From crobinso at redhat.com Sun Mar 1 00:54:19 2009 From: crobinso at redhat.com (Cole Robinson) Date: Sat, 28 Feb 2009 19:54:19 -0500 Subject: [et-mgmt-tools] Making labels selectable in properties dialog In-Reply-To: <5e5815d10902250902p1496e9c6y69d95cb8044c6a4d@mail.gmail.com> References: <5e5815d10902250403l780aae6bw11a4302e5c103adc@mail.gmail.com> <49A54BC2.3070403@redhat.com> <5e5815d10902250902p1496e9c6y69d95cb8044c6a4d@mail.gmail.com> Message-ID: <49A9DCBB.3030101@redhat.com> Rob James wrote: > Ah, OK, that seems better. I've attached the bundle for the patch. > Would you be able to review for me? i wasn't able to test against head > though as when compiling the hg source here it won't show the details > page ('NoneType' object is not iterable). > Not sure why you were getting an error with upstream checkout, but I applied your patch: http://hg.et.redhat.com/cgi-bin/hg-virt.cgi/applications/virt-manager--devel/rev/b8b3bf2c628c Thanks, Cole From crobinso at redhat.com Sun Mar 1 01:08:10 2009 From: crobinso at redhat.com (Cole Robinson) Date: Sat, 28 Feb 2009 20:08:10 -0500 Subject: [et-mgmt-tools] Heads up: new virtinst + virt-manager releases in coming weeks Message-ID: <49A9DFFA.1050000@redhat.com> Hi all, Just a heads up that I'm planning on pushing a new virtinst release this coming Tuesday, March 3rd. A new virt-manager release will be pushed out a week later (March 9th, maybe a few days earlier). All in time for Fedora 11 Beta :) So if you have anything you'd like to get resolved, please raise (or re-raise) it. Thanks, Cole From ben at nerp.net Mon Mar 2 02:03:53 2009 From: ben at nerp.net (Ben Kochie) Date: Sun, 1 Mar 2009 18:03:53 -0800 (PST) Subject: [et-mgmt-tools] Support for kvm/qemu cache mode In-Reply-To: <49A614A5.7060506@redhat.com> References: <49A614A5.7060506@redhat.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ok, thanks for all the off-list help. Here's my patch for adding cache support to virt-install. I'm not super happy with the xml if/elif bits at VitualDisk.py:681, but that part of the code is a bit tricky anyway. http://ben.nerp.net/virtinst/cache-support-v4.diff There are some other logs and output examples in the dir: http://ben.nerp.net/virtinst/ It would be great to get this in for the next release. - -ben "UNIX is user-friendly, it's just picky about its friends." On Wed, 25 Feb 2009, Cole Robinson wrote: > Ben Kochie wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Libvirt now supports an xml bit to select qemu cache types. >> >> >> >> >> >> >> >> It would be nice to select this with -disk in virt-install. > > Cool! Sounds good. > >> started looking at the VirtualDisk.py source to see where something like >> that would go. If anyone is working on this, or has ideas, please let >> me know. >> > > Following existing convention should be sufficient. Add cache as a > VirtualDisk __init__ parameter, setup a simple set_*/get_* property, and > if it's present, encorporate it the xml in get_config_xml. > > You could also add some validation tests in tests/validation.py. > >> I was thinking the virt-install command would look like: >> >> virt-install -disk "path=/path/to/dev,cache=writethrough" > > Looks good to me. Please be sure to include documentation in > man/en/virt-install.pod (don't worry about actually generating the man > page). > > Thanks, > Cole > -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFJqz6OfYg2asD844oRAiYzAJ9zsV7ysQGxi86fZWYdTpYyLQGzjwCfasE7 eeDiL6C8L7qlR/kELL47dHg= =QLKr -----END PGP SIGNATURE----- From crobinso at redhat.com Mon Mar 2 04:41:11 2009 From: crobinso at redhat.com (Cole Robinson) Date: Sun, 01 Mar 2009 23:41:11 -0500 Subject: [et-mgmt-tools] [PATCH 0/3] virt-install: support host device assignment Message-ID: <49AB6367.3040108@redhat.com> The following patches add support for host device assignment to virtinst and virt-install. The patches aren't complete, since I'm not entirely sure how managed or unmanaged devices coincide with HostDeviceDetach and HostDeviceReset, and how this relates to PCI vs. USB. But it shouldn't be a big change from what's posted. The patches are: 01 - Support for parsing libvirt nodedev xml 02 - Support for building domain 'hostdev' device xml 03 - virt-install --host-device option Thanks, Cole From crobinso at redhat.com Mon Mar 2 04:42:07 2009 From: crobinso at redhat.com (Cole Robinson) Date: Sun, 01 Mar 2009 23:42:07 -0500 Subject: [et-mgmt-tools] [PATCH 1/3]: Add NodeDeviceParser module In-Reply-To: <49AB6367.3040108@redhat.com> References: <49AB6367.3040108@redhat.com> Message-ID: <49AB639F.6080101@redhat.com> Add NodeDeviceParser module for parsing libvirt host device xml, associated helper functions and tests. Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virtinst-hostdev-01.patch Type: text/x-diff Size: 27358 bytes Desc: not available URL: From crobinso at redhat.com Mon Mar 2 04:42:58 2009 From: crobinso at redhat.com (Cole Robinson) Date: Sun, 01 Mar 2009 23:42:58 -0500 Subject: [et-mgmt-tools] [PATCH 2/3]: Add VirtualHostDevice classes In-Reply-To: <49AB6367.3040108@redhat.com> References: <49AB6367.3040108@redhat.com> Message-ID: <49AB63D2.4000102@redhat.com> Add VirtualHostDevice classes for build domain 'hostdev' device xml, along with associated tests. This is currently incomplete: I'm not entirely clear on how a managed vs. unmanaged device interacts with HostDeviceAttach/Reset, and if there are differences in whats applicable for USB vs. PCI devices. The libvirt API for all this hasn't officially landed yet either so this is blocked until then. Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virtinst-hostdev-02.patch Type: text/x-diff Size: 12538 bytes Desc: not available URL: From crobinso at redhat.com Mon Mar 2 04:43:30 2009 From: crobinso at redhat.com (Cole Robinson) Date: Sun, 01 Mar 2009 23:43:30 -0500 Subject: [et-mgmt-tools] [PATCH 3/3]: Add virt-install --host-device command In-Reply-To: <49AB6367.3040108@redhat.com> References: <49AB6367.3040108@redhat.com> Message-ID: <49AB63F2.10403@redhat.com> Add a --host-device command to virt-install. It takes a libvirt nodedev name as it's only argument, and attaches the associated device to the new VM. Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virtinst-hostdev-03.patch Type: text/x-diff Size: 2678 bytes Desc: not available URL: From berrange at redhat.com Mon Mar 2 08:32:52 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 2 Mar 2009 08:32:52 +0000 Subject: [et-mgmt-tools] [PATCH 0/3] virt-install: support host device assignment In-Reply-To: <49AB6367.3040108@redhat.com> References: <49AB6367.3040108@redhat.com> Message-ID: <20090302083252.GC15538@redhat.com> On Sun, Mar 01, 2009 at 11:41:11PM -0500, Cole Robinson wrote: > The following patches add support for host device assignment to virtinst > and virt-install. > > The patches aren't complete, since I'm not entirely sure how managed or > unmanaged devices coincide with HostDeviceDetach and HostDeviceReset, > and how this relates to PCI vs. USB. But it shouldn't be a big change > from what's posted. If 'managed=yes' then libvirt will automatically detach the device from the host driver. This is nice & simple, but also nice & easy to shoot yourself in the foot - don't mix up the PCI id for the NIC you want to pasthrough with your SATA controller ;-) If 'managed=no' then libvirt expects that the caller has already ensured the device is detached from the host before *ALL* attempts to start the guest, now & in the future. The app can use HostDeviceDetach todo this but remember it has to be manually run every time, which probably is not what you want. The more likely case for using managed=no, is if you have statically assigned the PCI device ID to pcistub.ko in /etc/modprobe.conf so that is guarenteed detached from the moment you boot. I'd strongly recommend virt-install only use managed=yes, and don't make the use pick PCI ids off the list, rather present them with the human readable vendor and product names. When picking PCI devices you can also use the node device APIs to query for any other devices which are children of that PCI device - so if there are 2 identical NICs in the machine, you can show the user that one is eth0 and the other is eth1 to help them pick the right one. Regards, Daniel -- |: Red Hat, Engineering, London -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 bkearney at redhat.com Mon Mar 2 12:31:47 2009 From: bkearney at redhat.com (Bryan Kearney) Date: Mon, 02 Mar 2009 07:31:47 -0500 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49A86E27.20605@redhat.com> References: <49A86E27.20605@redhat.com> Message-ID: <49ABD1B3.9030004@redhat.com> Cole Robinson wrote: > Hi all, > > > Page 5: Summary and Advanced Options > > http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-1.png > http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-2.png > http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-3.png > > The summary section is pretty straight forward, no surprises here. > The 'Advanced Options' section encompasses networking, hypervisor, and > architecture options. The hypervisor and arch defaults were explained above. > > For networking, the default is: > > - A bridge device if any exist, else > - Virtual Network 'default' (comes out of the box with libvirt), else > - First available virtual network, else > - no networking! Networking being "hidden" seems a bit odd here. In a decent number of the cases I have used the tool in (not all) i wanted to select this. It it possible to show this information on the summary, and then have them click into advanced to set it? -- bk From crobinso at redhat.com Mon Mar 2 14:09:14 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 02 Mar 2009 09:09:14 -0500 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49ABD1B3.9030004@redhat.com> References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> Message-ID: <49ABE88A.5030702@redhat.com> Bryan Kearney wrote: > Cole Robinson wrote: >> Hi all, >> >> >> Page 5: Summary and Advanced Options >> >> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-1.png >> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-2.png >> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-3.png >> >> The summary section is pretty straight forward, no surprises here. >> The 'Advanced Options' section encompasses networking, hypervisor, and >> architecture options. The hypervisor and arch defaults were explained >> above. >> >> For networking, the default is: >> >> - A bridge device if any exist, else >> - Virtual Network 'default' (comes out of the box with libvirt), else >> - First available virtual network, else >> - no networking! > > Networking being "hidden" seems a bit odd here. In a decent number of > the cases I have used the tool in (not all) i wanted to select this. It > it possible to show this information on the summary, and then have them > click into advanced to set it? > In the times you were changing the default network selection, what was the reason? Choosing from more than one virtual net or bridge? Just curious. The reasoning here is we want the UI to be as simple as possible for the out of the box user, which at most has 1 virtual network (and 1 bridge device if using xen networking). For these users, we don't want them to concern themselves with the default, we want it to 'just work' Anything beyond that case requires explicit user setup ahead of time, they should know full well when creating a new VM that they will have to select some non-default network setup. Forcing them to click the 'Advanced Options' expander doesn't seem like a loss, compared to the gain of allowing new users to not have to even think about networking (either by showing the network drop down, or having a summary field showing some choice they didn't explicitly make.) Thanks, Cole From mdehaan at redhat.com Mon Mar 2 14:17:07 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 02 Mar 2009 09:17:07 -0500 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49ABE88A.5030702@redhat.com> References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> <49ABE88A.5030702@redhat.com> Message-ID: <49ABEA63.8000707@redhat.com> Cole Robinson wrote: > Bryan Kearney wrote: > >> Cole Robinson wrote: >> >>> Hi all, >>> >>> >>> Page 5: Summary and Advanced Options >>> >>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-1.png >>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-2.png >>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-3.png >>> >>> The summary section is pretty straight forward, no surprises here. >>> The 'Advanced Options' section encompasses networking, hypervisor, and >>> architecture options. The hypervisor and arch defaults were explained >>> above. >>> >>> For networking, the default is: >>> >>> - A bridge device if any exist, else >>> - Virtual Network 'default' (comes out of the box with libvirt), else >>> - First available virtual network, else >>> - no networking! >>> >> Networking being "hidden" seems a bit odd here. In a decent number of >> the cases I have used the tool in (not all) i wanted to select this. It >> it possible to show this information on the summary, and then have them >> click into advanced to set it? >> >> > > In the times you were changing the default network selection, what was > the reason? Choosing from more than one virtual net or bridge? Just curious. > > The reasoning here is we want the UI to be as simple as possible for the > out of the box user, which at most has 1 virtual network (and 1 bridge > device if using xen networking). For these users, we don't want them to > concern themselves with the default, we want it to 'just work' > > Anything beyond that case requires explicit user setup ahead of time, > they should know full well when creating a new VM that they will have to > select some non-default network setup. Forcing them to click the > 'Advanced Options' expander doesn't seem like a loss, compared to the > gain of allowing new users to not have to even think about networking > (either by showing the network drop down, or having a summary field > showing some choice they didn't explicitly make.) > > Thanks, > Cole > > > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools > We still need to get a "install using data from remote cobbler server" in there :) --Michael From radek at eadresa.cz Mon Mar 2 14:56:24 2009 From: radek at eadresa.cz (Radek Hladik) Date: Mon, 02 Mar 2009 15:56:24 +0100 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49A86E27.20605@redhat.com> References: <49A86E27.20605@redhat.com> Message-ID: <49ABF398.4030201@eadresa.cz> Cole Robinson napsal(a): > Hi all, > > The current 'New VM' wizard has clearly started to show its age. The > original design was largely based on xen specific assumptions and the > state of libvirt/virtinst at the time: many of those assumptions don't > apply today, or require a bit more thought since we now support both xen > and qemu based VMs. > > You can find a patch for a new implementation of the 'New VM' wizard here: > > http://fedorapeople.org/~crobinso/virt-manager/newvm2/virt-manager-newvm-01.patch > I have to say I like this redesign. However I am missing two options in the wizard at all (and I missed them in the old one too). I would like to be able to just create a new machine and do not install anything on it and/or do not startup it. So I would like to see the install method "No installation" and an advanced option "do not start the VM now". There are situations when you do not want to start the VM right away. I can address both issues with installing from dummy ISO image (actually RIP linux :-) ) and stopping the machine, but I think it would be nice to have the options right in the wizard. Radek From berrange at redhat.com Mon Mar 2 14:53:34 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 2 Mar 2009 14:53:34 +0000 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49ABF398.4030201@eadresa.cz> References: <49A86E27.20605@redhat.com> <49ABF398.4030201@eadresa.cz> Message-ID: <20090302145334.GA23056@redhat.com> On Mon, Mar 02, 2009 at 03:56:24PM +0100, Radek Hladik wrote: > Cole Robinson napsal(a): > >Hi all, > > > >The current 'New VM' wizard has clearly started to show its age. The > >original design was largely based on xen specific assumptions and the > >state of libvirt/virtinst at the time: many of those assumptions don't > >apply today, or require a bit more thought since we now support both xen > >and qemu based VMs. > > > >You can find a patch for a new implementation of the 'New VM' wizard here: > > > >http://fedorapeople.org/~crobinso/virt-manager/newvm2/virt-manager-newvm-01.patch > > > > I have to say I like this redesign. However I am missing two options in > the wizard at all (and I missed them in the old one too). I would like > to be able to just create a new machine and do not install anything on > it and/or do not startup it. So I would like to see the install method > "No installation" and an advanced option "do not start the VM now". > There are situations when you do not want to start the VM right away. I > can address both issues with installing from dummy ISO image (actually > RIP linux :-) ) and stopping the machine, but I think it would be nice > to have the options right in the wizard. Yeah, this is the so called 'import existing disk image' or 'run livecd' use case. I can't remember off hand whether we were intending to put this in the 'new vm' wizard or have a separate option for doing that. It is probably easy enough to just have it all in one, now we've got the "install media" choice up-front. Daniel -- |: Red Hat, Engineering, London -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 bkearney at redhat.com Mon Mar 2 18:02:36 2009 From: bkearney at redhat.com (Bryan Kearney) Date: Mon, 02 Mar 2009 13:02:36 -0500 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49ABE88A.5030702@redhat.com> References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> <49ABE88A.5030702@redhat.com> Message-ID: <49AC1F3C.6070702@redhat.com> Cole Robinson wrote: > Bryan Kearney wrote: >> Cole Robinson wrote: >>> Hi all, >>> >>> >>> Page 5: Summary and Advanced Options >>> >>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-1.png >>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-2.png >>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-3.png >>> >>> The summary section is pretty straight forward, no surprises here. >>> The 'Advanced Options' section encompasses networking, hypervisor, and >>> architecture options. The hypervisor and arch defaults were explained >>> above. >>> >>> For networking, the default is: >>> >>> - A bridge device if any exist, else >>> - Virtual Network 'default' (comes out of the box with libvirt), else >>> - First available virtual network, else >>> - no networking! >> Networking being "hidden" seems a bit odd here. In a decent number of >> the cases I have used the tool in (not all) i wanted to select this. It >> it possible to show this information on the summary, and then have them >> click into advanced to set it? >> > > In the times you were changing the default network selection, what was > the reason? Choosing from more than one virtual net or bridge? Just curious. I had a specific bridge I wanted it to use. Normally.. I want it to use default, but in 2 use cases I have specific bridges which I want them to use. > > The reasoning here is we want the UI to be as simple as possible for the > out of the box user, which at most has 1 virtual network (and 1 bridge > device if using xen networking). For these users, we don't want them to > concern themselves with the default, we want it to 'just work' I agree. Which is why I wonder can you show that in the review step.. and then have them click to edit. > > Anything beyond that case requires explicit user setup ahead of time, > they should know full well when creating a new VM that they will have to > select some non-default network setup. Forcing them to click the > 'Advanced Options' expander doesn't seem like a loss, compared to the > gain of allowing new users to not have to even think about networking > (either by showing the network drop down, or having a summary field > showing some choice they didn't explicitly make.) > > Thanks, > Cole From agx at sigxcpu.org Mon Mar 2 19:12:03 2009 From: agx at sigxcpu.org (Guido =?iso-8859-1?Q?G=FCnther?