since the number of profiles is going to be very >> small, etc). >> > > That would be a good point, in this way, cobbler state could just use > "sync" for forcing a full cobbler files recreation just to be sure, > remove possible problems, as all the other operations would get > inmediate action on system . > > Regards > Pablo > > PD: Attached patch adds an extra check to the DHCP leases cleanup code > to check if we're doing ISC > > > ------------------------------------------------------------------------ > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From ssklar at stanford.edu Fri May 2 16:00:15 2008 From: ssklar at stanford.edu (Sandor W. Sklar) Date: Fri, 2 May 2008 09:00:15 -0700 Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? In-Reply-To: <481B3940.6090103@redhat.com> References: <481B2EAE.3050601@redhat.com> <20080502151319.GE5991@math.ohiou.edu> <481B3940.6090103@redhat.com> Message-ID: On May 2, 2008, at 8:54 AM, Michael DeHaan wrote: > > Still, more feedback welcome... as I'm curious as to how widely this > is being used (and also how > high dnsmasq is scaling) Since you said more feedback is welcome, I'll state that we in fact do not use any of the DNS features of cobbler; we've already got a large- scale, robust DNS service and management system, so, vis-a-vis this thread, I just hope that whatever decision is made, it doesn't have too much impact on those of us not using Cobbler for DNS. Thanks, -s- From mdehaan at redhat.com Fri May 2 16:05:15 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 02 May 2008 12:05:15 -0400 Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? In-Reply-To: References: <481B2EAE.3050601@redhat.com> <20080502151319.GE5991@math.ohiou.edu> <481B3940.6090103@redhat.com> Message-ID: <481B3BBB.6080409@redhat.com> Sandor W. Sklar wrote: > > On May 2, 2008, at 8:54 AM, Michael DeHaan wrote: > >> >> Still, more feedback welcome... as I'm curious as to how widely this >> is being used (and also how >> high dnsmasq is scaling) > > Since you said more feedback is welcome, I'll state that we in fact do > not use any of the DNS features of cobbler; we've already got a > large-scale, robust DNS service and management system, so, vis-a-vis > this thread, I just hope that whatever decision is made, it doesn't > have too much impact on those of us not using Cobbler for DNS. > > Thanks, > -s- It will not make any impact. Mainly I just needed feedback on dnsmasq. I know quite a few people are using the ISC support. The idea with Cobbler is that if you /do/ want to get out of the business of maintaining your DHCP/DNS/kickstart/system-definition info in 3 places, cobbler will let you do it. And while it is "batteries included", we don't require any of those to be used, nor are they on by default. > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From ssklar at stanford.edu Fri May 2 16:05:30 2008 From: ssklar at stanford.edu (Sandor W. Sklar) Date: Fri, 2 May 2008 09:05:30 -0700 Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? In-Reply-To: <481B3BBB.6080409@redhat.com> References: <481B2EAE.3050601@redhat.com> <20080502151319.GE5991@math.ohiou.edu> <481B3940.6090103@redhat.com> <481B3BBB.6080409@redhat.com> Message-ID: <46F4CFD8-D825-427D-B9CD-176E1ADAC8DB@stanford.edu> On May 2, 2008, at 9:05 AM, Michael DeHaan wrote: > Sandor W. Sklar wrote: >> >> On May 2, 2008, at 8:54 AM, Michael DeHaan wrote: >> >>> >>> Still, more feedback welcome... as I'm curious as to how widely >>> this is being used (and also how >>> high dnsmasq is scaling) >> >> Since you said more feedback is welcome, I'll state that we in fact >> do not use any of the DNS features of cobbler; we've already got a >> large-scale, robust DNS service and management system, so, vis-a- >> vis this thread, I just hope that whatever decision is made, it >> doesn't have too much impact on those of us not using Cobbler for >> DNS. >> >> Thanks, >> -s- > > It will not make any impact. Mainly I just needed feedback on > dnsmasq. I know quite a few people are using the ISC support. > > The idea with Cobbler is that if you /do/ want to get out of the > business of maintaining your DHCP/DNS/kickstart/system-definition > info in 3 places, cobbler will let you do it. And while it is > "batteries included", we don't require any of those to be used, nor > are they on by default. Good to hear. :-) I figured as much, but just wanted to make sure. Thanks! -s- From sghosh at redhat.com Fri May 2 21:17:07 2008 From: sghosh at redhat.com (Subhendu Ghosh) Date: Fri, 02 May 2008 17:17:07 -0400 Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? In-Reply-To: <481B2EAE.3050601@redhat.com> References: <481B2EAE.3050601@redhat.com> Message-ID: <481B84D3.4010509@redhat.com> Michael DeHaan wrote: > With John Eckersberg's recent patch (and likely future extensions), > Cobbler will be acquiring very good support for bind, so Cobbler can > manage DHCP and DNS at the same time using more scalable tools. > Similarly, we can now manage DHCP without restarts (Pablo Iranzo Gomez's > patch). So then, I'm wondering what are the reasons we would have for > continuing to support dnsmasq? > > If you would be affected by removal of dnsmasq support, I'd like to hear > from you, as to why you would like it to stay in place. This does not > neccessarily eliminate the ability to have seperate options in Cobbler > for (more modular) DHCP/DNS choices in the future. (Likely this is > something we could use /etc/cobbler/modules.conf for). > > --Michael > dnsmasq is useful for small test networks that are running isolated from the rest of the infrastructure. Having it available as an optional module simplifies quick deployment and removal. I am thinking along the lines of - project x poc x test - in a consulting toolbag. That said I would much rather have BIND support over dnsmaq. -sg -- Red Hat Summit Boston | June 18-20, 2008 Learn more: http://www.redhat.com/summit -------------- next part -------------- A non-text attachment was scrubbed... Name: sghosh.vcf Type: text/x-vcard Size: 266 bytes Desc: not available URL: From mdehaan at redhat.com Fri May 2 21:31:13 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 02 May 2008 17:31:13 -0400 Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? In-Reply-To: <481B84D3.4010509@redhat.com> References: <481B2EAE.3050601@redhat.com> <481B84D3.4010509@redhat.com> Message-ID: <481B8821.70902@redhat.com> Subhendu Ghosh wrote: > Michael DeHaan wrote: >> With John Eckersberg's recent patch (and likely future extensions), >> Cobbler will be acquiring very good support for bind, so Cobbler can >> manage DHCP and DNS at the same time using more scalable tools. >> Similarly, we can now manage DHCP without restarts (Pablo Iranzo >> Gomez's patch). So then, I'm wondering what are the reasons we >> would have for continuing to support dnsmasq? >> >> If you would be affected by removal of dnsmasq support, I'd like to >> hear from you, as to why you would like it to stay in place. This >> does not neccessarily eliminate the ability to have seperate options >> in Cobbler for (more modular) DHCP/DNS choices in the future. >> (Likely this is something we could use /etc/cobbler/modules.conf for). >> >> --Michael >> > > dnsmasq is useful for small test networks that are running isolated > from the rest of the infrastructure. Having it available as an > optional module simplifies quick deployment and removal. > > I am thinking along the lines of - project x poc x test - in a > consulting toolbag. > > That said I would much rather have BIND support over dnsmaq. > > -sg I decided to modularize the code more and am keeping dnsmasq (and supporting it in addition to BIND). For those not familiar, the reason dnsmasq is different is that it supports *both* dnsmasq and dhcp at the same time. It actually also contains a miniature TFTP server, but since the distro already has one we like, we don't use that. How it's working at the moment in 0.9.X (devel/git), FWIW: Go into /etc/cobbler/modules.conf [dns] module = manage_bind [dhcp] module = manage_isc (Or set both fields to manage_dnsmasq for dnsmasq) [dns] module = manage_dnsmasq [dhcp] module = manage_dnsmasq And in /var/lib/cobbler/settings, there are seperate booleans for: manage_dhcp: 0/1 manage_dns: 0/1 So you can use ISC without BIND, and BIND without ISC, or dnsmasq for only half of what it does, etc. These are still driven by templates in /etc/cobbler -- one each for BIND (named), ISC, and dnsmasq. I am sorely due for writing a "what's new" in 1.0 document for the Wiki, that will describe what the upgrade process is and what all the changelog features are. omapi is seperately enabled in /var/lib/cobbler/settings: omapi_enabled: 0/1 omapi_port: 647 This only applies when manage_dhcp is enabled and the module in use is ISC. Given this is going to require some good testing, so I'll do a test release before adding anything else. I think this cleans things up nicely and still allows people to add their own DHCP/DNS servers into modules fairly cleanly if they use something different, without having to have a bunch of "ifs" in the sync code. Folks are welcome to try this now, though there are no doubt a few bugs which we will be ironing out next week. The Wiki needs to be updated also to include more docs on this and sample template files. I'll also probably be moving more of the static HTML from cobbler.et.redhat.com to the Wiki so people can update it when they come up with interesting tricks as well as corrections. --Michael > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From sghosh at redhat.com Fri May 2 21:31:33 2008 From: sghosh at redhat.com (Subhendu Ghosh) Date: Fri, 02 May 2008 17:31:33 -0400 Subject: [et-mgmt-tools] Installing without PXE, and without the live CD -- "cobbler buildiso" In-Reply-To: <481B2553.3030609@redhat.com> References: <481B2553.3030609@redhat.com> Message-ID: <481B8835.7040407@redhat.com> Michael DeHaan wrote: > For a while Cobbler has had a solution based on koan and the live CD to > enable network provisioning of bare metal in environments that either do > not have DHCP, or do not have control over DHCP that is sufficient > enough to set up a PXE environment. Naturally, if you can set up a PXE > environment with Cobbler, it's very useful to have, and you'll want it. > > So, along came the live CD. > > The live CD was built using a recent Fedora, and it had the magic > ability to install any distro -- /except/ you had to use Fedora to build > the Live CD, and the build process was slow, and the live image was a > bit bigger than it needed to be. However, due to driver constraints, > it didn't always support the latest in storage technology -- that was a problem. > > We have finally implemented the low-tech solution, thanks to various > folks in Red Hat GPS, and some hacking I've done to integrate that > closer into cobbler. > > Using 0.9.X or later (it's checked in now into the git "devel" branch) > you can do: > > cobbler buildiso [--iso=] [--tempdir=] [--profiles=] > > This will automatically generate an ISO that allows for menus just like > Cobbler's PXE boot menus, for installing new systems. I also plan to > allow a configurable default profile for those who want to mass deploy > this image using remote management processors, etc. > > The menu will contain an entry for each bare metal bootable profile, > with the current data set in Cobbler. Given that Cobbler now > generates kickstarts in real time, changes to the kickstart templates > can be made without reburning the CD! > > Again, the live CD is more dynamic, but this is much smaller, faster, > and easier to build. > The one thing the live CD still offers is simulation of the MAC address > detection feature of PXE, but if you don't need to use cobbler system > records to provision your lab/datacenter/etc, this will also get you > there -- and is probably the perfect fit if the Live CD was not working > out for you. > > Test release will be available soon, git is available now. > > --Michael > If the iso(s) can be made available through the web infrastructure, and accessible via a predefined or calculated path, then a external rules engine could drive the provisioning process through the remote management processors's virtual media capabilities. Perhaps the profiles module could be extended to define whether a iso should be generated or not. An easy way to automatically generate boot isos. -sg -- Red Hat Summit Boston | June 18-20, 2008 Learn more: http://www.redhat.com/summit -------------- next part -------------- A non-text attachment was scrubbed... Name: sghosh.vcf Type: text/x-vcard Size: 266 bytes Desc: not available URL: From robin-lists at robinbowes.com Sat May 3 00:30:34 2008 From: robin-lists at robinbowes.com (Robin Bowes) Date: Sat, 03 May 2008 01:30:34 +0100 Subject: [et-mgmt-tools] New yum module: grouppackagetypes Message-ID: Hi all, I was trying to find out how I could install just the mandatory packages from a package group when using "yum groupinstall ...". I found that I could add a line to /etc/yum.conf: group_package_types=mandatory But I'd like to be able to do it from the command line, i.e. without editing any files, then having to edit them back. So, I asked on the #yum IRC channel and it was suggested that I could do this with a yum plugin. So, not letting the small matter that I don't really know any python get in the way, I came up with this: http://robinbowes.com/projects/yum-grouppackagetypes Hope someone else finds it useful. R. From Pablo.Iranzo at redhat.com Sat May 3 10:40:12 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Sat, 3 May 2008 12:40:12 +0200 (CEST) Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? In-Reply-To: <481B8821.70902@redhat.com> References: <481B2EAE.3050601@redhat.com> <481B84D3.4010509@redhat.com> <481B8821.70902@redhat.com> Message-ID: Hi This patch fixes a typo in sync code for removing dinamically leases (the function was renamed but this call was not renamed accordingly). There's also a patch for settings file to show isc_set_host_name And if interesting, changing ksdevice for bootif to use the one used in PXE Regards Pablo PD: buildiso works fine and found nothing wrong with the rest yet -- Pablo Iranzo G?mez (http://Alufis35.uv.es/~iranzo/) (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) -- Postulado de Boling sobre la Ley de Murphy: Si se encuentra bien, no se preocupe. Se le pasar? On Fri, 2 May 2008, Michael DeHaan wrote: > Subhendu Ghosh wrote: > > Michael DeHaan wrote: > >> With John Eckersberg's recent patch (and likely future extensions), > >> Cobbler will be acquiring very good support for bind, so Cobbler can > >> manage DHCP and DNS at the same time using more scalable tools. > >> Similarly, we can now manage DHCP without restarts (Pablo Iranzo > >> Gomez's patch). So then, I'm wondering what are the reasons we > >> would have for continuing to support dnsmasq? > >> > >> If you would be affected by removal of dnsmasq support, I'd like to > >> hear from you, as to why you would like it to stay in place. This > >> does not neccessarily eliminate the ability to have seperate options > >> in Cobbler for (more modular) DHCP/DNS choices in the future. > >> (Likely this is something we could use /etc/cobbler/modules.conf for). > >> > >> --Michael > >> > > > > dnsmasq is useful for small test networks that are running isolated > > from the rest of the infrastructure. Having it available as an > > optional module simplifies quick deployment and removal. > > > > I am thinking along the lines of - project x poc x test - in a > > consulting toolbag. > > > > That said I would much rather have BIND support over dnsmaq. > > > > -sg > > I decided to modularize the code more and am keeping dnsmasq (and supporting > it in addition to BIND). For those not familiar, the reason dnsmasq is > different > is that it supports *both* dnsmasq and dhcp at the same time. It > actually also contains > a miniature TFTP server, but since the distro already has one we like, > we don't use that. > > How it's working at the moment in 0.9.X (devel/git), FWIW: > > Go into /etc/cobbler/modules.conf > > [dns] > module = manage_bind > [dhcp] > module = manage_isc > > (Or set both fields to manage_dnsmasq for dnsmasq) > > [dns] > module = manage_dnsmasq > [dhcp] > module = manage_dnsmasq > > And in /var/lib/cobbler/settings, there are seperate booleans for: > > manage_dhcp: 0/1 > manage_dns: 0/1 > > So you can use ISC without BIND, and BIND without ISC, or dnsmasq for > only half > of what it does, etc. > > These are still driven by templates in /etc/cobbler -- one each for BIND > (named), ISC, and dnsmasq. > > I am sorely due for writing a "what's new" in 1.0 document for the Wiki, > that will describe what the > upgrade process is and what all the changelog features are. > > omapi is seperately enabled in /var/lib/cobbler/settings: > > omapi_enabled: 0/1 > omapi_port: 647 > > This only applies when manage_dhcp is enabled and the module in use is ISC. > > Given this is going to require some good testing, so I'll do a test > release before adding anything else. > I think this cleans things up nicely and still allows people to add > their own DHCP/DNS servers into > modules fairly cleanly if they use something different, without having > to have a bunch of "ifs" in the sync code. > > Folks are welcome to try this now, though there are no doubt a few bugs > which we will be ironing out next week. > > The Wiki needs to be updated also to include more docs on this and > sample template files. I'll also probably be moving > more of the static HTML from cobbler.et.redhat.com to the Wiki so people > can update it when they come up with > interesting tricks as well as corrections. > > --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 > -------------- next part -------------- A non-text attachment was scrubbed... Name: parche.diff Type: text/x-diff Size: 1841 bytes Desc: URL: From Pablo.Iranzo at redhat.com Sat May 3 10:51:26 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Sat, 3 May 2008 12:51:26 +0200 (CEST) Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? In-Reply-To: References: <481B2EAE.3050601@redhat.com> <481B84D3.4010509@redhat.com> <481B8821.70902@redhat.com> Message-ID: This patch makes the mac passed to OMSHELL to be lowercase Regards Pablo -- Pablo Iranzo G?mez (http://Alufis35.uv.es/~iranzo/) (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) -- Postulado de Boling sobre la Ley de Murphy: Si se encuentra bien, no se preocupe. Se le pasar? On Sat, 3 May 2008, Pablo Iranzo G?mez wrote: > Hi > This patch fixes a typo in sync code for removing dinamically > leases (the function was renamed but this call was not renamed > accordingly). > > There's also a patch for settings file to show isc_set_host_name > > And if interesting, changing ksdevice for bootif to use the one > used in PXE > > Regards > Pablo > > PD: buildiso works fine and found nothing wrong with the rest yet > > > -- > Pablo Iranzo G?mez > (http://Alufis35.uv.es/~iranzo/) > (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) > -- > Postulado de Boling sobre la Ley de Murphy: > > Si se encuentra bien, no se preocupe. Se le pasar? > > On Fri, 2 May 2008, Michael DeHaan wrote: > > > Subhendu Ghosh wrote: > > > Michael DeHaan wrote: > > >> With John Eckersberg's recent patch (and likely future extensions), > > >> Cobbler will be acquiring very good support for bind, so Cobbler can > > >> manage DHCP and DNS at the same time using more scalable tools. > > >> Similarly, we can now manage DHCP without restarts (Pablo Iranzo > > >> Gomez's patch). So then, I'm wondering what are the reasons we > > >> would have for continuing to support dnsmasq? > > >> > > >> If you would be affected by removal of dnsmasq support, I'd like to > > >> hear from you, as to why you would like it to stay in place. This > > >> does not neccessarily eliminate the ability to have seperate options > > >> in Cobbler for (more modular) DHCP/DNS choices in the future. > > >> (Likely this is something we could use /etc/cobbler/modules.conf for). > > >> > > >> --Michael > > >> > > > > > > dnsmasq is useful for small test networks that are running isolated > > > from the rest of the infrastructure. Having it available as an > > > optional module simplifies quick deployment and removal. > > > > > > I am thinking along the lines of - project x poc x test - in a > > > consulting toolbag. > > > > > > That said I would much rather have BIND support over dnsmaq. > > > > > > -sg > > > > I decided to modularize the code more and am keeping dnsmasq (and supporting > > it in addition to BIND). For those not familiar, the reason dnsmasq is > > different > > is that it supports *both* dnsmasq and dhcp at the same time. It > > actually also contains > > a miniature TFTP server, but since the distro already has one we like, > > we don't use that. > > > > How it's working at the moment in 0.9.X (devel/git), FWIW: > > > > Go into /etc/cobbler/modules.conf > > > > [dns] > > module = manage_bind > > [dhcp] > > module = manage_isc > > > > (Or set both fields to manage_dnsmasq for dnsmasq) > > > > [dns] > > module = manage_dnsmasq > > [dhcp] > > module = manage_dnsmasq > > > > And in /var/lib/cobbler/settings, there are seperate booleans for: > > > > manage_dhcp: 0/1 > > manage_dns: 0/1 > > > > So you can use ISC without BIND, and BIND without ISC, or dnsmasq for > > only half > > of what it does, etc. > > > > These are still driven by templates in /etc/cobbler -- one each for BIND > > (named), ISC, and dnsmasq. > > > > I am sorely due for writing a "what's new" in 1.0 document for the Wiki, > > that will describe what the > > upgrade process is and what all the changelog features are. > > > > omapi is seperately enabled in /var/lib/cobbler/settings: > > > > omapi_enabled: 0/1 > > omapi_port: 647 > > > > This only applies when manage_dhcp is enabled and the module in use is ISC. > > > > Given this is going to require some good testing, so I'll do a test > > release before adding anything else. > > I think this cleans things up nicely and still allows people to add > > their own DHCP/DNS servers into > > modules fairly cleanly if they use something different, without having > > to have a bunch of "ifs" in the sync code. > > > > Folks are welcome to try this now, though there are no doubt a few bugs > > which we will be ironing out next week. > > > > The Wiki needs to be updated also to include more docs on this and > > sample template files. I'll also probably be moving > > more of the static HTML from cobbler.et.redhat.com to the Wiki so people > > can update it when they come up with > > interesting tricks as well as corrections. > > > > --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 > > -------------- next part -------------- A non-text attachment was scrubbed... Name: lower.diff Type: text/x-diff Size: 589 bytes Desc: URL: From serwerr at gmail.com Sat May 3 12:13:16 2008 From: serwerr at gmail.com (serwer serwer) Date: Sat, 3 May 2008 14:13:16 +0200 Subject: [et-mgmt-tools] virt-manager snapshots Message-ID: Hi All, Can I expect virt-manager to support snapshots in such an easy way as VirtualBox for example? Or is this functionality already covered and I"m just simply missing something? Thanks a lot, serwerr From Pablo.Iranzo at redhat.com Sun May 4 10:59:27 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Sun, 4 May 2008 12:59:27 +0200 (CEST) Subject: [et-mgmt-tools] Do you use cobbler's support for dnsmasq? In-Reply-To: References: <481B2EAE.3050601@redhat.com> <481B84D3.4010509@redhat.com> <481B8821.70902@redhat.com> Message-ID: This patch: --------------------------------------------------------------------------- diff --git a/templates/pxedefault.template b/templates/pxedefault.template index bb09893..206c028 100644 --- a/templates/pxedefault.template +++ b/templates/pxedefault.template @@ -8,7 +8,7 @@ ONTIMEOUT local LABEL local MENU LABEL (local) MENU DEFAULT - LOCALBOOT 0 + LOCALBOOT -1 --------------------------------------------------------------------------- Returns control of booting process to BIOS. - LOCALBOOT 0 makes an attempt to boot from local device, like 0x80, etc - LOCALBOOT -1 informs BIOS that PXE has failed and then bios will process next entry into its boot sequence. For example, the boot sequence (BIOS) I follow is: 1st- Boot from PXE 2nd- HDD 3rd- CD-ROM So localboot 0 works for me ok, as "-1" does, but probably on some servers with iSCSI booting, the sequence could be something like: 1st PXE 2nd iSCSI 3rd CD-ROM And as that system doesn't have a local disk, this one would fail ?anone tested this? Regards Pablo -- Pablo Iranzo G?mez (http://Alufis35.uv.es/~iranzo/) (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) -- Postulado de Boling sobre la Ley de Murphy: Si se encuentra bien, no se preocupe. Se le pasar? On Sat, 3 May 2008, Pablo Iranzo G?mez wrote: > This patch makes the mac passed to OMSHELL to be lowercase > > Regards > Pablo > > > > -- > Pablo Iranzo G?mez > (http://Alufis35.uv.es/~iranzo/) > (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) > -- > Postulado de Boling sobre la Ley de Murphy: > > Si se encuentra bien, no se preocupe. Se le pasar? > > On Sat, 3 May 2008, Pablo Iranzo G?mez wrote: > > > Hi > > This patch fixes a typo in sync code for removing dinamically > > leases (the function was renamed but this call was not renamed > > accordingly). > > > > There's also a patch for settings file to show isc_set_host_name > > > > And if interesting, changing ksdevice for bootif to use the one > > used in PXE > > > > Regards > > Pablo > > > > PD: buildiso works fine and found nothing wrong with the rest yet > > > > > > -- > > Pablo Iranzo G?mez > > (http://Alufis35.uv.es/~iranzo/) > > (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) > > -- > > Postulado de Boling sobre la Ley de Murphy: > > > > Si se encuentra bien, no se preocupe. Se le pasar? > > > > On Fri, 2 May 2008, Michael DeHaan wrote: > > > > > Subhendu Ghosh wrote: > > > > Michael DeHaan wrote: > > > >> With John Eckersberg's recent patch (and likely future extensions), > > > >> Cobbler will be acquiring very good support for bind, so Cobbler can > > > >> manage DHCP and DNS at the same time using more scalable tools. > > > >> Similarly, we can now manage DHCP without restarts (Pablo Iranzo > > > >> Gomez's patch). So then, I'm wondering what are the reasons we > > > >> would have for continuing to support dnsmasq? > > > >> > > > >> If you would be affected by removal of dnsmasq support, I'd like to > > > >> hear from you, as to why you would like it to stay in place. This > > > >> does not neccessarily eliminate the ability to have seperate options > > > >> in Cobbler for (more modular) DHCP/DNS choices in the future. > > > >> (Likely this is something we could use /etc/cobbler/modules.conf for). > > > >> > > > >> --Michael > > > >> > > > > > > > > dnsmasq is useful for small test networks that are running isolated > > > > from the rest of the infrastructure. Having it available as an > > > > optional module simplifies quick deployment and removal. > > > > > > > > I am thinking along the lines of - project x poc x test - in a > > > > consulting toolbag. > > > > > > > > That said I would much rather have BIND support over dnsmaq. > > > > > > > > -sg > > > > > > I decided to modularize the code more and am keeping dnsmasq (and supporting > > > it in addition to BIND). For those not familiar, the reason dnsmasq is > > > different > > > is that it supports *both* dnsmasq and dhcp at the same time. It > > > actually also contains > > > a miniature TFTP server, but since the distro already has one we like, > > > we don't use that. > > > > > > How it's working at the moment in 0.9.X (devel/git), FWIW: > > > > > > Go into /etc/cobbler/modules.conf > > > > > > [dns] > > > module = manage_bind > > > [dhcp] > > > module = manage_isc > > > > > > (Or set both fields to manage_dnsmasq for dnsmasq) > > > > > > [dns] > > > module = manage_dnsmasq > > > [dhcp] > > > module = manage_dnsmasq > > > > > > And in /var/lib/cobbler/settings, there are seperate booleans for: > > > > > > manage_dhcp: 0/1 > > > manage_dns: 0/1 > > > > > > So you can use ISC without BIND, and BIND without ISC, or dnsmasq for > > > only half > > > of what it does, etc. > > > > > > These are still driven by templates in /etc/cobbler -- one each for BIND > > > (named), ISC, and dnsmasq. > > > > > > I am sorely due for writing a "what's new" in 1.0 document for the Wiki, > > > that will describe what the > > > upgrade process is and what all the changelog features are. > > > > > > omapi is seperately enabled in /var/lib/cobbler/settings: > > > > > > omapi_enabled: 0/1 > > > omapi_port: 647 > > > > > > This only applies when manage_dhcp is enabled and the module in use is ISC. > > > > > > Given this is going to require some good testing, so I'll do a test > > > release before adding anything else. > > > I think this cleans things up nicely and still allows people to add > > > their own DHCP/DNS servers into > > > modules fairly cleanly if they use something different, without having > > > to have a bunch of "ifs" in the sync code. > > > > > > Folks are welcome to try this now, though there are no doubt a few bugs > > > which we will be ironing out next week. > > > > > > The Wiki needs to be updated also to include more docs on this and > > > sample template files. I'll also probably be moving > > > more of the static HTML from cobbler.et.redhat.com to the Wiki so people > > > can update it when they come up with > > > interesting tricks as well as corrections. > > > > > > --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 lippold at gmail.com Mon May 5 02:35:49 2008 From: lippold at gmail.com (Aaron Lippold) Date: Sun, 4 May 2008 22:35:49 -0400 Subject: [et-mgmt-tools] Installing without PXE, and without the live CD -- "cobbler buildiso" In-Reply-To: <481B2553.3030609@redhat.com> References: <481B2553.3030609@redhat.com> Message-ID: <39d2723b0805041935y381c6d0dj39d94d7ba0a26158@mail.gmail.com> Hi Michael, This is a great feature! Thanks for putting the work in on this. I am working on the apache lockdown items for my organization and look forward to building a "Live Provisioning CD" that meets our requirements. Thanks again! Yours, Aaron On Fri, May 2, 2008 at 10:29 AM, Michael DeHaan wrote: > For a while Cobbler has had a solution based on koan and the live CD to > enable network provisioning of bare metal in environments that either do not > have DHCP, or do not have control over DHCP that is sufficient enough to set > up a PXE environment. Naturally, if you can set up a PXE environment with > Cobbler, it's very useful to have, and you'll want it. > > So, along came the live CD. > > The live CD was built using a recent Fedora, and it had the magic ability > to install any distro -- /except/ you had to use Fedora to build the Live > CD, and the build process was slow, and the live image was a bit bigger than > it needed to be. However, due to driver constraints, it didn't always > support the latest in storage technology -- that was a problem. > > We have finally implemented the low-tech solution, thanks to various folks > in Red Hat GPS, and some hacking I've done to integrate that closer into > cobbler. > > Using 0.9.X or later (it's checked in now into the git "devel" branch) you > can do: > > cobbler buildiso [--iso=] [--tempdir=] [--profiles=] > > This will automatically generate an ISO that allows for menus just like > Cobbler's PXE boot menus, for installing new systems. I also plan to allow > a configurable default profile for those who want to mass deploy this image > using remote management processors, etc. > > The menu will contain an entry for each bare metal bootable profile, with > the current data set in Cobbler. Given that Cobbler now generates > kickstarts in real time, changes to the kickstart templates can be made > without reburning the CD! > > Again, the live CD is more dynamic, but this is much smaller, faster, and > easier to build. > The one thing the live CD still offers is simulation of the MAC address > detection feature of PXE, but if you don't need to use cobbler system > records to provision your lab/datacenter/etc, this will also get you there > -- and is probably the perfect fit if the Live CD was not working out for > you. > > Test release will be available soon, git is available now. > > --Michael > > > > _______________________________________________ > 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 caglar at pardus.org.tr Mon May 5 14:31:50 2008 From: caglar at pardus.org.tr (=?utf-8?q?S=2E=C3=87a=C4=9Flar?= Onur) Date: Mon, 5 May 2008 17:31:50 +0300 Subject: [et-mgmt-tools] [PATCH] Use org.freedesktop.PolicyKit.AuthenticationAgent for PolicyKit Message-ID: <200805051731.51870.caglar@pardus.org.tr> Hi; Following patch replaces deprecated "ShowDialog" method with "ObtainAuthorization" one and also changes desktop dependent "org.gnome.PolicyKit" interface with "org.freedesktop.PolicyKit.AuthenticationAgent" one to permit other PolicyKit Authentication Agents (like PolicyKit-KDE). Signed-off-by: S.?a?lar Onur diff -r 5142197392f2 src/virtManager/connection.py --- a/src/virtManager/connection.py Fri May 02 09:36:36 2008 -0400 +++ b/src/virtManager/connection.py Mon May 05 17:21:05 2008 +0300 @@ -339,9 +339,9 @@ def _do_creds_polkit(self, action): logging.debug("Doing policykit for %s" % action) bus = dbus.SessionBus() - obj = bus.get_object("org.gnome.PolicyKit", "/org/gnome/PolicyKit/Manager") - pkit = dbus.Interface(obj, "org.gnome.PolicyKit.Manager") - pkit.ShowDialog(action, 0) + obj = bus.get_object("org.freedesktop.PolicyKit.AuthenticationAgent", "/") + pkit = dbus.Interface(obj, "org.freedesktop.PolicyKit.AuthenticationAgent") + pkit.ObtainAuthorization(action, 0, os.getpid()) return 0 def _do_creds_dialog(self, creds): Cheers -- S.?a?lar Onur http://cekirdek.pardus.org.tr/~caglar/ Linux is like living in a teepee. No Windows, no Gates and an Apache in house! From mdehaan at redhat.com Mon May 5 14:47:38 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 05 May 2008 10:47:38 -0400 Subject: [et-mgmt-tools] Installing without PXE, and without the live CD -- "cobbler buildiso" In-Reply-To: <481B8835.7040407@redhat.com> References: <481B2553.3030609@redhat.com> <481B8835.7040407@redhat.com> Message-ID: <481F1E0A.8010501@redhat.com> Subhendu Ghosh wrote: > Michael DeHaan wrote: >> For a while Cobbler has had a solution based on koan and the live CD >> to enable network provisioning of bare metal in environments that >> either do not have DHCP, or do not have control over DHCP that is >> sufficient enough to set up a PXE environment. Naturally, if you can >> set up a PXE environment with Cobbler, it's very useful to have, and >> you'll want it. >> >> So, along came the live CD. >> >> The live CD was built using a recent Fedora, and it had the magic >> ability to install any distro -- /except/ you had to use Fedora to >> build the Live CD, and the build process was slow, and the live image >> was a bit bigger than it needed to be. However, due to driver >> constraints, it didn't always support the latest in storage >> technology -- that was a > problem. >> >> We have finally implemented the low-tech solution, thanks to various >> folks in Red Hat GPS, and some hacking I've done to integrate that >> closer into cobbler. >> >> Using 0.9.X or later (it's checked in now into the git "devel" >> branch) you can do: >> >> cobbler buildiso [--iso=] [--tempdir=] [--profiles=] >> >> This will automatically generate an ISO that allows for menus just >> like Cobbler's PXE boot menus, for installing new systems. I also >> plan to allow a configurable default profile for those who want to >> mass deploy this image using remote management processors, etc. >> >> The menu will contain an entry for each bare metal bootable profile, >> with the current data set in Cobbler. Given that Cobbler now >> generates kickstarts in real time, changes to the kickstart templates >> can be made without reburning the CD! >> >> Again, the live CD is more dynamic, but this is much smaller, faster, >> and easier to build. >> The one thing the live CD still offers is simulation of the MAC >> address detection feature of PXE, but if you don't need to use >> cobbler system records to provision your lab/datacenter/etc, this >> will also get you there -- and is probably the perfect fit if the >> Live CD was not working out for you. >> >> Test release will be available soon, git is available now. >> >> --Michael >> > > If the iso(s) can be made available through the web infrastructure, > and accessible via a predefined or calculated path, then a external > rules engine could drive the provisioning process through the remote > management processors's virtual media capabilities. > It can :) mkdir /var/www/html/foo cobbler buildiso --iso=/var/www/html/foo/kickstart.iso I just need add the flag for setting what the default is and making the timeouts variable. > Perhaps the profiles module could be extended to define whether a iso > should be generated or not. An easy way to automatically generate boot > isos. > Right now there's some basic code to keep paravirt "-xen" distros from showing up on the ISO since they have no chance of being bootable. Though that is a good idea. Another possibility is to add it as "cobbler profile edit --name=foo ... --menu=foo" to do submenus, and if we really wanted that. We could allow --menu=hide as part of that syntax. > -sg > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From jeckersb at redhat.com Mon May 5 16:29:13 2008 From: jeckersb at redhat.com (John Eckersberg) Date: Mon, 05 May 2008 12:29:13 -0400 Subject: [et-mgmt-tools] [patch] Add zone-specific templating for BIND Message-ID: <481F35D9.107@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 http://jeckersb.fedorapeople.org/cobbler/0001-add-zone-specific-bind-templating.patch This most importantly (at least to me :D) lets you define things like MX/CNAME/SRV/TXT records for your zones. Yay. - -- John Eckersberg Production Operations Red Hat, Inc. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFIHzXZO5VSO/Qqm+gRAgarAJ9AXYTzTCPUVSKtdyyax/p6zZhbZgCeLq+a r9LUxMyBrgJOSKZo4NCzVjo= =9u0h -----END PGP SIGNATURE----- From mdehaan at redhat.com Mon May 5 16:43:24 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 05 May 2008 12:43:24 -0400 Subject: [et-mgmt-tools] [patch] Add zone-specific templating for BIND In-Reply-To: <481F35D9.107@redhat.com> References: <481F35D9.107@redhat.com> Message-ID: <481F392C.8020000@redhat.com> John Eckersberg wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > http://jeckersb.fedorapeople.org/cobbler/0001-add-zone-specific-bind-templating.patch > > > This most importantly (at least to me :D) lets you define things like > MX/CNAME/SRV/TXT records for your zones. Yay. > > - -- > John Eckersberg > Production Operations > Red Hat, Inc. > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.7 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iD8DBQFIHzXZO5VSO/Qqm+gRAgarAJ9AXYTzTCPUVSKtdyyax/p6zZhbZgCeLq+a > r9LUxMyBrgJOSKZo4NCzVjo= > =9u0h > -----END PGP SIGNATURE----- > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools John, This looks good -- make sure the Wiki info is also updated -- if you can share examples there that would be great. --Michael From jeckersb at redhat.com Mon May 5 17:49:40 2008 From: jeckersb at redhat.com (John Eckersberg) Date: Mon, 05 May 2008 13:49:40 -0400 Subject: [et-mgmt-tools] [patch] Add zone-specific templating for BIND In-Reply-To: <481F392C.8020000@redhat.com> References: <481F35D9.107@redhat.com> <481F392C.8020000@redhat.com> Message-ID: <481F48B4.70502@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Michael DeHaan wrote: | John Eckersberg wrote: | http://jeckersb.fedorapeople.org/cobbler/0001-add-zone-specific-bind-templating.patch | | | This most importantly (at least to me :D) lets you define things like | MX/CNAME/SRV/TXT records for your zones. Yay. | |> _______________________________________________ et-mgmt-tools mailing list et-mgmt-tools at redhat.com https://www.redhat.com/mailman/listinfo/et-mgmt-tools | John, | This looks good -- make sure the Wiki info is also updated -- if you can | share examples there that would be great. | --Michael | _______________________________________________ | et-mgmt-tools mailing list | et-mgmt-tools at redhat.com | https://www.redhat.com/mailman/listinfo/et-mgmt-tools Done: https://fedorahosted.org/cobbler/wiki/ManageDns - -- John Eckersberg Production Operations Red Hat, Inc. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFIH0i0O5VSO/Qqm+gRAg8vAJ4iaZH/VDVi6NE0CK8CsUq8B3urXgCgkGOz UHyDe2P7FaK+IrbmAjW5oAU= =O+R2 -----END PGP SIGNATURE----- From berrange at redhat.com Mon May 5 19:54:43 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 5 May 2008 20:54:43 +0100 Subject: [et-mgmt-tools] [PATCH] Use org.freedesktop.PolicyKit.AuthenticationAgent for PolicyKit In-Reply-To: <200805051731.51870.caglar@pardus.org.tr> References: <200805051731.51870.caglar@pardus.org.tr> Message-ID: <20080505195443.GG5881@redhat.com> On Mon, May 05, 2008 at 05:31:50PM +0300, S.?a??lar Onur wrote: > Hi; > > Following patch replaces deprecated "ShowDialog" method with > "ObtainAuthorization" one and also changes desktop dependent > "org.gnome.PolicyKit" interface with > "org.freedesktop.PolicyKit.AuthenticationAgent" one to permit > other PolicyKit Authentication Agents (like PolicyKit-KDE). > > Signed-off-by: S.?a??lar Onur I'm fine with this in principle, however, it needs to fallback to looking for "org.gnome.PolicyKit" if "org.freedesktop.PolicyKit.AuthenticationAgent" is not found, because the latter was only introduced in version 0.7 of PolicyKit Regards, Daniel -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From caglar at pardus.org.tr Mon May 5 19:57:13 2008 From: caglar at pardus.org.tr (=?utf-8?q?S=2E=C3=87a=C4=9Flar?