Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On Tue, Aug 12, 2014 at 10:09:52PM +0100, Mark McLoughlin wrote: On Wed, 2014-07-30 at 15:34 -0700, Clark Boylan wrote: On Wed, Jul 30, 2014, at 03:23 PM, Jeremy Stanley wrote: On 2014-07-30 13:21:10 -0700 (-0700), Joe Gordon wrote: While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? [...] The only dog I have in this fight is that using the split-out libvirt-python on PyPI means we finally get to run Nova unit tests in virtualenvs which aren't built with system-site-packages enabled. It's been a long-running headache which I'd like to see eradicated everywhere we can. I understand though if we have to go about it more slowly, I'm just excited to see it finally within our grasp. -- Jeremy Stanley We aren't quite forcing people to move to newer versions. Only those installing nova test-requirements need newer libvirt. Yeah, I'm a bit confused about the problem here. Is it that people want to satisfy test-requirements through packages rather than using a virtualenv? (i.e. if people just use virtualenvs for unit tests, there's no problem right?) If so, is it possible/easy to create new, alternate packages of the libvirt python bindings (from PyPI) on their own separately from the libvirt.so and libvirtd packages? The libvirt python API is (mostly) automatically generated from a description of the XML that is built from the C source files. In tree with have fakelibvirt which is a semi-crappy attempt to provide a pure python libvirt client API with the same signature. IIUC, what you are saying is that we should get a better fakelibvirt that is truely identical with same API coverage /signatures as real libvirt ? Regards, Daniel -- |: http://berrange.com -o-http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :| ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On Wed, 2014-08-13 at 10:26 +0100, Daniel P. Berrange wrote: On Tue, Aug 12, 2014 at 10:09:52PM +0100, Mark McLoughlin wrote: On Wed, 2014-07-30 at 15:34 -0700, Clark Boylan wrote: On Wed, Jul 30, 2014, at 03:23 PM, Jeremy Stanley wrote: On 2014-07-30 13:21:10 -0700 (-0700), Joe Gordon wrote: While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? [...] The only dog I have in this fight is that using the split-out libvirt-python on PyPI means we finally get to run Nova unit tests in virtualenvs which aren't built with system-site-packages enabled. It's been a long-running headache which I'd like to see eradicated everywhere we can. I understand though if we have to go about it more slowly, I'm just excited to see it finally within our grasp. -- Jeremy Stanley We aren't quite forcing people to move to newer versions. Only those installing nova test-requirements need newer libvirt. Yeah, I'm a bit confused about the problem here. Is it that people want to satisfy test-requirements through packages rather than using a virtualenv? (i.e. if people just use virtualenvs for unit tests, there's no problem right?) If so, is it possible/easy to create new, alternate packages of the libvirt python bindings (from PyPI) on their own separately from the libvirt.so and libvirtd packages? The libvirt python API is (mostly) automatically generated from a description of the XML that is built from the C source files. In tree with have fakelibvirt which is a semi-crappy attempt to provide a pure python libvirt client API with the same signature. IIUC, what you are saying is that we should get a better fakelibvirt that is truely identical with same API coverage /signatures as real libvirt ? No, I'm saying that people are installing packaged versions of recent releases of python libraries. But they're skeptical about upgrading their libvirt packages. With the work done to enable libvirt be uploaded to PyPI, can't the two be decoupled? Can't we have packaged versions of the recent python bindings on PyPI that are independent of the base packages containing libvirt.so and libvirtd? (Or I could be completely misunderstanding the issue people are seeing) Mark. ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On Wed, Aug 13, 2014 at 04:24:57PM +0100, Mark McLoughlin wrote: On Wed, 2014-08-13 at 10:26 +0100, Daniel P. Berrange wrote: On Tue, Aug 12, 2014 at 10:09:52PM +0100, Mark McLoughlin wrote: On Wed, 2014-07-30 at 15:34 -0700, Clark Boylan wrote: On Wed, Jul 30, 2014, at 03:23 PM, Jeremy Stanley wrote: On 2014-07-30 13:21:10 -0700 (-0700), Joe Gordon wrote: While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? [...] The only dog I have in this fight is that using the split-out libvirt-python on PyPI means we finally get to run Nova unit tests in virtualenvs which aren't built with system-site-packages enabled. It's been a long-running headache which I'd like to see eradicated everywhere we can. I understand though if we have to go about it more slowly, I'm just excited to see it finally within our grasp. -- Jeremy Stanley We aren't quite forcing people to move to newer versions. Only those installing nova test-requirements need newer libvirt. Yeah, I'm a bit confused about the problem here. Is it that people want to satisfy test-requirements through packages rather than using a virtualenv? (i.e. if people just use virtualenvs for unit tests, there's no problem right?) If so, is it possible/easy to create new, alternate packages of the libvirt python bindings (from PyPI) on their own separately from the libvirt.so and libvirtd packages? The libvirt python API is (mostly) automatically generated from a description of the XML that is built from the C source files. In tree with have fakelibvirt which is a semi-crappy attempt to provide a pure python libvirt client API with the same signature. IIUC, what you are saying is that we should get a better fakelibvirt that is truely identical with same API coverage /signatures as real libvirt ? No, I'm saying that people are installing packaged versions of recent releases of python libraries. But they're skeptical about upgrading their libvirt packages. With the work done to enable libvirt be uploaded to PyPI, can't the two be decoupled? Can't we have packaged versions of the recent python bindings on PyPI that are independent of the base packages containing libvirt.so and libvirtd? It is already de-coupled - the libvirt-python module up on PyPI is capable of building against any libvirt.so C library version 0.9.11 - $CURRENT. The problem with Ubuntu precise is that it is C library version 0.9.6 which we can't build against because that vintage libvirt never installed the libvirt-api.xml file that we use to auto-generated the python code from. Regards, Daniel -- |: http://berrange.com -o-http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :| ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On Wed, 2014-07-30 at 15:34 -0700, Clark Boylan wrote: On Wed, Jul 30, 2014, at 03:23 PM, Jeremy Stanley wrote: On 2014-07-30 13:21:10 -0700 (-0700), Joe Gordon wrote: While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? [...] The only dog I have in this fight is that using the split-out libvirt-python on PyPI means we finally get to run Nova unit tests in virtualenvs which aren't built with system-site-packages enabled. It's been a long-running headache which I'd like to see eradicated everywhere we can. I understand though if we have to go about it more slowly, I'm just excited to see it finally within our grasp. -- Jeremy Stanley We aren't quite forcing people to move to newer versions. Only those installing nova test-requirements need newer libvirt. Yeah, I'm a bit confused about the problem here. Is it that people want to satisfy test-requirements through packages rather than using a virtualenv? (i.e. if people just use virtualenvs for unit tests, there's no problem right?) If so, is it possible/easy to create new, alternate packages of the libvirt python bindings (from PyPI) on their own separately from the libvirt.so and libvirtd packages? Mark. ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On 8/5/2014 12:39 PM, Solly Ross wrote: Just to add my two cents, while I get that people need to run on older versions of software, at a certain point you have to bump the minimum version. Even libvirt 0.9.11 is from April 3rd 2012. That's two and a third years old at this point. I think at a certain point we need to say if you want to run OpenStack on an older platform, then you'll need to run an older OpenStack or backport the required packages. Best Regards, Solly Ross - Original Message - From: Joe Gordon joe.gord...@gmail.com To: OpenStack Development Mailing List openstack-dev@lists.openstack.org Sent: Wednesday, July 30, 2014 7:07:13 PM Subject: Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise? On Jul 30, 2014 3:36 PM, Clark Boylan cboy...