=) Date: Mon, 2 Mar 2009 20:12:03 +0100 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49A86E27.20605@redhat.com> References: <49A86E27.20605@redhat.com> Message-ID: <20090302191203.GA8388@bogon.ms20.nix> Hi Cole, this all looks great. On Fri, Feb 27, 2009 at 05:50:15PM -0500, Cole Robinson wrote: > The summary section is pretty straight forward, no surprises here. > The 'Advanced Options' section encompasses networking, hypervisor, and > architecture options. The hypervisor and arch defaults were explained above. An option to select disk (scsi, ide, virtio) and network adapter model (e1000, ...) as advanced options would be great (still defaulting to virtio if supported via the osdict) since there are some OSes that don't support all hardware out of the box and there are kvm/qemu versions that have problems with certain adapter types. Cheers, -- Guido From berrange at redhat.com Mon Mar 2 19:50:28 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 2 Mar 2009 19:50:28 +0000 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <20090302191203.GA8388@bogon.ms20.nix> References: <49A86E27.20605@redhat.com> <20090302191203.GA8388@bogon.ms20.nix> Message-ID: <20090302195028.GA1372@redhat.com> On Mon, Mar 02, 2009 at 08:12:03PM +0100, Guido G?nther wrote: > Hi Cole, > this all looks great. > On Fri, Feb 27, 2009 at 05:50:15PM -0500, Cole Robinson wrote: > > The summary section is pretty straight forward, no surprises here. > > The 'Advanced Options' section encompasses networking, hypervisor, and > > architecture options. The hypervisor and arch defaults were explained above. > An option to select disk (scsi, ide, virtio) and network adapter model > (e1000, ...) as advanced options would be great (still defaulting to > virtio if supported via the osdict) since there are some OSes that don't > support all hardware out of the box and there are kvm/qemu versions that > have problems with certain adapter types. If there are combinations of OS,Disk that don't work IMHO we should improve the OS type metadata so we don't use them. Choice of specific hardware models is something we really want to keep out of the new VM wizard, because the end user really isn't in a position to have the knowledge to make suitable choices. Daniel -- |: Red Hat, Engineering, London -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 Mon Mar 2 20:58:25 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 02 Mar 2009 15:58:25 -0500 Subject: [et-mgmt-tools] Support for kvm/qemu cache mode In-Reply-To: References: <49A614A5.7060506@redhat.com> Message-ID: <49AC4871.2020709@redhat.com> Ben Kochie wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Ok, thanks for all the off-list help. > > Here's my patch for adding cache support to virt-install. I'm not super > happy with the xml if/elif bits at VitualDisk.py:681, but that part of > the code is a bit tricky anyway. > > http://ben.nerp.net/virtinst/cache-support-v4.diff > > There are some other logs and output examples in the dir: > > http://ben.nerp.net/virtinst/ > > It would be great to get this in for the next release. > > - -ben > Fixed this up to work with some VirtualDisk xml tweaks I had done, and applied: http://hg.et.redhat.com/cgi-bin/hg-virt.cgi/applications/virtinst--devel/rev/52032ccd3372 Thanks! - Cole From crobinso at redhat.com Mon Mar 2 21:02:20 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 02 Mar 2009 16:02:20 -0500 Subject: [et-mgmt-tools] Configuration of the virt-manager In-Reply-To: <200902261044.12753.lugs@linuxmaker.de> References: <200902261044.12753.lugs@linuxmaker.de> Message-ID: <49AC495C.4030305@redhat.com> lugs at linuxmaker.de wrote: > Hello, > > I'm looking now already since days after the configuration of > virt-manager on Debian, but found nothing on the internet or any literature. > I now have a machine on Debian/Lenny with KVM and wants use here virtual > machines on logical volumes. With virt-manager should indeed be a breeze, but > it is not, because the configuration of virt-manager is still opaque to me. > > If I have virt-manager and the traces on the workstation installed the program > I want to start now, I get the messages > > Unable to open connection to hypervisor URI 'qemu+ssh://192.168.0.3/system': > Die Verbindung wurde vom Kommunikationspartner > zur?ckgesetzt > Traceback (most recent call last): > File "/usr/share/virt-manager/virtManager/connection.py", line 486, in > _open_thread > None], flags) > File "/usr/lib/python2.5/site-packages/libvirt.py", line 99, in openAuth > if ret is None:raise libvirtError('virConnectOpenAuth() failed') > libvirtError: Die Verbindung wurde vom Kommunikationspartner zur?ckgesetzt > > I assume that the /etc/libvirt/libvirtd.conf on the workstation and > possibly on the KVM-server must be configured accordingly. > But now I don't know how. Also tell me the URI syntax not so much. > Is what I intend "qemu + ssh: / /" the right thing? Have I "/ system" on > 192.168.0.3 to establish or it is a directory or it is a service? > For ssh access, you shouldn't need any changes to libvirtd.conf. The URI syntax looks correct as well. An easier way to test it all is to use virsh. So, try virsh --connect qemu+ssh://192.168.0.3/system (or) virsh --connect qemu+ssh://root at 192.168.0.3/system And see if that also gives errors. Thanks, Cole From sghosh at redhat.com Mon Mar 2 21:39:37 2009 From: sghosh at redhat.com (Subhendu Ghosh) Date: Mon, 02 Mar 2009 16:39:37 -0500 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49AC1F3C.6070702@redhat.com> References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> <49ABE88A.5030702@redhat.com> <49AC1F3C.6070702@redhat.com> Message-ID: <49AC5219.6000105@redhat.com> Bryan Kearney wrote: > Cole Robinson wrote: >> Bryan Kearney wrote: >>> Cole Robinson wrote: >>>> Hi all, >>>> >>>> >>>> Page 5: Summary and Advanced Options >>>> >>>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-1.png >>>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-2.png >>>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-3.png >>>> >>>> The summary section is pretty straight forward, no surprises here. >>>> The 'Advanced Options' section encompasses networking, hypervisor, and >>>> architecture options. The hypervisor and arch defaults were explained >>>> above. >>>> >>>> For networking, the default is: >>>> >>>> - A bridge device if any exist, else >>>> - Virtual Network 'default' (comes out of the box with libvirt), else >>>> - First available virtual network, else >>>> - no networking! >>> Networking being "hidden" seems a bit odd here. In a decent number of >>> the cases I have used the tool in (not all) i wanted to select this. It >>> it possible to show this information on the summary, and then have them >>> click into advanced to set it? >>> >> >> In the times you were changing the default network selection, what was >> the reason? Choosing from more than one virtual net or bridge? Just >> curious. > > I had a specific bridge I wanted it to use. Normally.. I want it to use > default, but in 2 use cases I have specific bridges which I want them to > use. > >> >> The reasoning here is we want the UI to be as simple as possible for the >> out of the box user, which at most has 1 virtual network (and 1 bridge >> device if using xen networking). For these users, we don't want them to >> concern themselves with the default, we want it to 'just work' > > I agree. Which is why I wonder can you show that in the review step.. > and then have them click to edit. > >> >> Anything beyond that case requires explicit user setup ahead of time, >> they should know full well when creating a new VM that they will have to >> select some non-default network setup. Forcing them to click the >> 'Advanced Options' expander doesn't seem like a loss, compared to the >> gain of allowing new users to not have to even think about networking >> (either by showing the network drop down, or having a summary field >> showing some choice they didn't explicitly make.) >> How dynamic is the network list under advanced? What happens if you don't find the network you are looking for and decide to go off an create it while leaving the wizard waiting on step 5? Do you have to restart the wizard? Also the wizard should support multiple NICs and disks - perhaps from step 5 - be able to launch sub-wizards to add nic and add disk? -regards Subhendu -- Subhendu Ghosh Red Hat Phone: +1-201-232-2851 Email: sghosh at redhat.com From ben at nerp.net Mon Mar 2 21:47:41 2009 From: ben at nerp.net (Ben Kochie) Date: Mon, 2 Mar 2009 13:47:41 -0800 (PST) Subject: [et-mgmt-tools] Ubuntu patches for next release In-Reply-To: <49AC5219.6000105@redhat.com> References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> <49ABE88A.5030702@redhat.com> <49AC1F3C.6070702@redhat.com> <49AC5219.6000105@redhat.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 There are a couple patches against 400.1 that are in the Ubuntu tree. These could probably be integrated upstream. http://packages.ubuntu.com/source/jaunty/virtinst In addition to these 2 patches, "ubuntuhardy" in osdict.py supports virtio disk, so that should be added. - -ben "UNIX is user-friendly, it's just picky about its friends." -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFJrFQBfYg2asD844oRAkiCAKCY4qd73ZoMeQ2bYysPbr1yZJZuzgCgjlnp DWi3szFYDaHRQ5xI+wYqxXM= =as/J -----END PGP SIGNATURE----- -------------- next part -------------- A non-text attachment was scrubbed... Name: 0004-manpage-typo.patch Type: text/x-diff Size: 1435 bytes Desc: URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 9001_Ubuntu.patch Type: text/x-diff Size: 1537 bytes Desc: URL: From crobinso at redhat.com Mon Mar 2 22:48:40 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 02 Mar 2009 17:48:40 -0500 Subject: [et-mgmt-tools] [PATCH 2/3]: Add VirtualHostDevice classes In-Reply-To: <49AB63D2.4000102@redhat.com> References: <49AB6367.3040108@redhat.com> <49AB63D2.4000102@redhat.com> Message-ID: <49AC6248.90606@redhat.com> Cole Robinson wrote: > Add VirtualHostDevice classes for build domain 'hostdev' device xml, > along with associated tests. > > This is currently incomplete: I'm not entirely clear on how a managed > vs. unmanaged device interacts with HostDeviceAttach/Reset, and if there > are differences in whats applicable for USB vs. PCI devices. The libvirt > API for all this hasn't officially landed yet either so this is blocked > until then. > Okay, I think this updated version should be sufficient now. Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virtinst-hostdev-02-2.patch Type: text/x-diff Size: 13910 bytes Desc: not available URL: From crobinso at redhat.com Mon Mar 2 22:49:24 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 02 Mar 2009 17:49:24 -0500 Subject: [et-mgmt-tools] [PATCH 3/3]: Add virt-install --host-device command In-Reply-To: <49AB63F2.10403@redhat.com> References: <49AB6367.3040108@redhat.com> <49AB63F2.10403@redhat.com> Message-ID: <49AC6274.3050100@redhat.com> Cole Robinson wrote: > Add a --host-device command to virt-install. It takes a libvirt nodedev > name as it's only argument, and attaches the associated device to the > new VM. > Updated patch which unbreaks the case when no host-devices are specified. Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virtinst-hostdev-03-2.patch Type: text/x-diff Size: 2716 bytes Desc: not available URL: From crobinso at redhat.com Mon Mar 2 22:50:59 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 02 Mar 2009 17:50:59 -0500 Subject: [et-mgmt-tools] [PATCH 0/3] virt-install: support host device assignment In-Reply-To: <20090302083252.GC15538@redhat.com> References: <49AB6367.3040108@redhat.com> <20090302083252.GC15538@redhat.com> Message-ID: <49AC62D3.4080204@redhat.com> Daniel P. Berrange wrote: > On Sun, Mar 01, 2009 at 11:41:11PM -0500, Cole Robinson wrote: >> The following patches add support for host device assignment to virtinst >> and virt-install. >> >> The patches aren't complete, since I'm not entirely sure how managed or >> unmanaged devices coincide with HostDeviceDetach and HostDeviceReset, >> and how this relates to PCI vs. USB. But it shouldn't be a big change >> from what's posted. > > If 'managed=yes' then libvirt will automatically detach the device > from the host driver. This is nice & simple, but also nice & easy > to shoot yourself in the foot - don't mix up the PCI id for the NIC > you want to pasthrough with your SATA controller ;-) > > If 'managed=no' then libvirt expects that the caller has already ensured > the device is detached from the host before *ALL* attempts to start the > guest, now & in the future. The app can use HostDeviceDetach todo this > but remember it has to be manually run every time, which probably is not > what you want. The more likely case for using managed=no, is if you have > statically assigned the PCI device ID to pcistub.ko in /etc/modprobe.conf > so that is guarenteed detached from the moment you boot. > > I'd strongly recommend virt-install only use managed=yes, and don't make > the use pick PCI ids off the list, rather present them with the human > readable vendor and product names. When picking PCI devices you can > also use the node device APIs to query for any other devices which are > children of that PCI device - so if there are 2 identical NICs in the > machine, you can show the user that one is eth0 and the other is eth1 > to help them pick the right one. > Thanks. I updated patch 2 to attempt a Detach and Reset simply as a sanity check (so if adding a PCI device via virt-manager the user can find out it will fail at attach time, rather then the next VM boot). - Cole From crobinso at redhat.com Mon Mar 2 23:00:54 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 02 Mar 2009 18:00:54 -0500 Subject: [et-mgmt-tools] [PATCH] Add --list-os-options to virt-install and virt-convert Message-ID: <49AC6526.70608@redhat.com> Add an option '--list-os-options' to virt-install and virt-convert to dump the valid os-type/variant values from our internal OS dictionary (then exit). Prior to this, the only way a user knew what to pass was via the quite outdated man pages, or via virt-manager (which reads the OS dictionary from virtinst). On one side I'm hesitant about this: we are going to expand this OS metadata in the future with a potentially more granular approach (ex. RHEL5.1/2/3 instead of just RHEL5), and giving the illusion that this is stable (and machine parseable) may make it harder to deviate later. That said, this solution is clearly much more useful and sustainable than the current situation, so until we work out something better I think this patch is the way to go. Example output: $ ./virt-install --list-os-options TYPE : VARIANT -------------- linux : debianetch linux : debianlenny linux : fedora5 linux : fedora6 linux : fedora7 linux : fedora8 linux : fedora9 linux : fedora10 linux : fedora11 linux : generic24 linux : generic26 linux : virtio26 linux : rhel2.1 linux : rhel3 linux : rhel4 linux : rhel5 linux : sles10 linux : ubuntuhardy other : generic other : msdos other : netware4 other : netware5 other : netware6 solaris : opensolaris solaris : solaris10 solaris : solaris9 unix : freebsd6 unix : freebsd7 unix : openbsd4 windows : vista windows : win2k windows : win2k3 windows : win2k8 windows : winxp windows : winxp64 Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virtinst-list-os-opts.patch Type: text/x-diff Size: 11363 bytes Desc: not available URL: From berrange at redhat.com Mon Mar 2 23:05:26 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 2 Mar 2009 23:05:26 +0000 Subject: [et-mgmt-tools] [PATCH] Add --list-os-options to virt-install and virt-convert In-Reply-To: <49AC6526.70608@redhat.com> References: <49AC6526.70608@redhat.com> Message-ID: <20090302230526.GG1372@redhat.com> On Mon, Mar 02, 2009 at 06:00:54PM -0500, Cole Robinson wrote: > Add an option '--list-os-options' to virt-install and virt-convert to > dump the valid os-type/variant values from our internal OS dictionary > (then exit). Prior to this, the only way a user knew what to pass was > via the quite outdated man pages, or via virt-manager (which reads the > OS dictionary from virtinst). > > On one side I'm hesitant about this: we are going to expand this OS > metadata in the future with a potentially more granular approach (ex. > RHEL5.1/2/3 instead of just RHEL5), and giving the illusion that this is > stable (and machine parseable) may make it harder to deviate later. > > That said, this solution is clearly much more useful and sustainable > than the current situation, so until we work out something better I > think this patch is the way to go. How about making the build process extract this list and stuff them into the POD man page, so it is always up2date ? Daniel -- |: Red Hat, Engineering, London -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 Mon Mar 2 23:03:50 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 02 Mar 2009 18:03:50 -0500 Subject: [et-mgmt-tools] [PATCH] Add --os-* options to virt-image Message-ID: <49AC65D6.9020001@redhat.com> The attached patched allows specifying --os-type and --os-variant for virt-image (as well as the --list-os-options command just posted). At some point we should support this in the virt-image format, but until then this will suffice. Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virtinst-virtimage-os-opts.patch Type: text/x-diff Size: 3813 bytes Desc: not available URL: From crobinso at redhat.com Tue Mar 3 00:24:41 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 02 Mar 2009 19:24:41 -0500 Subject: [et-mgmt-tools] [PATCH] Add --list-os-options to virt-install and virt-convert In-Reply-To: <20090302230526.GG1372@redhat.com> References: <49AC6526.70608@redhat.com> <20090302230526.GG1372@redhat.com> Message-ID: <49AC78C9.2070709@redhat.com> Daniel P. Berrange wrote: > On Mon, Mar 02, 2009 at 06:00:54PM -0500, Cole Robinson wrote: >> Add an option '--list-os-options' to virt-install and virt-convert to >> dump the valid os-type/variant values from our internal OS dictionary >> (then exit). Prior to this, the only way a user knew what to pass was >> via the quite outdated man pages, or via virt-manager (which reads the >> OS dictionary from virtinst). >> >> On one side I'm hesitant about this: we are going to expand this OS >> metadata in the future with a potentially more granular approach (ex. >> RHEL5.1/2/3 instead of just RHEL5), and giving the illusion that this is >> stable (and machine parseable) may make it harder to deviate later. >> >> That said, this solution is clearly much more useful and sustainable >> than the current situation, so until we work out something better I >> think this patch is the way to go. > > How about making the build process extract this list and stuff > them into the POD man page, so it is always up2date ? > That works too. Do you think it should be in all relevant man pages (virt-convert, virt-image, virt-install) or just have the other two link back to virt-install? - Cole From levon at movementarian.org Tue Mar 3 00:57:38 2009 From: levon at movementarian.org (John Levon) Date: Mon, 2 Mar 2009 19:57:38 -0500 Subject: [et-mgmt-tools] [PATCH] Add --list-os-options to virt-install and virt-convert In-Reply-To: <49AC78C9.2070709@redhat.com> References: <49AC6526.70608@redhat.com> <20090302230526.GG1372@redhat.com> <49AC78C9.2070709@redhat.com> Message-ID: <20090303005738.GA22376@movementarian.org> On Mon, Mar 02, 2009 at 07:24:41PM -0500, Cole Robinson wrote: > Do you think it should be in all relevant man pages (virt-convert, > virt-image, virt-install) or just have the other two link back to > virt-install? Just the one please :) Any thought been given to moving all of osdict into a system-installed XML file? The parameters of how different operating systems behave are only going to get more complex. Note that OVF will need a similar (in fact, most likely more complex) database for the virtual platforms. regards john From berrange at redhat.com Tue Mar 3 08:43:10 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Tue, 3 Mar 2009 08:43:10 +0000 Subject: [et-mgmt-tools] [PATCH] Add --list-os-options to virt-install and virt-convert In-Reply-To: <20090303005738.GA22376@movementarian.org> References: <49AC6526.70608@redhat.com> <20090302230526.GG1372@redhat.com> <49AC78C9.2070709@redhat.com> <20090303005738.GA22376@movementarian.org> Message-ID: <20090303084310.GA11646@redhat.com> On Mon, Mar 02, 2009 at 07:57:38PM -0500, John Levon wrote: > On Mon, Mar 02, 2009 at 07:24:41PM -0500, Cole Robinson wrote: > > > Do you think it should be in all relevant man pages (virt-convert, > > virt-image, virt-install) or just have the other two link back to > > virt-install? > > Just the one please :) > > Any thought been given to moving all of osdict into a system-installed > XML file? The parameters of how different operating systems behave are > only going to get more complex. Note that OVF will need a similar (in > fact, most likely more complex) database for the virtual platforms. Absolutely, this is the long term approach we want to take for a number of reasons. There's quite a few things I don't like about how we are currently handling it - It is not possible for local admins to change / extend it, eg to add new OS distribution release. It is guarenteed to be out of date almost as soon as each new virtinst release ships, due to frequency of new OS distros. This is particularly bad for RHEL with its 6 month cycle between update releases - It is tied into virt-install & apps based on it. This makes it hard to use from Cobbler, oVirt, VMWare, or any other virt provisioning tool that's not using virt-install python APIs - It does not include enough information. Type + variant are not really sufficient. RHEL-5 has updates where new capabilities appear like virt IO. We really need at least a 4 level grouping for some OS, eg family (Red Hat), distro (RHEL), release (RHEL5), update (RHEL-5.3). - It ignores the installation URL information. When doing net installs user has to specify URL to download.fedora.redhat.com and virtinstall has the fragile probing to validate this. The OS metadata could easily include the mirror URLs for any free distro Todo a better job there's several things I want todo with this: - An XML schema for defining all the information wrt to guest OS distros that is relevant to virt management tools. - A C library for querying the information in the XML file(s). - Bindings of the C library into Python/Ruby etc as needed - Ability for local admins to extend / override the information either by editing the XML files directly, or a pretty GUI The XML schema is the really important thing, but I think it'd be very important to have a API around it and not expect apps to read the XML directly. This is because to get sufficient flexibility without huge amounts of duplication, I think we'll need to try and keep the XML doc highly normalized, and not just a plain tree - think more of a structure like RDF where there are many tree's cross-referencing each other. I'm attaching an example XML doc I knocked up a few months ago, but I'm not really happy with it as is. In particular I think rather than trying todo a explicit 4 level (family, disto, release, update) hierarchy, it should just be doing an arbitrary generic nested structure, without imposing specific terminology on each level - there's just too much difference between the OS out there. The representation of supported drivers isn't all that great either - albeit better than current way. I was trying to keep two lists - sets of configs inheriting from each other, and refernce them from OS distros. This doesn't capture the full horror of the combinations in the real world though. I feel it might be better to try and maintain an explicit list of emulated devices per (hypervisor, release), and an explicit list of supported device drivers per (os distro, release), then compute the intersection. Daniel -- |: Red Hat, Engineering, London -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 :| -------------- next part -------------- A non-text attachment was scrubbed... Name: README.xml Type: text/xml Size: 1899 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: osinfo.xml Type: text/xml Size: 4653 bytes Desc: not available URL: From crobinso at redhat.com Tue Mar 3 17:16:31 2009 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 03 Mar 2009 12:16:31 -0500 Subject: [et-mgmt-tools] Ubuntu patches for next release In-Reply-To: References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> <49ABE88A.5030702@redhat.com> <49AC1F3C.6070702@redhat.com> <49AC5219.6000105@redhat.com> Message-ID: <49AD65EF.4020800@redhat.com> Ben Kochie wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > There are a couple patches against 400.1 that are in the Ubuntu tree. > These could probably be integrated upstream. > > http://packages.ubuntu.com/source/jaunty/virtinst > > In addition to these 2 patches, "ubuntuhardy" in osdict.py supports > virtio disk, so that should be added. > Thanks for raising these. The first patch was from the Debian maintainer and was pushed upstream a while ago. It's sad, but the second patch (from Ubuntu) looks like it's been carried in some form since November, and never mailed here. I'll apply it before cutting the release. Thanks again, Cole From crobinso at redhat.com Tue Mar 3 17:39:13 2009 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 03 Mar 2009 12:39:13 -0500 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49AC1F3C.6070702@redhat.com> References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> <49ABE88A.5030702@redhat.com> <49AC1F3C.6070702@redhat.com> Message-ID: <49AD6B41.3050304@redhat.com> Bryan Kearney wrote: > Cole Robinson wrote: >> Bryan Kearney wrote: >>> Cole Robinson wrote: >>>> Hi all, >>>> >>>> >>>> Page 5: Summary and Advanced Options >>>> >>>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-1.png >>>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-2.png >>>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-3.png >>>> >>>> The summary section is pretty straight forward, no surprises here. >>>> The 'Advanced Options' section encompasses networking, hypervisor, and >>>> architecture options. The hypervisor and arch defaults were explained >>>> above. >>>> >>>> For networking, the default is: >>>> >>>> - A bridge device if any exist, else >>>> - Virtual Network 'default' (comes out of the box with libvirt), else >>>> - First available virtual network, else >>>> - no networking! >>> Networking being "hidden" seems a bit odd here. In a decent number of >>> the cases I have used the tool in (not all) i wanted to select this. It >>> it possible to show this information on the summary, and then have them >>> click into advanced to set it? >>> >> >> In the times you were changing the default network selection, what was >> the reason? Choosing from more than one virtual net or bridge? Just >> curious. > > I had a specific bridge I wanted it to use. Normally.. I want it to use > default, but in 2 use cases I have specific bridges which I want them to > use. > >> >> The reasoning here is we want the UI to be as simple as possible for the >> out of the box user, which at most has 1 virtual network (and 1 bridge >> device if using xen networking). For these users, we don't want them to >> concern themselves with the default, we want it to 'just work' > > I agree. Which is why I wonder can you show that in the review step.. > and then have them click to edit. > I'm iffy on it. For users that expect it to just work, I'd rather not show anything, since they should have no reason to be concerned with it. For users that have specific networking needs, they will likely be changing the default anyways, so no need to explicitly show it outside of 'Advanced Options'. Also, the info would be a duplicate of what's listed below the 'Advanced Options' tab, which is needlessly redundant. That said it's a pretty minor thing to add, I guess we will just see if people trip up on it. I suppose if anything goes 'wrong' in choosing a default ('default' network isn't present or is inactive, no networks or bridge devs) we could expand the 'Advanced Options' tab automatically which would hopefully prompt the user to examine the choice. Thanks, Cole From crobinso at redhat.com Tue Mar 3 17:45:12 2009 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 03 Mar 2009 12:45:12 -0500 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49AC5219.6000105@redhat.com> References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> <49ABE88A.5030702@redhat.com> <49AC1F3C.6070702@redhat.com> <49AC5219.6000105@redhat.com> Message-ID: <49AD6CA8.3010707@redhat.com> Subhendu Ghosh wrote: > Bryan Kearney wrote: >> Cole Robinson wrote: >>> Bryan Kearney wrote: >>>> Cole Robinson wrote: >>>>> Hi all, >>>>> >>>>> >>>>> Page 5: Summary and Advanced Options >>>>> >>>>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-1.png >>>>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-2.png >>>>> http://fedorapeople.org/~crobinso/virt-manager/newvm2/newvm-pg5-3.png >>>>> >>>>> The summary section is pretty straight forward, no surprises here. >>>>> The 'Advanced Options' section encompasses networking, hypervisor, and >>>>> architecture options. The hypervisor and arch defaults were explained >>>>> above. >>>>> >>>>> For networking, the default is: >>>>> >>>>> - A bridge device if any exist, else >>>>> - Virtual Network 'default' (comes out of the box with libvirt), else >>>>> - First available virtual network, else >>>>> - no networking! >>>> Networking being "hidden" seems a bit odd here. In a decent number of >>>> the cases I have used the tool in (not all) i wanted to select this. It >>>> it possible to show this information on the summary, and then have them >>>> click into advanced to set it? >>>> >>> >>> In the times you were changing the default network selection, what was >>> the reason? Choosing from more than one virtual net or bridge? Just >>> curious. >> >> I had a specific bridge I wanted it to use. Normally.. I want it to >> use default, but in 2 use cases I have specific bridges which I want >> them to use. >> >>> >>> The reasoning here is we want the UI to be as simple as possible for the >>> out of the box user, which at most has 1 virtual network (and 1 bridge >>> device if using xen networking). For these users, we don't want them to >>> concern themselves with the default, we want it to 'just work' >> >> I agree. Which is why I wonder can you show that in the review step.. >> and then have them click to edit. >> >>> >>> Anything beyond that case requires explicit user setup ahead of time, >>> they should know full well when creating a new VM that they will have to >>> select some non-default network setup. Forcing them to click the >>> 'Advanced Options' expander doesn't seem like a loss, compared to the >>> gain of allowing new users to not have to even think about networking >>> (either by showing the network drop down, or having a summary field >>> showing some choice they didn't explicitly make.) >>> > > How dynamic is the network list under advanced? > > What happens if you don't find the network you are looking for and > decide to go off an create it while leaving the wizard waiting on step > 5? Do you have to restart the wizard? > The current code won't pick up the changes, you would have to restart the wizard. We have the mechanism in the backend to pick up the changes though, so it would be a minor thing to add. I'll look into it for the next patch spin. > Also the wizard should support multiple NICs and disks - perhaps from > step 5 - be able to launch sub-wizards to add nic and add disk? > I think such a process would add non-trivial complication to the UI for a rather niche case. For a user wanting a multi nic or disk configured VM, I would point them at virt-install which easily scales in this respect. Extra NICs could also be added post install, since I would think only one would be required for the install process. Thanks, Cole From crobinso at redhat.com Tue Mar 3 21:51:22 2009 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 03 Mar 2009 16:51:22 -0500 Subject: [et-mgmt-tools] [PATCH 0/3] virt-install: support host device assignment In-Reply-To: <49AB6367.3040108@redhat.com> References: <49AB6367.3040108@redhat.com> Message-ID: <49ADA65A.4020301@redhat.com> Cole Robinson wrote: > The following patches add support for host device assignment to virtinst > and virt-install. > > The patches aren't complete, since I'm not entirely sure how managed or > unmanaged devices coincide with HostDeviceDetach and HostDeviceReset, > and how this relates to PCI vs. USB. But it shouldn't be a big change > from what's posted. > > The patches are: > > 01 - Support for parsing libvirt nodedev xml > 02 - Support for building domain 'hostdev' device xml > 03 - virt-install --host-device option > I've committed this series with the updated 02 and 03. Thanks, Cole From crobinso at redhat.com Tue Mar 3 21:53:46 2009 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 03 Mar 2009 16:53:46 -0500 Subject: [et-mgmt-tools] [PATCH] Add --list-os-options to virt-install and virt-convert In-Reply-To: <20090303005738.GA22376@movementarian.org> References: <49AC6526.70608@redhat.com> <20090302230526.GG1372@redhat.com> <49AC78C9.2070709@redhat.com> <20090303005738.GA22376@movementarian.org> Message-ID: <49ADA6EA.7060809@redhat.com> John Levon wrote: > On Mon, Mar 02, 2009 at 07:24:41PM -0500, Cole Robinson wrote: > >> Do you think it should be in all relevant man pages (virt-convert, >> virt-image, virt-install) or just have the other two link back to >> virt-install? > > Just the one please :) > Cool, 'python setup.py sdist' will now stuff the current osdict list in virt-install.pod and remake the man files (I did this only for sdist, so there isn't a build requirement on pod2man). Thanks, Cole From crobinso at redhat.com Tue Mar 3 21:54:33 2009 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 03 Mar 2009 16:54:33 -0500 Subject: [et-mgmt-tools] [PATCH] Add --os-* options to virt-image In-Reply-To: <49AC65D6.9020001@redhat.com> References: <49AC65D6.9020001@redhat.com> Message-ID: <49ADA719.5080104@redhat.com> Cole Robinson wrote: > The attached patched allows specifying --os-type and --os-variant for > virt-image (as well as the --list-os-options command just posted). > > At some point we should support this in the virt-image format, but until > then this will suffice. > I committed this (minus the --list-os-options change which has been dropped outright). Thanks, Cole From crobinso at redhat.com Wed Mar 4 00:39:18 2009 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 03 Mar 2009 19:39:18 -0500 Subject: [et-mgmt-tools] [ANNOUNCE] New release virtinst 0.400.2 Message-ID: <49ADCDB6.3030204@redhat.com> I'm happy to announce a new virtinst release, version 0.400.2. The release can be downloaded from: http://virt-manager.org/download.html The direct download link is: http://virt-manager.org/download/sources/virtinst/virtinst-0.400.2.tar.gz This release includes: - New virt-clone option --original-xml, allows cloning a guest from an xml file, rather than require an existing, defined guest. - New virt-install option --import, allows creating a guest from an existing disk image, bypassing any OS install phase. - New virt-install option --host-device, for connecting a physical host device to the guest. - Allow specifying 'cache' value via virt-install's --disk options (Ben Kochie) - New virt-install option --nonetworks (John Levon) - Lots of backend cleanups and documentation improvements. Thanks to everyone who has contributed to this release through testing, bug reporting, submitting patches, and otherwise sending in feedback! Thanks, Cole From radek at eadresa.cz Wed Mar 4 02:56:30 2009 From: radek at eadresa.cz (Radek Hladik) Date: Wed, 04 Mar 2009 03:56:30 +0100 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49AD6CA8.3010707@redhat.com> References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> <49ABE88A.5030702@redhat.com> <49AC1F3C.6070702@redhat.com> <49AC5219.6000105@redhat.com> <49AD6CA8.3010707@redhat.com> Message-ID: <49ADEDDE.4040701@eadresa.cz> Cole Robinson napsal(a): > Subhendu Ghosh wrote: >> Bryan Kearney