= Onur) Date: Mon, 5 May 2008 22:57:13 +0300 Subject: [et-mgmt-tools] [PATCH] Use org.freedesktop.PolicyKit.AuthenticationAgent for PolicyKit In-Reply-To: <20080505195443.GG5881@redhat.com> References: <200805051731.51870.caglar@pardus.org.tr> <20080505195443.GG5881@redhat.com> Message-ID: <200805052257.15945.caglar@pardus.org.tr> Hi; 05 May 2008 Pts tarihinde, Daniel P. Berrange ?unlar? yazm??t?: > I'm fine with this in principle, however, it needs to fallback to looking > for "org.gnome.PolicyKit" if "org.freedesktop.PolicyKit.AuthenticationAgent" > is not found, because the latter was only introduced in version 0.7 of > PolicyKit Fair enough :), i'll update the patch with fallback and will re-send... > Regards, > Daniel Cheers -- S.?a?lar Onur http://cekirdek.pardus.org.tr/~caglar/ Linux is like living in a teepee. No Windows, no Gates and an Apache in house! -------------- 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 caglar at pardus.org.tr Mon May 5 20:24:40 2008 From: caglar at pardus.org.tr (=?utf-8?q?S=2E=C3=87a=C4=9Flar?= Onur) Date: Mon, 5 May 2008 23:24:40 +0300 Subject: [et-mgmt-tools] [PATCH] Use org.freedesktop.PolicyKit.AuthenticationAgent for PolicyKit In-Reply-To: <200805052257.15945.caglar@pardus.org.tr> References: <200805051731.51870.caglar@pardus.org.tr> <20080505195443.GG5881@redhat.com> <200805052257.15945.caglar@pardus.org.tr> Message-ID: <200805052324.41117.caglar@pardus.org.tr> 05 May 2008 Pts tarihinde, S.?a?lar Onur ?unlar? yazm??t?: > 05 May 2008 Pts tarihinde, Daniel P. Berrange ?unlar? yazm??t?: > > I'm fine with this in principle, however, it needs to fallback to looking > > for "org.gnome.PolicyKit" if "org.freedesktop.PolicyKit.AuthenticationAgent" > > is not found, because the latter was only introduced in version 0.7 of > > PolicyKit > > Fair enough :), i'll update the patch with fallback and will re-send... Following patch introduce usage of "org.freedesktop.PolicyKit.AuthenticationAgent" interface of PolicyKit to permit other PolicyKit Authentication Agents (like PolicyKit-KDE). If virt-manager runs with PolicyKit which not provides that interface, it fallbacks to old one as suggested by Daniel P. Berrange. Signed-off-by: S.?a?lar Onur diff -r 5142197392f2 src/virtManager/connection.py --- a/src/virtManager/connection.py Fri May 02 09:36:36 2008 -0400 +++ b/src/virtManager/connection.py Mon May 05 23:19:25 2008 +0300 @@ -339,9 +339,17 @@ def _do_creds_polkit(self, action): logging.debug("Doing policykit for %s" % action) bus = dbus.SessionBus() - obj = bus.get_object("org.gnome.PolicyKit", "/org/gnome/PolicyKit/Manager") - pkit = dbus.Interface(obj, "org.gnome.PolicyKit.Manager") - pkit.ShowDialog(action, 0) + + try: + # First try to use org.freedesktop.PolicyKit.AuthenticationAgent which is introduced with PolicyKit-0.7 + obj = bus.get_object("org.freedesktop.PolicyKit.AuthenticationAgent", "/") + pkit = dbus.Interface(obj, "org.freedesktop.PolicyKit.AuthenticationAgent") + pkit.ObtainAuthorization(action, 0, os.getpid()) + except org.freedesktop.DBus.Error.ServiceUnknown: + # If PolicyKit < 0.7, fallback to org.gnome.PolicyKit + obj = bus.get_object("org.gnome.PolicyKit", "/org/gnome/PolicyKit/Manager") + pkit = dbus.Interface(obj, "org.gnome.PolicyKit.Manager") + pkit.ShowDialog(action, 0) return 0 def _do_creds_dialog(self, creds): Cheers -- S.?a?lar Onur http://cekirdek.pardus.org.tr/~caglar/ Linux is like living in a teepee. No Windows, no Gates and an Apache in house! From berrange at redhat.com Mon May 5 23:38:47 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 6 May 2008 00:38:47 +0100 Subject: [et-mgmt-tools] [PATCH] Use org.freedesktop.PolicyKit.AuthenticationAgent for PolicyKit In-Reply-To: <200805052324.41117.caglar@pardus.org.tr> References: <200805051731.51870.caglar@pardus.org.tr> <20080505195443.GG5881@redhat.com> <200805052257.15945.caglar@pardus.org.tr> <200805052324.41117.caglar@pardus.org.tr> Message-ID: <20080505233847.GA17932@redhat.com> On Mon, May 05, 2008 at 11:24:40PM +0300, S.?a??lar Onur wrote: > 05 May 2008 Pts tarihinde, S.?a??lar Onur ??unlar?? yazm????t??: > > 05 May 2008 Pts tarihinde, Daniel P. Berrange ??unlar?? yazm????t??: > > > I'm fine with this in principle, however, it needs to fallback to looking > > > for "org.gnome.PolicyKit" if "org.freedesktop.PolicyKit.AuthenticationAgent" > > > is not found, because the latter was only introduced in version 0.7 of > > > PolicyKit > > > > Fair enough :), i'll update the patch with fallback and will re-send... > > Following patch introduce usage of "org.freedesktop.PolicyKit.AuthenticationAgent" > interface of PolicyKit to permit other PolicyKit Authentication Agents (like 5B > PolicyKit-KDE). If virt-manager runs with PolicyKit which not provides that > interface, it fallbacks to old one as suggested by Daniel P. Berrange. > > Signed-off-by: S.?a??lar Onur Thanks, I have committed this patch to the repo Regards, Daniel -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From lippold at gmail.com Tue May 6 03:34:55 2008 From: lippold at gmail.com (Aaron Lippold) Date: Mon, 5 May 2008 23:34:55 -0400 Subject: [et-mgmt-tools] Installing without PXE, and without the live CD -- "cobbler buildiso" In-Reply-To: <481F1E0A.8010501@redhat.com> References: <481B2553.3030609@redhat.com> <481B8835.7040407@redhat.com> <481F1E0A.8010501@redhat.com> Message-ID: <39d2723b0805052034g1e54fccaj9850e3155a7623dd@mail.gmail.com> Hi, One other option that would be useful is what if my cobbler server has a whole lot of profiles but I only want a few or the situation where - since we have module ownership now - I only want "my profiles" on the iso. >From what I read before, createiso basically mirrors the cobbler server onto the livecd right? ( if error, please correct ) That's cool, but what if I have a whole lot of media - like 10 distros - I may only want to put my "Fedora X x86 desktop profiles" on the cd right? Secondly, if I did have 10 distros, no way that all fits on my one dvd... Also, this might be a dumb question but does the livecd know how to "install to self" as well? Yours, Aaron On Mon, May 5, 2008 at 10:47 AM, Michael DeHaan wrote: > Subhendu Ghosh wrote: > > > Michael DeHaan wrote: > > > > > For a while Cobbler has had a solution based on koan and the live CD > > > to enable network provisioning of bare metal in environments that either do > > > not have DHCP, or do not have control over DHCP that is sufficient enough to > > > set up a PXE environment. Naturally, if you can set up a PXE environment > > > with Cobbler, it's very useful to have, and you'll want it. > > > > > > So, along came the live CD. > > > > > > The live CD was built using a recent Fedora, and it had the magic > > > ability to install any distro -- /except/ you had to use Fedora to build the > > > Live CD, and the build process was slow, and the live image was a bit bigger > > > than it needed to be. However, due to driver constraints, it didn't always > > > support the latest in storage technology -- that was a > > > > > problem. > > > > > > > > We have finally implemented the low-tech solution, thanks to various > > > folks in Red Hat GPS, and some hacking I've done to integrate that closer > > > into cobbler. > > > > > > Using 0.9.X or later (it's checked in now into the git "devel" branch) > > > you can do: > > > > > > cobbler buildiso [--iso=] [--tempdir=] [--profiles=] > > > > > > This will automatically generate an ISO that allows for menus just > > > like Cobbler's PXE boot menus, for installing new systems. I also plan to > > > allow a configurable default profile for those who want to mass deploy this > > > image using remote management processors, etc. > > > > > > The menu will contain an entry for each bare metal bootable profile, > > > with the current data set in Cobbler. Given that Cobbler now generates > > > kickstarts in real time, changes to the kickstart templates can be made > > > without reburning the CD! > > > > > > Again, the live CD is more dynamic, but this is much smaller, faster, > > > and easier to build. > > > The one thing the live CD still offers is simulation of the MAC > > > address detection feature of PXE, but if you don't need to use cobbler > > > system records to provision your lab/datacenter/etc, this will also get you > > > there -- and is probably the perfect fit if the Live CD was not working out > > > for you. > > > > > > Test release will be available soon, git is available now. > > > > > > --Michael > > > > > > > > If the iso(s) can be made available through the web infrastructure, and > > accessible via a predefined or calculated path, then a external rules engine > > could drive the provisioning process through the remote management > > processors's virtual media capabilities. > > > > It can :) > > mkdir /var/www/html/foo > cobbler buildiso --iso=/var/www/html/foo/kickstart.iso > I just need add the flag for setting what the default is and making the > timeouts variable. > > Perhaps the profiles module could be extended to define whether a iso > > should be generated or not. An easy way to automatically generate boot isos. > > > > > Right now there's some basic code to keep paravirt "-xen" distros from > showing up on the ISO since they have no chance of being bootable. Though > that is a good idea. > > Another possibility is to add it as "cobbler profile edit --name=foo ... > --menu=foo" to do submenus, and if we really wanted that. We could allow > --menu=hide as part of that syntax. > > -sg > > > > _______________________________________________ > > 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 > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Pablo.Iranzo at redhat.com Tue May 6 06:02:56 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Tue, 06 May 2008 06:02:56 +0000 Subject: [et-mgmt-tools] Installing without PXE, and without the live CD -- "cobbler buildiso" In-Reply-To: <39d2723b0805052034g1e54fccaj9850e3155a7623dd@mail.gmail.com> References: <481B2553.3030609@redhat.com> <481B8835.7040407@redhat.com> <481F1E0A.8010501@redhat.com> <39d2723b0805052034g1e54fccaj9850e3155a7623dd@mail.gmail.com> Message-ID: <1210053776.8766.164.camel@iranzo.users.redhat.com> Aaron, "buildiso" just copies all kernel and initrds not being for xen in an ISO for you to burn it and boot your machines, so you've plenty space for many many AFAIK it doesn't use the --user to show only profiles owned by you, but will need to test latest git code. Livecd != buildiso Regards Pablo El lun, 05-05-2008 a las 23:34 -0400, Aaron Lippold escribi?: > Hi, > > One other option that would be useful is what if my cobbler server has > a whole lot of profiles but I only want a few or the situation where - > since we have module ownership now - I only want "my profiles" on the > iso. > > From what I read before, createiso basically mirrors the cobbler > server onto the livecd right? ( if error, please correct ) That's > cool, but what if I have a whole lot of media - like 10 distros - I > may only want to put my "Fedora X x86 desktop profiles" on the cd > right? Secondly, if I did have 10 distros, no way that all fits on my > one dvd... > > Also, this might be a dumb question but does the livecd know how to > "install to self" as well? > > Yours, > > Aaron > > On Mon, May 5, 2008 at 10:47 AM, Michael DeHaan > wrote: > > Subhendu Ghosh wrote: > Michael DeHaan wrote: > For a while Cobbler has had a solution based > on koan and the live CD to enable network > provisioning of bare metal in environments > that either do not have DHCP, or do not have > control over DHCP that is sufficient enough to > set up a PXE environment. Naturally, if you > can set up a PXE environment with Cobbler, > it's very useful to have, and you'll want it. > > So, along came the live CD. > > The live CD was built using a recent Fedora, > and it had the magic ability to install any > distro -- /except/ you had to use Fedora to > build the Live CD, and the build process was > slow, and the live image was a bit bigger than > it needed to be. However, due to driver > constraints, it didn't always support the > latest in storage technology -- that was a > problem. > > We have finally implemented the low-tech > solution, thanks to various folks in Red Hat > GPS, and some hacking I've done to integrate > that closer into cobbler. > > Using 0.9.X or later (it's checked in now into > the git "devel" branch) you can do: > > cobbler buildiso [--iso=] [--tempdir=] > [--profiles=] > > This will automatically generate an ISO that > allows for menus just like Cobbler's PXE boot > menus, for installing new systems. I also > plan to allow a configurable default profile > for those who want to mass deploy this image > using remote management processors, etc. > > The menu will contain an entry for each bare > metal bootable profile, with the current data > set in Cobbler. Given that Cobbler now > generates kickstarts in real time, changes to > the kickstart templates can be made without > reburning the CD! > > Again, the live CD is more dynamic, but this > is much smaller, faster, and easier to build. > The one thing the live CD still offers is > simulation of the MAC address detection > feature of PXE, but if you don't need to use > cobbler system records to provision your > lab/datacenter/etc, this will also get you > there -- and is probably the perfect fit if > the Live CD was not working out for you. > > Test release will be available soon, git is > available now. > > --Michael > > > If the iso(s) can be made available through the web > infrastructure, and accessible via a predefined or > calculated path, then a external rules engine could > drive the provisioning process through the remote > management processors's virtual media capabilities. > > It can :) > > mkdir /var/www/html/foo > cobbler buildiso --iso=/var/www/html/foo/kickstart.iso > I just need add the flag for setting what the default is and > making the timeouts variable. > > > Perhaps the profiles module could be extended to > define whether a iso should be generated or not. An > easy way to automatically generate boot isos. > > > > Right now there's some basic code to keep paravirt "-xen" > distros from showing up on the ISO since they have no chance > of being bootable. Though that is a good idea. > > Another possibility is to add it as "cobbler profile edit > --name=foo ... --menu=foo" to do submenus, and if we really > wanted that. We could allow > --menu=hide as part of that syntax. > > -sg > > > _______________________________________________ > 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 > > > _______________________________________________ > 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 mdehaan at redhat.com Tue May 6 14:18:55 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 06 May 2008 10:18:55 -0400 Subject: [et-mgmt-tools] Installing without PXE, and without the live CD -- "cobbler buildiso" In-Reply-To: <1210053776.8766.164.camel@iranzo.users.redhat.com> References: <481B2553.3030609@redhat.com> <481B8835.7040407@redhat.com> <481F1E0A.8010501@redhat.com> <39d2723b0805052034g1e54fccaj9850e3155a7623dd@mail.gmail.com> <1210053776.8766.164.camel@iranzo.users.redhat.com> Message-ID: <482068CF.9080307@redhat.com> Pablo Iranzo G?mez wrote: > Aaron, > "buildiso" just copies all kernel and initrds not being for xen in an > ISO for you to burn it and boot your machines, so you've plenty space > for many many > Indeed you do, although if you want to add only certain profiles, that is what I have in mind regarding: cobbler buildiso --profiles=a,b,c,d,e Though this is only really necessary if you are making images for, say, 32 mb USB sticks from 1997 :) Putting a single distro on a CD makes a 24 Mb image or so. Also, if multiple profiles share the same distro, the kernels/initrds are only copied once for each distro, not once for each profile. So this is really a non-issue. > AFAIK it doesn't use the --user to show only profiles owned by you, but > will need to test latest git code. > Ownership affects who can make changes to objects using the WebUI, but anyone can always deploy any profile. > > Livecd != buildiso > > Regards > Pablo > > El lun, 05-05-2008 a las 23:34 -0400, Aaron Lippold escribi?: > >> Hi, >> >> One other option that would be useful is what if my cobbler server has >> a whole lot of profiles but I only want a few or the situation where - >> since we have module ownership now - I only want "my profiles" on the >> iso. >> >> From what I read before, createiso basically mirrors the cobbler >> server onto the livecd right? ( if error, please correct ) That's >> cool, but what if I have a whole lot of media - like 10 distros - I >> may only want to put my "Fedora X x86 desktop profiles" on the cd >> right? Secondly, if I did have 10 distros, no way that all fits on my >> one dvd... >> >> Also, this might be a dumb question but does the livecd know how to >> "install to self" as well? >> >> Yours, >> >> Aaron >> >> On Mon, May 5, 2008 at 10:47 AM, Michael DeHaan >> wrote: >> >> Subhendu Ghosh wrote: >> Michael DeHaan wrote: >> For a while Cobbler has had a solution based >> on koan and the live CD to enable network >> provisioning of bare metal in environments >> that either do not have DHCP, or do not have >> control over DHCP that is sufficient enough to >> set up a PXE environment. Naturally, if you >> can set up a PXE environment with Cobbler, >> it's very useful to have, and you'll want it. >> >> So, along came the live CD. >> >> The live CD was built using a recent Fedora, >> and it had the magic ability to install any >> distro -- /except/ you had to use Fedora to >> build the Live CD, and the build process was >> slow, and the live image was a bit bigger than >> it needed to be. However, due to driver >> constraints, it didn't always support the >> latest in storage technology -- that was a >> problem. >> >> We have finally implemented the low-tech >> solution, thanks to various folks in Red Hat >> GPS, and some hacking I've done to integrate >> that closer into cobbler. >> >> Using 0.9.X or later (it's checked in now into >> the git "devel" branch) you can do: >> >> cobbler buildiso [--iso=] [--tempdir=] >> [--profiles=] >> >> This will automatically generate an ISO that >> allows for menus just like Cobbler's PXE boot >> menus, for installing new systems. I also >> plan to allow a configurable default profile >> for those who want to mass deploy this image >> using remote management processors, etc. >> >> The menu will contain an entry for each bare >> metal bootable profile, with the current data >> set in Cobbler. Given that Cobbler now >> generates kickstarts in real time, changes to >> the kickstart templates can be made without >> reburning the CD! >> >> Again, the live CD is more dynamic, but this >> is much smaller, faster, and easier to build. >> The one thing the live CD still offers is >> simulation of the MAC address detection >> feature of PXE, but if you don't need to use >> cobbler system records to provision your >> lab/datacenter/etc, this will also get you >> there -- and is probably the perfect fit if >> the Live CD was not working out for you. >> >> Test release will be available soon, git is >> available now. >> >> --Michael >> >> >> If the iso(s) can be made available through the web >> infrastructure, and accessible via a predefined or >> calculated path, then a external rules engine could >> drive the provisioning process through the remote >> management processors's virtual media capabilities. >> >> It can :) >> >> mkdir /var/www/html/foo >> cobbler buildiso --iso=/var/www/html/foo/kickstart.iso >> I just need add the flag for setting what the default is and >> making the timeouts variable. >> >> >> Perhaps the profiles module could be extended to >> define whether a iso should be generated or not. An >> easy way to automatically generate boot isos. >> >> >> >> Right now there's some basic code to keep paravirt "-xen" >> distros from showing up on the ISO since they have no chance >> of being bootable. Though that is a good idea. >> >> Another possibility is to add it as "cobbler profile edit >> --name=foo ... --menu=foo" to do submenus, and if we really >> wanted that. We could allow >> --menu=hide as part of that syntax. >> >> -sg >> >> >> _______________________________________________ >> 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 >> >> >> _______________________________________________ >> 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 May 6 14:23:54 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 06 May 2008 10:23:54 -0400 Subject: [et-mgmt-tools] Installing without PXE, and without the live CD -- "cobbler buildiso" In-Reply-To: <39d2723b0805052034g1e54fccaj9850e3155a7623dd@mail.gmail.com> References: <481B2553.3030609@redhat.com> <481B8835.7040407@redhat.com> <481F1E0A.8010501@redhat.com> <39d2723b0805052034g1e54fccaj9850e3155a7623dd@mail.gmail.com> Message-ID: <482069FA.3000205@redhat.com> > > Also, this might be a dumb question but does the livecd know how to > "install to self" as well? The live CD (not to be confused with "cobbler buildiso", which is a simpler "dead" CD) works by detecting the first available HDD during boot up, then formats the drive and installs grub on it's MBR. Koan is then invoked to modify grub on the MBR. The livecd can be built to install to a specific profile name (as defined in Cobbler at the time of installation, not build time), or can be set to autodiscover. In this case, koan scans the interfaces for a mac address, and does a remote look up in Cobbler (via XMLRPC) to find the name of the matching Cobbler system object, if any. This way, the live CD simulates the assignment of a given system (by MAC address) to a given profile, in just the same way that PXE works. However, this requires building the live CD on F8, and is not as tolerant of finding "new" hardware (such as my SAS card), and the "dead" CD will always work in those scenarios -- you just won't have the potential for auto-discovery. The question then remains do we keep the live CD around, and just use the simpler "dead" CD, or what... the likely answer is that we end up having both, and explain that for 90% of the users, the "dead" CD is fine, and the live one is only needed if you want to do assignments between systems and profiles remotely, to achieve "PXE equivalence" in that respect. That is probably not critical for most folks, but might be nice to have. > > Yours, > > Aaron > > On Mon, May 5, 2008 at 10:47 AM, Michael DeHaan > wrote: > > Subhendu Ghosh wrote: > > Michael DeHaan wrote: > > For a while Cobbler has had a solution based on koan and > the live CD to enable network provisioning of bare metal > in environments that either do not have DHCP, or do not > have control over DHCP that is sufficient enough to set up > a PXE environment. Naturally, if you can set up a PXE > environment with Cobbler, it's very useful to have, and > you'll want it. > > So, along came the live CD. > > The live CD was built using a recent Fedora, and it had > the magic ability to install any distro -- /except/ you > had to use Fedora to build the Live CD, and the build > process was slow, and the live image was a bit bigger than > it needed to be. However, due to driver constraints, it > didn't always support the latest in storage technology -- > that was a > > problem. > > > We have finally implemented the low-tech solution, thanks > to various folks in Red Hat GPS, and some hacking I've > done to integrate that closer into cobbler. > > Using 0.9.X or later (it's checked in now into the git > "devel" branch) you can do: > > cobbler buildiso [--iso=] [--tempdir=] [--profiles=] > > This will automatically generate an ISO that allows for > menus just like Cobbler's PXE boot menus, for installing > new systems. I also plan to allow a configurable default > profile for those who want to mass deploy this image using > remote management processors, etc. > > The menu will contain an entry for each bare metal > bootable profile, with the current data set in Cobbler. > Given that Cobbler now generates kickstarts in real time, > changes to the kickstart templates can be made without > reburning the CD! > > Again, the live CD is more dynamic, but this is much > smaller, faster, and easier to build. > The one thing the live CD still offers is simulation of > the MAC address detection feature of PXE, but if you don't > need to use cobbler system records to provision your > lab/datacenter/etc, this will also get you there -- and is > probably the perfect fit if the Live CD was not working > out for you. > > Test release will be available soon, git is available now. > > --Michael > > > If the iso(s) can be made available through the web > infrastructure, and accessible via a predefined or calculated > path, then a external rules engine could drive the > provisioning process through the remote management > processors's virtual media capabilities. > > It can :) > > mkdir /var/www/html/foo > cobbler buildiso --iso=/var/www/html/foo/kickstart.iso > I just need add the flag for setting what the default is and > making the timeouts variable. > > > Perhaps the profiles module could be extended to define > whether a iso should be generated or not. An easy way to > automatically generate boot isos. > > > Right now there's some basic code to keep paravirt "-xen" distros > from showing up on the ISO since they have no chance of being > bootable. Though that is a good idea. > > Another possibility is to add it as "cobbler profile edit > --name=foo ... --menu=foo" to do submenus, and if we really wanted > that. We could allow > --menu=hide as part of that syntax. > > -sg > > > _______________________________________________ > 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 > > > ------------------------------------------------------------------------ > > _______________________________________________ > 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 Wed May 7 15:49:46 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 7 May 2008 16:49:46 +0100 Subject: [et-mgmt-tools] FYI: major UI changes in virt-manager Message-ID: <20080507154946.GG2963@redhat.com> I've just pushed a significant UI change to virt-manager. The per-VM details and console windows are now merged into a single tabbed window. The console is the primary tab, and overview/details tabs are as before. The goal is to reduce the number of windows on the desktop because with large numbers of VMs having two windows per VM was non-scalable. Basically we want at most the one main window, and then 1 window per hypervisor connection and 1 window per virtual machine. I expect we'll iterate over this UI change to tweak a few more things before the next release. In particular wrt to fullscreen mode. Eventually we'll hook up serial/parallel ports as additional tabs on the VM window too, so everything's in one place. Regards, Daniel. -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From berrange at redhat.com Wed May 7 15:59:39 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Wed, 7 May 2008 16:59:39 +0100 Subject: [et-mgmt-tools] FYI: major UI changes in virt-manager In-Reply-To: <539e5b310805070855r1e53bcc0m9b66bd06a4a35997@mail.gmail.com> References: <20080507154946.GG2963@redhat.com> <539e5b310805070855r1e53bcc0m9b66bd06a4a35997@mail.gmail.com> Message-ID: <20080507155939.GH2963@redhat.com> On Wed, May 07, 2008 at 05:55:19PM +0200, Maciej Borzecki wrote: > Hello Daniel, > > Is snapshot manager a planned feature? > Something similar to snapshots tab in virtualbox would be great... Snapshots are not an easy thing to provide unless you constrain which kind of storage a guest is allowed to use. I've no plans to work on this in the near future because there's more important things which are easier first. Dan. -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From mdehaan at redhat.com Wed May 7 22:35:53 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 07 May 2008 18:35:53 -0400 Subject: [et-mgmt-tools] Summary of things new in Cobbler 0.9/1.0 Message-ID: <48222EC9.3090200@redhat.com> I have a write up of what's new in Cobbler/Koan 0.9/1.0 here: https://fedorahosted.org/cobbler/wiki/WhatsNewInOhNine Comments welcome, and feel free to play with git if you like. I will be doing a test release likely this week (Friday?) with a source RPM, though you don't have to wait if you want to have at it. It's been a while coming, but this is a big release. It's probably time to stop and push new things into the next release when possible. Of possible interest to several folks: I had to (temporarily) remove the "--autonet" code in koan today (this is a feature intended to ease usage of --replace-self in DHCP-less environments), because of errors it was raising in Python 2.3 regarding some bit shifting and so forth. As I wasn't too familiar with the code and I don't have sufficient cycles to test it, it's out for now, but obviously we don't want to keep it that way if people need it. I will accept patches to get it cleaned up and added back -- though I'm not sure it worked on EL4 in the past. For patches, ideally it would not rely on direct parsing of commands and could get moved out to another file. Koan is getting big enough now to use some cleanup in general so I'll try to be doing some of that. Also of interest -- If you were asking about it, I believe I lost your email, but Xen Fullvirt in koan works again as I corrected some errors that kept it going into paravirt mode. Testing welcome. As indicated on the above Wiki page, there aren't too many features that are left to add. We're getting close! Thanks for everyone who has been providing testing so far, ideas, bug reports, and hanging out here and on IRC. You are much appreciated and I could not do this without you. Thanks! --Michael From jeckersb at redhat.com Thu May 8 01:26:30 2008 From: jeckersb at redhat.com (John Eckersberg) Date: Wed, 07 May 2008 21:26:30 -0400 Subject: [et-mgmt-tools] [PATCH] allow --mac=random to generate a random MAC address Message-ID: <482256C6.7040105@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 http://jeckersb.fedorapeople.org/cobbler/0001-allow-mac-random-to-generate-a-random-MAC-address.patch Pretty simple and self-explanatory. Only really noticeable thing is I took remote.CobblerXMLRPCInterface.get_random_mac and gutted it and moved the meat to utils.get_random_mac and then had the former just call the latter. I left the XMLRPC method since the webui calls into there and it doesn't have an api_handle lying around so I figured CobblerXMLRPCInterface can keep sharing :) - -- John Eckersberg Production Operations Red Hat, Inc. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFIIlbGO5VSO/Qqm+gRAv/RAJsGVuKU/ysS5PoFGIAS28PBL9R7WwCgtVgV +1pKxulN4xHb3KZfzzDr/pA= =tdx6 -----END PGP SIGNATURE----- From dimitri.henning at charite.de Thu May 8 07:17:03 2008 From: dimitri.henning at charite.de (Dimitri Henning) Date: Thu, 08 May 2008 09:17:03 +0200 Subject: [et-mgmt-tools] FYI: major UI changes in virt-manager In-Reply-To: <20080507154946.GG2963@redhat.com> References: <20080507154946.GG2963@redhat.com> Message-ID: <4822A8EF.3060203@charite.de> Daniel P. Berrange wrote: > Eventually we'll hook up serial/parallel ports as additional tabs on the > VM window too, so everything's in one place. > > Regards, > Daniel. > Is actually a great Idea. From tobert at gmail.com Thu May 8 08:07:23 2008 From: tobert at gmail.com (Al Tobey) Date: Thu, 8 May 2008 01:07:23 -0700 Subject: [et-mgmt-tools] Summary of things new in Cobbler 0.9/1.0 In-Reply-To: <48222EC9.3090200@redhat.com> References: <48222EC9.3090200@redhat.com> Message-ID: <5ac7acb10805080107x12cf8633h6dd4229176cd877e@mail.gmail.com> On Wed, May 7, 2008 at 3:35 PM, Michael DeHaan wrote: > Of possible interest to several folks: I had to (temporarily) remove the > "--autonet" code in koan today (this is a feature intended to ease usage of > --replace-self in DHCP-less environments), because of errors it was raising > in Python 2.3 regarding some bit shifting and so forth. As I wasn't too > familiar with the code and I don't have sufficient cycles to test it, it's > out for now, but obviously we don't want to keep it that way if people need > it. I will accept patches to get it cleaned up and added back -- though > I'm not sure it worked on EL4 in the past. For patches, ideally it would > not rely on direct parsing of commands and could get moved out to another > file. Koan is getting big enough now to use some cleanup in general so > I'll try to be doing some of that. That's too bad ... I was looking forward to not having to maintain a local version of koan any more. I'm not sure if it worked on EL4 since I mainly use koan -A for VM installs which run EL5 on dom0. My only request at this point is that either --ip, --netmask, and --gateway get added for setting network configuration on the kernel command line, or better yet a more generic option gets added, something like --cmdline "ip=x.x.x.x netmask=x.x.x.x gateway=x.x.x.x". That'd make it trivial to have a site-specific shell wrapper that sets up custom kernel command line arguments and solves a few other occasional problems (like pci=nommconf) in one fell swoop. Other neat stuff power users do could start working too. For example, koan --replace-self --cmdline "ksdevice=$macaddr" (handy for multi-homed Dell 1950's where you never know which interface will end up as eth0). I don't really have time to write the patch, or it'd be attached ;) It's a pretty simple modification though. -Al From mdehaan at redhat.com Thu May 8 13:24:49 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 08 May 2008 09:24:49 -0400 Subject: [et-mgmt-tools] [PATCH] allow --mac=random to generate a random MAC address In-Reply-To: <482256C6.7040105@redhat.com> References: <482256C6.7040105@redhat.com> Message-ID: <4822FF21.3030908@redhat.com> John Eckersberg wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > http://jeckersb.fedorapeople.org/cobbler/0001-allow-mac-random-to-generate-a-random-MAC-address.patch > > > Pretty simple and self-explanatory. > > Only really noticeable thing is I took > remote.CobblerXMLRPCInterface.get_random_mac and gutted it and moved the > meat to utils.get_random_mac and then had the former just call the > latter. I left the XMLRPC method since the webui calls into there and > it doesn't have an api_handle lying around so I figured > CobblerXMLRPCInterface can keep sharing :) Excellent.... I'll be applying this. Thanks! --Michael > > - -- > John Eckersberg > Production Operations > Red Hat, Inc. > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.7 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iD8DBQFIIlbGO5VSO/Qqm+gRAv/RAJsGVuKU/ysS5PoFGIAS28PBL9R7WwCgtVgV > +1pKxulN4xHb3KZfzzDr/pA= > =tdx6 > -----END PGP SIGNATURE----- > > _______________________________________________ > 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 May 8 15:57:01 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 08 May 2008 11:57:01 -0400 Subject: [et-mgmt-tools] Summary of things new in Cobbler 0.9/1.0 In-Reply-To: <5ac7acb10805080107x12cf8633h6dd4229176cd877e@mail.gmail.com> References: <48222EC9.3090200@redhat.com> <5ac7acb10805080107x12cf8633h6dd4229176cd877e@mail.gmail.com> Message-ID: <482322CD.5060600@redhat.com> Al Tobey wrote: > That's too bad ... I was looking forward to not having to maintain a > local version of koan any more. I'm not sure if it worked on EL4 > since I mainly use koan -A for VM installs which run EL5 on dom0. > Yep... I'm willing to have it back, it was just doing some math that I didn't have time to figure out and test. Basically the code in question returned different values for 2.3 and 2.5. > My only request at this point is that either --ip, --netmask, and > --gateway get added for setting network configuration on the kernel > command line, or better yet a more generic option gets added, > something like --cmdline "ip=x.x.x.x netmask=x.x.x.x gateway=x.x.x.x". > I like this. For consistancy, maybe --kopts to match cobbler. As there have been other requests for more cobbler overrides in koan, that can fit into that idea. Will add. From shenson at redhat.com Thu May 8 17:18:23 2008 From: shenson at redhat.com (Scott Henson) Date: Thu, 08 May 2008 13:18:23 -0400 Subject: [et-mgmt-tools] Cobbler replicate patch Message-ID: <482335DF.2030402@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I've added 'cobbler replicate' to replace cobbler_sync_master. The patch is available here[1]. Over all its pretty simple. You can add cobbler_master to your settings file and then call 'cobbler replicate'. It will then pull down all distro configs from the master and check if the kernel is already in its mirror store. If it is, it adds it. If not it tells you that its not here yet. It will then pull down any default profiles attached to a distro. You can also do 'cobbler replicate --master=cobbler.example.com' to pull from a different master than the one you have configured. I'm currently in the wiki adding some initial documentation for this feature. [1] http://shenson.fedorapeople.org/0001-cobbler-replicate-feature.patch - -- Scott Henson Red Hat Production Operations Release Engineer WVU Alum BSAE/BSME -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkgjNd8ACgkQcQOfHbe3DKAFRQCg4bXrgNl2FAni8kFvuAYpp+5T sTwAnjjlSLY99NnQ4Xr4teFlnqrr8EmA =Sw/d -----END PGP SIGNATURE----- From mdehaan at redhat.com Thu May 8 17:40:09 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 08 May 2008 13:40:09 -0400 Subject: [et-mgmt-tools] Cobbler replicate patch In-Reply-To: <482335DF.2030402@redhat.com> References: <482335DF.2030402@redhat.com> Message-ID: <48233AF9.7070602@redhat.com> Scott Henson wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I've added 'cobbler replicate' to replace cobbler_sync_master. The > patch is available here[1]. Over all its pretty simple. You can add > cobbler_master to your settings file and then call 'cobbler > replicate'. It will then pull down all distro configs from the master > and check if the kernel is already in its mirror store. If it is, it > adds it. If not it tells you that its not here yet. It will then > pull down any default profiles attached to a distro. > > You can also do 'cobbler replicate --master=cobbler.example.com' to > pull from a different master than the one you have configured. > > I'm currently in the wiki adding some initial documentation for this > feature. > > [1] http://shenson.fedorapeople.org/0001-cobbler-replicate-feature.patch Very cool. FYI -- Looks like this code no-ops when "cobbler_master" is set in settings and no --master is provided. Might want to make that fail. Applying this so it can get some testing out of git. I think it would be interesting to have code in there to optionally rsync content where appropriate (/var/www/cobbler/*mirror, /etc/cobbler, /var/lib/cobbler/snippets, /var/lib/cobbler/kickstarts) also for people who don't have other mirroring set up. Thanks! > > - -- > Scott Henson > Red Hat Production Operations Release Engineer > WVU Alum BSAE/BSME > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v2.0.9 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iEYEARECAAYFAkgjNd8ACgkQcQOfHbe3DKAFRQCg4bXrgNl2FAni8kFvuAYpp+5T > sTwAnjjlSLY99NnQ4Xr4teFlnqrr8EmA > =Sw/d > -----END PGP SIGNATURE----- > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From jeckersb at redhat.com Fri May 9 12:46:56 2008 From: jeckersb at redhat.com (John Eckersberg) Date: Fri, 09 May 2008 08:46:56 -0400 Subject: [et-mgmt-tools] [PATCH] fix zone config matching and minimize the number of zone files generated Message-ID: <482447C0.7030306@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 http://jeckersb.fedorapeople.org/cobbler/0001-fix-zone-config-matching-and-minimize-the-number-of.patch This fixes a few problems that I ran into as I sat down and actually started deploying my own feature :) Previously manage_forward_zones and manage_reverse_zones were optional and were supposed to restrict the zone files generated. Except it didn't really work all that well. For example... If my settings looked like this: manage_forward_zones: ~ - 'foo.example.com' Only hosts that matched 'hostname.foo.example.com' would go into the zone. If you had hostname.subdomain.foo.example.com it would not match. Furthermore if you did not specify manage_forward_zones then it would take every host, chop the hostname off the front and create a zone file for the remainder of the FQDN. As I started to use it I realized the number of zone files was quickly getting unmanageable. I have other nameservers to configure as slaves for these zones and I was needing to update them everytime somebody invented a new subdomain (which around here is about every 5 minutes :D). Which brings us to how things work now. manage_forward_zones and manage_reverse_zones are now mandatory for BIND management. The items listed in these values will directly map into zone files. Hosts will go into their "best fit" as far as zone files. To illustrate this: If my settings looked like this: manage_forward_zones: ~ - 'foo.example.com' ~ - 'bar.foo.example.com' The host 'baz.bar.foo.example.com' would go into the 'bar.foo.example.com' zone whereas the host 's1.foo.example.com' would go into the 'foo.example.com' zone. I hope everyone who is testing this out finds this to be much more useful than the way it previously functioned. I know I sure do. I'm going to go put most of this information onto the wiki. - -- John Eckersberg Production Operations Red Hat, Inc. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFIJEfAO5VSO/Qqm+gRAngqAJ9C9aL9fBYASJRQBAh+nw3mqrINDACgtGg9 E9KrC9JkAhg9tRka7vHiFuc= =93hk -----END PGP SIGNATURE----- From mdehaan at redhat.com Fri May 9 15:10:18 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 09 May 2008 11:10:18 -0400 Subject: [et-mgmt-tools] [PATCH] fix zone config matching and minimize the number of zone files generated In-Reply-To: <482447C0.7030306@redhat.com> References: <482447C0.7030306@redhat.com> Message-ID: <4824695A.3080902@redhat.com> John Eckersberg wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > http://jeckersb.fedorapeople.org/cobbler/0001-fix-zone-config-matching-and-minimize-the-number-of.patch > > > This fixes a few problems that I ran into as I sat down and actually > started deploying my own feature :) > > Previously manage_forward_zones and manage_reverse_zones were optional > and were supposed to restrict the zone files generated. Except it > didn't really work all that well. For example... > > If my settings looked like this: > > manage_forward_zones: > ~ - 'foo.example.com' > > Only hosts that matched 'hostname.foo.example.com' would go into the > zone. If you had hostname.subdomain.foo.example.com it would not match. > > Furthermore if you did not specify manage_forward_zones then it would > take every host, chop the hostname off the front and create a zone file > for the remainder of the FQDN. As I started to use it I realized the > number of zone files was quickly getting unmanageable. I have other > nameservers to configure as slaves for these zones and I was needing to > update them everytime somebody invented a new subdomain (which around > here is about every 5 minutes :D). > > Which brings us to how things work now. > > manage_forward_zones and manage_reverse_zones are now mandatory for BIND > management. The items listed in these values will directly map into > zone files. Hosts will go into their "best fit" as far as zone files. > To illustrate this: > > If my settings looked like this: > > manage_forward_zones: > ~ - 'foo.example.com' > ~ - 'bar.foo.example.com' > > The host 'baz.bar.foo.example.com' would go into the > 'bar.foo.example.com' zone whereas the host 's1.foo.example.com' would > go into the 'foo.example.com' zone. > > I hope everyone who is testing this out finds this to be much more > useful than the way it previously functioned. I know I sure do. > > I'm going to go put most of this information onto the wiki. > > - -- > John Eckersberg > Production Operations > Red Hat, Inc. > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.7 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iD8DBQFIJEfAO5VSO/Qqm+gRAngqAJ9C9aL9fBYASJRQBAh+nw3mqrINDACgtGg9 > E9KrC9JkAhg9tRka7vHiFuc= > =93hk > -----END PGP SIGNATURE----- > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Applied, thanks! --Michael From mdehaan at redhat.com Fri May 9 21:24:56 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 09 May 2008 17:24:56 -0400 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler and Koan 0.9.1-1 (Test1) Message-ID: <4824C128.4070006@redhat.com> Hi... I'm happy to announce that the first release of Cobbler and Koan 0.9.1-1 from the devel branch are available here for testing: http://cobbler.et.redhat.com/download/cobbler-0.9.1-1.src.rpm http://cobbler.et.redhat.com/download/koan-0.9.1-1.src.rpm In (short) summary, this release mainly features a lot of new auth options for the webapp (making it usable with FreeIPA and so forth), BIND support (thanks to John Eckersberg), and support for omapi for DHCPd (thanks to Pablo Iranzo Gomez). There's also some shiny nuts and bolts kinds of things like dynamic kickstart generation, and the new "cobbler buildiso" command -- plus lots of small stuff and bugfixes as well. There is a summary of everything that's new/changed on the Wiki, including a relatively simple set of instructions regarding some things that have changed in terms of configuration/setup. For those already using the git versions, this should be no suprise, but it's a good read: https://fedorahosted.org/cobbler/wiki/WhatsNewInOhNine The plan is to investigate/fix existing bugs in Trac over the next few weeks, while allowing people to test out this code before we push this into Fedora/EPEL as 1.0. I will accept patches for bugfixes before we get to 1.0 as well as small features, but we'll hold off on anything big until we have a 1.1 devel branch again and 1.0 is out. Then, as before, we'll continue to maintain 1.0 as stable and push fixes out to Fedora/EPEL while working on new things for 1.1. Thanks again to everyone who has been a part of this release! Happy testing! If you would like to submit items into Trac directly, you can get an account here: https://admin.fedoraproject.org/accounts/ Trac is here: https://fedorahosted.org/cobbler Also feel free to update any errant documenation on the Wiki you find along the way, as well as point out anything off/confusing in the manpages. One of the cool things I have planned after 1.0 is released is to come up with a nice deployment guide that collates and expands on all of the Wiki documentation and talks about all the various features of Cobbler and ways people are using it -- so having the Wiki nice and up to date is good for that in case I've missed something. Comments on future features or things you are interested in are always welcome as well. RFE's can be filed in Trac or you can share them here, whichever works better for you. --Michael From shenson at redhat.com Fri May 9 23:03:33 2008 From: shenson at redhat.com (Scott Henson) Date: Fri, 09 May 2008 19:03:33 -0400 Subject: [et-mgmt-tools] Replicate, now with docs and error on no master!! Message-ID: <4824D845.4020909@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 http://shenson.fedorapeople.org/cobbler/0001-Documentaiton-and-error-output-for-replicate.patch The above patch adds a blurb to the man page about the new replicate function. It also makes replicate error if you don't supply it with a master to replicate from. - -- Scott Henson Red Hat Production Operations Release Engineer WVU Alum BSAE/BSME -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkgk2EUACgkQcQOfHbe3DKCisACghApTTisMp/u0tVz/TygrJILX BJgAn2g5HTLnR+qT773+CXPwlSRvmktd =JpRB -----END PGP SIGNATURE----- From Pablo.Iranzo at redhat.com Sun May 11 01:38:39 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Sun, 11 May 2008 03:38:39 +0200 (CEST) Subject: [et-mgmt-tools] Cobbler [PATCH] for ticket 126 (createrepo) Message-ID: Hi Createrepo is not run by cobbler after each reposync if there is not a folder "RPMS" in it. As createrepo walks into subfolders, it's not needed to join that path that renders other repos unusable as there's no repodata folder for yum to use. This patch removes that requirement to make it work with repo's with no RPMS folder on it. Regards Pablo -- Pablo Iranzo G?mez (http://Alufis35.uv.es/~iranzo/) (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) -- Postulado de Boling sobre la Ley de Murphy: Si se encuentra bien, no se preocupe. Se le pasar? -------------- next part -------------- A non-text attachment was scrubbed... Name: createrepo.diff Type: text/x-diff Size: 581 bytes Desc: URL: From Pablo.Iranzo at redhat.com Sun May 11 01:40:20 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Sun, 11 May 2008 03:40:20 +0200 (CEST) Subject: [et-mgmt-tools] Cobbler [PATCH] for ticket 126 (createrepo) In-Reply-To: References: Message-ID: Forgot, this has been tested on cobbler GIT, should be aplied to 0.8.3 where the ticket was initially open Regards Pablo -- Pablo Iranzo G?mez (http://Alufis35.uv.es/~iranzo/) (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) -- Postulado de Boling sobre la Ley de Murphy: Si se encuentra bien, no se preocupe. Se le pasar? On Sun, 11 May 2008, Pablo Iranzo G?mez wrote: > Hi > > Createrepo is not run by cobbler after each reposync if there is > not a folder "RPMS" in it. > > As createrepo walks into subfolders, it's not needed to join that > path that renders other repos unusable as there's no repodata folder for > yum to use. > > This patch removes that requirement to make it work with repo's > with no RPMS folder on it. > > Regards > Pablo > > > > -- > Pablo Iranzo G?mez > (http://Alufis35.uv.es/~iranzo/) > (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) > -- > Postulado de Boling sobre la Ley de Murphy: > > Si se encuentra bien, no se preocupe. Se le pasar? From mdehaan at redhat.com Mon May 12 18:16:45 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 12 May 2008 14:16:45 -0400 Subject: [et-mgmt-tools] Cobbler [PATCH] for ticket 126 (createrepo) In-Reply-To: References: Message-ID: <4828898D.7070106@redhat.com> Pablo Iranzo G?mez wrote: > Hi > > Createrepo is not run by cobbler after each reposync if there is > not a folder "RPMS" in it. > > As createrepo walks into subfolders, it's not needed to join that > path that renders other repos unusable as there's no repodata folder for > yum to use. > > This patch removes that requirement to make it work with repo's > with no RPMS folder on it. > > Regards > Pablo > > Applied to devel/HEAD, thanks! Everyone can let me know if they see createrepo running in places it should not be, but it looks fine here. --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 Mon May 12 18:27:30 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 12 May 2008 14:27:30 -0400 Subject: [et-mgmt-tools] Replicate, now with docs and error on no master!! In-Reply-To: <4824D845.4020909@redhat.com> References: <4824D845.4020909@redhat.com> Message-ID: <48288C12.3060205@redhat.com> Scott Henson wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > http://shenson.fedorapeople.org/cobbler/0001-Documentaiton-and-error-output-for-replicate.patch > > > The above patch adds a blurb to the man page about the new replicate > function. It also makes replicate error if you don't supply it with a > master to replicate from. Applied, thanks! > > - -- > Scott Henson > Red Hat Production Operations Release Engineer > WVU Alum BSAE/BSME > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v2.0.9 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iEYEARECAAYFAkgk2EUACgkQcQOfHbe3DKCisACghApTTisMp/u0tVz/TygrJILX > BJgAn2g5HTLnR+qT773+CXPwlSRvmktd > =JpRB > -----END PGP SIGNATURE----- > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From rjones at redhat.com Tue May 13 16:42:38 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 13 May 2008 17:42:38 +0100 Subject: [et-mgmt-tools] [ANNOUNCE] virt-df 2.1.0 - a 'df' tool for virtual guests Message-ID: <20080513164238.GA10564@amd.home.annexia.org> I'm pleased to announce the most recent release of virt-df (2.1.0). Virt-df is 'df' for virtual guests. Run the program on the host / dom0 to display disk space used and available on all partitions on all guests. You don't need to run any sort of program/agent within the guest. Home page: http://et.redhat.com/~rjones/virt-df/ Source/binaries: http://et.redhat.com/~rjones/virt-df/files/ Developer repository: http://hg.et.redhat.com/virt/applications/virt-df--devel This version supports most common filesystems and partitioning schemes, including: - Linux ext2/3 - DOS FAT32 - Windows NTFS - Linux LVM2 (volume groups and logical volumes) - Primary and extended disk partitions - Linux swap - Linux suspend partition You can run it in your host / dom0 to display guest filesystems: # virt-df -c qemu:///system -h Filesystem Size Used Available Type rhel51x32kvm:hda1 96.8 MiB 14.6 MiB 82.2 MiB Linux ext2/3 rhel51x32kvm:VolGroup00/LogVol00 6.4 GiB 3.6 GiB 2.8 GiB Linux ext2/3 rhel51x32kvm:VolGroup00/LogVol01 992.0 MiB Linux swap You can also run it on general disk images, or disk devices: # virt-df -t /dev/sda Filesystem 1K-blocks Used Available Type /dev/sda:hda1 25599996 11309448 14290552 Windows NTFS /dev/sda:hda2 992016 93772 898244 Linux ext2/3 /dev/sda:F9VG/F9Root 23316072 7818164 15497908 Linux ext2/3 /dev/sda:F9VG/F9Swap 1015808 Linux swap /dev/sda:RHEL51VG/RHEL51Root 22382184 7796640 14585544 Linux ext2/3 /dev/sda:RHEL51VG/RHEL51Swap 2031616 Linux swap /dev/sda:VolGroup/FAT32Test 916736 4 914676 DOS/Windows You can write the output to a CSV file (use --csv option) in order to import the data easily into spreadsheets and databases. Included also is an experimental command line tool called 'diskzip' which intelligently compresses disk images by leaving out the bits which aren't actually used in the filesystems / partitions / volume groups contained within. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From sakaia at jp.fujitsu.com Thu May 15 02:51:17 2008 From: sakaia at jp.fujitsu.com (Atsushi SAKAI) Date: Thu, 15 May 2008 11:51:17 +0900 Subject: [et-mgmt-tools] incorrect link in virt-manager page Message-ID: <200805150251.m4F2pSpc006729@fjmscan501.ms.jp.fujitsu.com> Hi, I just inform an incorrect link information in virt-manager page. http://virt-manager.et.redhat.com/development.html Development -> Development roadmap cannot see. Following messages are appeared. =============== Not Found The requested URL /http://virt-manager.et.redhat.com/page/Roadmap was not found on this server. Apache/2.0.52 (Red Hat) Server at virt-manager.et.redhat.com Port 80 =============== Thanks Atsushi SAKAI From berrange at redhat.com Thu May 15 13:54:53 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 15 May 2008 14:54:53 +0100 Subject: [et-mgmt-tools] Re: [libvirt] [ANNOUNCE] virt-df 2.1.0 - a 'df' tool for virtual guests In-Reply-To: <20080513164238.GA10564@amd.home.annexia.org> References: <20080513164238.GA10564@amd.home.annexia.org> Message-ID: <20080515135453.GG5983@redhat.com> On Tue, May 13, 2008 at 05:42:38PM +0100, Richard W.M. Jones wrote: > I'm pleased to announce the most recent release of virt-df (2.1.0). > > Virt-df is 'df' for virtual guests. Run the program on the host / dom0 > to display disk space used and available on all partitions on all > guests. You don't need to run any sort of program/agent within the > guest. > > Home page: http://et.redhat.com/~rjones/virt-df/ > Source/binaries: http://et.redhat.com/~rjones/virt-df/files/ > Developer repository: > http://hg.et.redhat.com/virt/applications/virt-df--devel > > This version supports most common filesystems and partitioning > schemes, including: > - Linux ext2/3 > - DOS FAT32 > - Windows NTFS > - Linux LVM2 (volume groups and logical volumes) > - Primary and extended disk partitions > - Linux swap > - Linux suspend partition I'm assuming this only works for raw file & block devices ? Are you planning to support the funky QCow / VMDK formats too ? The other thing that could be annoying is that Fedora 9 support for encrypting all volumes - might need to prompt for a decryption key for that. > Included also is an experimental command line tool called 'diskzip' > which intelligently compresses disk images by leaving out the bits > which aren't actually used in the filesystems / partitions / volume > groups contained within. That's pretty neat. Which file systems does that work for ? VMWare have a funky guest tool which tries to let you get to a similar point. It works by basically openning a file inside the guest VM and filling it with zeros until the entire disk is full. THeir backend can then detect and discard all the sectors with zeros. Understanding the filesystem metdata is a much nicer way todo this :-) Regards, Daniel -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From crobinso at redhat.com Thu May 15 14:55:46 2008 From: crobinso at redhat.com (Cole Robinson) Date: Thu, 15 May 2008 10:55:46 -0400 Subject: [et-mgmt-tools] incorrect link in virt-manager page In-Reply-To: <200805150251.m4F2pSpc006729@fjmscan501.ms.jp.fujitsu.com> References: <200805150251.m4F2pSpc006729@fjmscan501.ms.jp.fujitsu.com> Message-ID: <482C4EF2.1010503@redhat.com> Atsushi SAKAI wrote: > Hi, > > I just inform an incorrect link information in virt-manager page. > http://virt-manager.et.redhat.com/development.html > > Development -> Development roadmap cannot see. > Following messages are appeared. > > =============== > Not Found > > The requested URL /http://virt-manager.et.redhat.com/page/Roadmap was not found on this server. > Apache/2.0.52 (Red Hat) Server at virt-manager.et.redhat.com Port 80 > =============== > > Thanks > Atsushi SAKAI > > > Whoops, that was my mistake. Fixed now, thanks for reporting. - Cole From rjones at redhat.com Thu May 15 18:53:55 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Thu, 15 May 2008 19:53:55 +0100 Subject: [et-mgmt-tools] Re: [libvirt] [ANNOUNCE] virt-df 2.1.0 - a 'df' tool for virtual guests In-Reply-To: <20080515135453.GG5983@redhat.com> References: <20080513164238.GA10564@amd.home.annexia.org> <20080515135453.GG5983@redhat.com> Message-ID: <20080515185355.GA32617@amd.home.annexia.org> On Thu, May 15, 2008 at 02:54:53PM +0100, Daniel P. Berrange wrote: > > This version supports most common filesystems and partitioning > > schemes, including: > > - Linux ext2/3 > > - DOS FAT32 > > - Windows NTFS > > - Linux LVM2 (volume groups and logical volumes) > > - Primary and extended disk partitions > > - Linux swap > > - Linux suspend partition > > I'm assuming this only works for raw file & block devices ? Are you planning > to support the funky QCow / VMDK formats too ? Yes, interesting point. Since I always use flat files or straight partitions for my guests I admit I hadn't given this much thought :-) However support for these formats is just a matter of decoding enough of the structure to enable the same mappings to be made in the 'virt-df' library, same as for LVM2 or indeed MBR partitions now. I'll take a look at it. IIRC there are several different undocumented variations on the QCow format? > The other thing that could be annoying is that Fedora 9 support for > encrypting all volumes - might need to prompt for a decryption key > for that. Yes -- any encrypted volumes aren't going to work at the moment, and couldn't work unless there was a way to access the passphrase. > > Included also is an experimental command line tool called 'diskzip' > > which intelligently compresses disk images by leaving out the bits > > which aren't actually used in the filesystems / partitions / volume > > groups contained within. > > That's pretty neat. Which file systems does that work for ? VMWare have > a funky guest tool which tries to let you get to a similar point. It works > by basically openning a file inside the guest VM and filling it with zeros > until the entire disk is full. THeir backend can then detect and discard > all the sectors with zeros. Understanding the filesystem metdata is a much > nicer way todo this :-) It works with most of the formats supported by virt-df. For MBR & LVM once you've parsed the partition tables / LVM metadata, then the information you need just falls out naturally. For NTFS & DOS FAT (you won't believe it but ...) you need to find the allocation bitmaps/tables in both cases in order to calculate blocks used/free for df anyway. For ext2 it's a little bit more tricky because one needs to additionally parse the group block free bitmaps [this bit doesn't work at the moment, but is in principle very simple to add]. 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 berrange at redhat.com Fri May 16 00:38:24 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Fri, 16 May 2008 01:38:24 +0100 Subject: [et-mgmt-tools] Re: [libvirt] [ANNOUNCE] virt-df 2.1.0 - a 'df' tool for virtual guests In-Reply-To: <20080515185355.GA32617@amd.home.annexia.org> References: <20080513164238.GA10564@amd.home.annexia.org> <20080515135453.GG5983@redhat.com> <20080515185355.GA32617@amd.home.annexia.org> Message-ID: <20080516003824.GB28467@redhat.com> On Thu, May 15, 2008 at 07:53:55PM +0100, Richard W.M. Jones wrote: > On Thu, May 15, 2008 at 02:54:53PM +0100, Daniel P. Berrange wrote: > > > This version supports most common filesystems and partitioning > > > schemes, including: > > > - Linux ext2/3 > > > - DOS FAT32 > > > - Windows NTFS > > > - Linux LVM2 (volume groups and logical volumes) > > > - Primary and extended disk partitions > > > - Linux swap > > > - Linux suspend partition > > > > I'm assuming this only works for raw file & block devices ? Are you planning > > to support the funky QCow / VMDK formats too ? > > Yes, interesting point. Since I always use flat files or straight > partitions for my guests I admit I hadn't given this much thought :-) > However support for these formats is just a matter of decoding enough > of the structure to enable the same mappings to be made in the > 'virt-df' library, same as for LVM2 or indeed MBR partitions now. > > I'll take a look at it. IIRC there are several different undocumented > variations on the QCow format? There's version 1, and version 2. And then the incompatible version 2 inflicted by Xen :-( > > The other thing that could be annoying is that Fedora 9 support for > > encrypting all volumes - might need to prompt for a decryption key > > for that. > > Yes -- any encrypted volumes aren't going to work at the moment, and > couldn't work unless there was a way to access the passphrase. The passphrase is in the user's brain. THe OS prompts for it at boot time, so virt-df would need todo similar if it wanted to support decryption. Perhaps its just not a important use case. In the ISP model, guest admins won't trust the host admin so you won't have the keys anyway. If the guest admin does trust the host admin, would they really be using encryption in the guest ? Dan. -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From mdehaan at redhat.com Fri May 16 18:16:38 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 16 May 2008 14:16:38 -0400 Subject: [et-mgmt-tools] So what do all those things in /var/lib/cobbler/settings do? Message-ID: <482DCF86.4000202@redhat.com> Someone pointed out on IRC that we can probably add comments to the settings file. It turns out we can. For those that don't want to wait for the next release, you can do some reading here: http://git.fedorahosted.org/git/?p=cobbler;a=blob_plain;f=config/settings;hb=devel Note some of these are specific to 0.9.X and later. --Michael From mdehaan at redhat.com Fri May 16 18:24:27 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 16 May 2008 14:24:27 -0400 Subject: [et-mgmt-tools] So what do all those things in /var/lib/cobbler/settings do? In-Reply-To: <482DCF86.4000202@redhat.com> References: <482DCF86.4000202@redhat.com> Message-ID: <482DD15B.9030407@redhat.com> Michael DeHaan wrote: > Someone pointed out on IRC that we can probably add comments to the > settings file. > > It turns out we can. For those that don't want to wait for the next > release, you can do some reading here: > > > http://git.fedorahosted.org/git/?p=cobbler;a=blob_plain;f=config/settings;hb=devel > > > Note some of these are specific to 0.9.X and later. > > --Michael > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools And yes, I'll finally be moving this to /etc/cobbler/settings too. I think there will be a friendly error message if you still have a copy in /var/lib from a upgrade which you'll have to resolve, but that will be a one time deal. --Michael From robin-lists at robinbowes.com Fri May 16 19:01:49 2008 From: robin-lists at robinbowes.com (Robin Bowes) Date: Fri, 16 May 2008 20:01:49 +0100 Subject: [et-mgmt-tools] Re: So what do all those things in /var/lib/cobbler/settings do? In-Reply-To: <482DD15B.9030407@redhat.com> References: <482DCF86.4000202@redhat.com> <482DD15B.9030407@redhat.com> Message-ID: Michael DeHaan wrote: > And yes, I'll finally be moving this to /etc/cobbler/settings too. > > I think there will be a friendly error message if you still have a copy > in /var/lib from a upgrade which you'll have to resolve, but that will > be a one time deal. Michael, I would leave it in /var/lib/cobbler. Symlink to /etc/cobbler if you must. IMHO, web-based programs like cobbler shouldn't be writing into /etc /var/lib is the right place. R. From mdehaan at redhat.com Fri May 16 19:10:46 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 16 May 2008 15:10:46 -0400 Subject: [et-mgmt-tools] Re: So what do all those things in /var/lib/cobbler/settings do? In-Reply-To: References: <482DCF86.4000202@redhat.com> <482DD15B.9030407@redhat.com> Message-ID: <482DDC36.6030804@redhat.com> Robin Bowes wrote: > Michael DeHaan wrote: > >> And yes, I'll finally be moving this to /etc/cobbler/settings too. >> >> I think there will be a friendly error message if you still have a >> copy in /var/lib from a upgrade which you'll have to resolve, but >> that will be a one time deal. > > Michael, > > I would leave it in /var/lib/cobbler. Symlink to /etc/cobbler if you > must. It's been moved. > > IMHO, web-based programs like cobbler shouldn't be writing into /etc > > /var/lib is the right place. > Nope, it's not the right place for a settings file. Cobbler hasn't written to /var/lib/cobbler/settings since we started packaging it in the RPM (and marked it as "config(noreplace)". It's read by the same loaders that read the other state files, but not written by them. > R. > > _______________________________________________ > 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 May 16 19:58:08 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 16 May 2008 15:58:08 -0400 Subject: [et-mgmt-tools] [TESTING] cobbler/koan 0.9.2 Message-ID: <482DE750.8010408@redhat.com> I've uploaded new cobbler/koan source RPMs for those who are interested in testing and commenting on the candidate for the 1.0 release. http://cobbler.et.redhat.com/download/ Testing is going very well, so this will likely be the last source rpm release we do before the real 1.0 release as we have a lot of folks already testing the latest out of git, etc. This contains fixes from items in the last test release, the move to /etc/cobbler/settings, and some improvements to cobbler check and other warning message cleanup. This page has been updated to explain a bit more about what is in: https://fedorahosted.org/cobbler/wiki/WhatsNewInOhNine Koan's Thanks! --Michael From mdehaan at redhat.com Fri May 16 20:23:05 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 16 May 2008 16:23:05 -0400 Subject: [et-mgmt-tools] [TESTING] cobbler/koan 0.9.2 In-Reply-To: <482DE750.8010408@redhat.com> References: <482DE750.8010408@redhat.com> Message-ID: <482DED29.4030101@redhat.com> Michael DeHaan wrote: > I've uploaded new cobbler/koan source RPMs for those who are > interested in testing and commenting on the candidate for the 1.0 > release. > > http://cobbler.et.redhat.com/download/ > > Testing is going very well, so this will likely be the last source rpm > release we do before the real 1.0 release as we have a lot of folks > already testing > the latest out of git, etc. > > This contains fixes from items in the last test release, the move to > /etc/cobbler/settings, and some improvements to cobbler check and > other warning message > cleanup. This page has been updated to explain a bit more about what > is in: https://fedorahosted.org/cobbler/wiki/WhatsNewInOhNine > > Koan's > > Thanks! > > --Michael > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Hold off on this one for a bit... checking something... From mdehaan at redhat.com Fri May 16 20:36:22 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 16 May 2008 16:36:22 -0400 Subject: [et-mgmt-tools] [TESTING] cobbler/koan 0.9.2 In-Reply-To: <482DED29.4030101@redhat.com> References: <482DE750.8010408@redhat.com> <482DED29.4030101@redhat.com> Message-ID: <482DF046.1000808@redhat.com> Michael DeHaan wrote: > Michael DeHaan wrote: >> I've uploaded new cobbler/koan source RPMs for those who are >> interested in testing and commenting on the candidate for the 1.0 >> release. >> >> http://cobbler.et.redhat.com/download/ >> >> Testing is going very well, so this will likely be the last source >> rpm release we do before the real 1.0 release as we have a lot of >> folks already testing >> the latest out of git, etc. >> >> This contains fixes from items in the last test release, the move to >> /etc/cobbler/settings, and some improvements to cobbler check and >> other warning message >> cleanup. This page has been updated to explain a bit more about >> what is in: https://fedorahosted.org/cobbler/wiki/WhatsNewInOhNine >> >> Koan's >> >> Thanks! >> >> --Michael >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > Hold off on this one for a bit... checking something... > > Ok, new versions pushed, same filenames... good to go. From lippold at gmail.com Sat May 17 16:15:33 2008 From: lippold at gmail.com (Aaron Lippold) Date: Sat, 17 May 2008 12:15:33 -0400 Subject: [et-mgmt-tools] Cobbler profile behavior when using same ks file name Message-ID: <39d2723b0805170915o4923e26atd2536ff145509555@mail.gmail.com> Hello All, While doing some installs this week I noticed a strange quirk in the cobbler profile add and or cobbler sync work flow. Here's what heepened. [aaronl at localhost ~]$ rpm -q cobbler cobbler-0.8.3-3.el5 If I add a profile "x" with distro "d" and kickstart "k". Cobbler sync. All Good. Now, if I test out my install notice some issues and edit my kickstart k to k' and keep the file name the same - let;s say d600-lab-workstation.ks, then add a new profile X1 with distro "d" and kickstart "k'" ( i.e. the modified ks file k' ) a cobbler sync will add the new profile and render k' in the ../kickstarts/X1/ks.cfg. However, it will also render k and replace ../kickstarts/X/ks.cfg as well. I have used this method before with cobbler - its a great way to keep a history of your build and tweaks during the refinement of a system - but now it seems that the profile object is linked to the ks file by path. Ok, I can see how this works given that cobbler sync most likely doesn't have an "activity log" on a profile. But I can see issues with this. Profile "x" didn't change since I didn't issue any commands to profile "x" so cobbler sync shouldn't modify it. At least, that is what I would expect. I can see issues with this - I think - if you had a distributed or even more than one SA working with profiles. If two folks took the baseline ks file for their company, made some changes, saved and created a profile, the last one in wins, This could cause issues right? Or like me where I really don't want to do a :w myname1 :w myname2 ... when I am doing quick changes and testing... Just thought I would bring it up for discussion. Yours, Aaron From et-mgmt-tools at clabbe.com Mon May 19 07:21:08 2008 From: et-mgmt-tools at clabbe.com (ClaesBas) Date: Mon, 19 May 2008 09:21:08 +0200 Subject: [et-mgmt-tools] [libvirt] [virt-manager] PATCH: Support nonstandard ssh-port for tunnels Message-ID: <48312A64.7060303@clabbe.com> If you have ssh moved/hidden somewhere else then port 22 this patch makes it possible to open the vnc/console. I have to add the entry from commandline like this: virt-manager -c qemu+ssh://username at some_machine:22222/system If ssh is running on port 22222. I've also submitet this on bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=447070 and by mistake to libvir-list (sorry). Regards Claes --- a/src/virtManager/details.py Tue May 13 15:59:25 2008 -0400 +++ b/src/virtManager/details.py Sun May 18 23:07:37 2008 +0200 @@ -975,7 +975,11 @@ class vmmDetails(gobject.GObject): os.close(1) os.dup(fds[1].fileno()) os.dup(fds[1].fileno()) - argv = ["ssh", "ssh", "-p", "22"] + if not server.count(":"): + sshport = 22 + else: + (server, sshport) = server.split(":") + argv = ["ssh", "ssh", "-p", sshport] if username: argv += ['-l', username] argv += [ server, "nc", vncaddr, str(vncport) ] From et-mgmt-tools at clabbe.com Mon May 19 10:17:20 2008 From: et-mgmt-tools at clabbe.com (ClaesBas) Date: Mon, 19 May 2008 12:17:20 +0200 Subject: [et-mgmt-tools] [virt-manager] TypeError: get_max_vcpus() takes exactly 1 argument (2 given), connection.py:520 Message-ID: <483153B0.6080906@clabbe.com> Could somebody who understand this bit of code look at connection.py:520. >From my ~/.virt-manager/virt-manager.log [Mon, 19 May 2008 10:40:55 virt-manager 28993] ERROR (virt-manager:141) Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/details.py", line 675, in refresh_resources self.page_refresh(details.get_current_page()) File "/usr/share/virt-manager/virtManager/details.py", line 693, in page_refresh self.refresh_config_cpu() File "/usr/share/virt-manager/virtManager/details.py", line 730, in refresh_config_cpu cpu_max = self.vm.get_connection().get_max_vcpus(self.vm.get_type()) File "/usr/share/virt-manager/virtManager/connection.py", line 520, in get_max_vcpus return virtinst.util.get_max_vcpus(self.vmm, type) TypeError: get_max_vcpus() takes exactly 1 argument (2 given) I think it should be (connection.py:520): return virtinst.util.get_max_vcpus(type) When I changed to the above, all errors was gone.... Regards Claes From crobinso at redhat.com Mon May 19 13:52:32 2008 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 19 May 2008 09:52:32 -0400 Subject: [et-mgmt-tools] [virt-manager] TypeError: get_max_vcpus() takes exactly 1 argument (2 given), connection.py:520 In-Reply-To: <483153B0.6080906@clabbe.com> References: <483153B0.6080906@clabbe.com> Message-ID: <48318620.9050208@redhat.com> ClaesBas wrote: > Could somebody who understand this bit of code > look at connection.py:520. > >>From my ~/.virt-manager/virt-manager.log > > [Mon, 19 May 2008 10:40:55 virt-manager 28993] ERROR (virt-manager:141) > Traceback (most recent call last): > File "/usr/share/virt-manager/virtManager/details.py", line 675, in > refresh_resources > self.page_refresh(details.get_current_page()) > File "/usr/share/virt-manager/virtManager/details.py", line 693, in > page_refresh > self.refresh_config_cpu() > File "/usr/share/virt-manager/virtManager/details.py", line 730, in > refresh_config_cpu > cpu_max = self.vm.get_connection().get_max_vcpus(self.vm.get_type()) > File "/usr/share/virt-manager/virtManager/connection.py", line 520, in > get_max_vcpus > return virtinst.util.get_max_vcpus(self.vmm, type) > TypeError: get_max_vcpus() takes exactly 1 argument (2 given) > > > I think it should be (connection.py:520): > return virtinst.util.get_max_vcpus(type) > > When I changed to the above, all errors was gone.... > > > Regards > Claes > A function changed in the virtinst library, and virt-manager is making use of the newer version. If you are using upstream virt-manager, its always a good idea to use upstream virtinst, as they are very closely tied. http://hg.et.redhat.com/virt/applications/virtinst--devel - Cole From dlutter at redhat.com Mon May 19 16:58:39 2008 From: dlutter at redhat.com (David Lutterkort) Date: Mon, 19 May 2008 16:58:39 +0000 Subject: [et-mgmt-tools] New yum module: grouppackagetypes In-Reply-To: References: Message-ID: <1211216319.29522.227.camel@localhost.localdomain> On Sat, 2008-05-03 at 01:30 +0100, Robin Bowes wrote: > I was trying to find out how I could install just the mandatory packages > from a package group when using "yum groupinstall ...". I found that I > could add a line to /etc/yum.conf: This is the wrong mailing list .. you'd probably get en even more enthusiastic response if you post this to the yum-devel mailing list[1] David [1] http://linux.duke.edu/projects/yum/ From robin-lists at robinbowes.com Mon May 19 21:29:50 2008 From: robin-lists at robinbowes.com (Robin Bowes) Date: Mon, 19 May 2008 22:29:50 +0100 Subject: [et-mgmt-tools] Re: New yum module: grouppackagetypes In-Reply-To: <1211216319.29522.227.camel@localhost.localdomain> References: <1211216319.29522.227.camel@localhost.localdomain> Message-ID: David Lutterkort wrote: > On Sat, 2008-05-03 at 01:30 +0100, Robin Bowes wrote: >> I was trying to find out how I could install just the mandatory packages >> from a package group when using "yum groupinstall ...". I found that I >> could add a line to /etc/yum.conf: > > This is the wrong mailing list .. you'd probably get en even more > enthusiastic response if you post this to the yum-devel mailing list[1] Thanks David - I actually posted it to this list and the "yum" list. I got a response from Seth, but I've only just noticed it! Thanks, R. From mdehaan at redhat.com Tue May 20 01:36:45 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 19 May 2008 21:36:45 -0400 Subject: [et-mgmt-tools] Cobbler profile behavior when using same ks file name In-Reply-To: <39d2723b0805170915o4923e26atd2536ff145509555@mail.gmail.com> References: <39d2723b0805170915o4923e26atd2536ff145509555@mail.gmail.com> Message-ID: <48322B2D.1070209@redhat.com> Aaron Lippold wrote: > Hello All, > > While doing some installs this week I noticed a strange quirk in the > cobbler profile add and or cobbler sync work flow. Here's what > heepened. > > [aaronl at localhost ~]$ rpm -q cobbler > cobbler-0.8.3-3.el5 > > If I add a profile "x" with distro "d" and kickstart "k". Cobbler > sync. All Good. > > Now, if I test out my install notice some issues and edit my > kickstart k to k' and keep the file name the same - let;s say > d600-lab-workstation.ks, then add a new profile X1 with distro "d" and > kickstart "k'" ( i.e. the modified ks file k' ) a cobbler sync will > add the new profile and render k' in the ../kickstarts/X1/ks.cfg. > However, it will also render k and replace ../kickstarts/X/ks.cfg as > well. > > Cobbler 0.9.X and later will not render any kickstarts to disk, just FYI... > I have used this method before with cobbler - its a great way to keep > a history of your build and tweaks during the refinement of a system - > but now it seems that the profile object is linked to the ks file by > path. Ok, I can see how this works given that cobbler sync most likely > doesn't have an "activity log" on a profile. But I can see issues with > this. Profile "x" didn't change since I didn't issue any commands to > profile "x" so cobbler sync shouldn't modify it. At least, that is > what I would expect. I can see issues with this - I think - if you had > a distributed or even more than one SA working with profiles. If two > folks took the baseline ks file for their company, made some changes, > saved and created a profile, the last one in wins, This could cause > issues right? Or like me where I really don't want to do a :w myname1 > :w myname2 ... when I am doing quick changes and testing... > I'm not sure I understand the question/problem/use-case/hack, but the purposes of those files are to serve for booting /existing/ profiles only, not to keep a history of how kickstarts were rendered in the past. In the future, there will be no physical rendered kickstarts at all, with these being dynamically generated by mod python. If two people are editing a kickstart template, the last one does win, because the kickstart template is a file on the filesystem, and this is how filesystems work. However, vim/other should warn that there are others editing the file. > Just thought I would bring it up for discussion. > > Yours, > > Aaron > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From ole.ersoy at gmail.com Tue May 20 02:00:38 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Mon, 19 May 2008 21:00:38 -0500 Subject: [et-mgmt-tools] Barebones Install Without Koan? Message-ID: <483230C6.60304@gmail.com> Hi, Currently I use a USB boot drive to do network installs on new hardware (The hardware is minimal, with no CD or DVD drive). I'd like to get around having to use the USB drive. Ideally I would just plug a raw machine into the network and be able to click through the install from there. Is that possible with Cobbler (Or do I have to have a kickstart profile for the hardware before starting)? Thanks, - Ole From zuber at puzzle.ch Tue May 20 08:56:45 2008 From: zuber at puzzle.ch (zuber at puzzle.ch) Date: Tue, 20 May 2008 10:56:45 +0200 (CEST) Subject: [et-mgmt-tools] Live- and rescue-system provisioning over pxe Message-ID: <52278.193.5.216.100.1211273805.squirrel@mail.worldweb2000.com> What do you think about providing live- and rescue systems for pxe with cobbler? With Fedora 9 there is a new script in the livecd-tools that turns a live-cd iso into a pxe bootable image. So we can now create a pxe live-system out of a kickstartfile. Providing rescue-system is already possible with cobbler. You only have to create a Kickstart file and a profile to provide it over pxe: http://wiki.centos.org/HowTos/PXE/Rescue_PXE My question is, does it makes sense to manage the build process of the live images with cobbler? It is always possible to do this manualy, but i think it will be cool if cobbler does that for me. Greetings Andy From crobinso at redhat.com Tue May 20 14:31:47 2008 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 20 May 2008 10:31:47 -0400 Subject: [et-mgmt-tools] [libvirt] [virt-manager] PATCH: Support nonstandard ssh-port for tunnels In-Reply-To: <48312A64.7060303@clabbe.com> References: <48312A64.7060303@clabbe.com> Message-ID: <4832E0D3.5000709@redhat.com> ClaesBas wrote: > If you have ssh moved/hidden somewhere else then port 22 this patch > makes it possible to open the vnc/console. > I have to add the entry from commandline like this: > virt-manager -c qemu+ssh://username at some_machine:22222/system > If ssh is running on port 