@sapwetik.org wrote: On Wed, Jul 30, 2014, at 03:23 PM, Jeremy Stanley wrote: On 2014-07-30 13:21:10 -0700 (-0700), Joe Gordon wrote: While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? [...] The only dog I have in this fight is that using the split-out libvirt-python on PyPI means we finally get to run Nova unit tests in virtualenvs which aren't built with system-site-packages enabled. It's been a long-running headache which I'd like to see eradicated everywhere we can. I understand though if we have to go about it more slowly, I'm just excited to see it finally within our grasp. -- Jeremy Stanley We aren't quite forcing people to move to newer versions. Only those installing nova test-requirements need newer libvirt. This does not include people using eg devstack. I think it is reasonable to expect people testing tip of nova master to have a reasonably newish test bed to test it (its not like the Infra team moves at a really fast pace :) ). Based on http://lists.openstack.org/pipermail/openstack-dev/2014-July/041457.html this patch is breaking people, which is the basis for my concerns. Perhaps we should get some further details from Salvatore. Avoiding system site packages in virtualenvs is a huge win particularly for consistency of test results. It avoids pollution of site packages that can happen differently across test machines. This particular type of inconsistency has been the cause of the previously mentioned headaches. I agree this is a huge win, but I am just concerned we don't have any deprecation cycle and just roll out a new requirement without a heads up. Clark ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev Yeah, I agree, I'm just, you know, a curmudgeon. I was doing a stable/havana backport though on my ubuntu precise + libvirt 1.2.2 from cloud-archive:icehouse and hit this bug: https://bugs.launchpad.net/nova/+bug/1266711 I guess I should just get off my ass and setup a Trusty VM for Juno+ development and leave my Precise one alone for stable branch work. -- Thanks, Matt Riedemann ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
Just to add my two cents, while I get that people need to run on older versions of software, at a certain point you have to bump the minimum version. Even libvirt 0.9.11 is from April 3rd 2012. That's two and a third years old at this point. I think at a certain point we need to say if you want to run OpenStack on an older platform, then you'll need to run an older OpenStack or backport the required packages. Best Regards, Solly Ross - Original Message - From: Joe Gordon joe.gord...@gmail.com To: OpenStack Development Mailing List openstack-dev@lists.openstack.org Sent: Wednesday, July 30, 2014 7:07:13 PM Subject: Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise? On Jul 30, 2014 3:36 PM, Clark Boylan cboy...@sapwetik.org wrote: On Wed, Jul 30, 2014, at 03:23 PM, Jeremy Stanley wrote: On 2014-07-30 13:21:10 -0700 (-0700), Joe Gordon wrote: While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? [...] The only dog I have in this fight is that using the split-out libvirt-python on PyPI means we finally get to run Nova unit tests in virtualenvs which aren't built with system-site-packages enabled. It's been a long-running headache which I'd like to see eradicated everywhere we can. I understand though if we have to go about it more slowly, I'm just excited to see it finally within our grasp. -- Jeremy Stanley We aren't quite forcing people to move to newer versions. Only those installing nova test-requirements need newer libvirt. This does not include people using eg devstack. I think it is reasonable to expect people testing tip of nova master to have a reasonably newish test bed to test it (its not like the Infra team moves at a really fast pace :) ). Based on http://lists.openstack.org/pipermail/openstack-dev/2014-July/041457.html this patch is breaking people, which is the basis for my concerns. Perhaps we should get some further details from Salvatore. Avoiding system site packages in virtualenvs is a huge win particularly for consistency of test results. It avoids pollution of site packages that can happen differently across test machines. This particular type of inconsistency has been the cause of the previously mentioned headaches. I agree this is a huge win, but I am just concerned we don't have any deprecation cycle and just roll out a new requirement without a heads up. Clark ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
[openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
This change: https://review.openstack.org/#/c/105501/ Tries to pull in libvirt-python = 1.2.5 for testing. I'm on Ubuntu Precise for development which has libvirt 0.9.8. The latest libvirt-python appears to require libvirt = 0.9.11. So do I have to move to Trusty? -- Thanks, Matt Riedemann ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On Wed, Jul 30, 2014 at 06:39:56AM -0700, Matt Riedemann wrote: This change: https://review.openstack.org/#/c/105501/ Tries to pull in libvirt-python = 1.2.5 for testing. I'm on Ubuntu Precise for development which has libvirt 0.9.8. The latest libvirt-python appears to require libvirt = 0.9.11. So do I have to move to Trusty? You can use the CloudArchive repository to get newer libvirt and qemu packages for Precise, which is what anyone deploying the Ubuntu provided OpenStack packages would be doing. Regards, Daniel -- |: http://berrange.com -o-http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :| ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On 7/30/2014 9:20 AM, Matt Riedemann wrote: On 7/30/2014 6:43 AM, Daniel P. Berrange wrote: On Wed, Jul 30, 2014 at 06:39:56AM -0700, Matt Riedemann wrote: This change: https://review.openstack.org/#/c/105501/ Tries to pull in libvirt-python = 1.2.5 for testing. I'm on Ubuntu Precise for development which has libvirt 0.9.8. The latest libvirt-python appears to require libvirt = 0.9.11. So do I have to move to Trusty? You can use the CloudArchive repository to get newer libvirt and qemu packages for Precise, which is what anyone deploying the Ubuntu provided OpenStack packages would be doing. Regards, Daniel Can we be more specific because this would also need to be updated in the devref docs for setting up your development environment with Ubuntu. Sorry for being a newb but I went here: https://wiki.ubuntu.com/ServerTeam/CloudArchive/ And tried doing: sudo add-apt-repository cloud-archive:icehouse Which failed, I guess it doesn't know about what that means or something? I added a repo to http://ubuntu-cloud.archive.canonical.com/ubuntu/dists/precise-updates/icehouse/ manually but it's not finding any newer libvirt packages. If I can get some help I can push a patch to update the docs since I'm assuming I won't be the only one that hits this and it sounds like minesweeper hit it recently too. [1] [1] http://lists.openstack.org/pipermail/openstack-dev/2014-July/041457.html Yay for docs team, I was missing this: apt-get install python-software-properties Found it here: http://docs.openstack.org/havana/install-guide/install/apt/content/basics-packages.html The devref env setup doc in nova should still probably be updated to say something like, 'hey if you're on juno using precise you need to enable cloud archive to update libvirt'. -- Thanks, Matt Riedemann ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On 7/30/2014 6:43 AM, Daniel P. Berrange wrote: On Wed, Jul 30, 2014 at 06:39:56AM -0700, Matt Riedemann wrote: This change: https://review.openstack.org/#/c/105501/ Tries to pull in libvirt-python = 1.2.5 for testing. I'm on Ubuntu Precise for development which has libvirt 0.9.8. The latest libvirt-python appears to require libvirt = 0.9.11. So do I have to move to Trusty? You can use the CloudArchive repository to get newer libvirt and qemu packages for Precise, which is what anyone deploying the Ubuntu provided OpenStack packages would be doing. Regards, Daniel Can we be more specific because this would also need to be updated in the devref docs for setting up your development environment with Ubuntu. Sorry for being a newb but I went here: https://wiki.ubuntu.com/ServerTeam/CloudArchive/ And tried doing: sudo add-apt-repository cloud-archive:icehouse Which failed, I guess it doesn't know about what that means or something? I added a repo to http://ubuntu-cloud.archive.canonical.com/ubuntu/dists/precise-updates/icehouse/ manually but it's not finding any newer libvirt packages. If I can get some help I can push a patch to update the docs since I'm assuming I won't be the only one that hits this and it sounds like minesweeper hit it recently