wrote: >>> Cole Robinson wrote: >>>> Bryan Kearney wrote: >>>>> Cole Robinson wrote: >>>>>> Hi all, > >> Also the wizard should support multiple NICs and disks - perhaps from >> step 5 - be able to launch sub-wizards to add nic and add disk? >> > > I think such a process would add non-trivial complication to the UI for > a rather niche case. For a user wanting a multi nic or disk configured > VM, I would point them at virt-install which easily scales in this > respect. Extra NICs could also be added post install, since I would > think only one would be required for the install process. > My opinion is that this can be solved by some sort of "do not start the VM and open VM details for fine tuning" option. I am not sure where I've seen this (VMWare ESX maybe?) but the last page of the wizard contained checkboxes like "start machine now" and "edit VM details after closing the wizard". With reasonable default values you can satisfy both advanced user and a "just to be done" user in a straight-forward way. Radek From sghosh at redhat.com Wed Mar 4 04:09:23 2009 From: sghosh at redhat.com (Subhendu Ghosh) Date: Tue, 03 Mar 2009 23:09:23 -0500 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49ADEDDE.4040701@eadresa.cz> References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> <49ABE88A.5030702@redhat.com> <49AC1F3C.6070702@redhat.com> <49AC5219.6000105@redhat.com> <49AD6CA8.3010707@redhat.com> <49ADEDDE.4040701@eadresa.cz> Message-ID: <49ADFEF3.6080301@redhat.com> Radek Hladik wrote: > Cole Robinson napsal(a): >> Subhendu Ghosh wrote: >>> Bryan Kearney wrote: >>>> Cole Robinson wrote: >>>>> Bryan Kearney wrote: >>>>>> Cole Robinson wrote: >>>>>>> Hi all, >> >>> Also the wizard should support multiple NICs and disks - perhaps from >>> step 5 - be able to launch sub-wizards to add nic and add disk? >>> >> >> I think such a process would add non-trivial complication to the UI for >> a rather niche case. For a user wanting a multi nic or disk configured >> VM, I would point them at virt-install which easily scales in this >> respect. Extra NICs could also be added post install, since I would >> think only one would be required for the install process. >> > My opinion is that this can be solved by some sort of "do not start the > VM and open VM details for fine tuning" option. I am not sure where I've > seen this (VMWare ESX maybe?) but the last page of the wizard contained > checkboxes like "start machine now" and "edit VM details after closing > the wizard". With reasonable default values you can satisfy both > advanced user and a "just to be done" user in a straight-forward way. > Radek > Radek's option seems to make good sense. It also has the added benefit of being able to create a VM but not starting it and containing the complexity of the wizard. This actually leads to another question: If we select PXE, do subsequent reboots also enable PXE? or is PXE only enable for the install? Can we have systems that will always run PXE? eg, stateless/diskless VMs? -subhendu -- Subhendu Ghosh Red Hat Phone: +1-201-232-2851 Email: sghosh at redhat.com From crobinso at redhat.com Wed Mar 4 14:41:43 2009 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 04 Mar 2009 09:41:43 -0500 Subject: [et-mgmt-tools] RFC: virt-manager: Redesigned 'New VM' wizard In-Reply-To: <49ADFEF3.6080301@redhat.com> References: <49A86E27.20605@redhat.com> <49ABD1B3.9030004@redhat.com> <49ABE88A.5030702@redhat.com> <49AC1F3C.6070702@redhat.com> <49AC5219.6000105@redhat.com> <49AD6CA8.3010707@redhat.com> <49ADEDDE.4040701@eadresa.cz> <49ADFEF3.6080301@redhat.com> Message-ID: <49AE9327.8070504@redhat.com> Subhendu Ghosh wrote: > Radek Hladik wrote: >> Cole Robinson napsal(a): >>> Subhendu Ghosh wrote: >>>> Bryan Kearney wrote: >>>>> Cole Robinson wrote: >>>>>> Bryan Kearney wrote: >>>>>>> Cole Robinson wrote: >>>>>>>> Hi all, >>> >>>> Also the wizard should support multiple NICs and disks - perhaps from >>>> step 5 - be able to launch sub-wizards to add nic and add disk? >>>> >>> >>> I think such a process would add non-trivial complication to the UI for >>> a rather niche case. For a user wanting a multi nic or disk configured >>> VM, I would point them at virt-install which easily scales in this >>> respect. Extra NICs could also be added post install, since I would >>> think only one would be required for the install process. >>> >> My opinion is that this can be solved by some sort of "do not start >> the VM and open VM details for fine tuning" option. I am not sure >> where I've seen this (VMWare ESX maybe?) but the last page of the >> wizard contained checkboxes like "start machine now" and "edit VM >> details after closing the wizard". With reasonable default values you >> can satisfy both advanced user and a "just to be done" user in a >> straight-forward way. >> Radek >> > > Radek's option seems to make good sense. It also has the added benefit > of being able to create a VM but not starting it and containing the > complexity of the wizard. > I agree. Offering to just define the domain and launch the details wizard for extra tweaking should address anyone with specific requirements. We would certainly need to fix up things a bit in the backend to make it work though, so this won't happen for this initial push. > This actually leads to another question: > If we select PXE, do subsequent reboots also enable PXE? or is PXE only > enable for the install? > > Can we have systems that will always run PXE? eg, stateless/diskless VMs? > If you do a PXE install and skip adding storage (there is a checkbox to that effect) the VM will boot PXE on every subsequent reboot. If you do add storage, it will fall back to a HD boot (as though you simply installed from PXE) but that is a simple change in the VM Details section. Thanks, Cole From quentusrex at gmail.com Wed Mar 4 20:14:25 2009 From: quentusrex at gmail.com (William King) Date: Wed, 04 Mar 2009 12:14:25 -0800 Subject: [et-mgmt-tools] There seems to be an issue with the 'file browse' button Message-ID: <49AEE121.9010905@gmail.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I have updated to virt-manager 6.1 and libvirt 6.1. On the screen where I am trying to provision a remote server, with local install media, the button where it should let me browse for the iso is greyed out. And if I manually put the location of the media in the box, it still says it can't find it. - -William -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkmu4R0ACgkQZ63Azu7rahr1kwCeKHul+3+oBocR2m/YitTy3Xlf oQYAoKPnIXi5Dctvth6DBZu2lMaqQ1RU =7ei9 -----END PGP SIGNATURE----- From crobinso at redhat.com Wed Mar 4 20:54:21 2009 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 04 Mar 2009 15:54:21 -0500 Subject: [et-mgmt-tools] There seems to be an issue with the 'file browse' button In-Reply-To: <49AEE121.9010905@gmail.com> References: <49AEE121.9010905@gmail.com> Message-ID: <49AEEA7D.1030205@redhat.com> William King wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I have updated to virt-manager 6.1 and libvirt 6.1. > > On the screen where I am trying to provision a remote server, with local > install media, the button where it should let me browse for the iso is > greyed out. And if I manually put the location of the media in the box, > it still says it can't find it. The latest virt-manager version does not have a browser dialog for remote systems yet. The reason it likely cant find the media on the remote host is that the media needs to be 'managed' by libvirt. If you are using an iso file, you'll have to add the parent directory as libvirt storage pool. See http://virt-manager.org/page/RemoteVMCreation for more info. - Cole From agx at sigxcpu.org Thu Mar 5 07:51:20 2009 From: agx at sigxcpu.org (Guido =?iso-8859-1?Q?G=FCnther?=) Date: Thu, 5 Mar 2009 08:51:20 +0100 Subject: [et-mgmt-tools] Configuration of the virt-manager In-Reply-To: <200902261044.12753.lugs@linuxmaker.de> References: <200902261044.12753.lugs@linuxmaker.de> Message-ID: <20090305075120.GA14945@bogon.ms20.nix> On Thu, Feb 26, 2009 at 10:44:12AM +0100, lugs at linuxmaker.de wrote: > Hello, > > I'm looking now already since days after the configuration of > virt-manager on Debian, but found nothing on the internet or any literature. > I now have a machine on Debian/Lenny with KVM and wants use here virtual > machines on logical volumes. With virt-manager should indeed be a breeze, but > it is not, because the configuration of virt-manager is still opaque to me. > > If I have virt-manager and the traces on the workstation installed the program > I want to start now, I get the messages > > Unable to open connection to hypervisor URI 'qemu+ssh://192.168.0.3/system': > Die Verbindung wurde vom Kommunikationspartner > zur?ckgesetzt > Traceback (most recent call last): > File "/usr/share/virt-manager/virtManager/connection.py", line 486, in > _open_thread > None], flags) > File "/usr/lib/python2.5/site-packages/libvirt.py", line 99, in openAuth > if ret is None:raise libvirtError('virConnectOpenAuth() failed') > libvirtError: Die Verbindung wurde vom Kommunikationspartner zur?ckgesetzt Did you setup ssh-keys to be able to login into the machine without prompting? -- Guido From crobinso at redhat.com Fri Mar 6 15:27:18 2009 From: crobinso at redhat.com (Cole Robinson) Date: Fri, 06 Mar 2009 10:27:18 -0500 Subject: [et-mgmt-tools] [PATCH] virt-manager: Storage aware delete dialog Message-ID: <49B140D6.9060301@redhat.com> Hi all, The attached patch adds a storage aware delete dialog to virt-manager. When deleting a VM, we are presented with a list of storage attached to it, with an option to remove individual disks as part of the delete process. Some screenshots: http://fedorapeople.org/~crobinso/virt-manager/delete/vmm-delete-2.1.png The dialog's look when it is launched (we make the user opt in to delete storage). http://fedorapeople.org/~crobinso/virt-manager/delete/vmm-delete-2.2.png Even when the user opts in, we don't select all disks for deletion by default. We won't select a disk by default if: It is marked as read-only in the xml It is marked as shareable in the xml It is in use by another VM on the same connection A tooltip to this effect is shown when hovering over the warning icon. http://fedorapeople.org/~crobinso/virt-manager/delete/vmm-delete-2.4.png If we think we can't delete certain storage, the check box is marked as inconsistent. There are multiple reasons we can't delete (no permissions, storage is an iSCSI volume, storage isn't managed and we are on a remote connection, etc.) The reasoning is also shown in a tooltip. If any storage deletion fails, we continue with the process, and just show all aggregated errors to the user at the end. The VM removal will be attempted regardless. That's about it. Questions or comments appreciated. Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-manager-delete-dialog.patch Type: text/x-diff Size: 25827 bytes Desc: not available URL: From crobinso at redhat.com Fri Mar 6 17:17:56 2009 From: crobinso at redhat.com (Cole Robinson) Date: Fri, 06 Mar 2009 12:17:56 -0500 Subject: [et-mgmt-tools] [PATCH] virt-manager: Redesigned 'New VM' wizard ver 2 Message-ID: <49B15AC4.1090206@redhat.com> Attached is an updated version of the redesigned 'New VM' wizard. Changes since the first posting: - Connection list shows a human readable string, rather than the full connection URI. Ex: http://fedorapeople.org/~crobinso/virt-manager/newvm3/vmm-newvm-3-connlist.png We explicitly show QEMU/KVM if the connection supports KVM. This helps clear up the confusion that a 'qemu' connection is needed for kvm usage. - Sparse vs. Non-sparse option added for storage. Ex: http://fedorapeople.org/~crobinso/virt-manager/newvm3/vmm-newvm-3-sparse.png - Distro Type/Variant detection implemented for URL installs (only fully works for RH distros at the moment). Ex: http://fedorapeople.org/~crobinso/virt-manager/newvm3/vmm-newvm-3-detect.png If the user selects a previously entered distro from the drop down, or press activates (presses 'enter' in) the text entry, it kicks of detection. Questions or comments appreciated. Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-manager-newvm-02.patch Type: text/x-diff Size: 449040 bytes Desc: not available URL: From lfarkas at lfarkas.org Sun Mar 8 11:47:38 2009 From: lfarkas at lfarkas.org (Farkas Levente) Date: Sun, 08 Mar 2009 12:47:38 +0100 Subject: [et-mgmt-tools] latest virt-manager and virtinst on centos-5 Message-ID: <49B3B05A.4030707@lfarkas.org> hi, the these packages: virt-manager-0.6.1-4 python-virtinst-0.400.2-3 libvirt-python-0.6.1-1 libvirt-0.6.1-1 on centos-5 gives this error when i try to install a new guest after press finish in virt manager: -------------------------------------------- Unable to complete install 'libvirt.libvirtError internal error unable to start guest: Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/create.py", line 717, in do_install dom = guest.start_install(False, meter = meter) File "/usr/lib/python2.4/site-packages/virtinst/Guest.py", line 536, in start_install return self._do_install(consolecb, meter, removeOld, wait) File "/usr/lib/python2.4/site-packages/virtinst/Guest.py", line 628, in _do_install self.domain = self.conn.createLinux(install_xml, 0) File "/usr/lib64/python2.4/site-packages/libvirt.py", line 973, in createLinux if ret is None:raise libvirtError('virDomainCreateLinux() failed', conn=self) libvirtError: internal error unable to start guest: ' -------------------------------------------- yours. -- Levente "Si vis pacem para bellum!" From crobinso at redhat.com Sun Mar 8 21:10:18 2009 From: crobinso at redhat.com (Cole Robinson) Date: Sun, 08 Mar 2009 17:10:18 -0400 Subject: [et-mgmt-tools] [PATCH] Add libvirt storage browser Message-ID: <49B4343A.4070900@redhat.com> The attached patch adds a libvirt-storage-aware browser, to replace the standard file chooser when adding storage on remote connections. Screenshot here: http://fedorapeople.org/~crobinso/virt-manager/vmm-storage-browse.png (Greyed out pools are inactive.) In the near future, this will be enabled for local connections as well, which will allow provisioning more complex storage on demand, as well as encourage all storage to go through the libvirt APIs (while still allowing the use of unmanaged storage through the 'Browse local' option, or manually entering a path). Before we turn it on in the local case though, default pool setup needs to be more configurable and robust. Questions or comments appreciated. Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-manager-storage-browse.patch Type: text/x-diff Size: 30697 bytes Desc: not available URL: From crobinso at redhat.com Sun Mar 8 23:34:46 2009 From: crobinso at redhat.com (Cole Robinson) Date: Sun, 08 Mar 2009 19:34:46 -0400 Subject: [et-mgmt-tools] [PATCH] virt-manager: 'Add Physical Host Device' wizard Message-ID: <49B45616.3090201@redhat.com> Hi all, The attached patch adds an option in the 'Add Hardware' wizard to attach a physical host device to a VM. Current explicitly supported options are USB, PCI, and Wired NICs (which really just attaches the parent PCI device, but this way provides a nicer interface). Some screenshots: http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev1.png http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev2.png http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev3.png It's all pretty straightforward: most of the work is done at the virtinst level to pull the useful info out of the nodedev xml. Questions or comments appreciated. Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-manager-add-hostdev.patch Type: text/x-diff Size: 141469 bytes Desc: not available URL: From berrange at redhat.com Mon Mar 9 10:55:18 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 9 Mar 2009 10:55:18 +0000 Subject: [et-mgmt-tools] [PATCH] Add libvirt storage browser In-Reply-To: <49B4343A.4070900@redhat.com> References: <49B4343A.4070900@redhat.com> Message-ID: <20090309105518.GQ4578@redhat.com> On Sun, Mar 08, 2009 at 05:10:18PM -0400, Cole Robinson wrote: > The attached patch adds a libvirt-storage-aware browser, to replace the > standard file chooser when adding storage on remote connections. > Screenshot here: > > http://fedorapeople.org/~crobinso/virt-manager/vmm-storage-browse.png > > (Greyed out pools are inactive.) > > In the near future, this will be enabled for local connections as well, > which will allow provisioning more complex storage on demand, as well as > encourage all storage to go through the libvirt APIs (while still > allowing the use of unmanaged storage through the 'Browse local' option, > or manually entering a path). Before we turn it on in the local case > though, default pool setup needs to be more configurable and robust. Personally I'd turn it on for both local & remote usage right away. With real users trying it out all the time, it'll quickly get debugged & more robust. I think having & documenting 2 different ways of doing storage will cause more trouble that you'll save. The dialog looks good btw - glad we're finally able todo this now. Daniel -- |: Red Hat, Engineering, London -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 Mon Mar 9 10:59:20 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Mon, 9 Mar 2009 10:59:20 +0000 Subject: [et-mgmt-tools] [PATCH] virt-manager: 'Add Physical Host Device' wizard In-Reply-To: <49B45616.3090201@redhat.com> References: <49B45616.3090201@redhat.com> Message-ID: <20090309105920.GR4578@redhat.com> On Sun, Mar 08, 2009 at 07:34:46PM -0400, Cole Robinson wrote: > Hi all, > > The attached patch adds an option in the 'Add Hardware' wizard to attach > a physical host device to a VM. Current explicitly supported options are > USB, PCI, and Wired NICs (which really just attaches the parent PCI > device, but this way provides a nicer interface). > > Some screenshots: > > http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev1.png > http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev2.png > http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev3.png > > It's all pretty straightforward: most of the work is done at the > virtinst level to pull the useful info out of the nodedev xml. I'm not sure I like having 'Wired NICs' as a separate option. How about listing the logic device type against each PCI / USB devices. eg, in the drop down for PCI devices, alongside the POCI device description, also include the nameof the logical device (eth1) associated with it. This lets you see that assigning PCI device 'X.y.z' means the host will no longer be able to use the device 'eth1'. I think it would help people to avoid shooting themselvs in the foot when choosing a PCI/USB device Regards, Daniel -- |: Red Hat, Engineering, London -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 Mon Mar 9 15:30:02 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 09 Mar 2009 11:30:02 -0400 Subject: [et-mgmt-tools] [PATCH] virt-manager: 'Add Physical Host Device' wizard In-Reply-To: <20090309105920.GR4578@redhat.com> References: <49B45616.3090201@redhat.com> <20090309105920.GR4578@redhat.com> Message-ID: <49B535FA.7060303@redhat.com> Daniel P. Berrange wrote: > On Sun, Mar 08, 2009 at 07:34:46PM -0400, Cole Robinson wrote: >> Hi all, >> >> The attached patch adds an option in the 'Add Hardware' wizard to attach >> a physical host device to a VM. Current explicitly supported options are >> USB, PCI, and Wired NICs (which really just attaches the parent PCI >> device, but this way provides a nicer interface). >> >> Some screenshots: >> >> http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev1.png >> http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev2.png >> http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev3.png >> >> It's all pretty straightforward: most of the work is done at the >> virtinst level to pull the useful info out of the nodedev xml. > > I'm not sure I like having 'Wired NICs' as a separate option. How about > listing the logic device type against each PCI / USB devices. > > eg, in the drop down for PCI devices, alongside the POCI device > description, also include the nameof the logical device (eth1) > associated with it. This lets you see that assigning PCI device > 'X.y.z' means the host will no longer be able to use the device > 'eth1'. I think it would help people to avoid shooting themselvs > in the foot when choosing a PCI/USB device > Sounds good, updated patch attached. Attaching a network card now looks like: http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev4.png Thanks, Cole -------------- next part -------------- A non-text attachment was scrubbed... Name: virt-manager-add-hostdev2.patch Type: text/x-diff Size: 141656 bytes Desc: not available URL: From crobinso at redhat.com Mon Mar 9 15:36:13 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 09 Mar 2009 11:36:13 -0400 Subject: [et-mgmt-tools] [PATCH] Add libvirt storage browser In-Reply-To: <20090309105518.GQ4578@redhat.com> References: <49B4343A.4070900@redhat.com> <20090309105518.GQ4578@redhat.com> Message-ID: <49B5376D.4030400@redhat.com> Daniel P. Berrange wrote: > On Sun, Mar 08, 2009 at 05:10:18PM -0400, Cole Robinson wrote: >> The attached patch adds a libvirt-storage-aware browser, to replace the >> standard file chooser when adding storage on remote connections. >> Screenshot here: >> >> http://fedorapeople.org/~crobinso/virt-manager/vmm-storage-browse.png >> >> (Greyed out pools are inactive.) >> >> In the near future, this will be enabled for local connections as well, >> which will allow provisioning more complex storage on demand, as well as >> encourage all storage to go through the libvirt APIs (while still >> allowing the use of unmanaged storage through the 'Browse local' option, >> or manually entering a path). Before we turn it on in the local case >> though, default pool setup needs to be more configurable and robust. > > Personally I'd turn it on for both local & remote usage right away. With > real users trying it out all the time, it'll quickly get debugged & more > robust. I think having & documenting 2 different ways of doing storage > will cause more trouble that you'll save. > Yeah, I'm unsure on the whole thing, but I'm leaning in your direction now. The situation with file browsing is just as confusing if running as nonroot at the moment, and the storage dialog will enable better functionality. The big issue is that this is new UI and there are no docs to point at when users have questions. More work for me I suppose :) I'll enable this for storage provisioning with the new release, not local install media though (since we really don't have anything storage pool related hooked up cleanly for this case). Thanks, Cole From crobinso at redhat.com Mon Mar 9 20:26:23 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 09 Mar 2009 16:26:23 -0400 Subject: [et-mgmt-tools] [PATCH] virt-manager: Redesigned 'New VM' wizard ver 2 In-Reply-To: <49B15AC4.1090206@redhat.com> References: <49B15AC4.1090206@redhat.com> Message-ID: <49B57B6F.5040009@redhat.com> Cole Robinson wrote: > Attached is an updated version of the redesigned 'New VM' wizard. > Changes since the first posting: > > - Connection list shows a human readable string, rather than the full > connection URI. Ex: > > http://fedorapeople.org/~crobinso/virt-manager/newvm3/vmm-newvm-3-connlist.png > > We explicitly show QEMU/KVM if the connection supports KVM. This helps > clear up the confusion that a 'qemu' connection is needed for kvm usage. > > > - Sparse vs. Non-sparse option added for storage. Ex: > > http://fedorapeople.org/~crobinso/virt-manager/newvm3/vmm-newvm-3-sparse.png > > > - Distro Type/Variant detection implemented for URL installs (only fully > works for RH distros at the moment). Ex: > > http://fedorapeople.org/~crobinso/virt-manager/newvm3/vmm-newvm-3-detect.png > > If the user selects a previously entered distro from the drop down, or > press activates (presses 'enter' in) the text entry, it kicks of detection. > This has been pushed now. Thanks, Cole From crobinso at redhat.com Mon Mar 9 20:26:56 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 09 Mar 2009 16:26:56 -0400 Subject: [et-mgmt-tools] [PATCH] virt-manager: Storage aware delete dialog In-Reply-To: <49B140D6.9060301@redhat.com> References: <49B140D6.9060301@redhat.com> Message-ID: <49B57B90.6060604@redhat.com> Cole Robinson wrote: > Hi all, > > The attached patch adds a storage aware delete dialog to virt-manager. > When deleting a VM, we are presented with a list of storage attached to > it, with an option to remove individual disks as part of the delete process. > > Some screenshots: > > http://fedorapeople.org/~crobinso/virt-manager/delete/vmm-delete-2.1.png > > The dialog's look when it is launched (we make the user opt in to delete > storage). > > http://fedorapeople.org/~crobinso/virt-manager/delete/vmm-delete-2.2.png > > Even when the user opts in, we don't select all disks for deletion by > default. We won't select a disk by default if: > > It is marked as read-only in the xml > It is marked as shareable in the xml > It is in use by another VM on the same connection > > A tooltip to this effect is shown when hovering over the warning icon. > > http://fedorapeople.org/~crobinso/virt-manager/delete/vmm-delete-2.4.png > > If we think we can't delete certain storage, the check box is marked as > inconsistent. There are multiple reasons we can't delete (no > permissions, storage is an iSCSI volume, storage isn't managed and we > are on a remote connection, etc.) The reasoning is also shown in a tooltip. > > If any storage deletion fails, we continue with the process, and just > show all aggregated errors to the user at the end. The VM removal will > be attempted regardless. > This has been pushed now. Thanks, Cole From crobinso at redhat.com Mon Mar 9 20:27:46 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 09 Mar 2009 16:27:46 -0400 Subject: [et-mgmt-tools] virt-manager: acquire kerberos tgt In-Reply-To: <20090228181909.GA7612@bogon.ms20.nix> References: <20090228181909.GA7612@bogon.ms20.nix> Message-ID: <49B57BC2.4070701@redhat.com> Guido G?nther wrote: > Hi, > recent krb5-auth-dialog svn has a dbus interface which can be used to > acquire a tgt. Here's a simple patch for virt-manager to use it: > http://honk.sigxcpu.org/unsorted-patches/virt-manager_acquire-tgt.diff > Cheers, > -- Guido > Thanks, applied (along with the ssh-askpass hint): http://hg.et.redhat.com/cgi-bin/hg-virt.cgi/applications/virt-manager--devel/rev/3f37d0519b17 - Cole From crobinso at redhat.com Mon Mar 9 20:29:28 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 09 Mar 2009 16:29:28 -0400 Subject: [et-mgmt-tools] latest virt-manager and virtinst on centos-5 In-Reply-To: <49B3B05A.4030707@lfarkas.org> References: <49B3B05A.4030707@lfarkas.org> Message-ID: <49B57C28.5010308@redhat.com> Farkas Levente wrote: > hi, > the these packages: > virt-manager-0.6.1-4 > python-virtinst-0.400.2-3 > libvirt-python-0.6.1-1 > libvirt-0.6.1-1 > on centos-5 gives this error when i try to install a new guest after > press finish in virt manager: > -------------------------------------------- > Unable to complete install 'libvirt.libvirtError internal error unable > to start guest: > Traceback (most recent call last): > File "/usr/share/virt-manager/virtManager/create.py", line 717, in > do_install > dom = guest.start_install(False, meter = meter) > File "/usr/lib/python2.4/site-packages/virtinst/Guest.py", line 536, > in start_install > return self._do_install(consolecb, meter, removeOld, wait) > File "/usr/lib/python2.4/site-packages/virtinst/Guest.py", line 628, > in _do_install > self.domain = self.conn.createLinux(install_xml, 0) > File "/usr/lib64/python2.4/site-packages/libvirt.py", line 973, in > createLinux > if ret is None:raise libvirtError('virDomainCreateLinux() failed', > conn=self) > libvirtError: internal error unable to start guest: > ' This was a libvirt bug that should have been fixed in the latest release. Could be some other variant of it since you are running centos. - Cole From crobinso at redhat.com Mon Mar 9 20:31:08 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 09 Mar 2009 16:31:08 -0400 Subject: [et-mgmt-tools] [PATCH] Add libvirt storage browser In-Reply-To: <49B5376D.4030400@redhat.com> References: <49B4343A.4070900@redhat.com> <20090309105518.GQ4578@redhat.com> <49B5376D.4030400@redhat.com> Message-ID: <49B57C8C.3060601@redhat.com> Cole Robinson wrote: > Daniel P. Berrange wrote: >> On Sun, Mar 08, 2009 at 05:10:18PM -0400, Cole Robinson wrote: >>> The attached patch adds a libvirt-storage-aware browser, to replace the >>> standard file chooser when adding storage on remote connections. >>> Screenshot here: >>> >>> http://fedorapeople.org/~crobinso/virt-manager/vmm-storage-browse.png >>> >>> (Greyed out pools are inactive.) >>> >>> In the near future, this will be enabled for local connections as well, >>> which will allow provisioning more complex storage on demand, as well as >>> encourage all storage to go through the libvirt APIs (while still >>> allowing the use of unmanaged storage through the 'Browse local' option, >>> or manually entering a path). Before we turn it on in the local case >>> though, default pool setup needs to be more configurable and robust. >> Personally I'd turn it on for both local & remote usage right away. With >> real users trying it out all the time, it'll quickly get debugged & more >> robust. I think having & documenting 2 different ways of doing storage >> will cause more trouble that you'll save. >> > > Yeah, I'm unsure on the whole thing, but I'm leaning in your direction > now. The situation with file browsing is just as confusing if running as > nonroot at the moment, and the storage dialog will enable better > functionality. The big issue is that this is new UI and there are no > docs to point at when users have questions. More work for me I suppose :) > > I'll enable this for storage provisioning with the new release, not > local install media though (since we really don't have anything storage > pool related hooked up cleanly for this case). > Applied now, with the above defaults change. Thanks, Cole From crobinso at redhat.com Mon Mar 9 20:31:31 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 09 Mar 2009 16:31:31 -0400 Subject: [et-mgmt-tools] [PATCH] virt-manager: 'Add Physical Host Device' wizard In-Reply-To: <49B535FA.7060303@redhat.com> References: <49B45616.3090201@redhat.com> <20090309105920.GR4578@redhat.com> <49B535FA.7060303@redhat.com> Message-ID: <49B57CA3.3000902@redhat.com> Cole Robinson wrote: > Daniel P. Berrange wrote: >> On Sun, Mar 08, 2009 at 07:34:46PM -0400, Cole Robinson wrote: >>> Hi all, >>> >>> The attached patch adds an option in the 'Add Hardware' wizard to attach >>> a physical host device to a VM. Current explicitly supported options are >>> USB, PCI, and Wired NICs (which really just attaches the parent PCI >>> device, but this way provides a nicer interface). >>> >>> Some screenshots: >>> >>> http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev1.png >>> http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev2.png >>> http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev3.png >>> >>> It's all pretty straightforward: most of the work is done at the >>> virtinst level to pull the useful info out of the nodedev xml. >> I'm not sure I like having 'Wired NICs' as a separate option. How about >> listing the logic device type against each PCI / USB devices. >> >> eg, in the drop down for PCI devices, alongside the POCI device >> description, also include the nameof the logical device (eth1) >> associated with it. This lets you see that assigning PCI device >> 'X.y.z' means the host will no longer be able to use the device >> 'eth1'. I think it would help people to avoid shooting themselvs >> in the foot when choosing a PCI/USB device >> > > Sounds good, updated patch attached. Attaching a network card now looks > like: > > http://fedorapeople.org/~crobinso/virt-manager/vmm-add-hostdev4.png > I've applied this version. Thanks, Cole From crobinso at redhat.com Tue Mar 10 03:47:21 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 09 Mar 2009 23:47:21 -0400 Subject: [et-mgmt-tools] [ANNOUNCE] New release virtinst 0.400.3 Message-ID: <49B5E2C9.5070600@redhat.com> I'm happy to announce a new virtinst release, version 0.400.3. The release can be downloaded from: http://virt-manager.org/download.html The direct download link is: http://virt-manager.org/download/sources/virtinst/virtinst-0.400.3.tar.gz This is largely a bug fix release, which fixes several annoying bugs present in virtinst-0.400.2: virt-install --file was busted virt-install --os-type windows was busted for --cdrom installs virt-install --os-variant values weren't used if installing from a URL Thanks to those who helped track down the bugs! - Cole From crobinso at redhat.com Tue Mar 10 03:47:58 2009 From: crobinso at redhat.com (Cole Robinson) Date: Mon, 09 Mar 2009 23:47:58 -0400 Subject: [et-mgmt-tools] [ANNOUNCE] New release virt-manager 0.7.0 Message-ID: <49B5E2EE.90402@redhat.com> I'm happy to announce a new virt-manager release, version 0.7.0. The release can be downloaded from: http://virt-manager.org/download.html The direct download link is: http://virt-manager.org/download/sources/virt-manager/virt-manager-0.7.0.tar.gz This release includes: - Redesigned 'New Virtual Machine' wizard (Jeremy Perry, Tim Allen, Cole Robinson) - Option to remove storage when deleting a virtual machine. - File browser for libvirt storage pools and volumes, for use when attaching storage to a new or existing guest. - Physical device assignment (PCI, USB) for existing virtual machines. - Bug fixes and minor improvements. Thanks to everyone who has contributed to this release through testing, bug reporting, submitting patches, and otherwise sending in feedback! Thanks, Cole From nick.haunold at det.wa.edu.au Tue Mar 10 05:03:20 2009 From: nick.haunold at det.wa.edu.au (Nick Haunold) Date: Tue, 10 Mar 2009 14:03:20 +0900 Subject: [et-mgmt-tools] P2V testing with redhat XEN Message-ID: <20090310140320419.00000004340@DET-1150672C331> Generator Microsoft Word 11 (filtered medium) Hi I was attempting to test the P2V iso to move some hosts into a Redhat xen env but when I boot the server from the p2v disk I get a kernel panic. It looks as tho it is unable to find a driver for the scsi controller on this particular server. It is a Dell 1855 blade server. Could you please let me know if the initrd contains the driver for this server? The OS running on this server looks to be using the mptbase driver. Thanks Nick -------------- next part -------------- An HTML attachment was scrubbed... URL: From zyphos at zyphos.be Tue Mar 10 10:13:32 2009 From: zyphos at zyphos.be (Zyphos) Date: Tue, 10 Mar 2009 11:13:32 +0100 Subject: [et-mgmt-tools] [BUG] Performance Memory usage graph show memory used even if guest is down. Message-ID: <49B63D4C.3090401@zyphos.be> Hi, I am using, libvirt 0.6.1, Virt-manager 0.7.0, KVM Bug description: The performance memory usage graph for guest show that memory is used even if the guest is not running. The numeric values are correct only the graph is wrong. PS: I didn't find any bug report system other than this Mailing list. Best regards Zyphos From takatom at jp.fujitsu.com Wed Mar 11 05:43:43 2009 From: takatom at jp.fujitsu.com (Takahashi Tomohiro) Date: Wed, 11 Mar 2009 14:43:43 +0900 Subject: [et-mgmt-tools] [Patch] [virtinst] Fix check keymap option in virt-install Message-ID: <49B74F8F.20400@jp.fujitsu.com> Hi, I made the patch that corrected the following problem. It is that virt-install hangs up, when I set keymap which doesn't exist in keytable. ex) virt-install ... --keymap=abc If you have any comments, please suggest me. Signed-off-by: Tomohiro Takahashi Thanks, Tomohiro Takahashi -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: fix_check_keymap_cli.patch URL: From dk at in-telegence.net Wed Mar 11 10:46:04 2009 From: dk at in-telegence.net (Dominik Klein) Date: Wed, 11 Mar 2009 11:46:04 +0100 Subject: [et-mgmt-tools] Some (probably) beginner questions Message-ID: <49B7966C.3070409@in-telegence.net> Hi I am trying to setup libvirt, virt-manager and xen. I successfully installed xen 3.3.1 and am able to run machines with "xm". libvirt is 0.6.1 virt-manager was pulled 30 minutes ago from the repo same for virtinst libvirt is installed, libvirtd is running. I can also start virt-manager and have it connect to xen. Now I want to define a new machine with virt-manager and face a couple of problems. * I can only select full vs. paravirtualization at the end of the installation dialog and (I think) therefore not select a .iso file as installation source. The installation source has to be chosen in the very first dialog and I can only select "network install" there. * When I try to create a new machine (pv with a network installation source), I get the following error message: ------------ Unable to complete install ' POST operation failed: xend_post: error from xen daemon: (xend.err "Error creating domain: (1, 'Internal error', 'Could not open kernel image (2 = No such file or directory)')") Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/create.py", line 1485, in do_install dom = guest.start_install(False, meter = meter) File "/usr/lib64/python2.6/site-packages/virtinst/Guest.py", line 541, in start_install return self._do_install(consolecb, meter, removeOld, wait) File "/usr/lib64/python2.6/site-packages/virtinst/Guest.py", line 633, in _do_install self.domain = self.conn.createLinux(install_xml, 0) File "/usr/lib/python2.6/site-packages/libvirt.py", line 973, in createLinux if ret is None:raise libvirtError('virDomainCreateLinux() failed', conn=self) libvirtError: POST operation failed: xend_post: error from xen daemon: (xend.err "Error creating domain: (1, 'Internal error', 'Could not open kernel image (2 = No such file or directory)')") ------------ If you need more information about the system, please let me know - I'll provide them asap. Regards Dominik From crobinso at redhat.com Wed Mar 11 15:03:36 2009 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 11 Mar 2009 11:03:36 -0400 Subject: [et-mgmt-tools] Some (probably) beginner questions In-Reply-To: <49B7966C.3070409@in-telegence.net> References: <49B7966C.3070409@in-telegence.net> Message-ID: <49B7D2C8.2010804@redhat.com> Dominik Klein wrote: > Hi > > I am trying to setup libvirt, virt-manager and xen. > > I successfully installed xen 3.3.1 and am able to run machines with "xm". > libvirt is 0.6.1 > virt-manager was pulled 30 minutes ago from the repo > same for virtinst > > libvirt is installed, libvirtd is running. I can also start virt-manager > and have it connect to xen. > > Now I want to define a new machine with virt-manager and face a couple > of problems. > > * I can only select full vs. paravirtualization at the end of the > installation dialog and (I think) therefore not select a .iso file as > installation source. The installation source has to be chosen in the > very first dialog and I can only select "network install" there. > Hmm, if the connection supports FV and PV, all install options should be enabled. At the end of the wizard, is 'xen (fullvirt)' and option listed in the 'Virt Type' drop down? Can you provide 'virsh --connect xen:/// capabilities' output > * When I try to create a new machine (pv with a network installation > source), I get the following error message: > > ------------ > Unable to complete install ' POST > operation failed: xend_post: error from xen daemon: (xend.err "Error > creating domain: (1, 'Internal error', 'Could not open kernel image (2 = > No such file or directory)')") > Traceback (most recent call last): > File "/usr/share/virt-manager/virtManager/create.py", line 1485, in > do_install > dom = guest.start_install(False, meter = meter) > File "/usr/lib64/python2.6/site-packages/virtinst/Guest.py", line 541, > in start_install > return self._do_install(consolecb, meter, removeOld, wait) > File "/usr/lib64/python2.6/site-packages/virtinst/Guest.py", line 633, > in _do_install > self.domain = self.conn.createLinux(install_xml, 0) > File "/usr/lib/python2.6/site-packages/libvirt.py", line 973, in > createLinux > if ret is None:raise libvirtError('virDomainCreateLinux() failed', > conn=self) > libvirtError: POST operation failed: xend_post: error from xen daemon: > (xend.err "Error creating domain: (1, 'Internal error', 'Could not open > kernel image (2 = No such file or directory)')") > ------------ > Hmm, not entirely sure what the problem could be here. ~/.virt-manager/virt-manager.log could help. Thanks for the report! - Cole From Jonathan.Pregler at cnetokc.com Wed Mar 11 18:10:06 2009 From: Jonathan.Pregler at cnetokc.com (Jonathan Pregler) Date: Wed, 11 Mar 2009 13:10:06 -0500 Subject: [et-mgmt-tools] libvirt Live CD Message-ID: I have downloaded your live cd and converted a PC to a Xen VM. It works great and it save A LOT OF TIME!!! Thank you!! When I go to convert a server that has RAID it does not see the card correctly because it does not have the RAID card drivers. How can I get the RAID card drivers compiled into the Live CD to see the RAID Card? Thanks again and this is a wonderful conversion tool! Jonathan Pregler Senior Network Engineer C-Net Inc. jonathan.pregler at cnetokc.com 405.376.6644 Fax: 405.376.3036 CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is being sent by Cnet Inc. for the sole use of the intended recipient(s) and may contain confidential, proprietary and/or privileged information. Any unauthorized review, use, disclosure and/or distribution by any recipient is prohibited. If you are not the intended recipient, please delete and/or destroy all copies of this message regardless of form and any included attachments and notify Cnet Inc. immediately by contacting the sender via reply e-mail or forwarding to support at cnetokc.com. -------------- next part -------------- An HTML attachment was scrubbed... URL: From john.levon at sun.com Wed Mar 11 23:20:57 2009 From: john.levon at sun.com (John Levon) Date: Wed, 11 Mar 2009 16:20:57 -0700 Subject: [et-mgmt-tools] [PATCH] Support NetWare PV installs Message-ID: # HG changeset patch # User John Levon # Date 1236813503 25200 # Node ID bc57f31980f9a408db53e83e177a00931be0a479 # Parent 143cad71a2a6e83bfff7f46a0f7105e3eeb43ab2 Support NetWare PV installs Signed-off-by: John Levon diff --git a/virtinst/DistroInstaller.py b/virtinst/DistroInstaller.py --- a/virtinst/DistroInstaller.py +++ b/virtinst/DistroInstaller.py @@ -196,7 +196,8 @@ class DistroInstaller(Installer.Installe self.install["extraargs"] = args self._tmpfiles.append(kernelfn) - self._tmpfiles.append(initrdfn) + if initrdfn: + self._tmpfiles.append(initrdfn) # If they're installing off a local file/device, we map it # through to a virtual CD or disk diff --git a/virtinst/OSDistro.py b/virtinst/OSDistro.py --- a/virtinst/OSDistro.py +++ b/virtinst/OSDistro.py @@ -86,6 +86,8 @@ def _storeForDistro(fetcher, baseuri, ty stores.append(SolarisDistro) if distro == "solaris" or distro is None: stores.append(OpenSolarisDistro) + if distro == "netware" or distro is None: + stores.append(NetWareDistro) stores.append(GenericDistro) @@ -1022,6 +1024,9 @@ class SolarisDistro(SunDistro): self.initrdpath) class OpenSolarisDistro(SunDistro): + + os_variant = "opensolaris" + kernelpath = "platform/i86xpv/kernel/unix" initrdpath = "boot/x86.microroot" @@ -1065,3 +1070,21 @@ class OpenSolarisDistro(SunDistro): os.unlink(kernel) raise RuntimeError(_("OpenSolaris microroot not found at %s") % self.initrdpath) + +# NetWare 6 PV +class NetWareDistro(Distro): + name = "NetWare" + os_type = "other" + os_variant = "netware6" + + loaderpath = "STARTUP/XNLOADER.SYS" + + def isValidStore(self, fetcher, progresscb): + if fetcher.hasFile(self.loaderpath): + logging.debug("Detected NetWare") + return True + return False + + def acquireKernel(self, guest, fetcher, progresscb): + loader = fetcher.acquireFile(self.loaderpath, progresscb) + return (loader, "", "") diff --git a/virtinst/osdict.py b/virtinst/osdict.py --- a/virtinst/osdict.py +++ b/virtinst/osdict.py @@ -211,7 +211,7 @@ OS_TYPES = {\ "msdos": { "label": "MS-DOS", "acpi": False, "apic": False }, "netware4": { "label": "Novell Netware 4" }, "netware5": { "label": "Novell Netware 5" }, - "netware6": { "label": "Novell Netware 6" }, + "netware6": { "label": "Novell Netware 6", "pv_cdrom_install": True, }, "generic": { "label": "Generic" }, }, },} From terry.mcintyre at gmail.com Thu Mar 12 17:06:47 2009 From: terry.mcintyre at gmail.com (Terry McIntyre) Date: Thu, 12 Mar 2009 10:06:47 -0700 Subject: [et-mgmt-tools] no substitution for ip_address in kickstart file Message-ID: <67f096ea0903121006l45ed66e6x5e05132fd3a3f177@mail.gmail.com> I have a cobbler 1.4.1 installation, and recently a kickstart failed. Looking at the http://my.host/cblr/svc/op/ks/system/sapcps kickstart file, I find this line: network --hostname=sapcps.mydomain --bootproto=static --netmask=255.255.254.0 --ip=$ip_address --gateway=192.168.62.40 --device=eth0 --onboot=on The source line in /etc/cobbler/sample.ks is: network --hostname=${hostname} --bootproto=static --netmask=255.255.254.0 --ip=$ip_address --gateway=$gateway --device=eth0 --onboot=on $hostname and $gateway are expanded; $ip_address is not. These values come from /var/lib/cobbler/config/system.d/sapcps: --- comment: '' ctime: 0 depth: 2 gateway: '192.168.62.40' hostname: sapcps.mydomain image: '' interfaces: eth0: bonding: '' bonding_master: '' bonding_opts: '' dhcp_tag: '' dns_name: sapcps.mydomain ip_address: '192.168.62.184' mac_address: '00:21:5E:40:9F:64' static: True static_routes: [] subnet: '255.255.254.0' virt_bridge: '' kernel_options: gateway: '192.168.62.40' ip_address: '192.168.62.184' ip: '192.168.62.184' netmask: '255.255.254.0' kernel_options_post: {} kickstart: <> ks_meta: {} mgmt_classes: [] mtime: 0 name: sapcps name_servers: <> netboot_enabled: True owners: - admin parent: '' power_address: '' power_id: '' power_pass: '' power_type: ipmitool power_user: '' profile: Redhat4x64-x86_64 redhat_management_key: <> server: <> template_files: {} uid: MTIzNjM2MjM5MS45MTc4NTk5NzUuNDkyOQ virt_bridge: <> virt_cpus: <> virt_file_size: <> virt_path: <> virt_ram: <> virt_type: <> What am I missing? Thanks for any assistance. -- Terry McIntyre UNIX for hire Software Development, Systems Administration, Security terry.mcintyre at gmail.com From nick.haunold at det.wa.edu.au Fri Mar 13 02:18:37 2009 From: nick.haunold at det.wa.edu.au (Nick Haunold) Date: Fri, 13 Mar 2009 11:18:37 +0900 Subject: [et-mgmt-tools] P2V boot problems Message-ID: <20090313111837472.00000005248@DET-1150672C331> An HTML attachment was scrubbed... URL: From terry.mcintyre at gmail.com Fri Mar 13 15:45:52 2009 From: terry.mcintyre at gmail.com (Terry McIntyre) Date: Fri, 13 Mar 2009 08:45:52 -0700 Subject: [et-mgmt-tools] cobbler - kickstart differs depending on whether accessed via web or command line Message-ID: <67f096ea0903130845m46447c34o978dbd1e1830631c@mail.gmail.com> When I use "cobbler system hetks --name=sapcps | grep url", the result is "url --url=http://192.168.63.57:80/cblr/links/Redhat4x64" When I use the url from the kickstart process, taken from the access_log file, I get "url --url=$tree" The url is: /cblr/svc/op/ks/system/sapcps Should be seeing the same kickstart text either way, right? Any cluse? This is cobbler 1.4.1 installed via yum on a CentOs 5.2 server. Thanks for any assistance. -- Terry McIntyre UNIX for hire Software Development, Systems Administration, Security terry.mcintyre at gmail.com From terry.mcintyre at gmail.com Fri Mar 13 16:00:29 2009 From: terry.mcintyre at gmail.com (Terry McIntyre) Date: Fri, 13 Mar 2009 09:00:29 -0700 Subject: [et-mgmt-tools] Re: cobbler - kickstart differs depending on whether accessed via web or command line In-Reply-To: <67f096ea0903130845m46447c34o978dbd1e1830631c@mail.gmail.com> References: <67f096ea0903130845m46447c34o978dbd1e1830631c@mail.gmail.com> Message-ID: <67f096ea0903130900h1a786fbal368b24a198435f1f@mail.gmail.com> Found a clue. When kickstarting, the actual url used is /./cblr/svc/op/ks/system/sapcps When using "cobbler system getks", the leading "/." is not used. This makes all the difference. Now I need to find a clue as to why it makes a difference. On Fri, Mar 13, 2009 at 8:45 AM, Terry McIntyre wrote: > When I use "cobbler system hetks --name=sapcps | grep url", the result > is "url --url=http://192.168.63.57:80/cblr/links/Redhat4x64" > > When I use the url from the kickstart process, taken from the > access_log file, I get "url --url=$tree" > > The url is: /cblr/svc/op/ks/system/sapcps > > Should be seeing the same kickstart text either way, right? > > Any cluse? > > This is cobbler 1.4.1 installed via yum on a CentOs 5.2 server. > > Thanks for any assistance. > > > > -- > Terry McIntyre > UNIX for hire > Software Development, Systems Administration, Security > terry.mcintyre at gmail.com > -- Terry McIntyre UNIX for hire Software Development, Systems Administration, Security terry.mcintyre at gmail.com From mdehaan at redhat.com Fri Mar 13 17:30:05 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 13 Mar 2009 13:30:05 -0400 Subject: [et-mgmt-tools] Re: cobbler - kickstart differs depending on whether accessed via web or command line In-Reply-To: <67f096ea0903130900h1a786fbal368b24a198435f1f@mail.gmail.com> References: <67f096ea0903130845m46447c34o978dbd1e1830631c@mail.gmail.com> <67f096ea0903130900h1a786fbal368b24a198435f1f@mail.gmail.com> Message-ID: <49BA981D.3070805@redhat.com> Terry McIntyre wrote: > Found a clue. When kickstarting, the actual url used is > /./cblr/svc/op/ks/system/sapcps > > When using "cobbler system getks", the leading "/." is not used. This > makes all the difference. > > Now I need to find a clue as to why it makes a difference. > I'm not sure I understand exactly what you are doing on the above, but the Cobbler list has moved here: https://fedorahosted.org/mailman/listinfo/cobbler Seems like you have something set wrong, can you repost this to cobbler-list? --Michael From niftyfedora at niftyegg.com Tue Mar 17 01:00:12 2009 From: niftyfedora at niftyegg.com (Nifty Fedora Mitch) Date: Mon, 16 Mar 2009 18:00:12 -0700 Subject: [et-mgmt-tools] no substitution for ip_address in kickstart file In-Reply-To: <67f096ea0903121006l45ed66e6x5e05132fd3a3f177@mail.gmail.com> References: <67f096ea0903121006l45ed66e6x5e05132fd3a3f177@mail.gmail.com> Message-ID: <20090317010012.GA3036@compegg.wr.niftyegg.com> On Thu, Mar 12, 2009 at 10:06:47AM -0700, Terry McIntyre wrote: > > I have a cobbler 1.4.1 installation, and recently a kickstart failed. > Looking at the > http://my.host/cblr/svc/op/ks/system/sapcps kickstart file, I find this line: > > network --hostname=sapcps.mydomain --bootproto=static > --netmask=255.255.254.0 --ip=$ip_address --gateway=192.168.62.40 > --device=eth0 --onboot=on > > The source line in /etc/cobbler/sample.ks is: > > network --hostname=${hostname} --bootproto=static > --netmask=255.255.254.0 --ip=$ip_address --gateway=$gateway > --device=eth0 --onboot=on > > $hostname and $gateway are expanded; $ip_address is not. > > These values come from /var/lib/cobbler/config/system.d/sapcps: I do not know but explore these three different versions. --netmask=255.255.254.0 --ip=ip_address --gateway=192.168.62.40 --netmask=255.255.254.0 --ip=$ip_address --gateway=192.168.62.40 --netmask=255.255.254.0 --ip=${ip_address} --gateway=192.168.62.40 It is not clear to me where you define "ip_address" and where you intend to assign it. See the part about an elephant: https://fedorahosted.org/cobbler/wiki/KickstartTemplating -- T o m M i t c h e l l Found me a new hat, now what? From john.levon at sun.com Tue Mar 17 02:46:11 2009 From: john.levon at sun.com (John Levon) Date: Mon, 16 Mar 2009 19:46:11 -0700 Subject: [et-mgmt-tools] [PATCH] Support OpenSolaris 2009.06 Message-ID: <7e7e8034601323d1b4f2.1237257971@girltalk> # HG changeset patch # User John Levon # Date 1237257954 25200 # Node ID 7e7e8034601323d1b4f21314e0ccf520e25d51b6 # Parent 202f8b4f8189e52a75d16930244274bda02c4af0 Support OpenSolaris 2009.06 This has unfortunately moved the location of the boot archive for SPARC support. Signed-off-by: John Levon diff --git a/virtinst/OSDistro.py b/virtinst/OSDistro.py --- a/virtinst/OSDistro.py +++ b/virtinst/OSDistro.py @@ -1040,7 +1040,7 @@ class OpenSolarisDistro(SunDistro): os_variant = "opensolaris" kernelpath = "platform/i86xpv/kernel/unix" - initrdpath = "boot/x86.microroot" + initrdpaths = [ "boot/boot_archive", "boot/x86.microroot" ] def isValidStore(self, fetcher, progresscb): if fetcher.hasFile(self.kernelpath): @@ -1076,12 +1076,16 @@ class OpenSolarisDistro(SunDistro): args = "/" + self.kernelpath + self.install_args(guest) try: - initrd = fetcher.acquireFile(self.initrdpath, progresscb) + initrd = fetcher.acquireFile(self.initrdpaths[0], progresscb) return (kernel, initrd, args) - except: - os.unlink(kernel) - raise RuntimeError(_("OpenSolaris microroot not found at %s") % - self.initrdpath) + except Exception, e: + try: + initrd = fetcher.acquireFile(self.initrdpaths[1], progresscb) + return (kernel, initrd, args) + except: + os.unlink(kernel) + raise Exception("No OpenSolaris boot archive found: %s\n" % e) + # NetWare 6 PV class NetWareDistro(Distro): From john.levon at sun.com Tue Mar 17 02:45:56 2009 From: john.levon at sun.com (John Levon) Date: Mon, 16 Mar 2009 19:45:56 -0700 Subject: [et-mgmt-tools] [PATCH] Ignore failures to kill the child console process Message-ID: <0eeb1929a03b7667bffe.1237257956@girltalk> # HG changeset patch # User John Levon # Date 1237257954 25200 # Node ID 0eeb1929a03b7667bffe55fed53f9546f81c736e # Parent 7e7e8034601323d1b4f21314e0ccf520e25d51b6 Ignore failures to kill the child console process For example, it may have quit already. Signed-off-by: John Levon diff --git a/virtinst/Guest.py b/virtinst/Guest.py --- a/virtinst/Guest.py +++ b/virtinst/Guest.py @@ -772,7 +772,10 @@ class Guest(object): def terminate_console(self): if self._consolechild: - os.kill(self._consolechild, signal.SIGKILL) + try: + os.kill(self._consolechild, signal.SIGKILL) + except: + pass def _wait_for_domain(conn, name): # sleep in .25 second increments until either a) we get running From veillard at redhat.com Tue Mar 17 10:27:49 2009 From: veillard at redhat.com (Daniel Veillard) Date: Tue, 17 Mar 2009 11:27:49 +0100 Subject: [et-mgmt-tools] [PATCH] The gui should drop the connection if libvirt stops for any reason like service libvirt stop. In-Reply-To: <49BEA17F.2050805@redhat.com> References: <49BEA17F.2050805@redhat.com> Message-ID: <20090317102749.GO6912@redhat.com> Since it's about virt-manager, let's use the proper list to track this. Also adding the [PATCH] suject keyword, It is also better to describe the patch in the email body rather than a 2 line long subject and no inner description, thanks, Daniel On Mon, Mar 16, 2009 at 02:59:11PM -0400, Daniel J Walsh wrote: > diff -u virtManager/engine.py~ virtManager/engine.py > --- virtManager/engine.py~ 2009-03-16 14:49:16.000000000 -0400 > +++ virtManager/engine.py 2009-03-16 14:58:05.000000000 -0400 > @@ -158,6 +158,7 @@ > logging.error(("Could not refresh connection %s\n" % > (uri)) + str(sys.exc_info()[0]) + \ > " " + str(sys.exc_info()[1]) + "\n" + \ > traceback.format_exc(sys.exc_info()[2])) > + self.connections[uri]["connection"].close() > return 1 > > def change_timer_interval(self,ignore1,ignore2,ignore3,ignore4): > > -- > Libvir-list mailing list > Libvir-list at redhat.com > https://www.redhat.com/mailman/listinfo/libvir-list -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel at veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/ From Jonathan.Pregler at cnetokc.com Tue Mar 17 12:33:17 2009 From: Jonathan.Pregler at cnetokc.com (Jonathan Pregler) Date: Tue, 17 Mar 2009 07:33:17 -0500 Subject: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Message-ID: Does anyone know if there is a module to compile into the virt-p2v cd that will support HP and Dell RAID controllers? Jonathan Pregler Senior Network Engineer C-Net Inc. jonathan.pregler at cnetokc.com 405.376.6644 Fax: 405.376.3036 CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is being sent by Cnet Inc. for the sole use of the intended recipient(s) and may contain confidential, proprietary and/or privileged information. Any unauthorized review, use, disclosure and/or distribution by any recipient is prohibited. If you are not the intended recipient, please delete and/or destroy all copies of this message