22222. > > I've also submitet this on bugzilla: > https://bugzilla.redhat.com/show_bug.cgi?id=447070 > and by mistake to libvir-list (sorry). > > Regards > Claes > > --- a/src/virtManager/details.py Tue May 13 15:59:25 2008 -0400 > +++ b/src/virtManager/details.py Sun May 18 23:07:37 2008 +0200 > @@ -975,7 +975,11 @@ class vmmDetails(gobject.GObject): > os.close(1) > os.dup(fds[1].fileno()) > os.dup(fds[1].fileno()) > - argv = ["ssh", "ssh", "-p", "22"] > + if not server.count(":"): > + sshport = 22 > + else: > + (server, sshport) = server.split(":") > + argv = ["ssh", "ssh", "-p", sshport] > if username: > argv += ['-l', username] > argv += [ server, "nc", vncaddr, str(vncport) ] > Thanks, I've committed this. http://hg.et.redhat.com/virt/applications/virt-manager--devel?cs=57ff06da8a73 - Cole From ole.ersoy at gmail.com Tue May 20 18:41:22 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Tue, 20 May 2008 13:41:22 -0500 Subject: [et-mgmt-tools] Troubleshooting Message-ID: <48331B52.7090702@gmail.com> Hi, I'm trying to boot some bare metal over the network. I followed all the steps in the tutorial: https://fedorahosted.org/cobbler/wiki/HowToCobblerServer When I run cobbler check it just tells me this: [root at localhost cobbler]# cobbler check The following potential problems were detected: #0: since iptables may be running, ensure 69, 80, 25150, and 25151 are unblocked And I did make sure that these ports are open. I then set the bare metal bios to boot from the LAN first, but I still get a "NO BOOT DISK FOUND ERROR...or something like that". Is there a way to simulate the request that boot client should be sending to cobbler? Thanks, - Ole From wright at imageworks.com Tue May 20 19:01:29 2008 From: wright at imageworks.com (Peter Wright) Date: Tue, 20 May 2008 12:01:29 -0700 Subject: [et-mgmt-tools] Troubleshooting In-Reply-To: <48331B52.7090702@gmail.com> References: <48331B52.7090702@gmail.com> Message-ID: <48332009.501@imageworks.com> Ole Ersoy wrote: > Hi, > > I'm trying to boot some bare metal over the network. I followed all the steps in the tutorial: > > https://fedorahosted.org/cobbler/wiki/HowToCobblerServer > > When I run cobbler check it just tells me this: > > [root at localhost cobbler]# cobbler check > The following potential problems were detected: > #0: since iptables may be running, ensure 69, 80, 25150, and 25151 are unblocked > > And I did make sure that these ports are open. I then set the bare metal bios to boot from the LAN first, but I still get a "NO BOOT DISK FOUND ERROR...or something like that". > > Is there a way to simulate the request that boot client should be sending to cobbler? > > There are a couple things you can check - during your pxeboot phase do you see a timeout or anything like that? I find it helpful to do a tcpdump on the cobbler box to see if infact the pxe and dhcp packets are making it from the host to the server. You can also tail /var/log/messages (by default) or /var/log/dhcpd (if you setup a syslogd rule for dhcpd) to see if there are requests getting through to the server. One common mistake I make is that I forget to add a subnet in my dhcp.template, and because of this my pxe requests will fail (as dhcpd won't be configured to offer leases on that subnet). This will usually show up in your log file. -pete -- Peter Wright Systems Engineer Sony Pictures Imageworks wright at imageworks.com www.imageworks.com From plc at novell.com Tue May 20 19:39:49 2008 From: plc at novell.com (Pat Campbell) Date: Tue, 20 May 2008 13:39:49 -0600 Subject: [et-mgmt-tools] [PATCH]: Make virt-viewer ungrab the mouse and keyboard on un-focus In-Reply-To: <480E0AED.6060208@redhat.com> References: <4809486B.2070600@redhat.com> <480DF97C.2080001@redhat.com> <20080422144957.GG27825@redhat.com> <480E0AED.6060208@redhat.com> Message-ID: <48332905.80401@novell.com> Chris Lalancette wrote: > Daniel P. Berrange wrote: > >> I struggle to understand how/what you managed todo here. If the keyboard >> grab is active nothing should be able to steal keyboard input focus, since >> this is the whole point of keyboard grab. If something did manage to steal >> keyboard input while grab was active, then this is a bug in the window >> manager and not something we should hack in virt-viewer - inded if this >> is possible, then every single app which uses grabs needs your patch which >> is clearly non-scalable >> > > This arguably is some sort of window manager bug, or possibly a bug in the way > keyboard grab is currently working. In any case, it is very easy to reproduce. > Just do something like: > > # virt-viewer guest1 & sleep 10 ; virt-viewer guest2 > > And make sure to click inside the guest1 virt-viewer window right after it pops > up. Assuming your keyboard is grabbed there, and that your window manager > placed the guest2 window on top of the guest1 window, you will be unable to > escape (with Ctrl-Alt) from the guest1 window. > > Chris Lalancette > > It is not a Window Manager bug, it is a virt-viewer bug. The Window Manager is not responsible for releasing keyboard or pointer grabs when a window looses focus. It is the applications job. Take a look at xorg/xc/Xserver/dix/event.c: ProcUngrabPointer(ClientPtr client) This function checks that the releasing client is the client that owns the grab. This behavior makes sense, would be a huge security hole otherwise. Not sure this will apply cleanly but I would suggest something like below to gtk-vnc/vncdisplay.c: --- vncdisplay.c.orig 2008-05-20 13:32:24.000000000 -0600 +++ vncdisplay.c 2008-05-20 13:33:22.000000000 -0600 @@ -492,6 +492,9 @@ static gboolean leave_event(GtkWidget *w if (priv->grab_keyboard) do_keyboard_ungrab(VNC_DISPLAY(widget), FALSE); + if (priv->grab_pointer) + do_pointer_ungrab(VNC_DISPLAY(widget), FALSE); + return TRUE; } From berrange at redhat.com Tue May 20 19:48:33 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 20 May 2008 20:48:33 +0100 Subject: [et-mgmt-tools] [PATCH]: Make virt-viewer ungrab the mouse and keyboard on un-focus In-Reply-To: <48332905.80401@novell.com> References: <4809486B.2070600@redhat.com> <480DF97C.2080001@redhat.com> <20080422144957.GG27825@redhat.com> <480E0AED.6060208@redhat.com> <48332905.80401@novell.com> Message-ID: <20080520194833.GI16499@redhat.com> On Tue, May 20, 2008 at 01:39:49PM -0600, Pat Campbell wrote: > Chris Lalancette wrote: > > Daniel P. Berrange wrote: > > > >> I struggle to understand how/what you managed todo here. If the keyboard > >> grab is active nothing should be able to steal keyboard input focus, since > >> this is the whole point of keyboard grab. If something did manage to steal > >> keyboard input while grab was active, then this is a bug in the window > >> manager and not something we should hack in virt-viewer - inded if this > >> is possible, then every single app which uses grabs needs your patch which > >> is clearly non-scalable > >> > > > > This arguably is some sort of window manager bug, or possibly a bug in the way > > keyboard grab is currently working. In any case, it is very easy to reproduce. > > Just do something like: > > > > # virt-viewer guest1 & sleep 10 ; virt-viewer guest2 > > > > And make sure to click inside the guest1 virt-viewer window right after it pops > > up. Assuming your keyboard is grabbed there, and that your window manager > > placed the guest2 window on top of the guest1 window, you will be unable to > > escape (with Ctrl-Alt) from the guest1 window. > > > > Chris Lalancette > > > > > It is not a Window Manager bug, it is a virt-viewer bug. The Window > Manager is not responsible for releasing keyboard or pointer grabs when > a window looses focus. It is the applications job. > > Take a look at xorg/xc/Xserver/dix/event.c: ProcUngrabPointer(ClientPtr > client) This function checks that the releasing client is the client > that owns the grab. This behavior makes sense, would be a huge security > hole otherwise. Ok, that makes a little more sense, though I'm still surprised its possible to loose keyboard focus while grabbed, and I'm fairly sure we always grab keyboard whenever the mouse is grabbed. > Not sure this will apply cleanly but I would suggest something like > below to gtk-vnc/vncdisplay.c: Looks reasonable to me - CC'ing the GTK-VNC mailing list.. > --- vncdisplay.c.orig 2008-05-20 13:32:24.000000000 -0600 > +++ vncdisplay.c 2008-05-20 13:33:22.000000000 -0600 > @@ -492,6 +492,9 @@ static gboolean leave_event(GtkWidget *w > if (priv->grab_keyboard) > do_keyboard_ungrab(VNC_DISPLAY(widget), FALSE); > > + if (priv->grab_pointer) > + do_pointer_ungrab(VNC_DISPLAY(widget), FALSE); > + > return TRUE; > } Dan -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From ole.ersoy at gmail.com Tue May 20 20:21:55 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Tue, 20 May 2008 15:21:55 -0500 Subject: [et-mgmt-tools] Troubleshooting In-Reply-To: <48332009.501@imageworks.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> Message-ID: <483332E3.9070901@gmail.com> Hi Pete, > There are a couple things you can check - during your pxeboot phase do > you see a timeout or anything like that? It does seem like the box pauses a little longer before it tells me it can't boot. > I find it helpful to do a > tcpdump on the cobbler box to see if infact the pxe and dhcp packets are > making it from the host to the server. I need to read up on tcpdump. I have a feeling it's going to be a valuable utility :-). > > You can also tail /var/log/messages (by default) or /var/log/dhcpd (if > you setup a syslogd rule for dhcpd) to see if there are requests getting > through to the server. That I can handle. This is what I get when I tail /var/log/messages: [root at localhost cobbler]# tail /var/log/messages May 20 13:59:02 localhost dhcpd: Internet Systems Consortium DHCP Server V3.0.6-Fedora May 20 13:59:02 localhost dhcpd: Copyright 2004-2007 Internet Systems Consortium. May 20 13:59:02 localhost dhcpd: All rights reserved. May 20 13:59:02 localhost dhcpd: For info, please visit http://www.isc.org/sw/dhcp/ May 20 13:59:02 localhost dhcpd: Not searching LDAP since ldap-server, ldap-username, ldap-password and ldap-base-dn were not specified in the config file May 20 13:59:02 localhost dhcpd: Wrote 0 leases to leases file. May 20 13:59:03 localhost dhcpd: Listening on LPF/eth0/00:0e:a6:65:57:99/192.168.1/24 May 20 13:59:03 localhost dhcpd: Sending on LPF/eth0/00:0e:a6:65:57:99/192.168.1/24 May 20 13:59:03 localhost dhcpd: Sending on Socket/fallback/fallback-net May 20 14:03:57 localhost xinetd[3381]: EXIT: tftp status=0 pid=3661 duration=1048(sec) So it looks like there was some activity right around the time I tried it, but I'm not sure exactly how to turn this log into further action. Any ideas? > One common mistake I make is that I forget to > add a subnet in my dhcp.template, and because of this my pxe requests > will fail (as dhcpd won't be configured to offer leases on that > subnet). This will usually show up in your log file. Hmmm - Now that you mention that I wonder if mine is goofed, because the log says "Wrote 0 leases to leases file". I let cobbler manage the DHCP configuration, and the configuration file looks like this: ddns-update-style interim; allow booting; allow bootp; ignore client-updates; set vendorclass = option vendor-class-identifier; subnet 192.168.1.0 netmask 255.255.255.0 { option routers 192.168.1.1; option subnet-mask 255.255.255.0; range dynamic-bootp 192.168.1.100 192.168.1.254; filename "/pxelinux.0"; default-lease-time 21600; max-lease-time 43200; next-server $next_server; } $insert_cobbler_system_definitions This seems ok, although I'm not an expert. My router has address 192.168.1.1 and this is a home environment so I only have machines 192.168.1.6, which is the cobbler server, and a few others 192.168.1.2, and 192.168.1.3...+ the machine that is the bare metal one. If anything jumps out, please let me know. I'll try to smarten up on tcpdump, and hopefully that will reveal more. Thanks again for all the tips! Ole > > > -pete > > > > From ole.ersoy at gmail.com Tue May 20 22:02:52 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Tue, 20 May 2008 17:02:52 -0500 Subject: [et-mgmt-tools] Troubleshooting In-Reply-To: <48332009.501@imageworks.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> Message-ID: <48334A8C.5070705@gmail.com> Hello again, I tried tcpdumping and looking at the traffic, but did not really see anything that looked like traffic from a machine on the local network. I figured I'd try turning back on the routers DHCP server to see whether it would assign an address to the bare metal machine. After doing this I still don't see it in the attached devices per the routers web admin tool. Anyone know whether it's supposed to show up? I'm just assuming that the first thing the bare metal machine should do is ask for an IP from the router (Or cobbler when that's being used). Actually do I need to do anything to the router to make sure that the bare metal machine can see cobbler's dhcp server? There's a router setting "Use External DHCP Server" that I tried setting to the cobbler server's IP, but then I get this message: Invalid External DHCP Server IP Address, and it must not be a LAN IP address. So I just assumed that I should turn off the router's DCHP server, and the bare metal machine would then automatically get it from the cobbler server. Am I going about this right? Thanks, - Ole Peter Wright wrote: > Ole Ersoy wrote: >> Hi, >> >> I'm trying to boot some bare metal over the network. I followed all >> the steps in the tutorial: >> >> https://fedorahosted.org/cobbler/wiki/HowToCobblerServer >> >> When I run cobbler check it just tells me this: >> >> [root at localhost cobbler]# cobbler check >> The following potential problems were detected: >> #0: since iptables may be running, ensure 69, 80, 25150, and 25151 are >> unblocked >> >> And I did make sure that these ports are open. I then set the bare >> metal bios to boot from the LAN first, but I still get a "NO BOOT DISK >> FOUND ERROR...or something like that". >> >> Is there a way to simulate the request that boot client should be >> sending to cobbler? >> >> > > There are a couple things you can check - during your pxeboot phase do > you see a timeout or anything like that? I find it helpful to do a > tcpdump on the cobbler box to see if infact the pxe and dhcp packets are > making it from the host to the server. > > You can also tail /var/log/messages (by default) or /var/log/dhcpd (if > you setup a syslogd rule for dhcpd) to see if there are requests getting > through to the server. One common mistake I make is that I forget to > add a subnet in my dhcp.template, and because of this my pxe requests > will fail (as dhcpd won't be configured to offer leases on that > subnet). This will usually show up in your log file. > > > -pete > > > > From ole.ersoy at gmail.com Wed May 21 01:12:33 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Tue, 20 May 2008 20:12:33 -0500 Subject: [et-mgmt-tools] Is a switch required? In-Reply-To: <48332009.501@imageworks.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> Message-ID: <48337701.8040509@gmail.com> Hi, I've been attempting to trouble shoot my attempt at pxe booting by reading through this: http://www.linux-sxs.org/internet_serving/pxeboot.html One of the things it says is a requirement is a switch. Right now I'm using a router, for which I just turned off DHCP. Anyone know if a switch is a requirement? Another thing I noticed is that this was entered in the DCHP config file: host llama0 { hardware ethernet 04:4B:80:80:80:03; fixed-address 192.168.0.254; option host-name "llama0"; filename "pxelinux.0"; } I don't have this in the DHCP config template that comes with cobbler. Is something like this required? Thanks, - Ole From rizwank at geekymedia.com Wed May 21 03:15:03 2008 From: rizwank at geekymedia.com (Rizwan Kassim) Date: Tue, 20 May 2008 20:15:03 -0700 Subject: [et-mgmt-tools] CFT Support for OSX In-Reply-To: References: Message-ID: I've started using CFT on some of our Linux machines - it's quite impressive. I've started to look into using CFT on OSX. There's nothing as simple and easy to use as FAM/Gamin on OSX, but there is the fseventer framework. http://fernlightning.com/doku.php?id=software:fseventer:start I'm not a skilled Ruby developer by any means, but iI'll give it a shot (along with my other projects - this probably won't be a fast thing to do). David suggested I email the list. There's also a Ruby module to make Ruby and FSEvents play together. http://www.dribin.org/dave/blog/archives/2008/01/04/fswatch/ I suppose a great way to start would be to better understand what CFT needs from ruby-fam? Does it just need file modification events or the content of the modification as well? Thanks! -- Rizwan Kassim Software and Systems Engineer, uWink Interactive Bistro http://www.geekymedia.com If there is any great secret of success in life, it lies in the ability to put yourself in the other's place and to see things from his point of view -- as well as your own. -- Henry Ford From sghosh at redhat.com Wed May 21 05:05:00 2008 From: sghosh at redhat.com (Subhendu Ghosh) Date: Wed, 21 May 2008 01:05:00 -0400 Subject: [et-mgmt-tools] Is a switch required? In-Reply-To: <48337701.8040509@gmail.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> <48337701.8040509@gmail.com> Message-ID: <4833AD7C.1080109@redhat.com> Ole Ersoy wrote: > Hi, > > I've been attempting to trouble shoot my attempt at pxe booting by > reading through this: > > http://www.linux-sxs.org/internet_serving/pxeboot.html > > One of the things it says is a requirement is a switch. Right now I'm > using a router, for which I just turned off DHCP. Anyone know if a > switch is a requirement? > > Another thing I noticed is that this was entered in the DCHP config file: > > host llama0 { > hardware ethernet 04:4B:80:80:80:03; > fixed-address 192.168.0.254; > option host-name "llama0"; > filename "pxelinux.0"; > } > > I don't have this in the DHCP config template that comes with cobbler. > Is something like this required? > Thanks, > - Ole > DHCP works via subnet broadcast - if you are doing DHCP across a router, the router needs to have dhcp-relay function turned on. Otherwise you are stuck. The filename is set in the main ISC dhcp definition if you are using the shipped /etc/cobbler/dhcp.template. Hence it is inherited by individual entries. No need to set it individually for each host entry. -cheers Subhendu From ole.ersoy at gmail.com Wed May 21 14:39:44 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Wed, 21 May 2008 09:39:44 -0500 Subject: [et-mgmt-tools] Is a switch required? In-Reply-To: <4833AD7C.1080109@redhat.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> <48337701.8040509@gmail.com> <4833AD7C.1080109@redhat.com> Message-ID: <48343430.2000400@gmail.com> > > DHCP works via subnet broadcast - if you are doing DHCP across a router, > the router needs to have dhcp-relay function turned on. Otherwise you > are stuck. Ah - That makes sense - I'll see if that's turned on - Thanks. > > The filename is set in the main ISC dhcp definition if you are using the > shipped /etc/cobbler/dhcp.template. I assume the "filename" corresponds to "option routers", which I set to the IP address of my router. My understanding is that ISC DHCP would broadcast to the router using that routers IP, and then the router would relay from there. Please correct me if I'm off. > Hence it is inherited by individual > entries. No need to set it individually for each host entry. Awesome - Thanks for all your help! Ole From ole.ersoy at gmail.com Wed May 21 15:37:13 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Wed, 21 May 2008 10:37:13 -0500 Subject: [et-mgmt-tools] Is a switch required? In-Reply-To: <4833AD7C.1080109@redhat.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> <48337701.8040509@gmail.com> <4833AD7C.1080109@redhat.com> Message-ID: <483441A9.1050205@gmail.com> Subhendu, Sorry for the stooopid filename clarification request. I see what you mean. The main entry has it, so the "host llama0" entry does not need it. I'm just reading up on DHCP now, so it's starting to come to me :-). Thanks, - Ole Subhendu Ghosh wrote: > Ole Ersoy wrote: >> Hi, >> >> I've been attempting to trouble shoot my attempt at pxe booting by >> reading through this: >> >> http://www.linux-sxs.org/internet_serving/pxeboot.html >> >> One of the things it says is a requirement is a switch. Right now I'm >> using a router, for which I just turned off DHCP. Anyone know if a >> switch is a requirement? >> >> Another thing I noticed is that this was entered in the DCHP config file: >> >> host llama0 { >> hardware ethernet 04:4B:80:80:80:03; >> fixed-address 192.168.0.254; >> option host-name "llama0"; >> filename "pxelinux.0"; >> } >> >> I don't have this in the DHCP config template that comes with >> cobbler. Is something like this required? Thanks, >> - Ole >> > > DHCP works via subnet broadcast - if you are doing DHCP across a router, > the router needs to have dhcp-relay function turned on. Otherwise you > are stuck. > > The filename is set in the main ISC dhcp definition if you are using the > shipped /etc/cobbler/dhcp.template. Hence it is inherited by individual > entries. No need to set it individually for each host entry. > > -cheers > Subhendu > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From sghosh at redhat.com Wed May 21 15:58:28 2008 From: sghosh at redhat.com (Subhendu Ghosh) Date: Wed, 21 May 2008 11:58:28 -0400 Subject: [et-mgmt-tools] Is a switch required? In-Reply-To: <48343430.2000400@gmail.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> <48337701.8040509@gmail.com> <4833AD7C.1080109@redhat.com> <48343430.2000400@gmail.com> Message-ID: <483446A4.8060101@redhat.com> Ole Ersoy wrote: >> >> DHCP works via subnet broadcast - if you are doing DHCP across a >> router, the router needs to have dhcp-relay function turned on. >> Otherwise you are stuck. > > Ah - That makes sense - I'll see if that's turned on - Thanks. >> >> The filename is set in the main ISC dhcp definition if you are using >> the shipped /etc/cobbler/dhcp.template. > > I assume the "filename" corresponds to "option routers", which I set to > the IP address of my router. My understanding is that ISC DHCP would > broadcast to the router using that routers IP, and then the router would > relay from there. Please correct me if I'm off. > Off slightly - filename corresponds to the /path/filename requested for pxe boot options routers - provides the default gateway. >> Hence it is inherited by individual entries. No need to set it >> individually for each host entry. > > Awesome - Thanks for all your help! > > Ole -- -regards Subhendu From ole.ersoy at gmail.com Wed May 21 16:37:18 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Wed, 21 May 2008 11:37:18 -0500 Subject: [et-mgmt-tools] Troubleshooting (Checked DHCPD) - Anything Special About PXE Booting BIOS Settings? In-Reply-To: <483446A4.8060101@redhat.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> <48337701.8040509@gmail.com> <4833AD7C.1080109@redhat.com> <48343430.2000400@gmail.com> <483446A4.8060101@redhat.com> Message-ID: <48344FBE.5000605@gmail.com> Hey Guys, I checked DHCP by switching the boot protocol of another machine on the network to dhcp to see whether it would obtain and IP and it did. I then tailed /var/log/messages to verify that DHCPD logged the lease, and that checked out as well. I then tried booting the bare metal again, but it still is not obtaining a lease. I assume that the only thing necessary for this is to set the first boot device to "LAN" in the bare metal bios? Anyone know if anything else is necessary? The network card light is on, and I took the cable from another machine on the network, so I know the connection is good. Thanks, - Ole From jensampol at yahoo.de Wed May 21 16:39:09 2008 From: jensampol at yahoo.de (Jens Ahrens) Date: Wed, 21 May 2008 16:39:09 +0000 (GMT) Subject: [et-mgmt-tools] Scientific Linux 4.6 paravirt install works - FC7 or FC8 not Message-ID: <199431.9404.qm@web25506.mail.ukl.yahoo.com> Hi everybody, I'm running FC7 as Dom0 and set up some paravirt SL 4.6 machines with cobbler-0.8.0-1.fc7 and koan-0.8.0-1.fc7. When I tried to setup a FC7 pvm the console output stops at a certain point: Write protecting the kernel read-only data: 938k Almost the same result for FC8 - it stops saying: Determining IP Adress for eth0 and hangs forever. When I set up a dhcp server on Dom0 manually for e.g. the FC8-Install the installation goes fairly automatically (some queries that a package can't be found - after retrying it is found). My question why does a SL4.6 install work without a dedicated dhcp server and FC7 or FC8 do not? Thanks for any help, Cheers, Jens __________________________________________________________ Gesendet von Yahoo! Mail. Dem pfiffigeren Posteingang. http://de.overview.mail.yahoo.com From dlutter at redhat.com Wed May 21 18:25:16 2008 From: dlutter at redhat.com (David Lutterkort) Date: Wed, 21 May 2008 18:25:16 +0000 Subject: [et-mgmt-tools] CFT Support for OSX In-Reply-To: References: Message-ID: <1211394316.29522.324.camel@localhost.localdomain> On Tue, 2008-05-20 at 20:15 -0700, Rizwan Kassim wrote: > I've started using CFT on some of our Linux machines - it's quite impressive. Thanks :) > I'm not a skilled Ruby developer by any means, but iI'll give it a > shot (along with my other projects - this probably won't be a fast > thing to do). David suggested I email the list. For file change notifications, have you looked if FAM is available for OS/X ? SGI's site[1] claims there is support for a number of *BSD's, though I have no idea whether any of that works on OS X. > I suppose a great way to start would be to better understand what CFT > needs from ruby-fam? Does it just need file modification events or the > content of the modification as well? It only needs notification that a modification has occurred; the FAM specific code is in lib/cft.rb. It uses the following features of FAM: * Notification that a file has been changed, added, or deleted * Tell FAM to watch a file and a directory; cancel those watches Besides getting your hands on a FAM-equivalent, there's a few RedHatisms in the code, most notably, that cft will hunt through the RPM database to notice package install/uninstall/update and to suppress notifications if a file has changed solely because of a package operation. That part doens't make much sense on OS/X. It should be relatively easy to suppress that functionality based on testing whether RPM is on the system or not. Most of the other things that cft notices, e.g. service restarts, are based on Puppet, and therefore should work on OS/X. David [1] http://oss.sgi.com/projects/fam/ From tomackell at duke-energy.com Thu May 22 14:32:04 2008 From: tomackell at duke-energy.com (Mackell, Thomas O) Date: Thu, 22 May 2008 10:32:04 -0400 Subject: [et-mgmt-tools] (no subject) In-Reply-To: <20080521160035.1E0FF61A21A@hormel.redhat.com> References: <20080521160035.1E0FF61A21A@hormel.redhat.com> Message-ID: <1B3160038A79B54DB674532195C68A586DAC72@IMCLTEXCP15.nam.ent.duke-energy.com> I am running the latest versions of cobbler and koan. cobbler-0.9.2-2.fc8 koan-0.9.2-2.fc8. I just love the buildiso option, works great for our environment. I have not found any major problems with the upgrade to the release 9 code. Other than a couple of my snippets not working correctly, until I tripped across this one: I figured it was something I have wrong, or set up incorrectly. I am running cobbler on Fedora FC 8, with all of the latest patches/updates. When I issue the cobbler status command: [root at linux1 bin]# cobbler status need more than 1 value to unpack File "/usr/lib/python2.5/site-packages/cobbler/cobbler.py", line 65, in main return BootCLI().run(sys.argv) File "/usr/lib/python2.5/site-packages/cobbler/cobbler.py", line 44, in run return self.loader.run(args) File "/usr/lib/python2.5/site-packages/cobbler/commands.py", line 69, in run return fn.run() File "/usr/lib/python2.5/site-packages/cobbler/modules/cli_misc.py", line 198, in run return self.api.status("text") # no other output modes supported yet File "/usr/lib/python2.5/site-packages/cobbler/api.py", line 354, in status return statusifier.run() File "/usr/lib/python2.5/site-packages/cobbler/action_status.py", line 152, in run self.scan_logfiles() File "/usr/lib/python2.5/site-packages/cobbler/action_status.py", line 60, in scan_logfiles (profile_or_system, name, ip, start_or_stop, ts) = tokens Any ideas of where to start in looking for the problem? I am not much of a python programmer, just started reading a book on it. Thank You, Thomas Mackell -------------- next part -------------- An HTML attachment was scrubbed... URL: From ole.ersoy at gmail.com Thu May 22 15:25:54 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Thu, 22 May 2008 10:25:54 -0500 Subject: [et-mgmt-tools] Configuration Testing Wiki Contribution In-Reply-To: <483446A4.8060101@redhat.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> <48337701.8040509@gmail.com> <4833AD7C.1080109@redhat.com> <48343430.2000400@gmail.com> <483446A4.8060101@redhat.com> Message-ID: <48359082.5060801@gmail.com> Hi, I managed to finally get a "raw box" to obtain a lease, thanks to Subhendu and Pete's very helpful comments. I'm thinking about adding a "Setup Verification Steps" section to the wiki that would have something like: "Verifying DHCP Configuration" "Selecting PXE Booting On Your Client" Those were the two trickiest things for me. The last one is probably hardware dependent, but I figured I'd throw some notes in since it might help others work through it on different types of hardware. Any thoughts / sound OK? I'd be happy to add this to the end of a pre-existing section if we think that's appropriate. Thanks, - Ole From pregier at ittc.ku.edu Thu May 22 17:52:22 2008 From: pregier at ittc.ku.edu (Phil Regier) Date: Thu, 22 May 2008 12:52:22 -0500 Subject: [et-mgmt-tools] What ever happened to "Stateless Linux"? Message-ID: <4835B2D6.6030007@ittc.ku.edu> Sorry if this is the wrong place to ask, but it would appear that the guides which popped up at the time of the FC6-FC7 transition have gone more or less defunct since then: http://fedoraproject.org/wiki/StatelessLinux Since this doesn't seem like the sort of thing that the Fedora-type families of distros would leave by the wayside (at the time it seemed highly relevant to the long-term direction of these distros), I was wondering if this type of configuration has been absorbed into any other emerging technology management projects. Any thoughts? Many thanks... Phil From berrange at redhat.com Thu May 22 17:55:03 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 22 May 2008 18:55:03 +0100 Subject: [et-mgmt-tools] What ever happened to "Stateless Linux"? In-Reply-To: <4835B2D6.6030007@ittc.ku.edu> References: <4835B2D6.6030007@ittc.ku.edu> Message-ID: <20080522175503.GS26567@redhat.com> On Thu, May 22, 2008 at 12:52:22PM -0500, Phil Regier wrote: > Sorry if this is the wrong place to ask, but it would appear that the > guides which popped up at the time of the FC6-FC7 transition have gone > more or less defunct since then: You are correct. THis is the wrong place. Pay a visit to the Fedora development list http://www.redhat.com/mailman/listinfo/fedora-devel-list Dan -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From mdehaan at redhat.com Thu May 22 19:23:41 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 22 May 2008 15:23:41 -0400 Subject: [et-mgmt-tools] Barebones Install Without Koan? In-Reply-To: <483230C6.60304@gmail.com> References: <483230C6.60304@gmail.com> Message-ID: <4835C83D.3010704@redhat.com> Ole Ersoy wrote: > Hi, > > Currently I use a USB boot drive to do network installs on new > hardware (The hardware is minimal, with no CD or DVD drive). I'd like > to get around having to use the USB drive. Ideally I would just plug > a raw machine into the network and be able to click through the > install from there. Is that possible with Cobbler (Or do I have to > have a kickstart profile for the hardware before starting)? Cobbler buildiso in 0.9.X/1.0 generates bootable images, and before you also have the koan live CD (read the Wiki for docs on this) that can be converted to a USB image. > > Thanks, > - Ole > > _______________________________________________ > 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 May 22 19:25:53 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 22 May 2008 15:25:53 -0400 Subject: [et-mgmt-tools] Live- and rescue-system provisioning over pxe In-Reply-To: <52278.193.5.216.100.1211273805.squirrel@mail.worldweb2000.com> References: <52278.193.5.216.100.1211273805.squirrel@mail.worldweb2000.com> Message-ID: <4835C8C1.6010705@redhat.com> zuber at puzzle.ch wrote: > What do you think about providing live- and rescue systems for pxe with > cobbler? With Fedora 9 there is a new script in the livecd-tools that > turns a live-cd iso into a pxe bootable image. So we can now create a pxe > live-system out of a kickstartfile. > See koan's livecd. It uses this to provide baremetal installations in PXE-less environments. > Providing rescue-system is already possible with cobbler. You only have to > create a Kickstart file and a profile to provide it over pxe: > http://wiki.centos.org/HowTos/PXE/Rescue_PXE > There's an email a while back from Jennifer Cranfill about managing rescue images in a neat way with Cobbler. We will probably be exploring this. > My question is, does it makes sense to manage the build process of the > live images with cobbler? It is always possible to do this manualy, but i > think it will be cool if cobbler does that for me. > The build process? Not really, IMHO. Does it make sense to make it easier for Cobbler to netboot non-installation images and handle abstracts around disk images and cloning? Yes. That's one of the things I want to explore next. > Greetings > Andy > > _______________________________________________ > 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 May 22 19:28:01 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 22 May 2008 15:28:01 -0400 Subject: [et-mgmt-tools] Scientific Linux 4.6 paravirt install works - FC7 or FC8 not In-Reply-To: <199431.9404.qm@web25506.mail.ukl.yahoo.com> References: <199431.9404.qm@web25506.mail.ukl.yahoo.com> Message-ID: <4835C941.2050704@redhat.com> Jens Ahrens wrote: > Hi everybody, > > I'm running FC7 as Dom0 and set up some paravirt SL 4.6 machines with cobbler-0.8.0-1.fc7 and koan-0.8.0-1.fc7. When I tried to setup a FC7 pvm the console output stops at a certain point: > > Write protecting the kernel read-only data: 938k > > Almost the same result for FC8 - it stops saying: > > Determining IP Adress for eth0 > > and hangs forever. When I set up a dhcp server on Dom0 manually for e.g. the FC8-Install the installation goes fairly automatically (some queries that a package can't be found - after retrying it is found). > My question why does a SL4.6 install work without a dedicated dhcp server and FC7 or FC8 do not? > Ten to one it's not actually hanging. What happens is that Xen has switched over from showing the console to showing via VNC. To suppress this behavior, use koan with --nogfx or just switch to using virt-manager at that point. > Thanks for any help, > Cheers, > Jens > > > > __________________________________________________________ > Gesendet von Yahoo! Mail. > Dem pfiffigeren Posteingang. > http://de.overview.mail.yahoo.com > > _______________________________________________ > 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 May 22 19:37:16 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 22 May 2008 15:37:16 -0400 Subject: [et-mgmt-tools] (no subject) In-Reply-To: <1B3160038A79B54DB674532195C68A586DAC72@IMCLTEXCP15.nam.ent.duke-energy.com> References: <20080521160035.1E0FF61A21A@hormel.redhat.com> <1B3160038A79B54DB674532195C68A586DAC72@IMCLTEXCP15.nam.ent.duke-energy.com> Message-ID: <4835CB6C.7030700@redhat.com> Mackell, Thomas O wrote: > > I am running the latest versions of cobbler and koan. cobbler-0.9.2-2.fc8 > > koan-0.9.2-2.fc8. I just love the buildiso option, works great for our > environment. > > I have not found any major problems with the upgrade to the release 9 > code. Other than a couple of my snippets not working correctly, until > I tripped across this one: I figured it was something I have wrong, or > set up incorrectly. > > I am running cobbler on Fedora FC 8, with all of the latest > patches/updates. > > When I issue the cobbler status command: > > [root at linux1 bin]# cobbler status > > need more than 1 value to unpack > > File "/usr/lib/python2.5/site-packages/cobbler/cobbler.py", line 65, > in main > > return BootCLI().run(sys.argv) > > File "/usr/lib/python2.5/site-packages/cobbler/cobbler.py", line > 44, in run > > return self.loader.run(args) > > File "/usr/lib/python2.5/site-packages/cobbler/commands.py", line > 69, in run > > return fn.run() > > File > "/usr/lib/python2.5/site-packages/cobbler/modules/cli_misc.py", line > 198, in run > > return self.api.status("text") # no other output modes supported yet > > File "/usr/lib/python2.5/site-packages/cobbler/api.py", line 354, > in status > > return statusifier.run() > > File "/usr/lib/python2.5/site-packages/cobbler/action_status.py", > line 152, in run > > self.scan_logfiles() > > File "/usr/lib/python2.5/site-packages/cobbler/action_status.py", > line 60, in scan_logfiles > > (profile_or_system, name, ip, start_or_stop, ts) = tokens > > > Any ideas of where to start in looking for the problem? > > I am not much of a python programmer, just started reading a book on it. > > > Thank You, > > Thomas Mackell > rm /var/lib/cobbler/install.log and see if that helps the problem. It may be you have some other noise in that file from a previous install. Failing that, email me the file and I'll check it out. (Actually, email me the file and then rm it). --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 May 22 19:39:54 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 22 May 2008 15:39:54 -0400 Subject: [et-mgmt-tools] Is a switch required? In-Reply-To: <483441A9.1050205@gmail.