too. [1] [1] http://lists.openstack.org/pipermail/openstack-dev/2014-July/041457.html -- Thanks, Matt Riedemann ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On 7/30/2014 9:57 AM, Matt Riedemann wrote: On 7/30/2014 9:20 AM, Matt Riedemann wrote: On 7/30/2014 6:43 AM, Daniel P. Berrange wrote: On Wed, Jul 30, 2014 at 06:39:56AM -0700, Matt Riedemann wrote: This change: https://review.openstack.org/#/c/105501/ Tries to pull in libvirt-python = 1.2.5 for testing. I'm on Ubuntu Precise for development which has libvirt 0.9.8. The latest libvirt-python appears to require libvirt = 0.9.11. So do I have to move to Trusty? You can use the CloudArchive repository to get newer libvirt and qemu packages for Precise, which is what anyone deploying the Ubuntu provided OpenStack packages would be doing. Regards, Daniel Can we be more specific because this would also need to be updated in the devref docs for setting up your development environment with Ubuntu. Sorry for being a newb but I went here: https://wiki.ubuntu.com/ServerTeam/CloudArchive/ And tried doing: sudo add-apt-repository cloud-archive:icehouse Which failed, I guess it doesn't know about what that means or something? I added a repo to http://ubuntu-cloud.archive.canonical.com/ubuntu/dists/precise-updates/icehouse/ manually but it's not finding any newer libvirt packages. If I can get some help I can push a patch to update the docs since I'm assuming I won't be the only one that hits this and it sounds like minesweeper hit it recently too. [1] [1] http://lists.openstack.org/pipermail/openstack-dev/2014-July/041457.html Yay for docs team, I was missing this: apt-get install python-software-properties Found it here: http://docs.openstack.org/havana/install-guide/install/apt/content/basics-packages.html The devref env setup doc in nova should still probably be updated to say something like, 'hey if you're on juno using precise you need to enable cloud archive to update libvirt'. Hopefully this helps people: https://review.openstack.org/#/c/110720/ -- Thanks, Matt Riedemann ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On Wed, Jul 30, 2014 at 6:43 AM, Daniel P. Berrange berra...@redhat.com wrote: On Wed, Jul 30, 2014 at 06:39:56AM -0700, Matt Riedemann wrote: This change: https://review.openstack.org/#/c/105501/ Tries to pull in libvirt-python = 1.2.5 for testing. I'm on Ubuntu Precise for development which has libvirt 0.9.8. The latest libvirt-python appears to require libvirt = 0.9.11. So do I have to move to Trusty? You can use the CloudArchive repository to get newer libvirt and qemu packages for Precise, which is what anyone deploying the Ubuntu provided OpenStack packages would be doing. I am not a fan of this approach the patch above along with [0], broke Minesweeper [1] and Matt, I am worried that we will be breaking other folks as well. I don't think we should force folks to upgrade to a newer version of libvirt just to do some code cleanup. I think we should revert these patches. Increase the min required libvirt version to 0.9.11 since we require that for libvirt-python from PyPI to build successfully. Kill off the legacy CPU model configuration and legacy OpenVSwitch setup code paths only required by libvirt 0.9.11 [0] https://review.openstack.org/#/c/58494/ [1] http://lists.openstack.org/pipermail/openstack-dev/2014-July/041457.html Regards, Daniel -- |: http://berrange.com -o-http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :| ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On 7/30/2014 11:49 AM, Joe Gordon wrote: On Wed, Jul 30, 2014 at 6:43 AM, Daniel P. Berrange berra...@redhat.com mailto:berra...@redhat.com wrote: On Wed, Jul 30, 2014 at 06:39:56AM -0700, Matt Riedemann wrote: This change: https://review.openstack.org/#/c/105501/ Tries to pull in libvirt-python = 1.2.5 for testing. I'm on Ubuntu Precise for development which has libvirt 0.9.8. The latest libvirt-python appears to require libvirt = 0.9.11. So do I have to move to Trusty? You can use the CloudArchive repository to get newer libvirt and qemu packages for Precise, which is what anyone deploying the Ubuntu provided OpenStack packages would be doing. I am not a fan of this approach the patch above along with [0], broke Minesweeper [1] and Matt, I am worried that we will be breaking other folks as well. I don't think we should force folks to upgrade to a newer