regardless of form and any included attachments and notify Cnet Inc. immediately by contacting the sender via reply e-mail or forwarding to support at cnetokc.com. -------------- next part -------------- An HTML attachment was scrubbed... URL: From nick.haunold at det.wa.edu.au Wed Mar 18 00:17:46 2009 From: nick.haunold at det.wa.edu.au (Nick Haunold) Date: Wed, 18 Mar 2009 09:17:46 +0900 Subject: [et-mgmt-tools] Virt-p2v and RAID Controller drivers In-Reply-To: Message-ID: <20090318091746947.00000005728@DET-1150672C331> Generator Microsoft Word 11 (filtered medium) Hey Jonathon I asked a similar question last week and haven' t heard a response. If you here something could you let me know? I am trying to use virt-p2v on Dell blades and get kernel panics during the boot up, I believe this is caused by raid driver issue. Thanks Nick From: et-mgmt-tools-bounces at redhat.com [mailto:et-mgmt-tools-bounces at redhat.com] On Behalf Of Jonathan Pregler Sent: Tuesday, 17 March 2009 9:33 PM To: et-mgmt-tools Subject: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Does anyone know if there is a module to compile into the virt-p2v cd that will support HP and Dell RAID controllers? Jonathan Pregler Senior Network Engineer C-Net Inc. jonathan.pregler at cnetokc.com 405.376.6644 Fax: 405.376.3036 CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is being sent by Cnet Inc. for the sole use of the intended recipient(s) and may contain confidential, proprietary and/or privileged information. Any unauthorized review, use, disclosure and/or distribution by any recipient is prohibited. If you are not the intended recipient, please delete and/or destroy all copies of this message regardless of form and any included attachments and notify Cnet Inc. immediately by contacting the sender via reply e-mail or forwarding to support at cnetokc.com. -------------- next part -------------- An HTML attachment was scrubbed... URL: From crobinso at redhat.com Wed Mar 18 14:45:02 2009 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 18 Mar 2009 10:45:02 -0400 Subject: [et-mgmt-tools] Some (probably) beginner questions In-Reply-To: <49B7D2C8.2010804@redhat.com> References: <49B7966C.3070409@in-telegence.net> <49B7D2C8.2010804@redhat.com> Message-ID: <49C108EE.4040801@redhat.com> Cole Robinson wrote: > Dominik Klein wrote: >> Hi >> >> I am trying to setup libvirt, virt-manager and xen. >> >> I successfully installed xen 3.3.1 and am able to run machines with "xm". >> libvirt is 0.6.1 >> virt-manager was pulled 30 minutes ago from the repo >> same for virtinst >> >> libvirt is installed, libvirtd is running. I can also start virt-manager >> and have it connect to xen. >> >> Now I want to define a new machine with virt-manager and face a couple >> of problems. >> >> * I can only select full vs. paravirtualization at the end of the >> installation dialog and (I think) therefore not select a .iso file as >> installation source. The installation source has to be chosen in the >> very first dialog and I can only select "network install" there. >> > > Hmm, if the connection supports FV and PV, all install options should be > enabled. At the end of the wizard, is 'xen (fullvirt)' and option listed > in the 'Virt Type' drop down? Can you provide 'virsh --connect xen:/// > capabilities' output > We followed up offline, and found this was a virt-manager bug, fixed here: http://hg.et.redhat.com/cgi-bin/hg-virt.cgi/applications/virt-manager--devel/rev/6374136c6247 - Cole From crobinso at redhat.com Wed Mar 18 15:54:35 2009 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 18 Mar 2009 11:54:35 -0400 Subject: [et-mgmt-tools] [Patch] [virtinst] Fix check keymap option in virt-install In-Reply-To: <49B74F8F.20400@jp.fujitsu.com> References: <49B74F8F.20400@jp.fujitsu.com> Message-ID: <49C1193B.1050804@redhat.com> Takahashi Tomohiro wrote: > Hi, > > I made the patch that corrected the following problem. > It is that virt-install hangs up, when I set keymap which doesn't > exist in keytable. > ex) > virt-install ... --keymap=abc > > If you have any comments, please suggest me. > > Signed-off-by: Tomohiro Takahashi > Applied now. Thanks, Cole From crobinso at redhat.com Wed Mar 18 15:55:11 2009 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 18 Mar 2009 11:55:11 -0400 Subject: [et-mgmt-tools] [PATCH] Support NetWare PV installs In-Reply-To: References: Message-ID: <49C1195F.5090304@redhat.com> John Levon wrote: > # HG changeset patch > # User John Levon > # Date 1236813503 25200 > # Node ID bc57f31980f9a408db53e83e177a00931be0a479 > # Parent 143cad71a2a6e83bfff7f46a0f7105e3eeb43ab2 > Support NetWare PV installs > > Signed-off-by: John Levon > Thanks, applied. - Cole From crobinso at redhat.com Wed Mar 18 15:55:31 2009 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 18 Mar 2009 11:55:31 -0400 Subject: [et-mgmt-tools] [PATCH] Support OpenSolaris 2009.06 In-Reply-To: <7e7e8034601323d1b4f2.1237257971@girltalk> References: <7e7e8034601323d1b4f2.1237257971@girltalk> Message-ID: <49C11973.6070400@redhat.com> John Levon wrote: > # HG changeset patch > # User John Levon > # Date 1237257954 25200 > # Node ID 7e7e8034601323d1b4f21314e0ccf520e25d51b6 > # Parent 202f8b4f8189e52a75d16930244274bda02c4af0 > Support OpenSolaris 2009.06 > > This has unfortunately moved the location of the boot archive for SPARC > support. > > Signed-off-by: John Levon > Thanks, applied. - Cole From crobinso at redhat.com Wed Mar 18 15:56:54 2009 From: crobinso at redhat.com (Cole Robinson) Date: Wed, 18 Mar 2009 11:56:54 -0400 Subject: [et-mgmt-tools] [PATCH] Ignore failures to kill the child console process In-Reply-To: <0eeb1929a03b7667bffe.1237257956@girltalk> References: <0eeb1929a03b7667bffe.1237257956@girltalk> Message-ID: <49C119C6.3040004@redhat.com> John Levon wrote: > # HG changeset patch > # User John Levon > # Date 1237257954 25200 > # Node ID 0eeb1929a03b7667bffe55fed53f9546f81c736e > # Parent 7e7e8034601323d1b4f21314e0ccf520e25d51b6 > Ignore failures to kill the child console process > > For example, it may have quit already. > > Signed-off-by: John Levon > Thanks, applied. - Cole From dwalsh at redhat.com Wed Mar 18 16:35:08 2009 From: dwalsh at redhat.com (Daniel J Walsh) Date: Wed, 18 Mar 2009 12:35:08 -0400 Subject: [et-mgmt-tools] [PATCH] Support NetWare PV installs In-Reply-To: <49C1195F.5090304@redhat.com> References: <49C1195F.5090304@redhat.com> Message-ID: <49C122BC.9060703@redhat.com> This patch allows the creating of seclabel in the XML. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: virtinst-svirt.patch URL: From scott at scott-phelps.com Wed Mar 18 19:19:58 2009 From: scott at scott-phelps.com (Scott Phelps) Date: Wed, 18 Mar 2009 15:19:58 -0400 Subject: [et-mgmt-tools] Is anyone working on... Message-ID: <32BDD0C3-3EF6-4E51-AAB1-86483A694DE5@scott-phelps.com> ..adding XFS as a FS type or adding the ability to tell libvirtd about serial consoles during a KVM JeOS install? If not, I was going possibly start working on a patch to add these features. Cheers, Scott From crobinso at redhat.com Tue Mar 17 19:26:30 2009 From: crobinso at redhat.com (Cole Robinson) Date: Tue, 17 Mar 2009 15:26:30 -0400 Subject: [et-mgmt-tools] Is anyone working on... In-Reply-To: <32BDD0C3-3EF6-4E51-AAB1-86483A694DE5@scott-phelps.com> References: <32BDD0C3-3EF6-4E51-AAB1-86483A694DE5@scott-phelps.com> Message-ID: <49BFF966.7050905@redhat.com> Scott Phelps wrote: > ..adding XFS as a FS type or adding the ability to tell libvirtd about > serial consoles during a KVM JeOS install? > If not, I was going possibly start working on a patch to add these > features. > Is this in regards to virt-manager? Do you mean XFS when provisioning a FS storage volume? If so, this would need to be done at the libvirt level if it's not already supported. WRT serial consoles, do you mean being able to attach serial devices to guests? I already have some work done locally to do this at the virtinst level, but no UI in virt-manager. Thanks, Cole From Jonathan.Pregler at cnetokc.com Wed Mar 18 19:58:02 2009 From: Jonathan.Pregler at cnetokc.com (Jonathan Pregler) Date: Wed, 18 Mar 2009 14:58:02 -0500 Subject: [et-mgmt-tools] Virt-p2v and RAID Controller drivers In-Reply-To: <20090318091746947.00000005728@DET-1150672C331> Message-ID: Nick, Thanks for the response. If I get a response to my question and a resolution I will let you know. Just to let you know I am in the process of building my own SUSE live CD with the virt-p2v conversion tool in it to have both HP and Dell RAID drivers compiled into the kernel. Have you had any luck with any RAID Cards yet? I am actually on site doing some VM conversions but I had to create my VM?s from scratch. Let me know Thanks. Jonathan Pregler Senior Network Engineer C-Net Inc. jonathan.pregler at cnetokc.com 405.376.6644 Fax: 405.376.3036 CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is being sent by Cnet Inc. for the sole use of the intended recipient(s) and may contain confidential, proprietary and/or privileged information. Any unauthorized review, use, disclosure and/or distribution by any recipient is prohibited. If you are not the intended recipient, please delete and/or destroy all copies of this message regardless of form and any included attachments and notify Cnet Inc. immediately by contacting the sender via reply e-mail or forwarding to support at cnetokc.com. _____ From: Nick Haunold [mailto:nick.haunold at det.wa.edu.au] Sent: Tuesday, March 17, 2009 7:18 PM To: Fedora/Linux Management Tools Subject: RE: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Hey Jonathon I asked a similar question last week and haven?t heard a response. If you here something could you let me know? I am trying to use virt-p2v on Dell blades and get kernel panics during the boot up, I believe this is caused by raid driver issue. Thanks Nick _____ From: et-mgmt-tools-bounces at redhat.com [mailto:et-mgmt-tools-bounces at redhat.com] On Behalf Of Jonathan Pregler Sent: Tuesday, 17 March 2009 9:33 PM To: et-mgmt-tools Subject: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Does anyone know if there is a module to compile into the virt-p2v cd that will support HP and Dell RAID controllers? Jonathan Pregler Senior Network Engineer C-Net Inc. jonathan.pregler at cnetokc.com 405.376.6644 Fax: 405.376.3036 CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is being sent by Cnet Inc. for the sole use of the intended recipient(s) and may contain confidential, proprietary and/or privileged information. Any unauthorized review, use, disclosure and/or distribution by any recipient is prohibited. If you are not the intended recipient, please delete and/or destroy all copies of this message regardless of form and any included attachments and notify Cnet Inc. immediately by contacting the sender via reply e-mail or forwarding to support at cnetokc.com. -------------- next part -------------- An HTML attachment was scrubbed... URL: From nick.haunold at det.wa.edu.au Wed Mar 18 22:50:10 2009 From: nick.haunold at det.wa.edu.au (Nick Haunold) Date: Thu, 19 Mar 2009 07:50:10 +0900 Subject: [et-mgmt-tools] Virt-p2v and RAID Controller drivers In-Reply-To: Message-ID: <20090319075010901.00000001568@DET-1150672C331> Generator Microsoft Word 11 (filtered medium) Jonathon Virt-p2v did work on a dell 6950 but I cannot get it to work on Dell blades (1855 and 1955). If you get that SUSE cd working do you mind sharing it? Thanks Nick From: et-mgmt-tools-bounces at redhat.com [mailto:et-mgmt-tools-bounces at redhat.com] On Behalf Of Jonathan Pregler Sent: Thursday, 19 March 2009 4:58 AM To: et-mgmt-tools Subject: RE: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Nick, Thanks for the response. If I get a response to my question and a resolution I will let you know. Just to let you know I am in the process of building my own SUSE live CD with the virt-p2v conversion tool in it to have both HP and Dell RAID drivers compiled into the kernel. Have you had any luck with any RAID Cards yet? I am actually on site doing some VM conversions but I had to create my VM' s from scratch. Let me know Thanks. Jonathan Pregler Senior Network Engineer C-Net Inc. jonathan.pregler at cnetokc.com 405.376.6644 Fax: 405.376.3036 CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is being sent by Cnet Inc. for the sole use of the intended recipient(s) and may contain confidential, proprietary and/or privileged information. Any unauthorized review, use, disclosure and/or distribution by any recipient is prohibited. If you are not the intended recipient, please delete and/or destroy all copies of this message regardless of form and any included attachments and notify Cnet Inc. immediately by contacting the sender via reply e-mail or forwarding to support at cnetokc.com. From: Nick Haunold [mailto:nick.haunold at det.wa.edu.au] Sent: Tuesday, March 17, 2009 7:18 PM To: Fedora/Linux Management Tools Subject: RE: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Hey Jonathon I asked a similar question last week and haven' t heard a response. If you here something could you let me know? I am trying to use virt-p2v on Dell blades and get kernel panics during the boot up, I believe this is caused by raid driver issue. Thanks Nick From: et-mgmt-tools-bounces at redhat.com [mailto:et-mgmt-tools-bounces at redhat.com] On Behalf Of Jonathan Pregler Sent: Tuesday, 17 March 2009 9:33 PM To: et-mgmt-tools Subject: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Does anyone know if there is a module to compile into the virt-p2v cd that will support HP and Dell RAID controllers? Jonathan Pregler Senior Network Engineer C-Net Inc. jonathan.pregler at cnetokc.com 405.376.6644 Fax: 405.376.3036 CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is being sent by Cnet Inc. for the sole use of the intended recipient(s) and may contain confidential, proprietary and/or privileged information. Any unauthorized review, use, disclosure and/or distribution by any recipient is prohibited. If you are not the intended recipient, please delete and/or destroy all copies of this message regardless of form and any included attachments and notify Cnet Inc. immediately by contacting the sender via reply e-mail or forwarding to support at cnetokc.com. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Jonathan.Pregler at cnetokc.com Fri Mar 20 12:59:23 2009 From: Jonathan.Pregler at cnetokc.com (Jonathan Pregler) Date: Fri, 20 Mar 2009 07:59:23 -0500 Subject: [et-mgmt-tools] Virt-p2v and RAID Controller drivers In-Reply-To: <20090319075010901.00000001568@DET-1150672C331> Message-ID: No I don?t mind at all. I have two of guys helping me with trying to compile drivers into a live cd. I hope that we can have it within a couple of weeks. Jonathan Pregler Senior Network Engineer C-Net Inc. jonathan.pregler at cnetokc.com 405.376.6644 Fax: 405.376.3036 CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is being sent by Cnet Inc. for the sole use of the intended recipient(s) and may contain confidential, proprietary and/or privileged information. Any unauthorized review, use, disclosure and/or distribution by any recipient is prohibited. If you are not the intended recipient, please delete and/or destroy all copies of this message regardless of form and any included attachments and notify Cnet Inc. immediately by contacting the sender via reply e-mail or forwarding to support at cnetokc.com. _____ From: Nick Haunold [mailto:nick.haunold at det.wa.edu.au] Sent: Wednesday, March 18, 2009 5:50 PM To: Fedora/Linux Management Tools Subject: RE: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Jonathon Virt-p2v did work on a dell 6950 but I cannot get it to work on Dell blades (1855 and 1955). If you get that SUSE cd working do you mind sharing it? Thanks Nick _____ From: et-mgmt-tools-bounces at redhat.com [mailto:et-mgmt-tools-bounces at redhat.com] On Behalf Of Jonathan Pregler Sent: Thursday, 19 March 2009 4:58 AM To: et-mgmt-tools Subject: RE: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Nick, Thanks for the response. If I get a response to my question and a resolution I will let you know. Just to let you know I am in the process of building my own SUSE live CD with the virt-p2v conversion tool in it to have both HP and Dell RAID drivers compiled into the kernel. Have you had any luck with any RAID Cards yet? I am actually on site doing some VM conversions but I had to create my VM?s from scratch. Let me know Thanks. Jonathan Pregler Senior Network Engineer C-Net Inc. jonathan.pregler at cnetokc.com 405.376.6644 Fax: 405.376.3036 CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is being sent by Cnet Inc. for the sole use of the intended recipient(s) and may contain confidential, proprietary and/or privileged information. Any unauthorized review, use, disclosure and/or distribution by any recipient is prohibited. If you are not the intended recipient, please delete and/or destroy all copies of this message regardless of form and any included attachments and notify Cnet Inc. immediately by contacting the sender via reply e-mail or forwarding to support at cnetokc.com. _____ From: Nick Haunold [mailto:nick.haunold at det.wa.edu.au] Sent: Tuesday, March 17, 2009 7:18 PM To: Fedora/Linux Management Tools Subject: RE: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Hey Jonathon I asked a similar question last week and haven?t heard a response. If you here something could you let me know? I am trying to use virt-p2v on Dell blades and get kernel panics during the boot up, I believe this is caused by raid driver issue. Thanks Nick _____ From: et-mgmt-tools-bounces at redhat.com [mailto:et-mgmt-tools-bounces at redhat.com] On Behalf Of Jonathan Pregler Sent: Tuesday, 17 March 2009 9:33 PM To: et-mgmt-tools Subject: [et-mgmt-tools] Virt-p2v and RAID Controller drivers Does anyone know if there is a module to compile into the virt-p2v cd that will support HP and Dell RAID controllers? Jonathan Pregler Senior Network Engineer C-Net Inc. jonathan.pregler at cnetokc.com 405.376.6644 Fax: 405.376.3036 CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is being sent by Cnet Inc. for the sole use of the intended recipient(s) and may contain confidential, proprietary and/or privileged information. Any unauthorized review, use, disclosure and/or distribution by any recipient is prohibited. If you are not the intended recipient, please delete and/or destroy all copies of this message regardless of form and any included attachments and notify Cnet Inc. immediately by contacting the sender via reply e-mail or forwarding to support at cnetokc.com. -------------- next part -------------- An HTML attachment was scrubbed... URL: From mdehaan at redhat.com Fri Mar 20 20:56:29 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 20 Mar 2009 16:56:29 -0400 Subject: [et-mgmt-tools] Win 2008 installs with virt-manager/python-virtinst