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> <48337701.8040509@gmail.com> <4833AD7C.1080109@redhat.com> <483441A9.1050205@gmail.com> Message-ID: <4835CC0A.2010608@redhat.com> Ole Ersoy wrote: > Subhendu, > > Sorry for the stooopid filename clarification request. I see what you > mean. The main entry has it, so the "host llama0" entry does not need > it. I'm just reading up on DHCP now, so it's starting to come to me :-). > > Thanks, > - Ole > > So, long thread, everything settled? FWIW, a good place for troubleshooting discussion is #cobbler on irc.freenode.net. I won't be online there until next week (vacation still) though there are a lot of other people there that can help if you're still trying to figure things out. --Michael > > Subhendu Ghosh wrote: >> Ole Ersoy wrote: >>> Hi, >>> >>> I've been attempting to trouble shoot my attempt at pxe booting by >>> reading through this: >>> >>> http://www.linux-sxs.org/internet_serving/pxeboot.html >>> >>> One of the things it says is a requirement is a switch. Right now >>> I'm using a router, for which I just turned off DHCP. Anyone know >>> if a switch is a requirement? >>> >>> Another thing I noticed is that this was entered in the DCHP config >>> file: >>> >>> host llama0 { >>> hardware ethernet 04:4B:80:80:80:03; >>> fixed-address 192.168.0.254; >>> option host-name "llama0"; >>> filename "pxelinux.0"; >>> } >>> >>> I don't have this in the DHCP config template that comes with >>> cobbler. Is something like this required? Thanks, >>> - Ole >>> >> >> DHCP works via subnet broadcast - if you are doing DHCP across a >> router, the router needs to have dhcp-relay function turned on. >> Otherwise you are stuck. >> >> The filename is set in the main ISC dhcp definition if you are using >> the shipped /etc/cobbler/dhcp.template. Hence it is inherited by >> individual entries. No need to set it individually for each host entry. >> >> -cheers >> Subhendu >> >> _______________________________________________ >> 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 ole.ersoy at gmail.com Thu May 22 19:58:46 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Thu, 22 May 2008 14:58:46 -0500 Subject: [et-mgmt-tools] Is a switch required? In-Reply-To: <4835CC0A.2010608@redhat.com> References: <48331B52.7090702@gmail.com> <48332009.501@imageworks.com> <48337701.8040509@gmail.com> <4833AD7C.1080109@redhat.com> <483441A9.1050205@gmail.com> <4835CC0A.2010608@redhat.com> Message-ID: <4835D076.6030102@gmail.com> Michael, Yes I got that part figured out. I'll also put my troubleshooting notes on the wiki a little later. It ended up being the original hardware that just did not know how to request a DHCP lease. It works fine when I try other hardware. Now I just got to get past the "TFTP Open Timeout". I'm sure it's something trivial in the config setup, so I'm just going to go through the tutorials again and make sure I buttoned everything up properly. Thanks for checking in - Especially on vacation and all! - Ole Michael DeHaan wrote: > Ole Ersoy wrote: >> Subhendu, >> >> Sorry for the stooopid filename clarification request. I see what you >> mean. The main entry has it, so the "host llama0" entry does not need >> it. I'm just reading up on DHCP now, so it's starting to come to me :-). >> >> Thanks, >> - Ole >> >> > > So, long thread, everything settled? > FWIW, a good place for troubleshooting discussion is #cobbler on > irc.freenode.net. > I won't be online there until next week (vacation still) though there > are a lot of other people there that can help if you're still trying > to figure things out. > > --Michael > >> >> Subhendu Ghosh wrote: >>> Ole Ersoy wrote: >>>> Hi, >>>> >>>> I've been attempting to trouble shoot my attempt at pxe booting by >>>> reading through this: >>>> >>>> http://www.linux-sxs.org/internet_serving/pxeboot.html >>>> >>>> One of the things it says is a requirement is a switch. Right now >>>> I'm using a router, for which I just turned off DHCP. Anyone know >>>> if a switch is a requirement? >>>> >>>> Another thing I noticed is that this was entered in the DCHP config >>>> file: >>>> >>>> host llama0 { >>>> hardware ethernet 04:4B:80:80:80:03; >>>> fixed-address 192.168.0.254; >>>> option host-name "llama0"; >>>> filename "pxelinux.0"; >>>> } >>>> >>>> I don't have this in the DHCP config template that comes with >>>> cobbler. Is something like this required? Thanks, >>>> - Ole >>>> >>> >>> DHCP works via subnet broadcast - if you are doing DHCP across a >>> router, the router needs to have dhcp-relay function turned on. >>> Otherwise you are stuck. >>> >>> The filename is set in the main ISC dhcp definition if you are using >>> the shipped /etc/cobbler/dhcp.template. Hence it is inherited by >>> individual entries. No need to set it individually for each host entry. >>> >>> -cheers >>> Subhendu >>> >>> _______________________________________________ >>> 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 > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From ole.ersoy at gmail.com Thu May 22 20:11:25 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Thu, 22 May 2008 15:11:25 -0500 Subject: [et-mgmt-tools] Barebones Install Without Koan? In-Reply-To: <4835C83D.3010704@redhat.com> References: <483230C6.60304@gmail.com> <4835C83D.3010704@redhat.com> Message-ID: <4835D36D.2090308@gmail.com> Michael, Just want to make sure I understand correctly. Michael DeHaan wrote: > Ole Ersoy wrote: >> Hi, >> >> Currently I use a USB boot drive to do network installs on new >> hardware (The hardware is minimal, with no CD or DVD drive). I'd like >> to get around having to use the USB drive. Ideally I would just plug >> a raw machine into the network and be able to click through the >> install from there. Is that possible with Cobbler (Or do I have to >> have a kickstart profile for the hardware before starting)? > > Cobbler buildiso in 0.9.X/1.0 generates bootable images, So would I create a bootable image and load that instead of pxelinux.0 in order to get the anaconda installer gui that I use to finish up the network install, or does pxelinux.0 provide that as well? > and before you > also have the koan live CD (read the Wiki for docs on this) that can be > converted to a USB image. Just want to make sure I presented the idea correctly - I'm trying to get around using a USB drive to get the initial anaconda linux prompt where I can specify either a kickstart install or a manual one over the network...so would I create a USB image, and load that via DHCP and TFTP instead of pxelinux.0? Thanks again for helping me out with these newbie questions...especially on vacation. No rush on any of this. I'm playing around and as soon as I get the pxelinux.0 to load, it will be a little clearer what I can do from that point onwards. Oh - And thanks for the IRC tip as well. I'll try that next if all my permutations don't work...I think I need to track them a little better :-). Cheers, - Ole From jensampol at yahoo.de Thu May 22 20:58:43 2008 From: jensampol at yahoo.de (Jens Ahrens) Date: Thu, 22 May 2008 20:58:43 +0000 (GMT) Subject: [et-mgmt-tools] Scientific Linux 4.6 paravirt install works - FC7 or FC8 not In-Reply-To: <4835C941.2050704@redhat.com> Message-ID: <640201.41119.qm@web25503.mail.ukl.yahoo.com> Hi Michael, thanks for your reply. I've already read in some other thread that the install just seems to be hanging. However I used the --nogfx flag with koan and the result is what I posted in my original message below about the last output I see on the console. When I try to connect to it via virt-manager, I just get a message saying "Console not configured for guest". Is there any other way to connect to it via vnc? Thanks, Jens --- Michael DeHaan schrieb am Do, 22.5.2008: > Von: Michael DeHaan > Betreff: Re: [et-mgmt-tools] Scientific Linux 4.6 paravirt install works - FC7 or FC8 not > An: jensampol at yahoo.de, "Fedora/Linux Management Tools" > Datum: Donnerstag, 22. Mai 2008, 21:28 > Jens Ahrens wrote: > > Hi everybody, > > > > I'm running FC7 as Dom0 and set up some paravirt > SL 4.6 machines with cobbler-0.8.0-1.fc7 and > koan-0.8.0-1.fc7. When I tried to setup a FC7 pvm the > console output stops at a certain point: > > > > Write protecting the kernel read-only data: 938k > > > > Almost the same result for FC8 - it stops saying: > > > > Determining IP Adress for eth0 > > > > and hangs forever. When I set up a dhcp server on Dom0 > manually for e.g. the FC8-Install the installation goes > fairly automatically (some queries that a package can't > be found - after retrying it is found). > > My question why does a SL4.6 install work without a > dedicated dhcp server and FC7 or FC8 do not? > > > > Ten to one it's not actually hanging. > > What happens is that Xen has switched over from showing the > console to > showing via VNC. > > To suppress this behavior, use koan with --nogfx or just > switch to using > virt-manager at that point. > > > Thanks for any help, > > Cheers, > > Jens > > > > > > > > > __________________________________________________________ > > Gesendet von Yahoo! Mail. > > Dem pfiffigeren Posteingang. > > http://de.overview.mail.yahoo.com > > > > _______________________________________________ > > et-mgmt-tools mailing list > > et-mgmt-tools at redhat.com > > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > __________________________________________________________ Gesendet von Yahoo! Mail. Dem pfiffigeren Posteingang. http://de.overview.mail.yahoo.com From fj7025cf at aa.jp.fujitsu.com Fri May 23 02:51:15 2008 From: fj7025cf at aa.jp.fujitsu.com (Hiroyuki Kaguchi) Date: Fri, 23 May 2008 11:51:15 +0900 Subject: [et-mgmt-tools] [PATCH] Enable to send F10 key to the guest OS. Message-ID: <48363123.605@aa.jp.fujitsu.com> The F10 key is assigned as the accelerator of the menu bar in the OS(Windows or Linux). On this account it is necessary to send the F10 key to the guest OS. However, with the existing specifications of virt-viewer, the F10 key is assigned to the accelerator of the menu bar of virt-viewer and we cannot send the F10 key to the guest OS. So, I change virt-viewer to be able to send F10 key to the guest OS when virt-viewer grabs the keyboard. Thanks, Signed-off-by: Hiroyuki Kaguchi From fj7025cf at aa.jp.fujitsu.com Fri May 23 03:04:38 2008 From: fj7025cf at aa.jp.fujitsu.com (Hiroyuki Kaguchi) Date: Fri, 23 May 2008 12:04:38 +0900 Subject: [et-mgmt-tools] [PATCH] Enable to send F10 key to the guest OS. In-Reply-To: <48363123.605@aa.jp.fujitsu.com> References: <48363123.605@aa.jp.fujitsu.com> Message-ID: <48363446.8020601@aa.jp.fujitsu.com> On 2008/05/23 11:51, Hiroyuki Kaguchi wrote: > The F10 key is assigned as the accelerator of the menu bar in the OS(Windows or > Linux). > On this account it is necessary to send the F10 key to the guest OS. > However, with the existing specifications of virt-viewer, the F10 key is > assigned to the accelerator > of the menu bar of virt-viewer and we cannot send the F10 key to the guest OS. > So, I change virt-viewer to be able to send F10 key to the guest OS when > virt-viewer grabs the keyboard. I forgot to attach the patch. I attach it. Thanks, Hiroyuki Kaguchi -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: menubar_accel.patch URL: From martin.pearson at japannext.co.jp Fri May 23 07:33:30 2008 From: martin.pearson at japannext.co.jp (Martin Pearson) Date: Fri, 23 May 2008 16:33:30 +0900 Subject: [et-mgmt-tools] Koan problem In-Reply-To: <4836728E.8070308@japannext.co.jp> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> Message-ID: <4836734A.5020207@japannext.co.jp> Hi All, Wonder if you can help me when I run Koan with the following inputs:- koan --virt --server=puppet.tk.japannext.co.jp --profile=RHEL4-ES-xen-x86_64 --virt-name=tkdvir1 I get the following error:- libvir: Xen error : Domain not found: xenUnifiedDomainLookupByUUID I am running it from the following cobbler profile:- RHEL4-ES-xen-x86_64 kernel : /var/www/cobbler/ks_mirror/RHEL4-ES-x86_64/images/xen/vmlinuz initrd : /var/www/cobbler/ks_mirror/RHEL4-ES-x86_64/images/xen/initrd.img kernel options : {} architecture : x86_64 ks metadata : {'tree': 'http://@@http_server@@/cblr/links/RHEL4-ES-xen-x86_64'} breed : redhat profile : RHEL4-ES-xen-x86_64 distro : RHEL4-ES-xen-x86_64 kickstart : /etc/cobbler/RHEL4-ES-xen-x86_64.cfg kernel options : {} ks metadata : {'os': 'RHEL4-ES-xen', 'arch': 'x86_64'} virt file size : 5 virt ram : 512 virt type : auto virt path : <> virt bridge : xenbr0 virt cpus : 1 repos : [] dhcp tag : default server : <> The virtual machine gets setup but is unusable after this. Kickstart file is as follows:- # System authorization information auth --useshadow --enablemd5 # System bootloader configuration bootloader --location=mbr # Partition clearing information clearpart --all --initlabel # Use text mode install text # Firewall configuration firewall --enabled # Run the Setup Agent on first boot firstboot --disable # System keyboard keyboard us # System language lang en_US # Use network installation url --url=http://10.145.100.35/cblr/links/RHEL4-ES-xen-x86_64 # Network information network --bootproto=dhcp --device=eth0 --onboot=on # Reboot after installation reboot #Root password rootpw --iscrypted $1$mF86/UHC$WvcIcX2t6crBz2onWxyac. # SELinux configuration selinux --disabled # Do not configure the X Window System skipx # System timezone timezone America/New_York # Install OS instead of upgrade install # Clear the Master Boot Record zerombr # Magically figure out how to partition this thing %include /tmp/partinfo %pre # Determine how many drives we have set $(list-harddrives) let numd=$#/2 d1=$1 d2=$3 cat << EOF > /tmp/partinfo part / --fstype ext3 --size=1024 --grow --ondisk=$d1 --asprimary part swap --size=1024 --ondisk=$d1 --asprimary EOF %packages %post wget "http://10.145.100.35/cobbler/kickstarts_sys/tkdvir1/ks.cfg" -O /root/cobbler It would be wonderful if someone could give me an idea of why this is going wrong. Also if anymore information is required to help solve the problem let me know. Thanks, Martin. From martin.pearson at japannext.co.jp Fri May 23 07:46:01 2008 From: martin.pearson at japannext.co.jp (Martin Pearson) Date: Fri, 23 May 2008 16:46:01 +0900 Subject: [et-mgmt-tools] Re: Koan problem In-Reply-To: <4836734A.5020207@japannext.co.jp> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> Message-ID: <48367639.6000108@japannext.co.jp> Martin Pearson wrote: > Hi All, > Wonder if you can help me when I run Koan with the following inputs:- > > koan --virt --server=puppet.tk.japannext.co.jp > --profile=RHEL4-ES-xen-x86_64 --virt-name=tkdvir1 > > I get the following error:- > > libvir: Xen error : Domain not found: xenUnifiedDomainLookupByUUID > > I am running it from the following cobbler profile:- > > RHEL4-ES-xen-x86_64 > kernel : > /var/www/cobbler/ks_mirror/RHEL4-ES-x86_64/images/xen/vmlinuz > initrd : > /var/www/cobbler/ks_mirror/RHEL4-ES-x86_64/images/xen/initrd.img > kernel options : {} > architecture : x86_64 > ks metadata : {'tree': > 'http://@@http_server@@/cblr/links/RHEL4-ES-xen-x86_64'} > breed : redhat > > profile : RHEL4-ES-xen-x86_64 > distro : RHEL4-ES-xen-x86_64 > kickstart : /etc/cobbler/RHEL4-ES-xen-x86_64.cfg > kernel options : {} > ks metadata : {'os': 'RHEL4-ES-xen', 'arch': 'x86_64'} > virt file size : 5 > virt ram : 512 > virt type : auto > virt path : <> > virt bridge : xenbr0 > virt cpus : 1 > repos : [] > dhcp tag : default > server : <> > > > The virtual machine gets setup but is unusable after this. > > Kickstart file is as follows:- > > # System authorization information > auth --useshadow --enablemd5 > # System bootloader configuration > bootloader --location=mbr > # Partition clearing information > clearpart --all --initlabel > # Use text mode install > text > # Firewall configuration > firewall --enabled > # Run the Setup Agent on first boot > firstboot --disable > # System keyboard > keyboard us > # System language > lang en_US > # Use network installation > url --url=http://10.145.100.35/cblr/links/RHEL4-ES-xen-x86_64 > # Network information > network --bootproto=dhcp --device=eth0 --onboot=on > # Reboot after installation > reboot > > #Root password > rootpw --iscrypted $1$mF86/UHC$WvcIcX2t6crBz2onWxyac. > # SELinux configuration > selinux --disabled > # Do not configure the X Window System > skipx > # System timezone > timezone America/New_York > # Install OS instead of upgrade > install > # Clear the Master Boot Record > zerombr > > # Magically figure out how to partition this thing > %include /tmp/partinfo > > %pre > # Determine how many drives we have > set $(list-harddrives) > let numd=$#/2 > d1=$1 > d2=$3 > > cat << EOF > /tmp/partinfo > part / --fstype ext3 --size=1024 --grow --ondisk=$d1 --asprimary > part swap --size=1024 --ondisk=$d1 --asprimary > EOF > > > > %packages > > %post > > > wget "http://10.145.100.35/cobbler/kickstarts_sys/tkdvir1/ks.cfg" -O > /root/cobbler > > It would be wonderful if someone could give me an idea of why this is > going wrong. > > Also if anymore information is required to help solve the problem let > me know. > > Thanks, > > Martin. > > P.S. XEN machines work fine when created with virt-install. Thanks, Martin. From steph33560 at gmail.com Fri May 23 08:32:35 2008 From: steph33560 at gmail.com (Steph) Date: Fri, 23 May 2008 10:32:35 +0200 Subject: [et-mgmt-tools] Strange behaviour of virt-manager ... bug ? Message-ID: <48368123.4010605@gmail.com> Hello, Using virt-manager 0.5.3 under Ubuntu Hardy Heron 8.04, I'm facing a strange behaviour. when using virt-manager as root user, everything's fine. when using virt-manager as non root user, everything's getting complex... I get this error : -------snip Error starting domain: virDomainCreate() failed _ Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/engine.py", line 480, in run_domain vm.startup() File "/usr/share/virt-manager/virtManager/domain.py", line 379, in startup self.vm.create() File "/usr/lib/python2.5/site-packages/libvirt.py", line 240, in create if ret == -1: raise libvirtError ('virDomainCreate() failed', dom=self) libvirtError: virDomainCreate() failed -------/snip But, with non root user, when using "virsh start myvm", the vm starts correctly. Afterwards : * as non root : virt-manager does not display the VM running * as root : the vm is displayed to be running vncviewer localhost confirm that the VM is correctly running. Is it a bug, or something else ? Thanks for your help ! Steph From atodorov at redhat.com Fri May 23 10:04:22 2008 From: atodorov at redhat.com (Alexander Todorov) Date: Fri, 23 May 2008 12:04:22 +0200 Subject: [et-mgmt-tools] How do remote connections through virt-manager work? Message-ID: <483696A6.6080301@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Hi all, I've tried to use the remote connection option in virt-manager but it doesn't work for me. Hypervisor: Xen Connection: Remote tunnel over SHH Hostname: the hostname I have root access to the host. Versions on host & client are the same (same OS, xen, libvirt, etc.) The connection is not working, it is asking me if libvirtd is started on the remote host. Am I doing something wrong or this is a bug? Thanks, Alexander. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Red Hat - http://enigmail.mozdev.org iD8DBQFINpamhmd3WOiFct4RCs3OAKCUYL/ioOrdo2X7581ADXuiVnU4rQCgkT/1 7rjDcl51tZ2HXUX9pYLnsTs= =a6If -----END PGP SIGNATURE----- From rjones at redhat.com Fri May 23 10:07:23 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Fri, 23 May 2008 11:07:23 +0100 Subject: [et-mgmt-tools] How do remote connections through virt-manager work? In-Reply-To: <483696A6.6080301@redhat.com> References: <483696A6.6080301@redhat.com> Message-ID: <20080523100723.GA18708@amd.home.annexia.org> On Fri, May 23, 2008 at 12:04:22PM +0200, Alexander Todorov wrote: > Hi all, > I've tried to use the remote connection option in virt-manager but it doesn't > work for me. > > Hypervisor: Xen > Connection: Remote tunnel over SHH > Hostname: the hostname > > I have root access to the host. Versions on host & client are the same (same OS, > xen, libvirt, etc.) > > The connection is not working, it is asking me if libvirtd is started on the > remote host. Am I doing something wrong or this is a bug? There are a number of reasons why this might not work. Libvirt over SSH requires passwordless logins (eg. using ssh agent). It also needs a netcat ("nc") program available on the remote side and the netcat program has to support UDP sockets -- this isn't true on Debian for example. Take a look at the virt-manager log file (in /root/.virt-manager) for more clues and also tell us what versions and operating systems you're actually running. Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones virt-p2v converts physical machines to virtual machines. Boot with a live CD or over the network (PXE) and turn machines into Xen guests. http://et.redhat.com/~rjones/virt-p2v From atodorov at redhat.com Fri May 23 12:53:47 2008 From: atodorov at redhat.com (Alexander Todorov) Date: Fri, 23 May 2008 14:53:47 +0200 Subject: [et-mgmt-tools] How do remote connections through virt-manager work? In-Reply-To: <20080523100723.GA18708@amd.home.annexia.org> References: <483696A6.6080301@redhat.com> <20080523100723.GA18708@amd.home.annexia.org> Message-ID: <4836BE5B.6010101@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Richard W.M. Jones wrote: | There are a number of reasons why this might not work. | | Libvirt over SSH requires passwordless logins (eg. using ssh agent). | It also needs a netcat ("nc") program available on the remote side and | the netcat program has to support UDP sockets -- this isn't true on | Debian for example. | Are these requirements somewhere documented? I couldn't find this. | Take a look at the virt-manager log file (in /root/.virt-manager) for | more clues and also tell us what versions and operating systems you're | actually running. | on the client side: virt-manager-0.5.3-8.el5 on the server side: nc-1.84-10.fc6 libvirt-0.3.3-7.el5 xen-3.0.3-64.el5 both machines run with RHEL5.2. On the server I can loging without a password (my pub key is in /root/.ssh/authorized_keys) Attached is the relevant part of virt-manager.log Any more info I can provide? Thanks, Alexander. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Red Hat - http://enigmail.mozdev.org iD8DBQFINr5bhmd3WOiFct4RCp+eAJwKj88KvvTy4U4COGP7YFTueTyJJgCgmGAP nibHf5WhqNn0fjWs1ZGrjgg= =AyuQ -----END PGP SIGNATURE----- -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-manager.log Type: text/x-log Size: 2515 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-manager.log.sig Type: application/pgp-signature Size: 65 bytes Desc: not available URL: From ole.ersoy at gmail.com Fri May 23 16:51:00 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Fri, 23 May 2008 11:51:00 -0500 Subject: [et-mgmt-tools] TFTP open timeout troubleshooting In-Reply-To: <4836734A.5020207@japannext.co.jp> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> Message-ID: <4836F5F4.6020505@gmail.com> Hi Guys, I've gone over my settings up and down several times per the tutorials, and I'm still gettinga PXE open timeout on the client. One difference between the tutorial and what I'm doing is that when I run (Fedora 8): yum -y install cobbler koan tftp-server dhcp cft facter puppet subversion httpd named I get: - No package named available. - No package cft available. Are these necessary? The DHCP Lease part works fine. Any ideas on how to troubleshoot this further? I did make sure all the services are running and that I did a cobbler sync before attempting the run. TIA, - Ole From drew.einhorn at gmail.com Fri May 23 17:09:16 2008 From: drew.einhorn at gmail.com (drew einhorn) Date: Fri, 23 May 2008 11:09:16 -0600 Subject: [et-mgmt-tools] TFTP open timeout troubleshooting In-Reply-To: <4836F5F4.6020505@gmail.com> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> Message-ID: Don't need named, if you use dnsmasq instead of isc for dhcp. Don't know what cft is, Is there a firewall on your Fedora 8 box? Maybe you need to open up some ports. On Fri, May 23, 2008 at 10:51 AM, Ole Ersoy wrote: > Hi Guys, > > I've gone over my settings up and down several times per the tutorials, and > I'm still gettinga PXE open timeout on the client. One difference between > the tutorial and what I'm doing is that when I run (Fedora 8): > > yum -y install cobbler koan tftp-server dhcp cft facter puppet subversion > httpd named > > I get: > > - No package named available. > - No package cft available. > > Are these necessary? > > The DHCP Lease part works fine. > Any ideas on how to troubleshoot this further? I did make sure all the > services are running and that I did a cobbler sync before attempting the > run. > > TIA, > - Ole > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -- Drew Einhorn From ole.ersoy at gmail.com Fri May 23 17:42:01 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Fri, 23 May 2008 12:42:01 -0500 Subject: [et-mgmt-tools] TFTP open timeout troubleshooting In-Reply-To: References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> Message-ID: <483701E9.7090703@gmail.com> drew einhorn wrote: > Don't need named, if you use dnsmasq instead of isc for dhcp. Hmmm - I'm using dhcp from isc. I'll see if I can find a repository with the "named" dependency. > Don't know what cft is, > > Is there a firewall on your Fedora 8 box? Maybe you need to > open up some ports. There is and I opened up the ports it mentions when during cobbler check. I'll try to see if I can find a repository with named in it, and after installing, if it still is not working, I'll try switching off iptables all together. All fingers crossed :-) Thanks, - Ole > > On Fri, May 23, 2008 at 10:51 AM, Ole Ersoy wrote: >> Hi Guys, >> >> I've gone over my settings up and down several times per the tutorials, and >> I'm still gettinga PXE open timeout on the client. One difference between >> the tutorial and what I'm doing is that when I run (Fedora 8): >> >> yum -y install cobbler koan tftp-server dhcp cft facter puppet subversion >> httpd named >> >> I get: >> >> - No package named available. >> - No package cft available. >> >> Are these necessary? >> >> The DHCP Lease part works fine. >> Any ideas on how to troubleshoot this further? I did make sure all the >> services are running and that I did a cobbler sync before attempting the >> run. >> >> TIA, >> - Ole >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools >> > > > From ole.ersoy at gmail.com Fri May 23 17:54:30 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Fri, 23 May 2008 12:54:30 -0500 Subject: [et-mgmt-tools] "named rpm" yum repository In-Reply-To: References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> Message-ID: <483704D6.5060304@gmail.com> Hi, Does anyone know of a yum repository that has the "named" rpm in it. I tried googleing for "named rpm yum repository" and "named rpm", but no love. TIA, - Ole From tim.verhoeven.be at gmail.com Fri May 23 17:58:30 2008 From: tim.verhoeven.be at gmail.com (Tim Verhoeven) Date: Fri, 23 May 2008 19:58:30 +0200 Subject: [et-mgmt-tools] "named rpm" yum repository In-Reply-To: <483704D6.5060304@gmail.com> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> <483704D6.5060304@gmail.com> Message-ID: <2a7fce340805231058x2afd4adbubb882acdb1f632db@mail.gmail.com> On Fri, May 23, 2008 at 7:54 PM, Ole Ersoy wrote: > Hi, > > Does anyone know of a yum repository that has the "named" rpm in it. I > tried googleing for "named rpm yum repository" and "named rpm", but no love. I guess you probably are looking for the bind package. It contains the named daemon. So doing "yum search bind" should turn up stuff 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 wright at imageworks.com Fri May 23 17:58:14 2008 From: wright at imageworks.com (Peter Wright) Date: Fri, 23 May 2008 10:58:14 -0700 Subject: [et-mgmt-tools] "named rpm" yum repository In-Reply-To: <483704D6.5060304@gmail.com> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> <483704D6.5060304@gmail.com> Message-ID: <483705B6.5010502@imageworks.com> Ole Ersoy wrote: > Hi, > > Does anyone know of a yum repository that has the "named" rpm in it. > I tried googleing for "named rpm yum repository" and "named rpm", but > no love. > you need to get BIND which contains the named daemon. -pete > TIA, > - Ole > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -- Peter Wright Systems Engineer Sony Pictures Imageworks wright at imageworks.com www.imageworks.com From tom at ng23.net Fri May 23 17:59:57 2008 From: tom at ng23.net (Tom Brown) Date: Fri, 23 May 2008 18:59:57 +0100 Subject: [et-mgmt-tools] "named rpm" yum repository In-Reply-To: <483704D6.5060304@gmail.com> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> <483704D6.5060304@gmail.com> Message-ID: <4837061D.9010500@ng23.net> > > > Does anyone know of a yum repository that has the "named" rpm in it. > I tried googleing for "named rpm yum repository" and "named rpm", but > no love. > you want bind not named ============================================================================= Package Arch Version Repository Size ============================================================================= Installing: bind i386 20:9.2.4-24.EL4 RHEL4-u5-i386 580 k From drew.einhorn at gmail.com Fri May 23 18:10:54 2008 From: drew.einhorn at gmail.com (drew einhorn) Date: Fri, 23 May 2008 12:10:54 -0600 Subject: [et-mgmt-tools] TFTP open timeout troubleshooting In-Reply-To: <483701E9.7090703@gmail.com> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> <483701E9.7090703@gmail.com> Message-ID: On Fri, May 23, 2008 at 11:42 AM, Ole Ersoy wrote: > drew einhorn wrote: >> >> Don't need named, if you use dnsmasq instead of isc for dhcp. > The nice thing about using dnsmasq instead of named and isc dhcp is that all your VMs automatically get added to your dns without any extra work on your part. > Hmmm - I'm using dhcp from isc. I'll see if I can find a repository with > the "named" dependency. > >> Don't know what cft is, >> >> Is there a firewall on your Fedora 8 box? Maybe you need to >> open up some ports. > > There is and I opened up the ports it mentions when during cobbler check. > I'll try to see if I can find a repository with named in it, and after > installing, if it still is not working, I'll try switching off iptables all > together. All fingers crossed :-) > > Thanks, > - Ole > > >> >> On Fri, May 23, 2008 at 10:51 AM, Ole Ersoy wrote: >>> >>> Hi Guys, >>> >>> I've gone over my settings up and down several times per the tutorials, >>> and >>> I'm still gettinga PXE open timeout on the client. One difference >>> between >>> the tutorial and what I'm doing is that when I run (Fedora 8): >>> >>> yum -y install cobbler koan tftp-server dhcp cft facter puppet subversion >>> httpd named >>> >>> I get: >>> >>> - No package named available. >>> - No package cft available. >>> >>> Are these necessary? >>> >>> The DHCP Lease part works fine. >>> Any ideas on how to troubleshoot this further? I did make sure all the >>> services are running and that I did a cobbler sync before attempting the >>> run. >>> >>> TIA, >>> - Ole >>> >>> _______________________________________________ >>> 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 > -- Drew Einhorn From ole.ersoy at gmail.com Fri May 23 18:33:38 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Fri, 23 May 2008 13:33:38 -0500 Subject: [et-mgmt-tools] TFTP open timeout troubleshooting In-Reply-To: References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> <483701E9.7090703@gmail.com> Message-ID: <48370E02.9080407@gmail.com> > The nice thing about using dnsmasq instead of named and isc dhcp > is that all your VMs automatically get added to your dns without any > extra work on your part. I'm sold! I'm actually reading through the man page right now to figure it how to replace ISC DHCP, since it helps with the host name configuration as well. I also got past the TFTP open timeout by turning off the firewall all together. I was using the Fedora KDE firewall configuration GUI, and did add the 25*** ports, but when I did an nmap -sU and nmap -sS I did not see them open, so that's probably what it was. I'll try defining a profile next. Hopefully the wheel turns full circle. Thanks again, - Ole From ole.ersoy at gmail.com Fri May 23 18:35:54 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Fri, 23 May 2008 13:35:54 -0500 Subject: [et-mgmt-tools] "named rpm" yum repository In-Reply-To: <4837061D.9010500@ng23.net> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> <483704D6.5060304@gmail.com> <4837061D.9010500@ng23.net> Message-ID: <48370E8A.8090602@gmail.com> Tom, Pete, Tim, Awesome - Thanks for the heads up. It's a lot clearer now :-). - Ole Tom Brown wrote: > >> >> >> Does anyone know of a yum repository that has the "named" rpm in it. >> I tried googleing for "named rpm yum repository" and "named rpm", but >> no love. >> > > you want bind not named > > ============================================================================= > > Package Arch Version Repository Size > ============================================================================= > > Installing: > bind i386 20:9.2.4-24.EL4 RHEL4-u5-i386 580 k > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From shenson at redhat.com Fri May 23 18:45:13 2008 From: shenson at redhat.com (Scott Henson) Date: Fri, 23 May 2008 14:45:13 -0400 Subject: [et-mgmt-tools] TFTP open timeout troubleshooting In-Reply-To: References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> <483701E9.7090703@gmail.com> Message-ID: <483710B9.80003@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 drew einhorn wrote: | On Fri, May 23, 2008 at 11:42 AM, Ole Ersoy wrote: |> drew einhorn wrote: |>> Don't need named, if you use dnsmasq instead of isc for dhcp. | | The nice thing about using dnsmasq instead of named and isc dhcp | is that all your VMs automatically get added to your dns without any | extra work on your part. Its my understanding that in the brave new cobbler world, named does the same thing. You just tell it to manage dns and what domains to manage and you are off. I say this as a person who sits 15 feet from the person who added that to cobbler but having never used it. - -- Scott Henson Red Hat Production Operations Release Engineer WVU Alum BSAE/BSME -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkg3ELgACgkQcQOfHbe3DKDqjgCg1riNTXQl9GloPBqp9tQID8fh LiwAoLl0KLdGg8jSu6bX+sAzm1mNFldK =vin6 -----END PGP SIGNATURE----- From ole.ersoy at gmail.com Fri May 23 22:20:52 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Fri, 23 May 2008 17:20:52 -0500 Subject: [et-mgmt-tools] TFTP open timeout troubleshooting In-Reply-To: <483710B9.80003@redhat.com> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> <483701E9.7090703@gmail.com> <483710B9.80003@redhat.com> Message-ID: <48374344.5080909@gmail.com> > Its my understanding that in the brave new cobbler world, named does the > same thing. You just tell it to manage dns and what domains to manage > and you are off. I say this as a person who sits 15 feet from the > person who added that to cobbler but having never used it. Proximity does matter :-). Thanks for the heads up, - Ole > > - -- > Scott Henson > Red Hat Production Operations Release Engineer > WVU Alum BSAE/BSME > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v2.0.9 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iEYEARECAAYFAkg3ELgACgkQcQOfHbe3DKDqjgCg1riNTXQl9GloPBqp9tQID8fh > LiwAoLl0KLdGg8jSu6bX+sAzm1mNFldK > =vin6 > -----END PGP SIGNATURE----- > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From dmourati at gmail.com Fri May 23 22:42:38 2008 From: dmourati at gmail.com (Demetri Mouratis) Date: Fri, 23 May 2008 15:42:38 -0700 Subject: [et-mgmt-tools] Cobbler PXE Menu Under CentOS 4.6 Message-ID: <3cc2cbc40805231542j4be3abacx2a249ffd9f0133b9@mail.gmail.com> Hi, I'm trying to setup PXE Menus under CentOS 4.6. My cobbler is: [root at lncbuild cobbler]# rpm -q cobbler cobbler-0.4.8-1.el4.centos I've rebuilt the syslinux package per some other notes I found. [root at lncbuild ~]# rpm -q syslinux syslinux-3.36-1 Still, I'm not able to see the "boot:" prompt while a system is PXE booting. It simply falls through to the default profile which is not what I want in this case. I'm reluctant to go setting up a bunch of systems at this point. Any idea if this should work? Do I need a newer Cobbler? Thanks. -D From sghosh at redhat.com Fri May 23 23:16:54 2008 From: sghosh at redhat.com (Subhendu Ghosh) Date: Fri, 23 May 2008 19:16:54 -0400 Subject: [et-mgmt-tools] Cobbler PXE Menu Under CentOS 4.6 In-Reply-To: <3cc2cbc40805231542j4be3abacx2a249ffd9f0133b9@mail.gmail.com> References: <3cc2cbc40805231542j4be3abacx2a249ffd9f0133b9@mail.gmail.com> Message-ID: <48375066.2080807@redhat.com> Demetri Mouratis wrote: > Hi, > > I'm trying to setup PXE Menus under CentOS 4.6. My cobbler is: > > [root at lncbuild cobbler]# rpm -q cobbler > cobbler-0.4.8-1.el4.centos Most definitely need a cobbler upgrade - 0.4 is ancient :) > I've rebuilt the syslinux package per some other notes I found. > > [root at lncbuild ~]# rpm -q syslinux > syslinux-3.36-1 > > Still, I'm not able to see the "boot:" prompt while a system is PXE booting. > > It simply falls through to the default profile which is not what I > want in this case. I'm reluctant to go setting up a bunch of systems > at this point. > > Any idea if this should work? > > Do I need a newer Cobbler? > > Thanks. > > -D > -cheers Subhendu From dmourati at gmail.com Fri May 23 23:47:39 2008 From: dmourati at gmail.com (Demetri Mouratis) Date: Fri, 23 May 2008 16:47:39 -0700 Subject: [et-mgmt-tools] Cobbler PXE Menu Under CentOS 4.6 In-Reply-To: <48375066.2080807@redhat.com> References: <3cc2cbc40805231542j4be3abacx2a249ffd9f0133b9@mail.gmail.com> <48375066.2080807@redhat.com> Message-ID: <3cc2cbc40805231647q7cfc6138r57aabb31aa51f36d@mail.gmail.com> On Fri, May 23, 2008 at 4:16 PM, Subhendu Ghosh wrote: > Demetri Mouratis wrote: >> >> Hi, >> >> I'm trying to setup PXE Menus under CentOS 4.6. My cobbler is: >> >> [root at lncbuild cobbler]# rpm -q cobbler >> cobbler-0.4.8-1.el4.centos > > Most definitely need a cobbler upgrade - 0.4 is ancient :) > Thanks Subhendu, I've upgraded to 0.8.3-3.el4. [root at lncbuild ~]# rpm -q cobbler cobbler-0.8.3-3.el4 Still, no joy on the PXE Menu. I'm also looking at just removing the "default" system. I think I'll wind up with just a local boot in this case. Thanks. -D From jeckersb at redhat.com Sat May 24 00:41:56 2008 From: jeckersb at redhat.com (John Eckersberg) Date: Fri, 23 May 2008 20:41:56 -0400 Subject: [et-mgmt-tools] TFTP open timeout troubleshooting In-Reply-To: <48374344.5080909@gmail.com> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> <483701E9.7090703@gmail.com> <483710B9.80003@redhat.com> <48374344.5080909@gmail.com> Message-ID: <48376454.2020401@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ole Ersoy wrote: |> Its my understanding that in the brave new cobbler world, named does |> the same thing. You just tell it to manage dns and what domains to |> manage and you are off. I say this as a person who sits 15 feet from |> the person who added that to cobbler but having never used it. | | Proximity does matter :-). | | Thanks for the heads up, | - Ole | | |> |> - -- |> Scott Henson |> Red Hat Production Operations Release Engineer |> WVU Alum BSAE/BSME |> -----BEGIN PGP SIGNATURE----- |> Version: GnuPG v2.0.9 (GNU/Linux) |> Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org |> |> iEYEARECAAYFAkg3ELgACgkQcQOfHbe3DKDqjgCg1riNTXQl9GloPBqp9tQID8fh |> LiwAoLl0KLdGg8jSu6bX+sAzm1mNFldK |> =vin6 |> -----END PGP SIGNATURE----- |> |> _______________________________________________ |> 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 I updated the documentation on the wiki just now. Before it was misleading and outright wrong in places. Check it out: https://fedorahosted.org/cobbler/wiki/ManageDns - -- John Eckersberg Production Operations Red Hat, Inc. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkg3ZFQACgkQO5VSO/Qqm+jmJQCfdl2iRa4BpvaOmhF197/PjVwm qFQAmwZPP9HLf6L1eDa+Izug5crD0Ir6 =tCYK -----END PGP SIGNATURE----- From mdehaan at redhat.com Sat May 24 02:04:48 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 23 May 2008 22:04:48 -0400 Subject: [et-mgmt-tools] Re: Koan problem In-Reply-To: <48367639.6000108@japannext.co.jp> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <48367639.6000108@japannext.co.jp> Message-ID: <483777C0.4080405@redhat.com> Martin Pearson wrote: > Martin Pearson wrote: >> Hi All, >> Wonder if you can help me when I run Koan with the following inputs:- >> >> koan --virt --server=puppet.tk.japannext.co.jp >> --profile=RHEL4-ES-xen-x86_64 --virt-name=tkdvir1 >> >> I get the following error:- >> >> libvir: Xen error : Domain not found: xenUnifiedDomainLookupByUUID >> I don't think I have seen this particular error before. Can you share the koan, libvirt/Xen, and OS versions you are using, and the full traceback (error message/output) if available? Koan also saves some additional logs in ~/.koan when creating VMs that may also be useful. --Michael From mdehaan at redhat.com Sat May 24 02:08:19 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 23 May 2008 22:08:19 -0400 Subject: [et-mgmt-tools] Cobbler PXE Menu Under CentOS 4.6 In-Reply-To: <3cc2cbc40805231647q7cfc6138r57aabb31aa51f36d@mail.gmail.com> References: <3cc2cbc40805231542j4be3abacx2a249ffd9f0133b9@mail.gmail.com> <48375066.2080807@redhat.com> <3cc2cbc40805231647q7cfc6138r57aabb31aa51f36d@mail.gmail.com> Message-ID: <48377893.4080308@redhat.com> Demetri Mouratis wrote: > On Fri, May 23, 2008 at 4:16 PM, Subhendu Ghosh wrote: > >> Demetri Mouratis wrote: >> >>> Hi, >>> >>> I'm trying to setup PXE Menus under CentOS 4.6. My cobbler is: >>> >>> [root at lncbuild cobbler]# rpm -q cobbler >>> cobbler-0.4.8-1.el4.centos >>> >> Most definitely need a cobbler upgrade - 0.4 is ancient :) >> >> > > Thanks Subhendu, > > I've upgraded to 0.8.3-3.el4. > That version of cobbler sub-packages a newer menu.c32 > [root at lncbuild ~]# rpm -q cobbler > cobbler-0.8.3-3.el4 > > Still, no joy on the PXE Menu. > > I'm also looking at just removing the "default" system. I think I'll > wind up with just a local boot in this case. > Do you mean you have a system named "default" in Cobbler? If so, that will override your menu behavior. Definitely delete it and then you should get the menu. Let me know if that's not the case. --Michael From ole.ersoy at gmail.com Sat May 24 05:04:26 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Sat, 24 May 2008 00:04:26 -0500 Subject: [et-mgmt-tools] TFTP open timeout troubleshooting In-Reply-To: <48376454.2020401@redhat.com> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <4836F5F4.6020505@gmail.com> <483701E9.7090703@gmail.com> <483710B9.80003@redhat.com> <48374344.5080909@gmail.com> <48376454.2020401@redhat.com> Message-ID: <4837A1DA.3090703@gmail.com> > > I updated the documentation on the wiki just now. Before it was > misleading and outright wrong in places. Check it out: > > https://fedorahosted.org/cobbler/wiki/ManageDns Sweet - This just keeps getting better and better - Thanks, - Ole From martin.pearson at japannext.co.jp Mon May 26 01:31:16 2008 From: martin.pearson at japannext.co.jp (Martin Pearson) Date: Mon, 26 May 2008 10:31:16 +0900 Subject: [et-mgmt-tools] Re: Koan problem In-Reply-To: <483777C0.4080405@redhat.com> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <48367639.6000108@japannext.co.jp> <483777C0.4080405@redhat.com> Message-ID: <483A12E4.7040109@japannext.co.jp> Michael DeHaan wrote: > Martin Pearson wrote: >> Martin Pearson wrote: >>> Hi All, >>> Wonder if you can help me when I run Koan with the following inputs:- >>> >>> koan --virt --server=puppet --profile=RHEL4-ES-xen-x86_64 >>> --virt-name=tkdvir1 >>> >>> I get the following error:- >>> >>> libvir: Xen error : Domain not found: xenUnifiedDomainLookupByUUID >>> > > I don't think I have seen this particular error before. > > Can you share the koan, libvirt/Xen, and OS versions you are using, > and the full traceback (error message/output) if available? > > Koan also saves some additional logs in ~/.koan when creating VMs that > may also be useful. > --Michael > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Hi Michael, Thanks very much for your reply. Please see below info requested:- [root at xxxxxx ~]# rpm -qa | grep -i libvirt libvirt-0.3.3-7.el5 libvirt-python-0.3.3-7.el5 [root at xxxxxx ~]# rpm -qa | grep -i xen xen-libs-3.0.3-64.el5_2.1 kernel-xen-devel-2.6.18-92.el5 kernel-xen-2.6.18-92.el5 xen-devel-3.0.3-64.el5_2.1 xen-3.0.3-64.el5_2.1 [root at xxxxxx ~]# uname -a Linux xxxxxxxx 2.6.18-92.el5xen #1 SMP Tue Apr 29 13:31:30 EDT 2008 x86_64 x86_64 x86_64 GNU/Linux Red Hat Enterprise Linux Server release 5 (Tikanga) Python info, just in case that helps. python-urlgrabber-3.1.0-2 python-numeric-23.7-2.2.2 gtk-vnc-python-0.3.2-3.el5 python-setuptools-0.6c7-1.el5 libxml2-python-2.6.26-2.1.2.1 python-virtinst-0.300.2-8.el5 python-elementtree-1.2.6-5 python-sqlite-1.1.7-1.2.1 dbus-python-0.70-7.el5 libvirt-python-0.3.3-7.el5 python-2.4.3-21.el5 python-devel-2.4.3-21.el5 libselinux-python-1.33.4-5.el5 audit-libs-python-1.6.5-9.el5 rpm-python-4.4.2-48.el5 python-devel-2.4.3-21.el5 Version of cobbler on the server:- cobbler-0.8.2-2 Output from koan command:- [root at xxxxxxxx .koan]# koan --virt --server=puppet --profile=RHEL4-ES-xen-x86_64 --virt-name=tkdvir1 - reading URL: http://10.145.100.35/cblr/kickstarts/RHEL4-ES-xen-x86_64/ks.cfg install_tree: http://10.145.100.35/cblr/links/RHEL4-ES-xen-x86_64 - no virt-type specified, auto-selecting xenpv libvirtd (pid 16816) is running... downloading initrd initrd.img to /var/lib/xen/initrd.img url=http://puppet/cobbler/images/RHEL4-ES-xen-x86_64/initrd.img - reading URL: http://puppet/cobbler/images/RHEL4-ES-xen-x86_64/initrd.img downloading kernel vmlinuz to /var/lib/xen/vmlinuz url=http://puppet/cobbler/images/RHEL4-ES-xen-x86_64/vmlinuz - reading URL: http://puppet/cobbler/images/RHEL4-ES-xen-x86_64/vmlinuz _*libvir: Xen error : Domain not found: xenUnifiedDomainLookupByUUID libvir: Xen error : Domain not found: xenUnifiedDomainLookupByName*_ Kernel arguments: ks=http://10.145.100.35/cblr/kickstarts/RHEL4-ES-xen-x86_64/ks.cfg ksdevice=eth0 lang= kssendmac syslog=10.145.100.35:25150 text use virt-manager or reconnect with virsh console tkdvir1 Sorry was not precise enough with my previous email. Does not seem to be Koan that is crashing it just appears to be having a problem creating the virtual machine, i.e. if I go into virsh I have a machine there I can connect to via console . Although the machine is stuck on boot see below:- RAMDISK driver initialized: 16 RAM disks of 16384K size 1024 blocksize Xen virtual console successfully installed as xvc0 Event-channel device installed. Console: switching to colour frame buffer device 100x37 mice: PS/2 mouse device common for all mice md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27 NET: Registered protocol family 2 IP route cache hash table entries: 32768 (order: 6, 262144 bytes) TCP established hash table entries: 131072 (order: 9, 2097152 bytes) TCP bind hash table entries: 131072 (order: 9, 2097152 bytes) TCP: Hash tables configured (established 131072 bind 131072) Initializing IPsec netlink socket NET: Registered protocol family 1 NET: Registered protocol family 17 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). _* *_Also the tail end of the /var/log/messages log just in case this is useful:- May 26 10:10:19 xxxxxxxx kernel: device vif28.0 entered promiscuous mode May 26 10:10:19 xxxxxxxx kernel: audit(1211764219.787:43): dev=vif28.0 prom=256 old_prom=0 auid=4294967295 ses=4294967295 May 26 10:10:19 xxxxxxxx kernel: ADDRCONF(NETDEV_UP): vif28.0: link is not ready May 26 10:10:20 xxxxxxxx kernel: ADDRCONF(NETDEV_CHANGE): vif28.0: link becomes ready May 26 10:10:20 xxxxxxxx kernel: xenbr0: port 3(vif28.0) entering learning state May 26 10:10:20 xxxxxxxx kernel: blkback: ring-ref 522, event-channel 9, protocol 1 (x86_64-abi) May 26 10:10:20 xxxxxxxx kernel: xenbr0: topology change detected, propagating May 26 10:10:20 xxxxxxxx kernel: xenbr0: port 3(vif28.0) entering forwarding state May 26 10:22:39 xxxxxxxx kernel: xenbr0: port 3(vif28.0) entering disabled state May 26 10:22:39 xxxxxxxx kernel: device vif28.0 left promiscuous mode May 26 10:22:39 xxxxxxxx kernel: audit(1211764959.982:44): dev=vif28.0 prom=0 old_prom=256 auid=4294967295 ses=4294967295 May 26 10:22:39 xxxxxxxx kernel: xenbr0: port 3(vif28.0) entering disabled state _* *_Interesting thing is I can create a VM fine manually with virt-install? Would be great if you could help out with this. Many thanks, Martin Pearson. -------------- next part -------------- A non-text attachment was scrubbed... Name: koan.log Type: text/x-log Size: 1875 bytes Desc: not available URL: From mmaccana at au1.ibm.com Mon May 26 04:32:32 2008 From: mmaccana at au1.ibm.com (Mike MacCana) Date: Mon, 26 May 2008 14:32:32 +1000 Subject: [et-mgmt-tools] Virt--manager website: adding link for Ubuntu on download section Message-ID: <1211776352.11406.20.camel@mmaccana-laptop> On http://virt-manager.org/download.html 1. Please add something like the following for Ubuntu: Ubuntu Users of Ubuntu 8.04 can click the following link to install Virtual Machine manager. Alternative, install the 'virt-manager' package using your preferred method. 