version of libvirt just to do some code cleanup. I think we should revert these patches. Increase the min required libvirt version to 0.9.11 since we require that for libvirt-python from PyPI to build successfully. Kill off the legacy CPU model configuration and legacy OpenVSwitch setup code paths only required by libvirt 0.9.11 [0] https://review.openstack.org/#/c/58494/ [1] http://lists.openstack.org/pipermail/openstack-dev/2014-July/041457.html Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :| ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org mailto:OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev So https://review.openstack.org/#/c/58494/ is new to me as of today. The 0.9.8 on ubuntu precise broke me (and our internal CI system which is running against precise images, but that's internal so meh). The gate is running against ubuntu trusty and I have a way forward on getting updated libvirt in ubuntu precise (with updated docs on how others can as well), which is a short-term fix until I move my dev environment to ubuntu trusty. My bigger concern here was how this impacts RHEL 6.5 which I'm running Juno on, but looks like that has libvirt 0.10.2 so I'm good. -- Thanks, Matt Riedemann ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On Wed, Jul 30, 2014 at 12:11 PM, Matt Riedemann mrie...@linux.vnet.ibm.com wrote: On 7/30/2014 11:49 AM, Joe Gordon wrote: On Wed, Jul 30, 2014 at 6:43 AM, Daniel P. Berrange berra...@redhat.com mailto:berra...@redhat.com wrote: On Wed, Jul 30, 2014 at 06:39:56AM -0700, Matt Riedemann wrote: This change: https://review.openstack.org/#/c/105501/ Tries to pull in libvirt-python = 1.2.5 for testing. I'm on Ubuntu Precise for development which has libvirt 0.9.8. The latest libvirt-python appears to require libvirt = 0.9.11. So do I have to move to Trusty? You can use the CloudArchive repository to get newer libvirt and qemu packages for Precise, which is what anyone deploying the Ubuntu provided OpenStack packages would be doing. I am not a fan of this approach the patch above along with [0], broke Minesweeper [1] and Matt, I am worried that we will be breaking other folks as well. I don't think we should force folks to upgrade to a newer version of libvirt just to do some code cleanup. I think we should revert these patches. Increase the min required libvirt version to 0.9.11 since we require that for libvirt-python from PyPI to build successfully. Kill off the legacy CPU model configuration and legacy OpenVSwitch setup code paths only required by libvirt 0.9.11 [0] https://review.openstack.org/#/c/58494/ [1] http://lists.openstack.org/pipermail/openstack-dev/2014- July/041457.html Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :| ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org mailto:OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev So https://review.openstack.org/#/c/58494/ is new to me as of today. The 0.9.8 on ubuntu precise broke me (and our internal CI system which is running against precise images, but that's internal so meh). The gate is running against ubuntu trusty and I have a way forward on getting updated libvirt in ubuntu precise (with updated docs on how others can as well), which is a short-term fix until I move my dev environment to ubuntu trusty. My bigger concern here was how this impacts RHEL 6.5 which I'm running Juno on, but looks like that has libvirt 0.10.2 so I'm good. While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? Is it ok to do without a deprecation cycle? Proposed revert patches: https://review.openstack.org/110773 https://review.openstack.org/110774 -- Thanks, Matt Riedemann ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On 2014-07-30 13:21:10 -0700 (-0700), Joe Gordon wrote: While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? [...] The only dog I have in this fight is that using the split-out libvirt-python on PyPI means we finally get to run Nova unit tests in virtualenvs which aren't built with system-site-packages enabled. It's been a long-running headache which I'd like to see eradicated everywhere we can. I understand though if we have to go about it more slowly, I'm just excited to see it finally within our grasp. -- Jeremy Stanley ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On Wed, Jul 30, 2014, at 03:23 PM, Jeremy Stanley wrote: On 2014-07-30 