Message-ID: <49C402FD.5050201@redhat.com> I notice that when I try to install Win 2008 Server I get a black installer error screen about a device being removed, and if I hit ENTER=Continue, I am thrown back to this screen. Some Google references seem to think the CD ROM drive is being removed, which I need to solve as I'm trying to see if this is doable from a fully automated installation. Any ideas? Is this perhaps fixed in a later libvirt and/or python-virtinst? --Michael From crobinso at redhat.com Sat Mar 21 15:14:08 2009 From: crobinso at redhat.com (Cole Robinson) Date: Sat, 21 Mar 2009 11:14:08 -0400 Subject: [et-mgmt-tools] Win 2008 installs with virt-manager/python-virtinst In-Reply-To: <49C402FD.5050201@redhat.com> References: <49C402FD.5050201@redhat.com> Message-ID: <49C50440.9080308@redhat.com> Michael DeHaan wrote: > I notice that when I try to install Win 2008 Server I get a black > installer error screen about a device being removed, and if I hit > ENTER=Continue, I am thrown back to this screen. Some Google > references seem to think the CD ROM drive is being removed, which I need > to solve as I'm trying to see if this is doable from a fully automated > installation. > > Any ideas? Is this perhaps fixed in a later libvirt and/or > python-virtinst? > What are the virtinst, libvirt, kvm, fedora versions you are using? - Cole From mdehaan at redhat.com Mon Mar 23 15:13:50 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 23 Mar 2009 11:13:50 -0400 Subject: [et-mgmt-tools] Win 2008 installs with virt-manager/python-virtinst In-Reply-To: <49C50440.9080308@redhat.com> References: <49C402FD.5050201@redhat.com> <49C50440.9080308@redhat.com> Message-ID: <49C7A72E.8030302@redhat.com> Cole Robinson wrote: > Michael DeHaan wrote: > >> I notice that when I try to install Win 2008 Server I get a black >> installer error screen about a device being removed, and if I hit >> ENTER=Continue, I am thrown back to this screen. Some Google >> references seem to think the CD ROM drive is being removed, which I need >> to solve as I'm trying to see if this is doable from a fully automated >> installation. >> >> Any ideas? Is this perhaps fixed in a later libvirt and/or >> python-virtinst? >> >> > > What are the virtinst, libvirt, kvm, fedora versions you are using? > > - Cole > > python-virtinst-0.400.0-5.fc10.noarch libvirt-0.5.1-2.fc10.i386 kvm-74-10.fc10.i386 Fedora-10 --Michael From georg at tropezien.de Tue Mar 24 05:53:34 2009 From: georg at tropezien.de (Georg Lohrer) Date: Tue, 24 Mar 2009 06:53:34 +0100 Subject: [et-mgmt-tools] Set keymap to local values and how to select my own NIC? Message-ID: <49C8755E.8070602@tropezien.de> Hi, I'm an absolutely rookie in VirtManager-affairs, therefore my question might sound silly, but it gives me some extra troubles. I'm using virt-manager 0.7.0 with virtinst-0.400.3 on may local configuration system. Using VirtManager creating an new virtual machine on my remote host-system (running Ubuntu 08.10 with Xeon-CPU, x86_64-kernel and kvm) always uses "keymap=en_us". Is it possible to set this value to my local "de" variant? It is possible to create the VM directly on the remote-host-system using virt-install and give the keymap-parameter directly. Then VirtManager correctly shows the "keymap=de" value. And another issue is the NIC. I want to use a bridge on the remote-host-system. But VirtManager does not give me another choice as NAT on the Hardware-dialog. Do I have missed anything? Doing something wrong? And, yes, using virt-install directly again makes it possible to setup the nic to br0 accordingly. Best regards, Georg From sascha.mail at web.de Tue Mar 24 08:03:18 2009 From: sascha.mail at web.de (Alexander Sascha) Date: Tue, 24 Mar 2009 09:03:18 +0100 Subject: [et-mgmt-tools] Need help for thesis on Xen - disabling HAP for benchmarking purposes In-Reply-To: <49C892F1.7070202@web.de> References: <49C892F1.7070202@web.de> Message-ID: <49C893C6.20904@web.de> Hi, I'm a German student, writing my thesis on "Virtualization with Xen. Analysis and Comparision of Different Techniques such as Paravirtualization, Full Virtualization, and Utilization of Hardware Support Provided by the Processor". I'd like to know how I can disable HAP/RVI/Nested Paging for benchmarking purposes. I read somewhere (http://markmail.org/message/bbnivuqx6vjz7jg4) that Xen developers decided to disable the global grub parameter introducing a per domain flag for HAP instead (hap=0/1). I used virt-manager built in CentOS for setting up all domUs; virt-manager uses libvirt so that no "old-style" configuration files are stored in /etc/xen. I actually don't know how if libvirt-XML-files already support somethin like 0. How would you proceed to modify an already existing domain using virt-manager? I really do apologize for my "newbie ignorance". Thanks in advance for any useful help. - Alex From jensiragh at hotmail.de Wed Mar 25 19:51:08 2009 From: jensiragh at hotmail.de (- -) Date: Wed, 25 Mar 2009 20:51:08 +0100 Subject: [et-mgmt-tools] virt-manager: starting up a virtual machine renders keyboard useless Message-ID: Hi *, I have a major (and very strange) problem using virt-manager (*): When trying to install a virtual machine (here: Ubuntu Server 8.10) the keyboard layout ist completely messed up which makes an installation impossible - i.e. I cannot enter any text, can hardly navigate in menus (**), can hardly press enter (***) and worst of it: cannot even release the mouse cursor because I have not yet found out which keys I have to press so that the virtual machine receives the intended "ctrl + alt" ("ctrl + alt" obviously does not work). Every key I press does either nothing or something completely different in the virtual machine. Before starting the virtual machine (i.e. setting it up via the Create Wizard) the keyboard works as expected. Any hints that might help me? (*) using v0.5.4 on Xubuntu 8.10 on an (headless) Intel Mac Mini being remotely connected to via LAN from an Intel iMac (**) When I press any of the cursor keys nothing happens. Pressing the "9" on the numeric keypad a "press the left cursor key" seems to be executed in the virtual machine. I have not yet found the other three directions though. (***) When I press the enter key nothing happens. I have found out though, that pressing "s" seems to execute an "enter" in the virtual machine. Thanks in advance Jens _________________________________________________________________ http://redirect.gimas.net/?n=M0903xHMMobile Nie wieder eine Mail verpassen mit Hotmail f?rs Handy! From mdehaan at redhat.com Thu Mar 26 15:17:34 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 26 Mar 2009 11:17:34 -0400 Subject: [et-mgmt-tools] Virt-manager feature suggestion Message-ID: <49CB9C8E.5040606@redhat.com> How about a snapshot button for paused/off guests? It could copy the file if it's on regular storage, though if the guest was on LVM it could offer to do a LVM snapshot? --Michael From berrange at redhat.com Thu Mar 26 16:18:39 2009 From: berrange at redhat.com (Daniel P. Berrange) Date: Thu, 26 Mar 2009 16:18:39 +0000 Subject: [et-mgmt-tools] Virt-manager feature suggestion In-Reply-To: <49CB9C8E.5040606@redhat.com> References: <49CB9C8E.5040606@redhat.com> Message-ID: <20090326161839.GA16725@redhat.com> On Thu, Mar 26, 2009 at 11:17:34AM -0400, Michael DeHaan wrote: > How about a snapshot button for paused/off guests? > > It could copy the file if it's on regular storage, though if the guest > was on LVM it could offer to do a LVM snapshot? There's two things in this are we'd like to do - Full guest cloning (aka GUI for virt-clone). I believe that Cole has got some UI designs for this being finalized, so hopefully get this in the not too distant future - Snapshotting. For this we want to improve the API coverage in libvirt so we can more easily track the fact that some 'save' images may be snapshots http://www.redhat.com/archives/libvir-list/2009-March/msg00205.html Daniel -- |: Red Hat, Engineering, London -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 jahrens at uni-mainz.de Thu Mar 26 17:17:08 2009 From: jahrens at uni-mainz.de (Ahrens, Jens) Date: Thu, 26 Mar 2009 18:17:08 +0100 Subject: [et-mgmt-tools] Scientific Linux -No distro signature for kernel Message-ID: <49CBB894.3010500@uni-mainz.de> Hi everybody, I know the problem has already been addressed in Ticket #374 (closed defect: fixed) at: https://fedorahosted.org/cobbler/ticket/374 I use cobbler 1.4.3-4 - at least in this version it is not yet fixed. Michel wrote that a workaround would be to use "cobbler distro add". Could someone that has used this workaround point out in a more specific way how to get the distros and profiles that I e.g. get for a FC9 import like: bash> cobbler list distro FC9-x86_64 profile FC9-x86_64 profile rescue-FC9-x86_64 distro FC9-xen-x86_64 profile FC9-xen-x86_64 Thanks and cheers, Jens From mdehaan at redhat.com Thu Mar 26 18:03:34 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 26 Mar 2009 14:03:34 -0400 Subject: [et-mgmt-tools] Win 2008 installs with virt-manager/python-virtinst In-Reply-To: <49C7A72E.8030302@redhat.com> References: <49C402FD.5050201@redhat.com> <49C50440.9080308@redhat.com> <49C7A72E.8030302@redhat.com> Message-ID: <49CBC376.5020003@redhat.com> Michael DeHaan wrote: > Cole Robinson wrote: >> Michael DeHaan wrote: >> >>> I notice that when I try to install Win 2008 Server I get a black >>> installer error screen about a device being removed, and if I hit >>> ENTER=Continue, I am thrown back to this screen. Some Google >>> references seem to think the CD ROM drive is being removed, which I >>> need >>> to solve as I'm trying to see if this is doable from a fully automated >>> installation. >>> >>> Any ideas? Is this perhaps fixed in a later libvirt and/or >>> python-virtinst? >>> >>> >> >> What are the virtinst, libvirt, kvm, fedora versions you are using? >> >> - Cole >> >> > python-virtinst-0.400.0-5.fc10.noarch > libvirt-0.5.1-2.fc10.i386 > kvm-74-10.fc10.i386 > Fedora-10 > > --Michael > > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools FWIW, I haven't gotten around to checking rawhide for this -- trying to get a Cobbler release out -- expect something Friday or early next week. I see we can create virtual floppies and such via libvirt so I have pretty much all I need for scripting Windows installs provided we have a way to not get the CD ejected 5 seconds into the install :) (The merits of scripting Windows installs are still TBD, seeing how little you can actually script in them) This is different from the F-9 bug where it ejects halfway through the install, this is very early on (this was using 2008 and Vista -- perhaps it might not happen with older installers). --Michael From mdehaan at redhat.com Thu Mar 26 18:21:56 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 26 Mar 2009 14:21:56 -0400 Subject: [et-mgmt-tools] Scientific Linux -No distro signature for kernel In-Reply-To: <49CBB894.3010500@uni-mainz.de> References: <49CBB894.3010500@uni-mainz.de> Message-ID: <49CBC7C4.3070503@redhat.com> Ahrens, Jens wrote: > Hi everybody, > > I know the problem has already been addressed in Ticket #374 (closed > defect: fixed) at: > https://fedorahosted.org/cobbler/ticket/374 > > I use cobbler 1.4.3-4 - at least in this version it is not yet fixed. > Michel wrote that a workaround would be to use "cobbler distro add". > Could someone that has used this workaround point out in a more specific > way how to get the distros and profiles that I e.g. get for a FC9 import > like: > bash> cobbler list > distro FC9-x86_64 > profile FC9-x86_64 > profile rescue-FC9-x86_64 > distro FC9-xen-x86_64 > profile FC9-xen-x86_64 > > Thanks and cheers, > Jens > > > _______________________________________________ > et-mgmt-tools mailing list > et-mgmt-tools at redhat.com > https://www.redhat.com/mailman/listinfo/et-mgmt-tools Hi Jens, Cobbler's mailing list is: https://fedorahosted.org/mailman/listinfo/cobbler You'd want to post that there in the future. This should be fixed in 1.6 which is going out tomorrow if all goes according to plan. Thanks! --Michael From mdehaan at redhat.com Tue Mar 31 15:49:44 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 31 Mar 2009 11:49:44 -0400 Subject: [et-mgmt-tools] Win 2008 installs with virt-manager/python-virtinst In-Reply-To: <49CBC376.5020003@redhat.com> References: <49C402FD.5050201@redhat.com> <49C50440.9080308@redhat.com> <49C7A72E.8030302@redhat.com> <49CBC376.5020003@redhat.com> Message-ID: <49D23B98.9010401@redhat.com> Michael DeHaan wrote: > Michael DeHaan wrote: >> Cole Robinson wrote: >>> Michael DeHaan wrote: >>> >>>> I notice that when I try to install Win 2008 Server I get a black >>>> installer error screen about a device being removed, and if I hit >>>> ENTER=Continue, I am thrown back to this screen. Some Google >>>> references seem to think the CD ROM drive is being removed, which I >>>> need >>>> to solve as I'm trying to see if this is doable from a fully automated >>>> installation. >>>> >>>> Any ideas? Is this perhaps fixed in a later libvirt and/or >>>> python-virtinst? >>>> >>>> >>> >>> What are the virtinst, libvirt, kvm, fedora versions you are using? >>> >>> - Cole >>> >>> >> python-virtinst-0.400.0-5.fc10.noarch >> libvirt-0.5.1-2.fc10.i386 >> kvm-74-10.fc10.i386 >> Fedora-10 >> >> --Michael >> >> >> >> _______________________________________________ >> et-mgmt-tools mailing list >> et-mgmt-tools at redhat.com >> https://www.redhat.com/mailman/listinfo/et-mgmt-tools Trying this with rawhide, the initial install step works -- it does not immediately eject the CD-ROM. The installation does stop the guest and not restart it, however, so you have to manually restart it (and, as we want, the install still has the CD mounted). Should we consider the need to toggle the VM power state in virt-manager to complete the installation a bug or is that something it did before? I'd be glad to write this up if it is (installing against Win 2008 Server) --Michael From mdehaan at redhat.com Tue Mar 31 20:15:44 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 31 Mar 2009 16:15:44 -0400 Subject: [et-mgmt-tools] virtinst ... proper API way to add a floppy drive? Message-ID: <49D279F0.7080607@redhat.com> I'm working on adding scripted Windows install support (ISO + answer file on floppy) support to koan. I'm getting the following when trying to attach a floppy image. ======================= Here's the error: libvir: QEMU error : Domain not found libvir: QEMU error : Domain not found libvir: Domain Config error : internal error Invalid floppy device name: hda internal error Invalid floppy device name: hda File "app.py", line 193, in main k.run() File "app.py", line 298, in run self.virt() File "app.py", line 586, in virt return self.net_install(after_download) File "app.py", line 505, in net_install after_download(self, profile_data) File "app.py", line 584, in after_download self.virt_net_install(profile_data) File "app.py", line 1055, in virt_net_install virt_auto_boot = virt_auto_boot File "/home/mdehaan/cg/koan/koan/qcreate.py", line 203, in start_install guest.start_install() File "/usr/lib/python2.5/site-packages/virtinst/Guest.py", line 732, in start_install return self._do_install(consolecb, meter, removeOld, wait) File "/usr/lib/python2.5/site-packages/virtinst/Guest.py", line 767, in _do_install self.domain = self.conn.createLinux(install_xml, 0) File "/usr/lib/python2.5/site-packages/libvirt.py", line 915, in createLinux if ret is None:raise libvirtError('virDomainCreateLinux() failed', conn=self) ======================== Here's the extended log: 9 16:12:12 koan 31039] DEBUG (DistroManager:152) DistroInstaller location is a local file/path: /opt/images/win2008.iso [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:397) Using self.path for VirtualDisk. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:407) VirtualDisk storage exists. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:271) Detected storage as type 'file' [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:397) Using self.path for VirtualDisk. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:415) VirtualDisk storage does not exist. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:397) Using self.path for VirtualDisk. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:415) VirtualDisk storage does not exist. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:348) Didn't find path '/opt/images/win2008.iso' managed on connection: None [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:397) Using self.path for VirtualDisk. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:407) VirtualDisk storage exists. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:271) Detected storage as type 'file' [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:397) Using self.path for VirtualDisk. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:407) VirtualDisk storage exists. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:271) Detected storage as type 'file' [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:397) Using self.path for VirtualDisk. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:407) VirtualDisk storage exists. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:271) Detected storage as type 'file' [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:397) Using self.path for VirtualDisk. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:407) VirtualDisk storage exists. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:271) Detected storage as type 'file' [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:348) Didn't find path '/opt/images/win2008.iso' managed on connection: None [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:397) Using self.path for VirtualDisk. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:407) VirtualDisk storage exists. [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (VirtualDisk:271) Detected storage as type 'file' [Tue, 31 Mar 2009 16:12:13 koan 31039] DEBUG (Guest:765) Creating guest from ' Tue_Mar_31_16_12_12_2009 524288 524288 ada7fd37-949a-7511-e41a-33da4ef6fe14 hvm destroy destroy destroy 1 /usr/bin/qemu-kvm ' ========================== Here's the code I'm using to add the floppy: guest.disks.append(virtinst.VirtualDisk(device=virtinst.VirtualDisk.DEVICE_FLOPPY, path=floppy_path)) ============================= Here's what version of everything I have: [mdehaan at mdehaan koan]$ rpm -q libvirt python-virtinst qemu kvm libvirt-0.5.1-2.fc10.i386 python-virtinst-0.400.0-5.fc10.noarch qemu-0.9.1-10.fc10.i386 kvm-74-10.fc10.i386 ============================== Any idea on why it may be trying to call it "hda" ? This is obviously not right for a floppy but when reading "pydoc virtinst.VirtualDisk" I didn't see what I was missing. Help would be much appreciated. Thanks! --Michael