2. Doesn't RHEL include virt-manager? Is there a reason RHEL isn't listed? 3. It would be better to put the info on packages before most of the other content. Cheers, Mike ________________________________________________ Mike MacCana Technical Specialist Linux Services IBM Global Services Level 14, 60 City Rd Southgate Vic 3000 Phone: +61-3-8656-2138 Fax: +61-3-?8656-2423 Email: mmaccana at au1.ibm.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From fj7025cf at aa.jp.fujitsu.com Mon May 26 05:19:09 2008 From: fj7025cf at aa.jp.fujitsu.com (Hiroyuki Kaguchi) Date: Mon, 26 May 2008 14:19:09 +0900 Subject: [et-mgmt-tools] [PATCH] Enable to send the mnemonic of the menu to the guest OS Message-ID: <483A484D.6070105@aa.jp.fujitsu.com> We cannot send "Alt+M" and "Alt+V" and "Alt+H" to the guest OS because the mnemonic of the menu of the graphical console is enabled. So, I change the graphical console to be able to send these keys to the guest OS. This change is that the mnemonic of the menu is disabled when the graphical console grabs the keyboard. Moreover, the patch includes the bugfix of the following bugs. 1) Two times of _disable_modifiers() is called consecutively. In that case, the value of self.gtk_settings_accel is always "None", and the setting of the accelerator is not restored. 2) Because the information that self.accel_groups has is not updated in the latest state, the assertion occurs when the topwin.remove_accel_group() is called. Thanks, Signed-off-by: Hiroyuki Kaguchi -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: menubar_accel.patch URL: From atodorov at redhat.com Mon May 26 08:37:52 2008 From: atodorov at redhat.com (Alexander Todorov) Date: Mon, 26 May 2008 10:37:52 +0200 Subject: [et-mgmt-tools] [PATCH] Enable to send the mnemonic of the menu to the guest OS In-Reply-To: <483A484D.6070105@aa.jp.fujitsu.com> References: <483A484D.6070105@aa.jp.fujitsu.com> Message-ID: <483A76E0.8020901@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Hiroyuki Kaguchi wrote: | We cannot send "Alt+M" and "Alt+V" and "Alt+H" to the guest OS because the | mnemonic of the menu of the graphical console is enabled. Even better idea: why not have a menu item "Send arbitrary key" which will open a dialog and the user can enter arbitrary combination of keys + modifiers and this will be sent to the guest os. What do you think? -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Red Hat - http://enigmail.mozdev.org iD8DBQFIOnbfhmd3WOiFct4RCuJdAJ9kLVbKN2/+o4qqof9mFg/sG9MPIACgiTWf zbwkLRNC239z85QBtlwKeww= =3UgJ -----END PGP SIGNATURE----- From Pablo.Iranzo at redhat.com Mon May 26 08:42:31 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Mon, 26 May 2008 10:42:31 +0200 Subject: [et-mgmt-tools] [PATCH] Enable to send the mnemonic of the menu to the guest OS In-Reply-To: <483A76E0.8020901@redhat.com> References: <483A484D.6070105@aa.jp.fujitsu.com> <483A76E0.8020901@redhat.com> Message-ID: <1211791351.5034.23.camel@iranzo.users.redhat.com> Hi El lun, 26-05-2008 a las 10:37 +0200, Alexander Todorov escribi?: > Hiroyuki Kaguchi wrote: > | We cannot send "Alt+M" and "Alt+V" and "Alt+H" to the guest OS because the > | mnemonic of the menu of the graphical console is enabled. > > Even better idea: > why not have a menu item "Send arbitrary key" which will open a dialog and the > user can enter arbitrary combination of keys + modifiers and this will be sent > to the guest os. What do you think? Sounds very very good :) > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.5 (GNU/Linux) > Comment: Using GnuPG with Red Hat - http://enigmail.mozdev.org > > iD8DBQFIOnbfhmd3WOiFct4RCuJdAJ9kLVbKN2/+o4qqof9mFg/sG9MPIACgiTWf > zbwkLRNC239z85QBtlwKeww= > =3UgJ > -----END PGP SIGNATURE----- > > _______________________________________________ > 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 Feichtinger.Guenter at bartelt.at Tue May 27 09:45:49 2008 From: Feichtinger.Guenter at bartelt.at (=?iso-8859-1?Q?Feichtinger_G=FCnter?=) Date: Tue, 27 May 2008 11:45:49 +0200 Subject: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't connect to Xen- Host Message-ID: Hello Richard, I intstalled the ocaml-libvirt-0.4.0.1.exe on MS- Vista without problems. Also the virt-ctrl.exe starts without problems. But when I try to connect to a Xen-Host I get the messages: libvir: Remote error : Cannot access CA certificate 'C:/msys/1.0/local/etc/pki/C A/cacert.pem': No such file or directory (2) Please be so kind and help. Regard, G?nter * From: "Richard W.M. Jones" * To: Fedora/Linux Management Tools * Subject: [et-mgmt-tools] Fwd: OCaml bindings - Windows installer * Date: Tue, 08 Jan 2008 11:18:39 +0000 Might interest some people on this list ... --- Begin Message --- * From: "Richard W.M. Jones" * To: libvir-list * Subject: [Libvir] OCaml bindings - Windows installer * Date: Tue, 08 Jan 2008 11:15:19 +0000 This is a Windows installer for the OCaml libvirt bindings and programs: http://libvirt.org/sources/ocaml/ocaml-libvirt-0.4.0.1.exe If someone has a 'virgin' Windows system and could test that the above installer works, particularly for Windows < Vista and for Windows which has not had any GTK/MinGW development tools installed. About 90% of the size of the installer is down to the GTK DLLs and configuration files which need to be carried along to support the GTK graphical program (Virt Control). It's possible that I haven't placed the GTK files in exactly the right place, so instead it is using my own GTK development environment. Screenshots showing the installer in action: http://annexia.org/tmp/wininstaller-1-desktop.png Desktop, nothing installed, showing the installer icon. http://annexia.org/tmp/wininstaller-2-packages.png List of required, recommended and optional subpackages. http://annexia.org/tmp/wininstaller-3-instpath.png Select the installation path. http://annexia.org/tmp/wininstaller-4-installdone.png Install complete. Notice the desktop icons, ... http://annexia.org/tmp/wininstaller-5-menu.png ... and the menu entries and uninstaller program. http://annexia.org/tmp/wininstaller-6-virtctrl.png Running Virt Control. http://annexia.org/tmp/wininstaller-7-uninstall.png After uninstall. Notice that the desktop icons have gone. The installer uses NSIS (the Nullsoft Scriptable Install System, http://nsis.sf.net/) and is reasonably well integrated into the autoconf / make build system. In particular if you are compiling under Windows and NSIS is installed, then a simple './configure --with-nsis; make wininstaller' will build a Windows installer. This is something to look at if libvirt wants/needs a Windows installer. 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 Attachment: smime.p7s Description: S/MIME Cryptographic Signature -- Libvir-list mailing list Libvir-list redhat com https://www.redhat.com/mailman/listinfo/libvir-list --- End Message --- From rjones at redhat.com Tue May 27 09:45:16 2008 From: rjones at redhat.com (Richard W.M. Jones) Date: Tue, 27 May 2008 10:45:16 +0100 Subject: [et-mgmt-tools] Re: ocaml-libvirt-0.4.0.1: Can't connect to Xen- Host In-Reply-To: References: Message-ID: <20080527094516.GC29698@amd.home.annexia.org> On Tue, May 27, 2008 at 11:37:10AM +0200, Feichtinger G?nter wrote: > I intstalled the ocaml-libvirt-0.4.0.1.exe on MS- Vista without problems. > Also the virt-ctrl.exe starts without problems. > But when I try to connect to a Xen-Host I get the messages: > libvir: Remote error : Cannot access CA certificate 'C:/msys/1.0/local/etc/pki/C > A/cacert.pem': No such file or directory (2) > > Please be so kind and help. It's not very clear to me what you are trying to connect to what, but you'll probably want to read about remote connections, here: http://libvirt.org/remote.html Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my OCaml programming blog: http://camltastic.blogspot.com/ Fedora now supports 59 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora From tomackell at duke-energy.com Tue May 27 12:58:13 2008 From: tomackell at duke-energy.com (Mackell, Thomas O) Date: Tue, 27 May 2008 08:58:13 -0400 Subject: [et-mgmt-tools] Cobbler Log status command Failure Message-ID: <1B3160038A79B54DB674532195C68A586DB033@IMCLTEXCP15.nam.ent.duke-energy.com> Here is the two log files. I was unable to send them to you.. I received the following message: Your message did not reach some or all of the intended recipients. Subject: Cobbler status log files Sent: 5/23/2008 2:33 PM The following recipient(s) could not be reached: mdehann at redhat.com on 5/23/2008 2:34 PM The message reached the recipient's e-mail system, but delivery was refused. Attempt to resend the message. If it still fails, contact your system administrator. Thank You, Thomas Mackell 704-382-9731 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: install.log Type: application/octet-stream Size: 98 bytes Desc: install.log URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: cobbler.log Type: application/octet-stream Size: 247568 bytes Desc: cobbler.log URL: From crobinso at redhat.com Tue May 27 14:17:59 2008 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 27 May 2008 10:17:59 -0400 Subject: [et-mgmt-tools] [PATCH] Enable to send the mnemonic of the menu to the guest OS In-Reply-To: <483A76E0.8020901@redhat.com> References: <483A484D.6070105@aa.jp.fujitsu.com> <483A76E0.8020901@redhat.com> Message-ID: <483C1817.5090703@redhat.com> Alexander Todorov wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA512 > > Hiroyuki Kaguchi wrote: > | We cannot send "Alt+M" and "Alt+V" and "Alt+H" to the guest OS because the > | mnemonic of the menu of the graphical console is enabled. > > Even better idea: > why not have a menu item "Send arbitrary key" which will open a dialog and the > user can enter arbitrary combination of keys + modifiers and this will be sent > to the guest os. What do you think? I think that is the only way to go. The current situation is certainly non scalable. Someone proposed that a while ago as well but no one has moved on it. - Cole From lfarkas at bppiac.hu Tue May 27 14:56:06 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Tue, 27 May 2008 16:56:06 +0200 Subject: [et-mgmt-tools] virt-manager and pxe install Message-ID: <483C2106.20300@bppiac.hu> hi, we try again to install a new guest from virt-manager through pxe boot without much success:-( out setup: - host: - Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz - Intel S3000AHV - 8GB RAM - CentOS-5.1 - kernel-2.6.18-53.1.21.el5 x86_64 64bit - packages from http://www.lfarkas.org/linux/packages/centos/5/x86_64/ libvirt-0.4.2-5 libvirt-python-0.4.2-5 python-virtinst-0.300.3-6 virt-manager-0.5.4-4 virt-top-0.4.1.0-2.el5 virt-viewer-0.0.3-1 kmod-kvm-69-1.2.6.18_53.1.21.el5 kvm-69-1 we try to create a new file based virtual guest fedora-8 through pxe install on the kvm host machine. the problems: - if i select cpu architecture i386, the i can't select hypervisor kvm. why? imho it's a bug! - after select i386 creating the disk image virt-manager crash with some python backtrace as can't connect the guest. - if i select x86_64 then in a bridged network it can't get an ip address (searching for dhcp server). so anyone has any success on kvm with pxe boot? what should i have to do? is there any way to be able to use pxe boot in kvm? it's a virt-manager/libvirt or kvm bug? thanks in advance. yours. -- Levente "Si vis pacem para bellum!" From mdehaan at redhat.com Tue May 27 14:59:15 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 27 May 2008 10:59:15 -0400 Subject: [et-mgmt-tools] Cobbler Log status command Failure In-Reply-To: <1B3160038A79B54DB674532195C68A586DB033@IMCLTEXCP15.nam.ent.duke-energy.com> References: <1B3160038A79B54DB674532195C68A586DB033@IMCLTEXCP15.nam.ent.duke-energy.com> Message-ID: <483C21C3.5010901@redhat.com> Mackell, Thomas O wrote: > > Here is the two log files. > > I was unable to send them to you.. I received the following message: > > Your message did not reach some or all of the intended recipients. > > > > Subject: Cobbler status log files > > Sent: 5/23/2008 2:33 PM > > > > The following recipient(s) could not be reached: > > > > mdehann at redhat.com on 5/23/2008 2:34 PM > > The message reached the recipient's e-mail system, but > delivery was refused. Attempt to resend the message. If it still > fails, contact your system administrator. > > > > > > Thank You, > > Thomas Mackell > > 704-382-9731 > > > > ------------------------------------------------------------------------ > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools I'll check this out. FWIW, your email error is from a typo in my username. --Michael From berrange at redhat.com Tue May 27 15:03:54 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 27 May 2008 16:03:54 +0100 Subject: [et-mgmt-tools] virt-manager and pxe install In-Reply-To: <483C2106.20300@bppiac.hu> References: <483C2106.20300@bppiac.hu> Message-ID: <20080527150354.GG9310@redhat.com> On Tue, May 27, 2008 at 04:56:06PM +0200, Farkas Levente wrote: > hi, > we try again to install a new guest from virt-manager through pxe boot > without much success:-( > out setup: > - host: > - Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz > - Intel S3000AHV > - 8GB RAM > - CentOS-5.1 > - kernel-2.6.18-53.1.21.el5 x86_64 64bit > - packages from http://www.lfarkas.org/linux/packages/centos/5/x86_64/ > libvirt-0.4.2-5 > libvirt-python-0.4.2-5 > python-virtinst-0.300.3-6 > virt-manager-0.5.4-4 > virt-top-0.4.1.0-2.el5 > virt-viewer-0.0.3-1 > kmod-kvm-69-1.2.6.18_53.1.21.el5 > kvm-69-1 > > we try to create a new file based virtual guest fedora-8 through pxe > install on the kvm host machine. the problems: > - if i select cpu architecture i386, the i can't select hypervisor kvm. > why? imho it's a bug! KVM only virtualizes a CPU that matches the host. If you're on a 64-bit host, then it only virtualizes a 64-bit CPU. It doesn't virtualize a 32-bit CPU - if you want a true 32-bit cpu on a 64-bit host you need to emulate with QEMU instead. That said its rather pointless because you can run a 32-bit guest OS in a 64-kvm KVM vm just fine. > - after select i386 creating the disk image virt-manager crash with some > python backtrace as can't connect the guest. Please file a BZ with the details > - if i select x86_64 then in a bridged network it can't get an ip > address (searching for dhcp server). There's many things which can go wrong - some versions of KVM have broken PXE support. Start off by tracing the network traffic in the host using tcpdump. It may also be an incompatability with your DHCP server and etherboot Dan -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From lfarkas at bppiac.hu Tue May 27 15:54:12 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Tue, 27 May 2008 17:54:12 +0200 Subject: [et-mgmt-tools] virt-manager and pxe install In-Reply-To: <20080527150354.GG9310@redhat.com> References: <483C2106.20300@bppiac.hu> <20080527150354.GG9310@redhat.com> Message-ID: <483C2EA4.4070900@bppiac.hu> Daniel P. Berrange wrote: > On Tue, May 27, 2008 at 04:56:06PM +0200, Farkas Levente wrote: >> hi, >> we try again to install a new guest from virt-manager through pxe boot >> without much success:-( >> out setup: >> - host: >> - Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz >> - Intel S3000AHV >> - 8GB RAM >> - CentOS-5.1 >> - kernel-2.6.18-53.1.21.el5 x86_64 64bit >> - packages from http://www.lfarkas.org/linux/packages/centos/5/x86_64/ >> libvirt-0.4.2-5 >> libvirt-python-0.4.2-5 >> python-virtinst-0.300.3-6 >> virt-manager-0.5.4-4 >> virt-top-0.4.1.0-2.el5 >> virt-viewer-0.0.3-1 >> kmod-kvm-69-1.2.6.18_53.1.21.el5 >> kvm-69-1 >> >> we try to create a new file based virtual guest fedora-8 through pxe >> install on the kvm host machine. the problems: >> - if i select cpu architecture i386, the i can't select hypervisor kvm. >> why? imho it's a bug! > > KVM only virtualizes a CPU that matches the host. If you're on a 64-bit > host, then it only virtualizes a 64-bit CPU. It doesn't virtualize a > 32-bit CPU - if you want a true 32-bit cpu on a 64-bit host you need to > emulate with QEMU instead. That said its rather pointless because you > can run a 32-bit guest OS in a 64-kvm KVM vm just fine. huu i learn again something:-) until now (ie. in previous version of virt-manager) i can select 32bit for my guest so i assume kvm can virtualize 32bit cpu on 64bit host. but if it can't it'd better to not allow to choose 32bit in case of kvm. and of course virt-manager already know the host has only kvm hypervisor. >> - after select i386 creating the disk image virt-manager crash with some >> python backtrace as can't connect the guest. > > Please file a BZ with the details it's more then half year old: https://www.redhat.com/archives/et-mgmt-tools/2007-October/msg00062.html >> - if i select x86_64 then in a bridged network it can't get an ip >> address (searching for dhcp server). > > There's many things which can go wrong - some versions of KVM have > broken PXE support. Start off by tracing the network traffic in the > host using tcpdump. It may also be an incompatability with your > DHCP server and etherboot the dhcp server is a centos-5.1 with default dhcp server, default pxe boot etc. and the pxe boot works for fedora and centos with real hardware (we did dozens of such installs). and at the same kvm host we've got 5 other guest using the same shared network adapter. i'm just attached the screenshot of the install screen. and when we reach this screen virt-manager's shutdown not working, the only way to stop the guest is a kill. here is a process list (the last one should have to be the new guest): --------------------------------------------------------------- root 3278 181 25.9 2285204 2119796 ? Sl 14:32 350:50 /usr/bin/qemu-kvm -M pc -m 2048 -smp 4 -monitor pty -boot c -hda /dev/VolGroup00/centos-i386-root -hdb /dev/VolGroup00/centos-i386-swap -hdc /dev/md2 -net nic,macaddr=00:16:3e:58:66:7b,vlan=0 -net tap,fd=11,script=,vlan=0,ifname=vnet0 -usb -usbdevice tablet -vnc 127.0.0.1:0 root 3314 134 5.9 707332 490444 ? Sl 14:32 260:41 /usr/bin/qemu-kvm -M pc -m 512 -smp 4 -monitor pty -boot c -hda /dev/VolGroup00/centos-x86_64-root -hdb /dev/VolGroup00/centos-x86_64-swap -net nic,macaddr=00:16:3e:2e:99:e2,vlan=0 -net tap,fd=14,script=,vlan=0,ifname=vnet1 -usb -usbdevice tablet -vnc 127.0.0.1:1 root 3347 3.3 6.7 686688 548920 ? Sl 14:32 6:28 /usr/bin/qemu-kvm -M pc -m 512 -smp 2 -monitor pty -no-acpi -boot c -hda /dev/VolGroup00/windows-xp -net nic,macaddr=00:16:3e:18:81:69,vlan=0 -net tap,fd=18,script=,vlan=0,ifname=vnet2 -usb -usbdevice tablet -vnc 127.0.0.1:2 root 3378 3.0 1.5 676440 129008 ? Sl 14:32 5:53 /usr/bin/qemu-kvm -M pc -m 512 -smp 1 -monitor pty -boot c -hda /dev/VolGroup00/mandrake-9-root -hdb /dev/VolGroup00/mandrake-9-swap -net nic,macaddr=00:16:3e:37:68:fd,vlan=0 -net tap,fd=22,script=,vlan=0,ifname=vnet3 -usb -usbdevice tablet -vnc 127.0.0.1:3 root 3409 0.2 0.8 676440 68260 ? Sl 14:32 0:33 /usr/bin/qemu-kvm -M pc -m 512 -smp 1 -monitor pty -boot c -hda /dev/VolGroup00/mandrake-10 -net nic,macaddr=00:16:3e:03:37:eb,vlan=0 -net tap,fd=26,script=,vlan=0,ifname=vnet4 -usb -vnc 127.0.0.1:4 root 4334 6.7 0.3 494440 32536 ? Ss 17:43 0:11 python /usr/share/virt-manager/virt-manager.py --connect qemu:///system root 4345 85.7 0.1 676672 15344 ? Sl 17:45 0:30 /usr/bin/qemu-kvm -M pc -m 512 -smp 1 -monitor pty -no-reboot -boot n -hda /var/lib/libvirt/images/dfgdsfg.img -net nic,macaddr=00:16:3e:76:03:2c,vlan=0 -net tap,fd=30,script=,vlan=0,ifname=vnet5 -usb -vnc 127.0.0.1:5 -k en-us --------------------------------------------------------------- but anyway. my question is anybody ever be able to install a guest from virt-manager using pxe boot on kvm? since my assumption is that it's never been tried and never worked. -- Levente "Si vis pacem para bellum!" -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-Fedora-8VirtualMachineConsole.png Type: image/png Size: 24636 bytes Desc: not available URL: From dmourati at gmail.com Tue May 27 18:51:31 2008 From: dmourati at gmail.com (Demetri Mouratis) Date: Tue, 27 May 2008 11:51:31 -0700 Subject: [et-mgmt-tools] Cobbler PXE Menu Under CentOS 4.6 In-Reply-To: <48377893.4080308@redhat.com> References: <3cc2cbc40805231542j4be3abacx2a249ffd9f0133b9@mail.gmail.com> <48375066.2080807@redhat.com> <3cc2cbc40805231647q7cfc6138r57aabb31aa51f36d@mail.gmail.com> <48377893.4080308@redhat.com> Message-ID: <3cc2cbc40805271151hec19077n294163635fecb2d2@mail.gmail.com> On Fri, May 23, 2008 at 7:08 PM, Michael DeHaan wrote: > Do you mean you have a system named "default" in Cobbler? If so, that will > override your menu behavior. > Definitely delete it and then you should get the menu. > > Let me know if that's not the case. Oh, I see. I wanted to have the default in there as a fall back in the event no input was detected during the menu phase. It's not imperative, so I've deleted the cobbler system "default". The behavior now is that the system does not PXE boot. It simply falls through to the local grub config and boots up the old Centos 4.6 image installed on the machine. My environment was previously homogenous installation of Centos 4.x but now I'm having to support Centos 4.x/5 in parallel. Thanks. From mdehaan at redhat.com Tue May 27 18:59:05 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 27 May 2008 14:59:05 -0400 Subject: [et-mgmt-tools] Cobbler PXE Menu Under CentOS 4.6 In-Reply-To: <3cc2cbc40805271151hec19077n294163635fecb2d2@mail.gmail.com> References: <3cc2cbc40805231542j4be3abacx2a249ffd9f0133b9@mail.gmail.com> <48375066.2080807@redhat.com> <3cc2cbc40805231647q7cfc6138r57aabb31aa51f36d@mail.gmail.com> <48377893.4080308@redhat.com> <3cc2cbc40805271151hec19077n294163635fecb2d2@mail.gmail.com> Message-ID: <483C59F9.3070505@redhat.com> Demetri Mouratis wrote: > On Fri, May 23, 2008 at 7:08 PM, Michael DeHaan wrote: > > >> Do you mean you have a system named "default" in Cobbler? If so, that will >> override your menu behavior. >> Definitely delete it and then you should get the menu. >> >> Let me know if that's not the case. >> > > Oh, I see. I wanted to have the default in there as a fall back in > the event no input was detected during the menu phase. It's not > imperative, so I've deleted the cobbler system "default". The > behavior now is that the system does not PXE boot. It simply falls > through to the local grub config and boots up the old Centos 4.6 image > installed on the machine. > > My environment was previously homogenous installation of Centos 4.x > but now I'm having to support Centos 4.x/5 in parallel. > Perhaps it does. With the 0.8.X and before implementation, you have to type "menu" to get to the PXE menu, because that menu did not time out automatically. So there is a good chance it /is/ PXE booting, but syslinux times out because you didn't type "menu" at the prompt. The other PXE behavior would be achieved by something like: cobbler system add --name=foo --mac=AA:BB:CC:DD:EE:FF --profile=bar Where then you pin that mac address to a specific installation target. Could that explain what you are seeing? --Michael > Thanks. > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > From crobinso at redhat.com Tue May 27 19:22:22 2008 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 27 May 2008 15:22:22 -0400 Subject: [et-mgmt-tools] virt-manager and pxe install In-Reply-To: <483C2EA4.4070900@bppiac.hu> References: <483C2106.20300@bppiac.hu> <20080527150354.GG9310@redhat.com> <483C2EA4.4070900@bppiac.hu> Message-ID: <483C5F6E.6050907@redhat.com> Farkas Levente wrote: > >>> - after select i386 creating the disk image virt-manager crash with some >>> python backtrace as can't connect the guest. >> Please file a BZ with the details > > it's more then half year old: > https://www.redhat.com/archives/et-mgmt-tools/2007-October/msg00062.html > Email is not an option for tracking bugs. If you have been hitting this for months, file a bug in bugzilla.redhat.com. Please include ~/.virt-manager/virt-manager.log with the bug report, and all applicable application versions. > > but anyway. my question is anybody ever be able to install a guest from > virt-manager using pxe boot on kvm? since my assumption is that it's > never been tried and never worked. > That's a poor assumption. We don't make a habit of including features and never testing them. I just kicked off a PXE install on F9 for an x86_64 kvm guest using bridged networking. # rpm -q libvirt kvm qemu virt-manager libvirt-0.4.2-3.fc9.x86_64 kvm-65-1.fc9.x86_64 qemu-0.9.1-5.fc9.x86_64 virt-manager-0.5.4-4.fc9.x86_64 - Cole From dmourati at gmail.com Tue May 27 19:28:39 2008 From: dmourati at gmail.com (Demetri Mouratis) Date: Tue, 27 May 2008 12:28:39 -0700 Subject: [et-mgmt-tools] Cobbler PXE Menu Under CentOS 4.6 In-Reply-To: <483C59F9.3070505@redhat.com> References: <3cc2cbc40805231542j4be3abacx2a249ffd9f0133b9@mail.gmail.com> <48375066.2080807@redhat.com> <3cc2cbc40805231647q7cfc6138r57aabb31aa51f36d@mail.gmail.com> <48377893.4080308@redhat.com> <3cc2cbc40805271151hec19077n294163635fecb2d2@mail.gmail.com> <483C59F9.3070505@redhat.com> Message-ID: <3cc2cbc40805271228t12198e2em47d0afcd1ad89eb0@mail.gmail.com> On Tue, May 27, 2008 at 11:59 AM, Michael DeHaan wrote: > Perhaps it does. > > With the 0.8.X and before implementation, you have to type "menu" to get to > the PXE menu, because that menu did not time out automatically. > So there is a good chance it /is/ PXE booting, but syslinux times out > because you didn't type "menu" at the prompt. > > The other PXE behavior would be achieved by something like: > > cobbler system add --name=foo --mac=AA:BB:CC:DD:EE:FF --profile=bar > > Where then you pin that mac address to a specific installation target. > > Could that explain what you are seeing? > > --Michael > >> Thanks. Hi Michael, Thanks for all the help so far. I've deleted the "default" system and now I think I'm running into a timeout issue or something similar. I'm looking at a hung boot process and no prompt at which to enter "menu". Part of the problem is that I'm doing this all over serial console (the systems are in London, I'm in California). Are there any peculiarities you are aware of with respect to serial console and PXE menu booting? I know I have set this up manually prior to being introduced to cobbler. The thing is, I'm somewhat stuck unless I take your suggestion and starting adding systems by MAC address which I'd like to avoid in favor of the menu approach. Thanks. -D From berrange at redhat.com Tue May 27 20:06:02 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 27 May 2008 21:06:02 +0100 Subject: [et-mgmt-tools] virt-manager and pxe install In-Reply-To: <483C2EA4.4070900@bppiac.hu> References: <483C2106.20300@bppiac.hu> <20080527150354.GG9310@redhat.com> <483C2EA4.4070900@bppiac.hu> Message-ID: <20080527200602.GA14298@redhat.com> On Tue, May 27, 2008 at 05:54:12PM +0200, Farkas Levente wrote: > Daniel P. Berrange wrote: > >On Tue, May 27, 2008 at 04:56:06PM +0200, Farkas Levente wrote: > >>hi, > > > >KVM only virtualizes a CPU that matches the host. If you're on a 64-bit > >host, then it only virtualizes a 64-bit CPU. It doesn't virtualize a > >32-bit CPU - if you want a true 32-bit cpu on a 64-bit host you need to > >emulate with QEMU instead. That said its rather pointless because you > >can run a 32-bit guest OS in a 64-kvm KVM vm just fine. > > huu i learn again something:-) until now (ie. in previous version of > virt-manager) i can select 32bit for my guest so i assume kvm can > virtualize 32bit cpu on 64bit host. Then you have likely been running your VMs with plain QEMU and no acceleration > but if it can't it'd better to not > allow to choose 32bit in case of kvm. and of course virt-manager already > know the host has only kvm hypervisor. That is already the case - if you select KVM then it won't let you select i386 and vica-verca. > >>- after select i386 creating the disk image virt-manager crash with some > >>python backtrace as can't connect the guest. > > > >Please file a BZ with the details > > it's more then half year old: > https://www.redhat.com/archives/et-mgmt-tools/2007-October/msg00062.html Note where I said *BZ*, and that is not enough information anyway. It needs the full virt-manager log file, and the VM log from /var/log/libvirt > but anyway. my question is anybody ever be able to install a guest from > virt-manager using pxe boot on kvm? since my assumption is that it's > never been tried and never worked. It has been tried and does work. As I said in my previous mail there are sometimes problems with some versions of KVM, and also incompatabilities with certain DHCP/PXE servers. There are also frequently mis-configured hosts by the sysadmin which block the traffic in the firewall. Dan. -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From mdehaan at redhat.com Tue May 27 22:08:42 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 27 May 2008 18:08:42 -0400 Subject: [et-mgmt-tools] Cobbler Log status command Failure In-Reply-To: <483C21C3.5010901@redhat.com> References: <1B3160038A79B54DB674532195C68A586DB033@IMCLTEXCP15.nam.ent.duke-energy.com> <483C21C3.5010901@redhat.com> Message-ID: <483C866A.1090502@redhat.com> Michael DeHaan wrote: > Mackell, Thomas O wrote: >> >> Here is the two log files. >> >> I was unable to send them to you.. I received the following message: >> >> Your message did not reach some or all of the intended recipients. >> >> >> >> Subject: Cobbler status log files >> >> Sent: 5/23/2008 2:33 PM >> >> >> >> The following recipient(s) could not be reached: >> >> >> >> mdehann at redhat.com on 5/23/2008 2:34 PM >> >> The message reached the recipient's e-mail system, but >> delivery was refused. Attempt to resend the message. If it still >> fails, contact your system administrator. >> >> >> >> >> >> Thank You, >> >> Thomas Mackell >> >> 704-382-9731 >> >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools > > I'll check this out. > > FWIW, your email error is from a typo in my username. > > --Michael > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Running "cobbler status" from devel-HEAD in version control, I see the following output using your install.log: [root at mdehaan tmp]# cobbler status ip |target |start |state 10.250.48.174 |profile:ossnts100 |Wed Dec 31 18:59:59 1969|finished 127.0.0.1 |system:server |Wed Dec 31 18:59:59 1969|finished 139.46.28.39 |profile:linux2 |Wed Dec 31 18:59:59 1969|finished 172.16.57.100 |profile:F8-xen-x86_64|Wed May 14 16:47:55 2008|unknown/stalled [root at mdehaan tmp]# Actually it's much nicer than that, though my mail reader is not mono-spacing it :) If this is not working for you now, it seems happy in what will be 1.0. --Michael From Feichtinger.Guenter at bartelt.at Wed May 28 10:22:53 2008 From: Feichtinger.Guenter at bartelt.at (=?iso-8859-1?Q?Feichtinger_G=FCnter?=) Date: Wed, 28 May 2008 12:22:53 +0200 Subject: [et-mgmt-tools] AW: ocaml-libvirt-0.4.0.1: Can't connect to Xen- Host In-Reply-To: <20080527094516.GC29698@amd.home.annexia.org> Message-ID: > -----Urspr?ngliche Nachricht----- > Von: Richard W.M. Jones [mailto:rjones at redhat.com] > Gesendet: Dienstag, 27. Mai 2008 11:45 > An: Feichtinger G?nter > Cc: et-mgmt-tools at redhat.com > Betreff: Re: ocaml-libvirt-0.4.0.1: Can't connect to Xen- Host > > On Tue, May 27, 2008 at 11:37:10AM +0200, Feichtinger G?nter wrote: > > I intstalled the ocaml-libvirt-0.4.0.1.exe on MS- Vista > without problems. > > Also the virt-ctrl.exe starts without problems. > > But when I try to connect to a Xen-Host I get the messages: > > libvir: Remote error : Cannot access CA certificate > > 'C:/msys/1.0/local/etc/pki/C > > A/cacert.pem': No such file or directory (2) > > > > Please be so kind and help. > > It's not very clear to me what you are trying to connect to > what, but you'll probably want to read about remote connections, here: > > http://libvirt.org/remote.html > > Rich. Hello Richard, thanks for your prompt answer. I see, I have to explain it more excatly. I have 3 Xen-Hosts (CentOS 5.1) and use virt-managaer local an this hosts. I'm looking for a possibility to manages xen-hosts from MS-Windows-Clients like VMWareServerConsole. So I was looking around and found your port of the virt-manager. I tried to connect with xen://xen-host/ and get the message above. Which possibility to I have for remote connections? I think I read that with your virt-manager port only support TLS, isn't it? So do I have to work with certificates? In my test-envirment a "low" security solution is also fine :-) regard, G?nter > > -- > Richard Jones, Emerging Technologies, Red Hat > http://et.redhat.com/~rjones Read my OCaml programming blog: > http://camltastic.blogspot.com/ Fedora now supports 59 OCaml > packages (the OPEN alternative to F#) > http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora > From martin.pearson at japannext.co.jp Wed May 28 11:05:15 2008 From: martin.pearson at japannext.co.jp (Martin Pearson) Date: Wed, 28 May 2008 20:05:15 +0900 Subject: [et-mgmt-tools] Re: Koan problem In-Reply-To: <483A12E4.7040109@japannext.co.jp> References: <48366493.4010802@japannext.co.jp> <4836728E.8070308@japannext.co.jp> <4836734A.5020207@japannext.co.jp> <48367639.6000108@japannext.co.jp> <483777C0.4080405@redhat.com> <483A12E4.7040109@japannext.co.jp> Message-ID: <483D3C6B.5080406@japannext.co.jp> Martin Pearson wrote: > Michael DeHaan wrote: >> Martin Pearson wrote: >>> Martin Pearson wrote: >>>> Hi All, >>>> Wonder if you can help me when I run Koan with the following inputs:- >>>> >>>> koan --virt --server=puppet --profile=RHEL4-ES-xen-x86_64 >>>> --virt-name=tkdvir1 >>>> >>>> I get the following error:- >>>> >>>> libvir: Xen error : Domain not found: xenUnifiedDomainLookupByUUID >>>> >> >> I don't think I have seen this particular error before. >> >> Can you share the koan, libvirt/Xen, and OS versions you are using, >> and the full traceback (error message/output) if available? >> >> Koan also saves some additional logs in ~/.koan when creating VMs >> that may also be useful. >> --Michael >> >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools > Hi Michael, > Thanks very much for your reply. Please see below info requested:- > > [root at xxxxxx ~]# rpm -qa | grep -i libvirt > libvirt-0.3.3-7.el5 > libvirt-python-0.3.3-7.el5 > [root at xxxxxx ~]# rpm -qa | grep -i xen > xen-libs-3.0.3-64.el5_2.1 > kernel-xen-devel-2.6.18-92.el5 > kernel-xen-2.6.18-92.el5 > xen-devel-3.0.3-64.el5_2.1 > xen-3.0.3-64.el5_2.1 > [root at xxxxxx ~]# uname -a > Linux xxxxxxxx 2.6.18-92.el5xen #1 SMP Tue Apr 29 13:31:30 EDT 2008 > x86_64 x86_64 x86_64 GNU/Linux > Red Hat Enterprise Linux Server release 5 (Tikanga) > > Python info, just in case that helps. > > python-urlgrabber-3.1.0-2 > python-numeric-23.7-2.2.2 > gtk-vnc-python-0.3.2-3.el5 > python-setuptools-0.6c7-1.el5 > libxml2-python-2.6.26-2.1.2.1 > python-virtinst-0.300.2-8.el5 > python-elementtree-1.2.6-5 > python-sqlite-1.1.7-1.2.1 > dbus-python-0.70-7.el5 > libvirt-python-0.3.3-7.el5 > python-2.4.3-21.el5 > python-devel-2.4.3-21.el5 > libselinux-python-1.33.4-5.el5 > audit-libs-python-1.6.5-9.el5 > rpm-python-4.4.2-48.el5 > python-devel-2.4.3-21.el5 > > Version of cobbler on the server:- > > cobbler-0.8.2-2 > > Output from koan command:- > > [root at xxxxxxxx .koan]# koan --virt --server=puppet > --profile=RHEL4-ES-xen-x86_64 --virt-name=tkdvir1 > - reading URL: > http://10.145.100.35/cblr/kickstarts/RHEL4-ES-xen-x86_64/ks.cfg > install_tree: http://10.145.100.35/cblr/links/RHEL4-ES-xen-x86_64 > - no virt-type specified, auto-selecting xenpv > libvirtd (pid 16816) is running... > downloading initrd initrd.img to /var/lib/xen/initrd.img > url=http://puppet/cobbler/images/RHEL4-ES-xen-x86_64/initrd.img > - reading URL: > http://puppet/cobbler/images/RHEL4-ES-xen-x86_64/initrd.img > downloading kernel vmlinuz to /var/lib/xen/vmlinuz > url=http://puppet/cobbler/images/RHEL4-ES-xen-x86_64/vmlinuz > - reading URL: http://puppet/cobbler/images/RHEL4-ES-xen-x86_64/vmlinuz > _*libvir: Xen error : Domain not found: xenUnifiedDomainLookupByUUID > libvir: Xen error : Domain not found: xenUnifiedDomainLookupByName*_ > Kernel arguments: > ks=http://10.145.100.35/cblr/kickstarts/RHEL4-ES-xen-x86_64/ks.cfg > ksdevice=eth0 lang= kssendmac syslog=10.145.100.35:25150 text > use virt-manager or reconnect with virsh console tkdvir1 > > Sorry was not precise enough with my previous email. Does not seem to > be Koan that is crashing it just appears to be having a problem > creating the virtual machine, i.e. if I go into virsh I have a machine > there I can connect to via console . Although the > machine is stuck on boot see below:- > > RAMDISK driver initialized: 16 RAM disks of 16384K size 1024 blocksize > Xen virtual console successfully installed as xvc0 > Event-channel device installed. > Console: switching to colour frame buffer device 100x37 > mice: PS/2 mouse device common for all mice > md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27 > NET: Registered protocol family 2 > IP route cache hash table entries: 32768 (order: 6, 262144 bytes) > TCP established hash table entries: 131072 (order: 9, 2097152 bytes) > TCP bind hash table entries: 131072 (order: 9, 2097152 bytes) > TCP: Hash tables configured (established 131072 bind 131072) > Initializing IPsec netlink socket > NET: Registered protocol family 1 > NET: Registered protocol family 17 > 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). _* > > *_Also the tail end of the /var/log/messages log just in case this is > useful:- > > May 26 10:10:19 xxxxxxxx kernel: device vif28.0 entered promiscuous mode > May 26 10:10:19 xxxxxxxx kernel: audit(1211764219.787:43): dev=vif28.0 > prom=256 old_prom=0 auid=4294967295 ses=4294967295 > May 26 10:10:19 xxxxxxxx kernel: ADDRCONF(NETDEV_UP): vif28.0: link is > not ready > May 26 10:10:20 xxxxxxxx kernel: ADDRCONF(NETDEV_CHANGE): vif28.0: > link becomes ready > May 26 10:10:20 xxxxxxxx kernel: xenbr0: port 3(vif28.0) entering > learning state > May 26 10:10:20 xxxxxxxx kernel: blkback: ring-ref 522, event-channel > 9, protocol 1 (x86_64-abi) > May 26 10:10:20 xxxxxxxx kernel: xenbr0: topology change detected, > propagating > May 26 10:10:20 xxxxxxxx kernel: xenbr0: port 3(vif28.0) entering > forwarding state > May 26 10:22:39 xxxxxxxx kernel: xenbr0: port 3(vif28.0) entering > disabled state > May 26 10:22:39 xxxxxxxx kernel: device vif28.0 left promiscuous mode > May 26 10:22:39 xxxxxxxx kernel: audit(1211764959.982:44): dev=vif28.0 > prom=0 old_prom=256 auid=4294967295 ses=4294967295 > May 26 10:22:39 xxxxxxxx kernel: xenbr0: port 3(vif28.0) entering > disabled state > _* > *_Interesting thing is I can create a VM fine manually with virt-install? > > Would be great if you could help out with this. > > > Many thanks, > > > Martin Pearson. > > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Hi, Sorry I know understand what the problem was with this error message. It was not important and the output from the VM was going trying to go to a graphics terminal therefore I could not see anything as I was running it in text mode. Sorry!! When I was using the old version of koan I did not need the no graphics switch. Thanks for your help. Regards, Martin. From mdehaan at redhat.com Wed May 28 16:35:01 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 28 May 2008 12:35:01 -0400 Subject: [et-mgmt-tools] [ANNOUNCE] Cobbler and Koan 1.0 Message-ID: <483D89B5.6050909@redhat.com> Hello Cobbler land. Cobbler and koan 1.0 are now built and pushed out to the mirrors for F-7 through F-9, and EL-4/5. If you can't wait, there are source RPMs at http://cobbler.et.redhat.com/download What's new in 1.0? If you've been reading my emails about 0.9, it's the same thing. This is the merge of the development branch (0.9) work to stable so we can start doing more "unstable" development again. If any major issues come up with 1.0, there can be a 1.0.1, 1.0.2 and so forth as new features will be rolled into 1.1 (devel) for future inclusion in 1.2 stable. Should there be any problems with this release, please let me know either through Trac, email, or IRC so we can keep this release working for everyone. This has had a few test releases and people have been very good about playing with source versions, so I do not expect any major problems, knock on silicon :) I've taken the opportunity to streamline a few things -- namely the settings file(s). All users upgraded are strongly encouraged to read the link below, as your Cobbler configuration will yell at you until you fix at least some basic things, and if you've customized it, you need to also pay attention to the parts about modules.conf. Since 1.0 is the "big" release, it was a good time to clean some things up so we don't have to do this kind of thing as much later. https://fedorahosted.org/cobbler/wiki/WhatsNewInOhNine What's in a name? 1.0 is technically just the release after 0.8. However at this point, with all of the security features and so on going into this release, the core of features we are going to want are there and things are looking pretty well rounded to me. This doesn't mean there isn't a lot of territory to conquer and other things to clean up. There are still things we could possibly do around power management, image cloning (maybe), and definitely a lot of things we can do to polish up the web interface. There are also a whole lot of great RFEs in Trac on things to improve, so likely the near future is about getting to these small/miscellaneous items, especially as there were a lot of big items in 0.9 (primarily in the Web/security area). As 0.9/1.0 targeted a lot of features that are useful to some larger install bases, 1.1 will likely hit all the little items everyone has been asking for. The roadmap is always kept here and suggestions are always welcome: https://fedorahosted.org/cobbler/wiki/TheRoadmap Once again, and I say this a lot but I really mean it -- thanks very much for everyone who has contributed testing, ideas, feedback, bug reports, and patches. Cobbler is really about building a community around administrators being able to share and build common tooling together, and lately this is clearly showing that this sort of collaboration works. Thank you all very much. Also thanks very much to everyone who has shared Cobbler with friends, coworkers, LUGs, and so on. You are very helpful in expanding the community and bringing more ideas to the table. --Michael From Pablo.Iranzo at redhat.com Wed May 28 19:01:20 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Wed, 28 May 2008 21:01:20 +0200 (CEST) Subject: [et-mgmt-tools] COBBLER patch for url consistency in validateks Message-ID: Hi all! This patch is for adding url consistency with the changes we did for work-arrounding the libvirt problems. It "fixes" the url used by validateks action for use the slashed one. Doesn't add or remove features :) Regards Pablo PD: Congratulations to all of you in your collaboration for making Cobbler 1.0 real :) -- Pablo Iranzo G?mez (http://Alufis35.uv.es/~iranzo/) (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) -- Postulado de Boling sobre la Ley de Murphy: Si se encuentra bien, no se preocupe. Se le pasar? -------------- next part -------------- A non-text attachment was scrubbed... Name: validate-url.diff Type: text/x-diff Size: 723 bytes Desc: URL: From mdehaan at redhat.com Wed May 28 19:50:16 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 28 May 2008 15:50:16 -0400 Subject: [et-mgmt-tools] COBBLER patch for url consistency in validateks In-Reply-To: References: Message-ID: <483DB778.8080909@redhat.com> Pablo Iranzo G?mez wrote: > Hi all! > > This patch is for adding url consistency with the changes we did > for work-arrounding the libvirt problems. > > It "fixes" the url used by validateks action for use the slashed > one. Doesn't add or remove features :) > > Regards > Pablo > > PD: Congratulations to all of you in your collaboration for making Cobbler > 1.0 real :) > op=ks should be op/ks, though I've applied it with that tweak. Thanks! > > > > ------------------------------------------------------------------------ > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From fj7025cf at aa.jp.fujitsu.com Thu May 29 02:21:10 2008 From: fj7025cf at aa.jp.fujitsu.com (Hiroyuki Kaguchi) Date: Thu, 29 May 2008 11:21:10 +0900 Subject: [et-mgmt-tools] [PATCH][virt-inst] add error message when the number of vcpus is out of range. Message-ID: <483E1316.3030000@aa.jp.fujitsu.com> Hi, When the number of vcpus is out of range, the domain boots with vcpus=1 successfully. This patch changes the policy that virt-install exits and shows error message. Thanks, Signed-off-by: Hiroyuki Kaguchi diff -r 4776c61e0fa7 virtinst/cli.py --- a/virtinst/cli.py Thu May 15 16:13:55 2008 -0400 +++ b/virtinst/cli.py Wed May 28 15:57:32 2008 +0900 @@ -183,11 +183,12 @@ def get_vcpus(vcpus, check_cpu, guest, c vcpus = int(prompt_for_input(_("How many VCPUs should be attached?"))) except ValueError, e: print _("ERROR: "), e - if vcpus: + if vcpus is not None: try: guest.vcpus = vcpus except ValueError, e: print _("ERROR: "), e + sys.exit(1) def get_cpuset(cpuset, mem, guest, conn): if cpuset and cpuset != "auto": From fj7025cf at aa.jp.fujitsu.com Thu May 29 02:38:25 2008 From: fj7025cf at aa.jp.fujitsu.com (Hiroyuki Kaguchi) Date: Thu, 29 May 2008 11:38:25 +0900 Subject: [et-mgmt-tools] [PATCH] Enable to send F10 key to the guest OS. In-Reply-To: <48363446.8020601@aa.jp.fujitsu.com> References: <48363123.605@aa.jp.fujitsu.com> <48363446.8020601@aa.jp.fujitsu.com> Message-ID: <483E1721.6000203@aa.jp.fujitsu.com> > On 2008/05/23 11:51, Hiroyuki Kaguchi wrote: >> The F10 key is assigned as the accelerator of the menu bar in the OS(Windows or >> Linux). >> On this account it is necessary to send the F10 key to the guest OS. >> However, with the existing specifications of virt-viewer, the F10 key is >> assigned to the accelerator >> of the menu bar of virt-viewer and we cannot send the F10 key to the guest OS. >> So, I change virt-viewer to be able to send F10 key to the guest OS when >> virt-viewer grabs the keyboard. > Would you give me a comment on this patch? If not, please apply it. Thanks, Hiroyuki Kaguchi From fj7025cf at aa.jp.fujitsu.com Thu May 29 05:44:30 2008 From: fj7025cf at aa.jp.fujitsu.com (Hiroyuki Kaguchi) Date: Thu, 29 May 2008 14:44:30 +0900 Subject: [et-mgmt-tools] [PATCH][virt-viewer] Add error message when it connects to the stopping domain. Message-ID: <483E42BE.9040907@aa.jp.fujitsu.com> Hi, When the guest domain is stopping, virt-viewer exits without error message. The patch adds error message when it fails to connect stopping domain. Thanks, Signed-off-by: Hiroyuki Kaguchi diff -r 5299a3aaebb5 src/main.c --- a/src/main.c Sun Apr 27 23:41:26 2008 -0400 +++ b/src/main.c Thu May 29 13:58:00 2008 +0900 @@ -785,6 +785,7 @@ viewer_start (const char *uri, const cha GtkWidget *vnc; virConnectPtr conn = NULL; virDomainPtr dom = NULL; + virDomainInfo domInfo; char *host = NULL; char *vncport = NULL; char *transport = NULL; @@ -813,11 +814,24 @@ viewer_start (const char *uri, const cha usleep(500*1000); } while (!dom); - viewer_extract_vnc_graphics(dom, &vncport); - if (!vncport && !waitvnc) { - fprintf(stderr, "unable to find vnc graphics for %s\n", name); - return 4; + if (virDomainGetInfo(dom, &domInfo) != 0) { + fprintf(stderr, "unable to get information for %s\n", name); + return 6; } + + if (domInfo.state == VIR_DOMAIN_SHUTOFF && !waitvnc) { + fprintf(stderr, "%s is not running\n", name); + return 7; + } + + if (domInfo.state != VIR_DOMAIN_SHUTOFF) { + viewer_extract_vnc_graphics(dom, &vncport); + if (!vncport) { + fprintf(stderr, "unable to find vnc graphics for %s\n", name); + return 4; + } + } + if (!vncport) { virDomainFree(dom); usleep(300*1000); From tom at ng23.net Thu May 29 08:55:07 2008 From: tom at ng23.net (Tom Brown) Date: Thu, 29 May 2008 09:55:07 +0100 Subject: [et-mgmt-tools] koan 1.0 - RHEL 5.2 x86_64 - Full Virt Message-ID: <483E6F6B.3040205@ng23.net> Hi Does anyone know if a RHEL5.2 x86_64 dom0 using the default Xen which is 3.0.3 can do full virt of a 32bit domU ? I want to avoid using Fedora if at all possible so i wonder if recent a RHEL can handle it. thanks From lhaig at haigmail.com Thu May 29 09:53:56 2008 From: lhaig at haigmail.com (Lance Haig) Date: Thu, 29 May 2008 10:53:56 +0100 Subject: [et-mgmt-tools] error connecting using virt-manager Message-ID: <483E7D34.5010801@haigmail.com> Hey guys, I have packaged virt-manager for Foresight Linux and during testing I getthis error when connecting to the libvirtd daemon Unable to open connection to hypervisor URI 'qemu:///system': libvirt.libvirtError virConnectOpenAuth() failed Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/connection.py", line 427, in _open_thread None], flags) File "/usr/lib/python2.4/site-packages/libvirt.py", line 94, in openAuth if ret is None:raise libvirtError('virConnectOpenAuth() failed') libvirtError: virConnectOpenAuth() failed I see there is an auth error but am not sure where to start looking to fix it. Regards Lance From robin-lists at robinbowes.com Thu May 29 10:05:56 2008 From: robin-lists at robinbowes.com (Robin Bowes) Date: Thu, 29 May 2008 11:05:56 +0100 Subject: [et-mgmt-tools] Re: koan 1.0 - RHEL 5.2 x86_64 - Full Virt In-Reply-To: <483E6F6B.3040205@ng23.net> References: <483E6F6B.3040205@ng23.net> Message-ID: Tom Brown wrote: > Hi > > Does anyone know if a RHEL5.2 x86_64 dom0 using the default Xen which is > 3.0.3 can do full virt of a 32bit domU ? > > I want to avoid using Fedora if at all possible so i wonder if recent a > RHEL can handle it. This is the wrong list to ask on - try the xen list, or fedora-xen. That said, I'm running CentOS 5.1 with xen 3.1.x that I rebuilt from the Fedora RPMS. I believe the general feeling is that 32-bit guests are unstable on x86_64 Dom0. R. From clalance at redhat.com Thu May 29 10:12:18 2008 From: clalance at redhat.com (Chris Lalancette) Date: Thu, 29 May 2008 12:12:18 +0200 Subject: [et-mgmt-tools] Re: koan 1.0 - RHEL 5.2 x86_64 - Full Virt In-Reply-To: References: <483E6F6B.3040205@ng23.net> Message-ID: <483E8182.6020006@redhat.com> Robin Bowes wrote: > This is the wrong list to ask on - try the xen list, or fedora-xen. > > That said, I'm running CentOS 5.1 with xen 3.1.x that I rebuilt from the > Fedora RPMS. I believe the general feeling is that 32-bit guests are > unstable on x86_64 Dom0. They shouldn't be; 32-bit fully virtualized guests have been fully supported on 64-bit dom0 since 5.0. 5.2 has added 32-bit paravirtualized on 64-bit dom0 as a tech preview, although it still has a couple of bugs to work out. If there are problems with running 32-bit fully virtualized guests on 64-bit dom0, please report them to BZ. Chris Lalancette From haerry+et at puzzle.ch Thu May 29 12:38:50 2008 From: haerry+et at puzzle.ch (haerry+et at puzzle.ch) Date: Thu, 29 May 2008 14:38:50 +0200 Subject: [et-mgmt-tools] cobbler 1.0 profiles/repo inheritance check problem Message-ID: <483EA3DA.4060201@puzzle.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi beside switching our cobbler setup to a new server we upgraded to 1.0. However we then had a problem running cobbler check: # cobbler check The following potential problems were detected: #0: since iptables may be running, ensure 69, 80, 25150, and 25151 are unblocked #1: One or more repos referenced by profile objects is no longer defined in cobbler: <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, > So first I thought of having a problem due to our migration, however I then could say for sure that the setup is the same. Talking then on irc with iranzo, we figured out that it have to be a problem in the profiles at the repos: part. I had 27 profiles with 7 being the top and all the others inheriting from these 7. Makes 20 repos which inherits and have therefor repos: <> in the yaml. Which is exactly the amount of <,<,i,n... in the error above. Wit dumpvars everything looked ok and they had the repos inherited from the parent. However when I now added the repos for each profile without inheriting from the parent it works fine and there's only the firewall warning left. So for me it looks like cobbler check have a problem with inheriting repos in profiles. Anybody can confirm this behaviour? Or what else might cause the problem? thanks and greets Marcel -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFIPqPZxqnHThA4pEkRArMJAJ0Qf+9ybElm0uQApI32OPrYsayEnQCfZDmD vAGXUqbA1V8jz+nCDA0KVsY= =dtvu -----END PGP SIGNATURE----- From pasik at iki.fi Thu May 29 13:37:15 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Thu, 29 May 2008 16:37:15 +0300 Subject: [et-mgmt-tools] koan 1.0 - RHEL 5.2 x86_64 - Full Virt In-Reply-To: <483E6F6B.3040205@ng23.net> References: <483E6F6B.3040205@ng23.net> Message-ID: <20080529133715.GP8042@edu.joroinen.fi> On Thu, May 29, 2008 at 09:55:07AM +0100, Tom Brown wrote: > Hi > > Does anyone know if a RHEL5.2 x86_64 dom0 using the default Xen which is > 3.0.3 can do full virt of a 32bit domU ? > RHEL5.2 has Xen 3.1.2. The package names might say 3.0.3, but it is 3.1.2. Check RHEL5.2 release notes for more information. -- Pasi From berrange at redhat.com Thu May 29 13:44:52 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 29 May 2008 14:44:52 +0100 Subject: [et-mgmt-tools] koan 1.0 - RHEL 5.2 x86_64 - Full Virt In-Reply-To: <20080529133715.GP8042@edu.joroinen.fi> References: <483E6F6B.3040205@ng23.net> <20080529133715.GP8042@edu.joroinen.fi> Message-ID: <20080529134452.GB2433@redhat.com> On Thu, May 29, 2008 at 04:37:15PM +0300, Pasi K?rkk?inen wrote: > On Thu, May 29, 2008 at 09:55:07AM +0100, Tom Brown wrote: > > Hi > > > > Does anyone know if a RHEL5.2 x86_64 dom0 using the default Xen which is > > 3.0.3 can do full virt of a 32bit domU ? > > > > RHEL5.2 has Xen 3.1.2. The package names might say 3.0.3, but it is 3.1.2. No, it is not actually that simple. - Hypervisor - 3.1.2 - Kernel - 3.0.3 + important patchs from 3.1.0 + 3.1.2 - Userspace - 3.0.3 XenD/xm + 3.1.0 QEMU + important patches In particular anything which is a user visible API / command has remained only 3.0.3 since we need to maintain stable API / CLI syntax throughout the lifetime of RHEL-5. This is why the packages are still called 3.0.3; their user facing API / cli is 3.0.3 based. Dan. -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From lfarkas at bppiac.hu Thu May 29 13:50:36 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Thu, 29 May 2008 15:50:36 +0200 Subject: [et-mgmt-tools] virt-manager and pxe install In-Reply-To: <20080527200602.GA14298@redhat.com> References: <483C2106.20300@bppiac.hu> <20080527150354.GG9310@redhat.com> <483C2EA4.4070900@bppiac.hu> <20080527200602.GA14298@redhat.com> Message-ID: <483EB4AC.5030800@bppiac.hu> Daniel P. Berrange wrote: >> but if it can't it'd better to not >> allow to choose 32bit in case of kvm. and of course virt-manager already >> know the host has only kvm hypervisor. > > That is already the case - if you select KVM then it won't let you select > i386 and vica-verca. currently even if i select kvm i can select 32. >> but anyway. my question is anybody ever be able to install a guest from >> virt-manager using pxe boot on kvm? since my assumption is that it's >> never been tried and never worked. > > It has been tried and does work. As I said in my previous mail there are > sometimes problems with some versions of KVM, and also incompatabilities > with certain DHCP/PXE servers. There are also frequently mis-configured > hosts by the sysadmin which block the traffic in the firewall. as it can be seen from my screenshot it (the kvm bios) can't found the hdcp server. so it's nothing to do with pxe. on the same kvm host other guest are able to use dhcp, but of course in that case the os's (fedora and centos) dhcp client, while in my case kvm bios emulation's dhcp client. boht the host and guest is on our labs without firewall and selinux disabled. dhcp server is on another server on the same lan. if i run on the host: tcpdump -i peth0 icmp i didn't got any kind of packages. so what else can i do? how can i find the reason? -- Levente "Si vis pacem para bellum!" From pasik at iki.fi Thu May 29 14:07:50 2008 From: pasik at iki.fi (Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?=) Date: Thu, 29 May 2008 17:07:50 +0300 Subject: [et-mgmt-tools] koan 1.0 - RHEL 5.2 x86_64 - Full Virt In-Reply-To: <20080529134452.GB2433@redhat.com> References: <483E6F6B.3040205@ng23.net> <20080529133715.GP8042@edu.joroinen.fi> <20080529134452.GB2433@redhat.com> Message-ID: <20080529140750.GQ8042@edu.joroinen.fi> On Thu, May 29, 2008 at 02:44:52PM +0100, Daniel P. Berrange wrote: > On Thu, May 29, 2008 at 04:37:15PM +0300, Pasi K?rkk?inen wrote: > > On Thu, May 29, 2008 at 09:55:07AM +0100, Tom Brown wrote: > > > Hi > > > > > > Does anyone know if a RHEL5.2 x86_64 dom0 using the default Xen which is > > > 3.0.3 can do full virt of a 32bit domU ? > > > > > > > RHEL5.2 has Xen 3.1.2. The package names might say 3.0.3, but it is 3.1.2. > > No, it is not actually that simple. > > - Hypervisor - 3.1.2 > - Kernel - 3.0.3 + important patchs from 3.1.0 + 3.1.2 > - Userspace - 3.0.3 XenD/xm + 3.1.0 QEMU + important patches > > In particular anything which is a user visible API / command has remained > only 3.0.3 since we need to maintain stable API / CLI syntax throughout > the lifetime of RHEL-5. This is why the packages are still called 3.0.3; > their user facing API / cli is 3.0.3 based. > Yep, thanks for explaining in in detail. I meant the hypervisor itself is 3.1.2 :) -- Pasi From mdehaan at redhat.com Thu May 29 14:26:18 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 29 May 2008 10:26:18 -0400 Subject: [et-mgmt-tools] cobbler 1.0 profiles/repo inheritance check problem In-Reply-To: <483EA3DA.4060201@puzzle.ch> References: <483EA3DA.4060201@puzzle.ch> Message-ID: <483EBD0A.3090103@redhat.com> haerry+et at puzzle.ch wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi > > beside switching our cobbler setup to a new server we upgraded to 1.0. > However we then had a problem running cobbler check: > > # cobbler check > The following potential problems were detected: > #0: since iptables may be running, ensure 69, 80, 25150, and 25151 are > unblocked > #1: One or more repos referenced by profile objects is no longer defined > in cobbler: <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, > >> , >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, >> > <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, > i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, > h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, > r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, > t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, > >> , <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, >> > <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, > n, h, e, r, i, t, >, > > > So first I thought of having a problem due to our migration, however I > then could say for sure that the setup is the same. Talking then on irc > with iranzo, we figured out that it have to be a problem in the profiles > at the repos: part. I had 27 profiles with 7 being the top and all the > others inheriting from these 7. Makes 20 repos which inherits and have > therefor repos: <> in the yaml. Which is exactly the amount of > <,<,i,n... in the error above. > Wit dumpvars everything looked ok and they had the repos inherited from > the parent. However when I now added the repos for each profile without > inheriting from the parent it works fine and there's only the firewall > warning left. > > So for me it looks like cobbler check have a problem with inheriting > repos in profiles. Anybody can confirm this behaviour? Or what else > might cause the problem? > > thanks and greets Marcel > Thankfully it's just "cobbler check" so you can ignore that message for now. It should not be causing problems elsewhere, but yes, that should not be happening :) I'll teach it to ignore inherits and this can go into the next 1.0.X maintaince release. Patch should be in shortly. --Michael > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.6 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > iD8DBQFIPqPZxqnHThA4pEkRArMJAJ0Qf+9ybElm0uQApI32OPrYsayEnQCfZDmD > vAGXUqbA1V8jz+nCDA0KVsY= > =dtvu > -----END PGP SIGNATURE----- > > _______________________________________________ > 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 May 29 16:23:51 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 29 May 2008 12:23:51 -0400 Subject: [et-mgmt-tools] cobbler 1.0 profiles/repo inheritance check problem In-Reply-To: <483EBD0A.3090103@redhat.com> References: <483EA3DA.4060201@puzzle.ch> <483EBD0A.3090103@redhat.com> Message-ID: <483ED897.7050206@redhat.com> Michael DeHaan wrote: > haerry+et at puzzle.ch wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Hi >> >> beside switching our cobbler setup to a new server we upgraded to 1.0. >> However we then had a problem running cobbler check: >> >> # cobbler check >> The following potential problems were detected: >> #0: since iptables may be running, ensure 69, 80, 25150, and 25151 are >> unblocked >> #1: One or more repos referenced by profile objects is no longer defined >> in cobbler: <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >> >>> , >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, >>> >> <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, >> i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, >> h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, >> r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, >> t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >> >>> , <, <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, >>> >> <, i, n, h, e, r, i, t, >, >, <, <, i, n, h, e, r, i, t, >, >, <, <, i, >> n, h, e, r, i, t, >, > >> >> So first I thought of having a problem due to our migration, however I >> then could say for sure that the setup is the same. Talking then on irc >> with iranzo, we figured out that it have to be a problem in the profiles >> at the repos: part. I had 27 profiles with 7 being the top and all the >> others inheriting from these 7. Makes 20 repos which inherits and have >> therefor repos: <> in the yaml. Which is exactly the amount of >> <,<,i,n... in the error above. >> Wit dumpvars everything looked ok and they had the repos inherited from >> the parent. However when I now added the repos for each profile without >> inheriting from the parent it works fine and there's only the firewall >> warning left. >> >> So for me it looks like cobbler check have a problem with inheriting >> repos in profiles. Anybody can confirm this behaviour? Or what else >> might cause the problem? >> >> thanks and greets Marcel >> > > Thankfully it's just "cobbler check" so you can ignore that message > for now. It should not be causing problems elsewhere, but yes, that > should not be happening :) > > I'll teach it to ignore inherits and this can go into the next 1.0.X > maintaince release. Patch should be in shortly. > > --Michael > I've pushed a patch out to git for this, and it will be applied to the next release whether 1.0.X or 1.1 devel (both). I've also removed two items of directory ownership in tftpboot. F9 uses /var/lib/tftpboot, and while I check for it, the RPM should not be creating directories there. cobbler itself will create the intermediate directories it needs if they do not exist. If you need the patch and don't want to rebuild from source, it's here: http://git.fedorahosted.org/git/?p=cobbler;a=commitdiff;h=eafedd4d33a63107f9a382ab613b5e5d2f18f98d;hp=b4ccb85ecaa9da26203baa3f27e90f177fd677f0 This seems to work for me, please let me know if it doesn't for you. --Michael > > >> -----BEGIN PGP SIGNATURE----- >> Version: GnuPG v1.4.6 (GNU/Linux) >> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org >> >> iD8DBQFIPqPZxqnHThA4pEkRArMJAJ0Qf+9ybElm0uQApI32OPrYsayEnQCfZDmD >> vAGXUqbA1V8jz+nCDA0KVsY= >> =dtvu >> -----END PGP SIGNATURE----- >> >> _______________________________________________ >> 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 josir at globo.com Thu May 29 16:54:54 2008 From: josir at globo.com (Josir Gomes) Date: Thu, 29 May 2008 13:54:54 -0300 Subject: [et-mgmt-tools] New happy user / Differences between virsh and virt-manager Message-ID: <483EDFDE.8000008@globo.com> Hi folks, I am new on this virtualization business and I need some tips from you. I've already acomplished to install and run a Ubuntu 0.4 guest under a Ubuntu 8.04 Host (both 64bits) using the above tutorials: https://help.ubuntu.com/community/KVM https://help.ubuntu.com/8.04/serverguide/C/libvirt.html I can ssh from other machines in the vm and I managed to add new drives too using only virt-manager. Everything works fine and the vm is running very well. As I expected the qemu/kvm performance is far better than vmware. My problem: when I try to monitor my vms using the virsh prompt virsh --connect qemu:///system >>list I got no vm status. virsh isn't only another way to look at the vm info ? Or am I having a concept misunderstanding ? Where can I find more info on this and learn how to clone vms? I tried wiki (http://virt-manager.et.redhat.com/page/Main_Page) but I couldn't get any info. Thanks in advance, Josir Gomes From berrange at redhat.com Thu May 29 17:55:39 2008 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 29 May 2008 18:55:39 +0100 Subject: [et-mgmt-tools] New happy user / Differences between virsh and virt-manager In-Reply-To: <483EDFDE.8000008@globo.com> References: <483EDFDE.8000008@globo.com> Message-ID: <20080529175537.GD2433@redhat.com> On Thu, May 29, 2008 at 01:54:54PM -0300, Josir Gomes wrote: > Hi folks, > > I am new on this virtualization business and I need some tips from you. > > I've already acomplished to install and run a Ubuntu 0.4 guest under a > Ubuntu 8.04 Host (both 64bits) using the above tutorials: > > https://help.ubuntu.com/community/KVM > https://help.ubuntu.com/8.04/serverguide/C/libvirt.html > > I can ssh from other machines in the vm and I managed to add new drives > too using only virt-manager. > Everything works fine and the vm is running very well. > As I expected the qemu/kvm performance is far better than vmware. > > My problem: when I try to monitor my vms using the virsh prompt > > virsh --connect qemu:///system > >>list > > I got no vm status. virsh isn't only another way to look at the vm info ? It only shows running VMs by default - try adding --inactive, or --all if you want idle VMs too. > Where can I find more info on this and learn how to clone vms? There is a virt-clone command - the man page has examples. Dan. -- |: Red Hat, Engineering, Boston -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| From robin-lists at robinbowes.com Thu May 29 20:16:54 2008 From: robin-lists at robinbowes.com (Robin Bowes) Date: Thu, 29 May 2008 21:16:54 +0100 Subject: [et-mgmt-tools] Re: koan 1.0 - RHEL 5.2 x86_64 - Full Virt In-Reply-To: <483E8182.6020006@redhat.com> References: <483E6F6B.3040205@ng23.net> <483E8182.6020006@redhat.com> Message-ID: Chris Lalancette wrote: > Robin Bowes wrote: >> This is the wrong list to ask on - try the xen list, or fedora-xen. >> >> That said, I'm running CentOS 5.1 with xen 3.1.x that I rebuilt from the >> Fedora RPMS. I believe the general feeling is that 32-bit guests are >> unstable on x86_64 Dom0. > > They shouldn't be; 32-bit fully virtualized guests have been fully supported on > 64-bit dom0 since 5.0. 5.2 has added 32-bit paravirtualized on 64-bit dom0 as a > tech preview, although it still has a couple of bugs to work out. If there are > problems with running 32-bit fully virtualized guests on 64-bit dom0, please > report them to BZ. /me needs to read more carefully - I missed the "full" bit. I am running paravirt, and see occasional crashes. R. From sakaia at jp.fujitsu.com Fri May 30 05:16:13 2008 From: sakaia at jp.fujitsu.com (Atsushi SAKAI) Date: Fri, 30 May 2008 14:16:13 +0900 Subject: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't connect to Xen- Host In-Reply-To: References: Message-ID: <200805300516.m4U5GEvw023603@fjmscan502.ms.jp.fujitsu.com> Hello, You should put on x509 files to that path. Thanks Atsushi SAKAI Feichtinger G$B > Hello Richard, > > I intstalled the ocaml-libvirt-0.4.0.1.exe on MS- Vista without problems. > Also the virt-ctrl.exe starts without problems. > But when I try to connect to a Xen-Host I get the messages: > libvir: Remote error : Cannot access CA certificate 'C:/msys/1.0/local/etc/pki/C > A/cacert.pem': No such file or directory (2) > > Please be so kind and help. > > > Regard, > G$B > > > > > > > * From: "Richard W.M. Jones" > * To: Fedora/Linux Management Tools > * Subject: [et-mgmt-tools] Fwd: OCaml bindings - Windows installer > * Date: Tue, 08 Jan 2008 11:18:39 +0000 > > Might interest some people on this list ... > > --- Begin Message --- > > * From: "Richard W.M. Jones" > * To: libvir-list > * Subject: [Libvir] OCaml bindings - Windows installer > * Date: Tue, 08 Jan 2008 11:15:19 +0000 > > This is a Windows installer for the OCaml libvirt bindings and programs: > > http://libvirt.org/sources/ocaml/ocaml-libvirt-0.4.0.1.exe > > If someone has a 'virgin' Windows system and could test that the above installer works, particularly for Windows < Vista and for Windows which has not had any GTK/MinGW development tools installed. About 90% of the size of the installer is down to the GTK DLLs and configuration files which need to be carried along to support the GTK graphical program (Virt Control). It's possible that I haven't placed the GTK files in exactly the right place, so instead it is using my own GTK development environment. > > Screenshots showing the installer in action: > > http://annexia.org/tmp/wininstaller-1-desktop.png > > Desktop, nothing installed, showing the installer icon. > > http://annexia.org/tmp/wininstaller-2-packages.png > > List of required, recommended and optional subpackages. > > http://annexia.org/tmp/wininstaller-3-instpath.png > > Select the installation path. > > http://annexia.org/tmp/wininstaller-4-installdone.png > > Install complete. Notice the desktop icons, ... > > http://annexia.org/tmp/wininstaller-5-menu.png > > ... and the menu entries and uninstaller program. > > http://annexia.org/tmp/wininstaller-6-virtctrl.png > > Running Virt Control. > > http://annexia.org/tmp/wininstaller-7-uninstall.png > > After uninstall. Notice that the desktop icons have gone. > > > The installer uses NSIS (the Nullsoft Scriptable Install System, http://nsis.sf.net/) and is reasonably well integrated into the autoconf / make build system. In particular if you are compiling under Windows and NSIS is installed, then a simple './configure --with-nsis; make wininstaller' will build a Windows installer. > > This is something to look at if libvirt wants/needs a Windows installer. > > 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 > > Attachment: smime.p7s > Description: S/MIME Cryptographic Signature > > -- > Libvir-list mailing list > Libvir-list redhat com > https://www.redhat.com/mailman/listinfo/libvir-list > > > --- End Message --- > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From Feichtinger.Guenter at bartelt.at Fri May 30 06:54:03 2008 From: Feichtinger.Guenter at bartelt.at (=?iso-8859-1?Q?Feichtinger_G=FCnter?=) Date: Fri, 30 May 2008 08:54:03 +0200 Subject: AW: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't connect to Xen- Host In-Reply-To: <200805300516.m4U5GEvw023603@fjmscan502.ms.jp.fujitsu.com> Message-ID: > Von: et-mgmt-tools-bounces at redhat.com > [mailto:et-mgmt-tools-bounces at redhat.com] Im Auftrag von Atsushi SAKAI > Gesendet: Freitag, 30. Mai 2008 07:16 > An: Fedora/Linux Management Tools > Betreff: Re: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't > connect to Xen- Host > > Hello, > > You should put on x509 files to that path. I'm sorry, I don't know what x509 files means. Thanks, Guenter > > Thanks > Atsushi SAKAI > > Feichtinger G > > > Hello Richard, > > > > I intstalled the ocaml-libvirt-0.4.0.1.exe on MS- Vista > without problems. > > Also the virt-ctrl.exe starts without problems. > > But when I try to connect to a Xen-Host I get the messages: > > libvir: Remote error : Cannot access CA certificate > > 'C:/msys/1.0/local/etc/pki/C > > A/cacert.pem': No such file or directory (2) > > > > Please be so kind and help. > > > > > > Regard, > > G > > > > > > > > > > > > > > * From: "Richard W.M. Jones" > > * To: Fedora/Linux Management Tools > > * Subject: [et-mgmt-tools] Fwd: OCaml bindings - > Windows installer > > * Date: Tue, 08 Jan 2008 11:18:39 +0000 > > > > Might interest some people on this list ... > > > > --- Begin Message --- > > > > * From: "Richard W.M. Jones" > > * To: libvir-list > > * Subject: [Libvir] OCaml bindings - Windows installer > > * Date: Tue, 08 Jan 2008 11:15:19 +0000 > > > > This is a Windows installer for the OCaml libvirt > bindings and programs: > > > > http://libvirt.org/sources/ocaml/ocaml-libvirt-0.4.0.1.exe > > > > If someone has a 'virgin' Windows system and could test > that the above installer works, particularly for Windows < > Vista and for Windows which has not had any GTK/MinGW > development tools installed. About 90% of the size of the > installer is down to the GTK DLLs and configuration files > which need to be carried along to support the GTK graphical > program (Virt Control). It's possible that I haven't placed > the GTK files in exactly the right place, so instead it is > using my own GTK development environment. > > > > Screenshots showing the installer in action: > > > > http://annexia.org/tmp/wininstaller-1-desktop.png > > > > Desktop, nothing installed, showing the installer icon. > > > > http://annexia.org/tmp/wininstaller-2-packages.png > > > > List of required, recommended and optional subpackages. > > > > http://annexia.org/tmp/wininstaller-3-instpath.png > > > > Select the installation path. > > > > http://annexia.org/tmp/wininstaller-4-installdone.png > > > > Install complete. Notice the desktop icons, ... > > > > http://annexia.org/tmp/wininstaller-5-menu.png > > > > ... and the menu entries and uninstaller program. > > > > http://annexia.org/tmp/wininstaller-6-virtctrl.png > > > > Running Virt Control. > > > > http://annexia.org/tmp/wininstaller-7-uninstall.png > > > > After uninstall. Notice that the desktop icons have gone. > > > > > > The installer uses NSIS (the Nullsoft Scriptable > Install System, http://nsis.sf.net/) and is reasonably well > integrated into the autoconf / make build system. In > particular if you are compiling under Windows and NSIS is > installed, then a simple './configure --with-nsis; make > wininstaller' will build a Windows installer. > > > > This is something to look at if libvirt wants/needs a > Windows installer. > > > > 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 > > > > Attachment: smime.p7s > > Description: S/MIME Cryptographic Signature > > > > -- > > Libvir-list mailing list > > Libvir-list redhat com > > https://www.redhat.com/mailman/listinfo/libvir-list > > > > > > --- End Message --- > > > > > > > > > > _______________________________________________ > > 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 Pablo.Iranzo at redhat.com Fri May 30 07:13:39 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Fri, 30 May 2008 09:13:39 +0200 Subject: AW: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't connect to Xen- Host In-Reply-To: References: Message-ID: <1212131619.16221.39.camel@iranzo.users.redhat.com> The certificates El vie, 30-05-2008 a las 08:54 +0200, Feichtinger G?nter escribi?: > > Von: et-mgmt-tools-bounces at redhat.com > > [mailto:et-mgmt-tools-bounces at redhat.com] Im Auftrag von Atsushi SAKAI > > Gesendet: Freitag, 30. Mai 2008 07:16 > > An: Fedora/Linux Management Tools > > Betreff: Re: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't > > connect to Xen- Host > > > > Hello, > > > > You should put on x509 files to that path. > > I'm sorry, I don't know what x509 files means. > > Thanks, > Guenter > > > > > Thanks > > Atsushi SAKAI > > > > Feichtinger G > > > > > Hello Richard, > > > > > > I intstalled the ocaml-libvirt-0.4.0.1.exe on MS- Vista > > without problems. > > > Also the virt-ctrl.exe starts without problems. > > > But when I try to connect to a Xen-Host I get the messages: > > > libvir: Remote error : Cannot access CA certificate > > > 'C:/msys/1.0/local/etc/pki/C > > > A/cacert.pem': No such file or directory (2) > > > > > > Please be so kind and help. > > > > > > > > > Regard, > > > G > > > > > > > > > > > > > > > > > > > > > * From: "Richard W.M. Jones" > > > * To: Fedora/Linux Management Tools > > > * Subject: [et-mgmt-tools] Fwd: OCaml bindings - > > Windows installer > > > * Date: Tue, 08 Jan 2008 11:18:39 +0000 > > > > > > Might interest some people on this list ... > > > > > > --- Begin Message --- > > > > > > * From: "Richard W.M. Jones" > > > * To: libvir-list > > > * Subject: [Libvir] OCaml bindings - Windows installer > > > * Date: Tue, 08 Jan 2008 11:15:19 +0000 > > > > > > This is a Windows installer for the OCaml libvirt > > bindings and programs: > > > > > > http://libvirt.org/sources/ocaml/ocaml-libvirt-0.4.0.1.exe > > > > > > If someone has a 'virgin' Windows system and could test > > that the above installer works, particularly for Windows < > > Vista and for Windows which has not had any GTK/MinGW > > development tools installed. About 90% of the size of the > > installer is down to the GTK DLLs and configuration files > > which need to be carried along to support the GTK graphical > > program (Virt Control). It's possible that I haven't placed > > the GTK files in exactly the right place, so instead it is > > using my own GTK development environment. > > > > > > Screenshots showing the installer in action: > > > > > > http://annexia.org/tmp/wininstaller-1-desktop.png > > > > > > Desktop, nothing installed, showing the installer icon. > > > > > > http://annexia.org/tmp/wininstaller-2-packages.png > > > > > > List of required, recommended and optional subpackages. > > > > > > http://annexia.org/tmp/wininstaller-3-instpath.png > > > > > > Select the installation path. > > > > > > http://annexia.org/tmp/wininstaller-4-installdone.png > > > > > > Install complete. Notice the desktop icons, ... > > > > > > http://annexia.org/tmp/wininstaller-5-menu.png > > > > > > ... and the menu entries and uninstaller program. > > > > > > http://annexia.org/tmp/wininstaller-6-virtctrl.png > > > > > > Running Virt Control. > > > > > > http://annexia.org/tmp/wininstaller-7-uninstall.png > > > > > > After uninstall. Notice that the desktop icons have gone. > > > > > > > > > The installer uses NSIS (the Nullsoft Scriptable > > Install System, http://nsis.sf.net/) and is reasonably well > > integrated into the autoconf / make build system. In > > particular if you are compiling under Windows and NSIS is > > installed, then a simple './configure --with-nsis; make > > wininstaller' will build a Windows installer. > > > > > > This is something to look at if libvirt wants/needs a > > Windows installer. > > > > > > 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 > > > > > > Attachment: smime.p7s > > > Description: S/MIME Cryptographic Signature > > > > > > -- > > > Libvir-list mailing list > > > Libvir-list redhat com > > > https://www.redhat.com/mailman/listinfo/libvir-list > > > > > > > > > --- End Message --- > > > > > > > > > > > > > > > _______________________________________________ > > > 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 > > > > _______________________________________________ > 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 sakaia at jp.fujitsu.com Fri May 30 07:42:22 2008 From: sakaia at jp.fujitsu.com (Atsushi SAKAI) Date: Fri, 30 May 2008 16:42:22 +0900 Subject: AW: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't connect to Xen- Host In-Reply-To: References: Message-ID: <200805300742.m4U7gNsj011738@fjmscan501.ms.jp.fujitsu.com> Hi, Guenter Please see following page. http://libvirt.org/remote.html The section of "Generating TLS certificates" answers your question. P.S. Now I realized. ocaml-libvirt should be discussed on libvirt ML. https://www.redhat.com/mailman/listinfo/libvir-list Thanks Atsushi SAKAI Feichtinger G$B > > > Von: et-mgmt-tools-bounces at redhat.com > > [mailto:et-mgmt-tools-bounces at redhat.com] Im Auftrag von Atsushi SAKAI > > Gesendet: Freitag, 30. Mai 2008 07:16 > > An: Fedora/Linux Management Tools > > Betreff: Re: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't > > connect to Xen- Host > > > > Hello, > > > > You should put on x509 files to that path. > > I'm sorry, I don't know what x509 files means. > > Thanks, > Guenter > > > > > Thanks > > Atsushi SAKAI > > > > Feichtinger G > > > > > Hello Richard, > > > > > > I intstalled the ocaml-libvirt-0.4.0.1.exe on MS- Vista > > without problems. > > > Also the virt-ctrl.exe starts without problems. > > > But when I try to connect to a Xen-Host I get the messages: > > > libvir: Remote error : Cannot access CA certificate > > > 'C:/msys/1.0/local/etc/pki/C > > > A/cacert.pem': No such file or directory (2) > > > > > > Please be so kind and help. > > > > > > > > > Regard, > > > G > > > > > > > > > > > > > > > > > > > > > * From: "Richard W.M. Jones" > > > * To: Fedora/Linux Management Tools > > > * Subject: [et-mgmt-tools] Fwd: OCaml bindings - > > Windows installer > > > * Date: Tue, 08 Jan 2008 11:18:39 +0000 > > > > > > Might interest some people on this list ... > > > > > > --- Begin Message --- > > > > > > * From: "Richard W.M. Jones" > > > * To: libvir-list > > > * Subject: [Libvir] OCaml bindings - Windows installer > > > * Date: Tue, 08 Jan 2008 11:15:19 +0000 > > > > > > This is a Windows installer for the OCaml libvirt > > bindings and programs: > > > > > > http://libvirt.org/sources/ocaml/ocaml-libvirt-0.4.0.1.exe > > > > > > If someone has a 'virgin' Windows system and could test > > that the above installer works, particularly for Windows < > > Vista and for Windows which has not had any GTK/MinGW > > development tools installed. About 90% of the size of the > > installer is down to the GTK DLLs and configuration files > > which need to be carried along to support the GTK graphical > > program (Virt Control). It's possible that I haven't placed > > the GTK files in exactly the right place, so instead it is > > using my own GTK development environment. > > > > > > Screenshots showing the installer in action: > > > > > > http://annexia.org/tmp/wininstaller-1-desktop.png > > > > > > Desktop, nothing installed, showing the installer icon. > > > > > > http://annexia.org/tmp/wininstaller-2-packages.png > > > > > > List of required, recommended and optional subpackages. > > > > > > http://annexia.org/tmp/wininstaller-3-instpath.png > > > > > > Select the installation path. > > > > > > http://annexia.org/tmp/wininstaller-4-installdone.png > > > > > > Install complete. Notice the desktop icons, ... > > > > > > http://annexia.org/tmp/wininstaller-5-menu.png > > > > > > ... and the menu entries and uninstaller program. > > > > > > http://annexia.org/tmp/wininstaller-6-virtctrl.png > > > > > > Running Virt Control. > > > > > > http://annexia.org/tmp/wininstaller-7-uninstall.png > > > > > > After uninstall. Notice that the desktop icons have gone. > > > > > > > > > The installer uses NSIS (the Nullsoft Scriptable > > Install System, http://nsis.sf.net/) and is reasonably well > > integrated into the autoconf / make build system. In > > particular if you are compiling under Windows and NSIS is > > installed, then a simple './configure --with-nsis; make > > wininstaller' will build a Windows installer. > > > > > > This is something to look at if libvirt wants/needs a > > Windows installer. > > > > > > 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 > > > > > > Attachment: smime.p7s > > > Description: S/MIME Cryptographic Signature > > > > > > -- > > > Libvir-list mailing list > > > Libvir-list redhat com > > > https://www.redhat.com/mailman/listinfo/libvir-list > > > > > > > > > --- End Message --- > > > > > > > > > > > > > > > _______________________________________________ > > > 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 > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From Feichtinger.Guenter at bartelt.at Fri May 30 08:04:03 2008 From: Feichtinger.Guenter at bartelt.at (=?iso-8859-1?Q?Feichtinger_G=FCnter?=) Date: Fri, 30 May 2008 10:04:03 +0200 Subject: AW: AW: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't connect to Xen-Host In-Reply-To: <200805300742.m4U7gNsj011738@fjmscan501.ms.jp.fujitsu.com> Message-ID: > Von: et-mgmt-tools-bounces at redhat.com > [mailto:et-mgmt-tools-bounces at redhat.com] Im Auftrag von Atsushi SAKAI > Gesendet: Freitag, 30. Mai 2008 09:42 > An: Fedora/Linux Management Tools > Betreff: Re: AW: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't > connect to Xen-Host > > Hi, Guenter > > Please see following page. > http://libvirt.org/remote.html > The section of "Generating TLS certificates" > answers your question. Is there any possibility to use it without certificates? > > P.S. > Now I realized. > ocaml-libvirt should be discussed on libvirt ML. > https://www.redhat.com/mailman/listinfo/libvir-list Okay, I change to the libvirt ML... Thanks, Guenter > > Thanks > Atsushi SAKAI > > > Feichtinger G > > > > > > Von: et-mgmt-tools-bounces at redhat.com > > > [mailto:et-mgmt-tools-bounces at redhat.com] Im Auftrag von Atsushi > > > SAKAI > > > Gesendet: Freitag, 30. Mai 2008 07:16 > > > An: Fedora/Linux Management Tools > > > Betreff: Re: [et-mgmt-tools] ocaml-libvirt-0.4.0.1: Can't > connect to > > > Xen- Host > > > > > > Hello, > > > > > > You should put on x509 files to that path. > > > > I'm sorry, I don't know what x509 files means. > > > > Thanks, > > Guenter > > > > > > > > Thanks > > > Atsushi SAKAI > > > > > > Feichtinger G > > > > > > > Hello Richard, > > > > > > > > I intstalled the ocaml-libvirt-0.4.0.1.exe on MS- Vista > > > without problems. > > > > Also the virt-ctrl.exe starts without problems. > > > > But when I try to connect to a Xen-Host I get the messages: > > > > libvir: Remote error : Cannot access CA certificate > > > > 'C:/msys/1.0/local/etc/pki/C > > > > A/cacert.pem': No such file or directory (2) > > > > > > > > Please be so kind and help. > > > > > > > > > > > > Regard, > > > > G > > > > > > > > > > > > > > > > > > > > > > > > > > > > * From: "Richard W.M. Jones" > > > > * To: Fedora/Linux Management Tools redhat com> > > > > * Subject: [et-mgmt-tools] Fwd: OCaml bindings - > > > Windows installer > > > > * Date: Tue, 08 Jan 2008 11:18:39 +0000 > > > > > > > > Might interest some people on this list ... > > > > > > > > --- Begin Message --- > > > > > > > > * From: "Richard W.M. Jones" > > > > * To: libvir-list > > > > * Subject: [Libvir] OCaml bindings - Windows installer > > > > * Date: Tue, 08 Jan 2008 11:15:19 +0000 > > > > > > > > This is a Windows installer for the OCaml libvirt > > > bindings and programs: > > > > > > > > http://libvirt.org/sources/ocaml/ocaml-libvirt-0.4.0.1.exe > > > > > > > > If someone has a 'virgin' Windows system and could test > > > that the above installer works, particularly for Windows > < Vista and > > > for Windows which has not had any GTK/MinGW development tools > > > installed. About 90% of the size of the installer is down > to the GTK > > > DLLs and configuration files which need to be carried along to > > > support the GTK graphical program (Virt Control). It's > possible that > > > I haven't placed the GTK files in exactly the right place, so > > > instead it is using my own GTK development environment. > > > > > > > > Screenshots showing the installer in action: > > > > > > > > http://annexia.org/tmp/wininstaller-1-desktop.png > > > > > > > > Desktop, nothing installed, showing the installer icon. > > > > > > > > http://annexia.org/tmp/wininstaller-2-packages.png > > > > > > > > List of required, recommended and optional subpackages. > > > > > > > > http://annexia.org/tmp/wininstaller-3-instpath.png > > > > > > > > Select the installation path. > > > > > > > > http://annexia.org/tmp/wininstaller-4-installdone.png > > > > > > > > Install complete. Notice the desktop icons, ... > > > > > > > > http://annexia.org/tmp/wininstaller-5-menu.png > > > > > > > > ... and the menu entries and uninstaller program. > > > > > > > > http://annexia.org/tmp/wininstaller-6-virtctrl.png > > > > > > > > Running Virt Control. > > > > > > > > http://annexia.org/tmp/wininstaller-7-uninstall.png > > > > > > > > After uninstall. Notice that the desktop icons have gone. > > > > > > > > > > > > The installer uses NSIS (the Nullsoft Scriptable > > > Install System, http://nsis.sf.net/) and is reasonably well > > > integrated into the autoconf / make build system. In > particular if > > > you are compiling under Windows and NSIS is installed, > then a simple > > > './configure --with-nsis; make wininstaller' will build a Windows > > > installer. > > > > > > > > This is something to look at if libvirt wants/needs a > > > Windows installer. > > > > > > > > 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 > > > > > > > > Attachment: smime.p7s > > > > Description: S/MIME Cryptographic Signature > > > > > > > > -- > > > > Libvir-list mailing list > > > > Libvir-list redhat com > > > > https://www.redhat.com/mailman/listinfo/libvir-list > > > > > > > > > > > > --- End Message --- > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > > > 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 > > > > > > > _______________________________________________ > > 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 rainer at ultra-secure.de Fri May 30 14:41:55 2008 From: rainer at ultra-secure.de (Rainer Duffner) Date: Fri, 30 May 2008 16:41:55 +0200 Subject: [et-mgmt-tools] cobbler upgrade 0.6.something to 1.0.2 on RHEL5 Message-ID: <48401233.80204@ultra-secure.de> Hi, did the upgrade: now, what does that mean? [root at zeus1 noarch]# cobbler check exceptions.IOError [Errno 21] Is a directory File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 65, in main return BootCLI().run(sys.argv) File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 44, 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 53, in run status = self.api.check() File "/usr/lib/python2.4/site-packages/cobbler/api.py", line 303, in check return check.run() File "/usr/lib/python2.4/site-packages/cobbler/action_check.py", line 68, in run self.check_for_default_password(status) File "/usr/lib/python2.4/site-packages/cobbler/action_check.py", line 130, in check_for_default_password fd = open(t) Thanks in advance, Rainer From hyclak at math.ohiou.edu Fri May 30 14:54:38 2008 From: hyclak at math.ohiou.edu (Matt Hyclak) Date: Fri, 30 May 2008 10:54:38 -0400 Subject: [et-mgmt-tools] cobbler upgrade 0.6.something to 1.0.2 on RHEL5 In-Reply-To: <48401233.80204@ultra-secure.de> References: <48401233.80204@ultra-secure.de> Message-ID: <20080530145438.GI7087@math.ohiou.edu> On Fri, May 30, 2008 at 04:41:55PM +0200, Rainer Duffner enlightened us: > did the upgrade: > > now, what does that mean? > > [root at zeus1 noarch]# cobbler check > exceptions.IOError > [Errno 21] Is a directory > File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 65, > in main > return BootCLI().run(sys.argv) > File "/usr/lib/python2.4/site-packages/cobbler/cobbler.py", line 44, > 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 53, in run > status = self.api.check() > File "/usr/lib/python2.4/site-packages/cobbler/api.py", line 303, in > check > return check.run() > File "/usr/lib/python2.4/site-packages/cobbler/action_check.py", line > 68, in run > self.check_for_default_password(status) > File "/usr/lib/python2.4/site-packages/cobbler/action_check.py", line > 130, in check_for_default_password > fd = open(t) > > > Did you follow the upgrade instructions? https://fedorahosted.org/cobbler/wiki/WhatsNewInOhNine Matt -- Matt Hyclak Department of Mathematics Department of Social Work Ohio University (740) 593-1263 From rainer at ultra-secure.de Fri May 30 15:01:35 2008 From: rainer at ultra-secure.de (Rainer Duffner) Date: Fri, 30 May 2008 17:01:35 +0200 Subject: [et-mgmt-tools] cobbler upgrade 0.6.something to 1.0.2 on RHEL5 In-Reply-To: <20080530145438.GI7087@math.ohiou.edu> References: <48401233.80204@ultra-secure.de> <20080530145438.GI7087@math.ohiou.edu> Message-ID: <484016CF.5070407@ultra-secure.de> Matt Hyclak schrieb: > On Fri, May 30, 2008 at 04:41:55PM +0200, Rainer Duffner enlightened us: > >> > > Did you follow the upgrade instructions? > > https://fedorahosted.org/cobbler/wiki/WhatsNewInOhNine > > Well, I thought so. I'd really like to know what file he wants to read which is a directory on my system. If the error-message is actually representative of the error... Rainer From mdehaan at redhat.com Fri May 30 15:17:58 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 30 May 2008 11:17:58 -0400 Subject: [et-mgmt-tools] cobbler upgrade 0.6.something to 1.0.2 on RHEL5 In-Reply-To: <484016CF.5070407@ultra-secure.de> References: <48401233.80204@ultra-secure.de> <20080530145438.GI7087@math.ohiou.edu> <484016CF.5070407@ultra-secure.de> Message-ID: <48401AA6.8040201@redhat.com> Rainer Duffner wrote: > Matt Hyclak schrieb: >> On Fri, May 30, 2008 at 04:41:55PM +0200, Rainer Duffner enlightened us: >> >>> >> >> Did you follow the upgrade instructions? >> >> https://fedorahosted.org/cobbler/wiki/WhatsNewInOhNine >> > > Well, I thought so. > I'd really like to know what file he wants to read which is a > directory on my system. > If the error-message is actually representative of the error... > > > Rainer > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Cobbler check is being dumb and doesn't know that all the files in /var/lib/cobbler/kickstarts are not files. I suppose you have a directory under version control and it's made a subdirectory in there. I already have a fix for this in git and it will be in 1.0.1 From rainer at ultra-secure.de Fri May 30 15:24:51 2008 From: rainer at ultra-secure.de (Rainer Duffner) Date: Fri, 30 May 2008 17:24:51 +0200 Subject: [et-mgmt-tools] cobbler upgrade 0.6.something to 1.0.2 on RHEL5 In-Reply-To: <48401AA6.8040201@redhat.com> References: <48401233.80204@ultra-secure.de> <20080530145438.GI7087@math.ohiou.edu> <484016CF.5070407@ultra-secure.de> <48401AA6.8040201@redhat.com> Message-ID: <48401C43.4040007@ultra-secure.de> Michael DeHaan schrieb: > > Cobbler check is being dumb and doesn't know that all the files in > /var/lib/cobbler/kickstarts are not files. I suppose you have a > directory under version control and it's made a subdirectory in there. > > I already have a fix for this in git and it will be in 1.0.1 > Indeed: [root at zeus1 cobbler]# ll /var/lib/cobbler/kickstarts total 48 -rw-r--r-- 1 root root 1640 May 23 12:13 kickstart-centos5-two-disk-opt.ks -rw-r--r-- 1 root root 1508 Feb 22 09:52 kickstart-rhel4.4-cciss-vmware-server.ks -rw-r--r-- 1 root root 1537 Mar 14 12:18 kickstart-rhel5-cciss-big-opt.ks -rw-r--r-- 1 root root 1553 Feb 5 15:27 kickstart-rhel5-cciss-big-usr-local.ks -rw-r--r-- 1 root root 1537 Apr 15 09:36 kickstart-rhel5-cciss-big-var.ks -rw-r--r-- 1 root root 1412 Feb 5 15:27 kickstart-rhel5-generic-big-usr-local.ks -rw-r--r-- 1 root root 1417 Feb 25 17:05 kickstart-rhel5-generic-vz-cciss.ks -rw-r--r-- 1 root root 1396 Feb 25 15:51 kickstart-rhel5-generic-vz.ks -rw-r--r-- 1 root root 1414 Feb 5 15:27 kickstart-rhel5-mailserver.ks -rw-r--r-- 1 root root 1489 Feb 5 15:27 kickstart-rhel5-one-disk-hsphere.ks -rw-r--r-- 1 root root 1582 Feb 5 15:28 kickstart-rhel5-one-disk-hsphere-qmail.ks drwxr-xr-x 2 root root 4096 Apr 28 12:48 non-template Though the bit about version-control was slightly over-optimistic... Can I pull that fix from the web somehow? The cobbler should start working again on monday ;-) So, I need to fix it over the weekend ;-) cheers, Rainer From mdehaan at redhat.com Fri May 30 15:37:46 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 30 May 2008 11:37:46 -0400 Subject: [et-mgmt-tools] cobbler upgrade 0.6.something to 1.0.2 on RHEL5 In-Reply-To: <48401C43.4040007@ultra-secure.de> References: <48401233.80204@ultra-secure.de> <20080530145438.GI7087@math.ohiou.edu> <484016CF.5070407@ultra-secure.de> <48401AA6.8040201@redhat.com> <48401C43.4040007@ultra-secure.de> Message-ID: <48401F4A.7020900@redhat.com> Rainer Duffner wrote: > Michael DeHaan schrieb: >> >> Cobbler check is being dumb and doesn't know that all the files in >> /var/lib/cobbler/kickstarts are not files. I suppose you have a >> directory under version control and it's made a subdirectory in there. >> >> I already have a fix for this in git and it will be in 1.0.1 >> > > > Indeed: > > [root at zeus1 cobbler]# ll /var/lib/cobbler/kickstarts > total 48 > -rw-r--r-- 1 root root 1640 May 23 12:13 > kickstart-centos5-two-disk-opt.ks > -rw-r--r-- 1 root root 1508 Feb 22 09:52 > kickstart-rhel4.4-cciss-vmware-server.ks > -rw-r--r-- 1 root root 1537 Mar 14 12:18 kickstart-rhel5-cciss-big-opt.ks > -rw-r--r-- 1 root root 1553 Feb 5 15:27 > kickstart-rhel5-cciss-big-usr-local.ks > -rw-r--r-- 1 root root 1537 Apr 15 09:36 kickstart-rhel5-cciss-big-var.ks > -rw-r--r-- 1 root root 1412 Feb 5 15:27 > kickstart-rhel5-generic-big-usr-local.ks > -rw-r--r-- 1 root root 1417 Feb 25 17:05 > kickstart-rhel5-generic-vz-cciss.ks > -rw-r--r-- 1 root root 1396 Feb 25 15:51 kickstart-rhel5-generic-vz.ks > -rw-r--r-- 1 root root 1414 Feb 5 15:27 kickstart-rhel5-mailserver.ks > -rw-r--r-- 1 root root 1489 Feb 5 15:27 > kickstart-rhel5-one-disk-hsphere.ks > -rw-r--r-- 1 root root 1582 Feb 5 15:28 > kickstart-rhel5-one-disk-hsphere-qmail.ks > drwxr-xr-x 2 root root 4096 Apr 28 12:48 non-template > > > Though the bit about version-control was slightly over-optimistic... > > > Can I pull that fix from the web somehow? > The cobbler should start working again on monday ;-) > So, I need to fix it over the weekend ;-) https://fedorahosted.org/cobbler/wiki/QuickDevelRpmBuildGuide or: http://git.fedorahosted.org/git/?p=cobbler;a=blobdiff;f=cobbler/utils.py;h=8ddf6af2ea5d3fa4c4844acb2ca16d2c6fa62e11;hp=de00150c18ac5a027fa5f5d93bdcfed5c99fdedb;hb=a63789be503d43b6f4fa42794a18cfe9b8f2c586;hpb=230efa2eba9f6578cfbc7c3905c8c0295518c3e5 --Michael > > > cheers, > Rainer > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From rainer at ultra-secure.de Fri May 30 15:52:07 2008 From: rainer at ultra-secure.de (Rainer Duffner) Date: Fri, 30 May 2008 17:52:07 +0200 Subject: [et-mgmt-tools] cobbler upgrade 0.6.something to 1.0.2 on RHEL5 In-Reply-To: <48401F4A.7020900@redhat.com> References: <48401233.80204@ultra-secure.de> <20080530145438.GI7087@math.ohiou.edu> <484016CF.5070407@ultra-secure.de> <48401AA6.8040201@redhat.com> <48401C43.4040007@ultra-secure.de> <48401F4A.7020900@redhat.com> Message-ID: <484022A7.2050003@ultra-secure.de> Michael DeHaan schrieb: > > https://fedorahosted.org/cobbler/wiki/QuickDevelRpmBuildGuide > > or: > > http://git.fedorahosted.org/git/?p=cobbler;a=blobdiff;f=cobbler/utils.py;h=8ddf6af2ea5d3fa4c4844acb2ca16d2c6fa62e11;hp=de00150c18ac5a027fa5f5d93bdcfed5c99fdedb;hb=a63789be503d43b6f4fa42794a18cfe9b8f2c586;hpb=230efa2eba9f6578cfbc7c3905c8c0295518c3e5 > > > --Michael > OK, that fixed it. However, I now had to create a new user (for the digest-file). My old user doesn't seem to work anymore. Strange. Rainer From mdehaan at redhat.com Fri May 30 16:16:12 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 30 May 2008 12:16:12 -0400 Subject: [et-mgmt-tools] cobbler upgrade 0.6.something to 1.0.2 on RHEL5 In-Reply-To: <484022A7.2050003@ultra-secure.de> References: <48401233.80204@ultra-secure.de> <20080530145438.GI7087@math.ohiou.edu> <484016CF.5070407@ultra-secure.de> <48401AA6.8040201@redhat.com> <48401C43.4040007@ultra-secure.de> <48401F4A.7020900@redhat.com> <484022A7.2050003@ultra-secure.de> Message-ID: <4840284C.3090207@redhat.com> Rainer Duffner wrote: > Michael DeHaan schrieb: >> >> https://fedorahosted.org/cobbler/wiki/QuickDevelRpmBuildGuide >> >> or: >> >> http://git.fedorahosted.org/git/?p=cobbler;a=blobdiff;f=cobbler/utils.py;h=8ddf6af2ea5d3fa4c4844acb2ca16d2c6fa62e11;hp=de00150c18ac5a027fa5f5d93bdcfed5c99fdedb;hb=a63789be503d43b6f4fa42794a18cfe9b8f2c586;hpb=230efa2eba9f6578cfbc7c3905c8c0295518c3e5 >> >> >> --Michael >> > > > OK, that fixed it. > However, I now had to create a new user (for the digest-file). > My old user doesn't seem to work anymore. > > Strange. > Not strange. The Webapp was pretty new in 0.6, and security has been streamlined a lot in 0.8/1.0 Go through these instructions again as if you haven't before: https://fedorahosted.org/cobbler/wiki/CobblerWebInterface From SealeM at ag.state.la.us Fri May 30 18:43:48 2008 From: SealeM at ag.state.la.us (Seale, Mark) Date: Fri, 30 May 2008 13:43:48 -0500 Subject: [et-mgmt-tools] Virt-Install options Message-ID: <1E61D08F64B6BB4AB0EC91C72FBDB81613AB3740@exchange.ag.state.la.us> Would anyone happen to know if there is a method to specify different NIC models on startup or install of a guest OS? I know Qemu and KVM are capable of it with the '-net nic,model=', but I wasn't sure if there is a way to do this when using virt-install. Every OpenBSD VM I make is giving watchdog timeout errors for the NIC (re0)... Thanks for any help! The information contained in this transmission may contain privileged and confidential information. It is intended only for the use of the person(s) named above. If you are not the intended recipient, you are hereby notified that any review, dissemination, distribution or duplication of this communication is strictly prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. To reply to our e-mail administrator directly, please send an e-mail to postmaster at ag.state.la.us -------------- next part -------------- An HTML attachment was scrubbed... URL: From jboggs at redhat.com Fri May 30 19:43:14 2008 From: jboggs at redhat.com (Joey Boggs) Date: Fri, 30 May 2008 15:43:14 -0400 Subject: [et-mgmt-tools] [PATCH] virt-inst Package an image for VMware distribution Message-ID: <484058D2.6030006@redhat.com> These patches were originally written by David Lutterkort and I've tweaked the conversion process a little bit to work out some staging issues virt-pack will tar all the files for the image into a tarball. The name and toplevel directory are derived from the name and version of the image descriptor. Scratch disks are omitted, since they can be rebuilt when the image is deployed. A VMware format configuration file is produced with vmdk descriptors pointing at the disk images rather than converting the raw images. For scratch disks, sparse VMDK files are generated, since VMWare has no notion of scratch disks. The generated VMX file is very conservative, so that it's widely usable, but there are many things that could be improved in it if we had more information about the image (one example: are vmware-tools installed ?) The image.xml file must contain two new elements: version and release example-image Example running: [jboggs at localhost RHX_Sugar]$ sudo ./virt-pack i1.xml Checking /home/jboggs/RHX_Sugar/rhx_sugar.img Writing /home/jboggs/RHX_Sugar/../example-image-1.tgz example-image-1/ example-image-1/i1.xml example-image-1/rhx_sugar.vmdk example-image-1/rhx_sugar.img example-image-1/example-image.vmx When completed you will have an virtual instance capable of being run under VMWare or virt-image. Prior to conversion the image must contain VMWare disk controller drivers enabling it to boot, otherwise will result in a kernel panic: no init found. -------------- next part -------------- A non-text attachment was scrubbed... Name: example-image.xml Type: text/xml Size: 935 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: abspath.patch Type: text/x-patch Size: 1257 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: drives.patch Type: text/x-patch Size: 1805 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: filename.patch Type: text/x-patch Size: 2737 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: package.patch Type: text/x-patch Size: 16625 bytes Desc: not available URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: series URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: version.patch Type: text/x-patch Size: 2029 bytes Desc: not available URL: From ole.ersoy at gmail.com Fri May 30 20:13:03 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Fri, 30 May 2008 15:13:03 -0500 Subject: [et-mgmt-tools] Changing the name of an imported distribution Message-ID: <48405FCF.1080402@gmail.com> Hi, I imported Fedora 9 and gave it a really long name. I'd like to change it from: fedora_reallyreallylong_i386 >> fedora9_i386 Is that possible? Thanks, - Ole From mdehaan at redhat.com Fri May 30 20:17:30 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 30 May 2008 16:17:30 -0400 Subject: [et-mgmt-tools] Changing the name of an imported distribution In-Reply-To: <48405FCF.1080402@gmail.com> References: <48405FCF.1080402@gmail.com> Message-ID: <484060DA.8040703@redhat.com> Ole Ersoy wrote: > Hi, > > I imported Fedora 9 and gave it a really long name. I'd like to > change it from: > > fedora_reallyreallylong_i386 >> fedora9_i386 cobbler profile rename --name=fedora_reallyreallylong_i386 --newname=fedora9_i386 > > Is that possible? > > Thanks, > - Ole > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools From ole.ersoy at gmail.com Fri May 30 20:37:27 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Fri, 30 May 2008 15:37:27 -0500 Subject: [et-mgmt-tools] Changing the name of an imported distribution In-Reply-To: <484060DA.8040703@redhat.com> References: <48405FCF.1080402@gmail.com> <484060DA.8040703@redhat.com> Message-ID: <48406587.9050504@gmail.com> That was easy :-), Thanks, - Ole Michael DeHaan wrote: > Ole Ersoy wrote: >> Hi, >> >> I imported Fedora 9 and gave it a really long name. I'd like to >> change it from: >> >> fedora_reallyreallylong_i386 >> fedora9_i386 > > cobbler profile rename --name=fedora_reallyreallylong_i386 > --newname=fedora9_i386 > >> >> Is that possible? >> >> Thanks, >> - Ole >> >> _______________________________________________ >> 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 dmourati at gmail.com Fri May 30 21:36:01 2008 From: dmourati at gmail.com (Demetri Mouratis) Date: Fri, 30 May 2008 14:36:01 -0700 Subject: [et-mgmt-tools] CentOS 4 Cobbler Server Hosting CentOS 5 Repository Message-ID: <3cc2cbc40805301436g3fe78536tcd29f93bb207f2f4@mail.gmail.com> Hi, I'm running into some additional problems getting setup to support CentOS 5 from my cobber server running CentOS 4. When I PXE a new CentOS 5 host, I get some error messages like: Unable to read group information from repositories. This is a problem with the generation of your install tree. Looking a bit closer, I noticed that the repostiory structure seems to have changed between 4 and 5. # find centos4-i386 centos5-i386 -type d centos4-i386 centos4-i386/headers centos4-i386/isolinux centos4-i386/.olddata centos4-i386/NOTES centos4-i386/cache centos4-i386/images centos4-i386/images/xen centos4-i386/images/pxeboot centos4-i386/repodata centos4-i386/CentOS centos4-i386/CentOS/base centos4-i386/CentOS/RPMS centos5-i386 centos5-i386/isolinux centos5-i386/NOTES centos5-i386/cache centos5-i386/images centos5-i386/images/xen centos5-i386/images/pxeboot centos5-i386/repodata centos5-i386/CentOS centos5-i386/CentOS/repodata Under 5, there are no longer RPMS or base directories. What is going on here? Thanks. -D I think that the createrepo command under CentOS4 is having trouble doing the right thing. From smooge at gmail.com Fri May 30 21:39:29 2008 From: smooge at gmail.com (Stephen John Smoogen) Date: Fri, 30 May 2008 15:39:29 -0600 Subject: [et-mgmt-tools] CentOS 4 Cobbler Server Hosting CentOS 5 Repository In-Reply-To: <3cc2cbc40805301436g3fe78536tcd29f93bb207f2f4@mail.gmail.com> References: <3cc2cbc40805301436g3fe78536tcd29f93bb207f2f4@mail.gmail.com> Message-ID: <80d7e4090805301439u658ae9a4xa2ee1e18170ab897@mail.gmail.com> On Fri, May 30, 2008 at 3:36 PM, Demetri Mouratis wrote: > Hi, > > I'm running into some additional problems getting setup to support > CentOS 5 from my cobber server running CentOS 4. When I PXE a new > CentOS 5 host, I get some error messages like: > > Unable to read group information from repositories. This > is a problem with the generation of your install tree. > > Looking a bit closer, I noticed that the repostiory structure seems to > have changed between 4 and 5. > > # find centos4-i386 centos5-i386 -type d > centos4-i386 > centos4-i386/headers > centos4-i386/isolinux > centos4-i386/.olddata > centos4-i386/NOTES > centos4-i386/cache > centos4-i386/images > centos4-i386/images/xen > centos4-i386/images/pxeboot > centos4-i386/repodata > centos4-i386/CentOS > centos4-i386/CentOS/base > centos4-i386/CentOS/RPMS > centos5-i386 > centos5-i386/isolinux > centos5-i386/NOTES > centos5-i386/cache > centos5-i386/images > centos5-i386/images/xen > centos5-i386/images/pxeboot > centos5-i386/repodata > centos5-i386/CentOS > centos5-i386/CentOS/repodata > > Under 5, there are no longer RPMS or base directories. > That is a change in how EL-4 and EL-5 layed out their packages. The older anaconda used oneway and the newer one did not. > What is going on here? > > Thanks. > > -D > > I think that the createrepo command under CentOS4 is having trouble > doing the right thing. Unknown at this time.. what version of createrepo are you using? -- Stephen J Smoogen. -- BSD/GNU/Linux How far that little candle throws his beams! So shines a good deed in a naughty world. = Shakespeare. "The Merchant of Venice" From dmourati at gmail.com Fri May 30 21:44:08 2008 From: dmourati at gmail.com (Demetri Mouratis) Date: Fri, 30 May 2008 14:44:08 -0700 Subject: [et-mgmt-tools] CentOS 4 Cobbler Server Hosting CentOS 5 Repository In-Reply-To: <80d7e4090805301439u658ae9a4xa2ee1e18170ab897@mail.gmail.com> References: <3cc2cbc40805301436g3fe78536tcd29f93bb207f2f4@mail.gmail.com> <80d7e4090805301439u658ae9a4xa2ee1e18170ab897@mail.gmail.com> Message-ID: <3cc2cbc40805301444m6cb134d8m57f4153efb5e3a2@mail.gmail.com> On Fri, May 30, 2008 at 2:39 PM, Stephen John Smoogen wrote: >> >> Under 5, there are no longer RPMS or base directories. >> > > That is a change in how EL-4 and EL-5 layed out their packages. The > older anaconda used oneway and the newer one did not. > > >> What is going on here? >> >> Thanks. >> >> -D >> >> I think that the createrepo command under CentOS4 is having trouble >> doing the right thing. > > Unknown at this time.. what version of createrepo are you using? > # rpm -q createrepo createrepo-0.4.4-2 From smooge at gmail.com Fri May 30 22:22:41 2008 From: smooge at gmail.com (Stephen John Smoogen) Date: Fri, 30 May 2008 16:22:41 -0600 Subject: [et-mgmt-tools] CentOS 4 Cobbler Server Hosting CentOS 5 Repository In-Reply-To: <3cc2cbc40805301444m6cb134d8m57f4153efb5e3a2@mail.gmail.com> References: <3cc2cbc40805301436g3fe78536tcd29f93bb207f2f4@mail.gmail.com> <80d7e4090805301439u658ae9a4xa2ee1e18170ab897@mail.gmail.com> <3cc2cbc40805301444m6cb134d8m57f4153efb5e3a2@mail.gmail.com> Message-ID: <80d7e4090805301522p432b320as8b9348f885303584@mail.gmail.com> On Fri, May 30, 2008 at 3:44 PM, Demetri Mouratis wrote: > On Fri, May 30, 2008 at 2:39 PM, Stephen John Smoogen wrote: >>> >>> Under 5, there are no longer RPMS or base directories. >>> >> >> That is a change in how EL-4 and EL-5 layed out their packages. The >> older anaconda used oneway and the newer one did not. >> >> >>> What is going on here? >>> >>> Thanks. >>> >>> -D >>> >>> I think that the createrepo command under CentOS4 is having trouble >>> doing the right thing. >> >> Unknown at this time.. what version of createrepo are you using? >> > > # rpm -q createrepo > createrepo-0.4.4-2 > Hmmm my brain thinks it remembers something about that. I have 0.4.11 on my EL-5 servers for some reason. Could you try the following rpm to see if it helps? http://smooge.fedorapeople.org/alpha/EL-4/createrepo-0.4.11-1.el4.noarch.rpm and let me know if that fixes the issue? -- Stephen J Smoogen. -- BSD/GNU/Linux How far that little candle throws his beams! So shines a good deed in a naughty world. = Shakespeare. "The Merchant of Venice" From ole.ersoy at gmail.com Fri May 30 22:41:02 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Fri, 30 May 2008 17:41:02 -0500 Subject: [et-mgmt-tools] dhcpctl_connect: connection refused In-Reply-To: <80d7e4090805301522p432b320as8b9348f885303584@mail.gmail.com> References: <3cc2cbc40805301436g3fe78536tcd29f93bb207f2f4@mail.gmail.com> <80d7e4090805301439u658ae9a4xa2ee1e18170ab897@mail.gmail.com> <3cc2cbc40805301444m6cb134d8m57f4153efb5e3a2@mail.gmail.com> <80d7e4090805301522p432b320as8b9348f885303584@mail.gmail.com> Message-ID: <4840827E.7010000@gmail.com> Hi, I added a profile and now when I attempt to sync I get: [root at localhost cobbler]# cobbler sync dhcpctl_connect: connection refused dhcpctl_connect: connection refused Any ideas? Running cobbler check shows this: [root at localhost cobbler]# cobbler check The following potential problems were detected: #0: One or more kickstart templates references default password 'cobbler' and should be changed for security reasons: /etc/cobbler/sample.ks, /etc/cobbler/legacy.ks, /etc/cobbler/sample_end.ks TIA, - Ole From dmourati at gmail.com Fri May 30 23:03:17 2008 From: dmourati at gmail.com (Demetri Mouratis) Date: Fri, 30 May 2008 16:03:17 -0700 Subject: [et-mgmt-tools] CentOS 4 Cobbler Server Hosting CentOS 5 Repository In-Reply-To: <80d7e4090805301522p432b320as8b9348f885303584@mail.gmail.com> References: <3cc2cbc40805301436g3fe78536tcd29f93bb207f2f4@mail.gmail.com> <80d7e4090805301439u658ae9a4xa2ee1e18170ab897@mail.gmail.com> <3cc2cbc40805301444m6cb134d8m57f4153efb5e3a2@mail.gmail.com> <80d7e4090805301522p432b320as8b9348f885303584@mail.gmail.com> Message-ID: <3cc2cbc40805301603j2d841861vcb965b0c235fe295@mail.gmail.com> On Fri, May 30, 2008 at 3:22 PM, Stephen John Smoogen wrote: > Hmmm my brain thinks it remembers something about that. I have 0.4.11 > on my EL-5 servers for some reason. Could you try the following rpm to > see if it helps? > > http://smooge.fedorapeople.org/alpha/EL-4/createrepo-0.4.11-1.el4.noarch.rpm > > and let me know if that fixes the issue? Hi Stephen, Thanks for your help. Same result here after installing the above createrepo RPM. "Unable to read group information from repositories. This is a problem with the generation of your install tree." The dialog box says simply "Reboot." At the next box, I get: | You have specified that the group 'virtualization' should | | be installed. This group does not exist. Would you like to | | continue or abort your installation? If I say continue, the install seems to proceed okay. At boot time, however, the system hangs at Mounting Other Filesystems. I suspect that the initial install was not clean and so I am chasing that down before moving on to the hung CentOS 5 boot. Thanks. -D From smooge at gmail.com Fri May 30 23:11:55 2008 From: smooge at gmail.com (Stephen John Smoogen) Date: Fri, 30 May 2008 17:11:55 -0600 Subject: [et-mgmt-tools] CentOS 4 Cobbler Server Hosting CentOS 5 Repository In-Reply-To: <3cc2cbc40805301603j2d841861vcb965b0c235fe295@mail.gmail.com> References: <3cc2cbc40805301436g3fe78536tcd29f93bb207f2f4@mail.gmail.com> <80d7e4090805301439u658ae9a4xa2ee1e18170ab897@mail.gmail.com> <3cc2cbc40805301444m6cb134d8m57f4153efb5e3a2@mail.gmail.com> <80d7e4090805301522p432b320as8b9348f885303584@mail.gmail.com> <3cc2cbc40805301603j2d841861vcb965b0c235fe295@mail.gmail.com> Message-ID: <80d7e4090805301611u66a1f342qb9ab0e34ad94e086@mail.gmail.com> On Fri, May 30, 2008 at 5:03 PM, Demetri Mouratis wrote: > On Fri, May 30, 2008 at 3:22 PM, Stephen John Smoogen wrote: > >> Hmmm my brain thinks it remembers something about that. I have 0.4.11 >> on my EL-5 servers for some reason. Could you try the following rpm to >> see if it helps? >> >> http://smooge.fedorapeople.org/alpha/EL-4/createrepo-0.4.11-1.el4.noarch.rpm >> >> and let me know if that fixes the issue? > > Hi Stephen, > > Thanks for your help. > > Same result here after installing the above createrepo RPM. > > "Unable to read group information from repositories. This > is a problem with the generation of your install tree." > > The dialog box says simply "Reboot." At the next box, I get: > > | You have specified that the group 'virtualization' should | > | be installed. This group does not exist. Would you like to | > | continue or abort your installation? > > If I say continue, the install seems to proceed okay. > Ok this sounds more like the kickstart you are using is not pointing to the correct repo. If you loaded centos in via a profile.. you should be able to look at it like: cobbler profile report --name=centos-5.1-server-i386 or some such thing.. the default kickstart is then: # cobbler profile report --name=rhel-5.1-server-i386 profile : rhel-5.1-server-i386 distro : rhel-5.1-server-i386 kickstart : /var/lib/cobbler/kickstarts/rhel5/rhel5-hardware-default.cfg kernel options : {} ks metadata : {} virt file size : 32 virt ram : 512 virt type : auto virt path : virt bridge : xenbr0 virt cpus : 1 repos : [] dhcp tag : default server : <> # cobbler system report --name=bakeneko_smoogespace_com system : bakeneko_smoogespace_com profile : rhel-5.1-server-i386 kernel options : {} ks metadata : {} netboot enabled? : True kickstart : /var/lib/cobbler/kickstarts/rhel5/rhel5-hardware-xendom0.cfg virt type : <> virt path : <> server : <> interface : intf0 mac address : 00:19:B9:BB:8F:9E ip address : 192.168.2.1 hostname : bakeneko.smoogespace.com gateway : 192.168.2.255 subnet : 255.255.255.0 virt bridge : dhcp tag : net2 Here I am using kickstarts that have rhel5 items in them. My guess is that you might be trying one with CentOS which is looking for other names. > At boot time, however, the system hangs at Mounting Other Filesystems. > Thats usually a network issue where its trying to get IPs that don't exist. > I suspect that the initial install was not clean and so I am chasing > that down before moving on to the hung CentOS 5 boot. > > Thanks. > > -D > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > -- Stephen J Smoogen. -- BSD/GNU/Linux How far that little candle throws his beams! So shines a good deed in a naughty world. = Shakespeare. "The Merchant of Venice" From ole.ersoy at gmail.com Sat May 31 19:59:22 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Sat, 31 May 2008 14:59:22 -0500 Subject: [et-mgmt-tools] Re: dhcpctl_connect: connection refused In-Reply-To: <4840827E.7010000@gmail.com> References: <3cc2cbc40805301436g3fe78536tcd29f93bb207f2f4@mail.gmail.com> <80d7e4090805301439u658ae9a4xa2ee1e18170ab897@mail.gmail.com> <3cc2cbc40805301444m6cb134d8m57f4153efb5e3a2@mail.gmail.com> <80d7e4090805301522p432b320as8b9348f885303584@mail.gmail.com> <4840827E.7010000@gmail.com> Message-ID: <4841AE1A.8070000@gmail.com> I think I got it and created a ticket for it in case someone else runs into the same thing: https://fedorahosted.org/cobbler/ticket/160 Cheers, - Ole Ole Ersoy wrote: > Hi, > > I added a profile and now when I attempt to sync I get: > > [root at localhost cobbler]# cobbler sync > dhcpctl_connect: connection refused > dhcpctl_connect: connection refused > > Any ideas? > > Running cobbler check shows this: > > [root at localhost cobbler]# cobbler check > The following potential problems were detected: > #0: One or more kickstart templates references default password > 'cobbler' and should be changed for security reasons: > /etc/cobbler/sample.ks, /etc/cobbler/legacy.ks, /etc/cobbler/sample_end.ks > > TIA, > - Ole > > > From ole.ersoy at gmail.com Sat May 31 20:42:46 2008 From: ole.ersoy at gmail.com (Ole Ersoy) Date: Sat, 31 May 2008 15:42:46 -0500 Subject: [et-mgmt-tools] Booting Hangs on "Trying to Load: pxelinux.cfg/default" Message-ID: <4841B846.3090101@gmail.com> Any trouble shooting tips? I used to get a "boot" menu prior to adding a system... TIA, - Ole