13:21:10 -0700 (-0700), Joe Gordon wrote: While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? [...] The only dog I have in this fight is that using the split-out libvirt-python on PyPI means we finally get to run Nova unit tests in virtualenvs which aren't built with system-site-packages enabled. It's been a long-running headache which I'd like to see eradicated everywhere we can. I understand though if we have to go about it more slowly, I'm just excited to see it finally within our grasp. -- Jeremy Stanley We aren't quite forcing people to move to newer versions. Only those installing nova test-requirements need newer libvirt. This does not include people using eg devstack. I think it is reasonable to expect people testing tip of nova master to have a reasonably newish test bed to test it (its not like the Infra team moves at a really fast pace :) ). Avoiding system site packages in virtualenvs is a huge win particularly for consistency of test results. It avoids pollution of site packages that can happen differently across test machines. This particular type of inconsistency has been the cause of the previously mentioned headaches. Clark ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On 07/30/2014 03:34 PM, Clark Boylan wrote: On Wed, Jul 30, 2014, at 03:23 PM, Jeremy Stanley wrote: On 2014-07-30 13:21:10 -0700 (-0700), Joe Gordon wrote: While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? [...] The only dog I have in this fight is that using the split-out libvirt-python on PyPI means we finally get to run Nova unit tests in virtualenvs which aren't built with system-site-packages enabled. It's been a long-running headache which I'd like to see eradicated everywhere we can. I understand though if we have to go about it more slowly, I'm just excited to see it finally within our grasp. -- Jeremy Stanley We aren't quite forcing people to move to newer versions. Only those installing nova test-requirements need newer libvirt. This does not include people using eg devstack. I think it is reasonable to expect people testing tip of nova master to have a reasonably newish test bed to test it (its not like the Infra team moves at a really fast pace :) ). Avoiding system site packages in virtualenvs is a huge win particularly for consistency of test results. It avoids pollution of site packages that can happen differently across test machines. This particular type of inconsistency has been the cause of the previously mentioned headaches. We, as a project, MUST be able to push forward to new versions of software. libvirt is 100% backportable without screwing precise, so it totally fits well within the scope of our current policy. If there are additional hardships this causes devs, then we should investigate fixing them rather than avoiding upgrading because the _previous_ ubuntu LTS doesn't happen to have a new enough library. ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [nova] so what do i do about libvirt-python if i'm on precise?
On Jul 30, 2014 3:36 PM, Clark Boylan cboy...@sapwetik.org wrote: On Wed, Jul 30, 2014, at 03:23 PM, Jeremy Stanley wrote: On 2014-07-30 13:21:10 -0700 (-0700), Joe Gordon wrote: While forcing people to move to a newer version of libvirt is doable on most environments, do we want to do that now? What is the benefit of doing so? [...] The only dog I have in this fight is that using the split-out libvirt-python on PyPI means we finally get to run Nova unit tests in virtualenvs which aren't built with system-site-packages enabled. It's been a long-running headache which I'd like to see eradicated everywhere we can. I understand though if we have to go about it more slowly, I'm just excited to see it finally within our grasp. -- Jeremy Stanley We aren't quite forcing people to move to newer versions. Only those installing nova test-requirements need newer libvirt. This does not include people using eg devstack. I think it is reasonable to expect people testing tip of nova master to have a reasonably newish test bed to test it (its not like the Infra team moves at a really fast pace :) ). Based on http://lists.openstack.org/pipermail/openstack-dev/2014-July/041457.html this patch is breaking people, which is the basis for my concerns. Perhaps we should get some further details from Salvatore. Avoiding system site packages in virtualenvs is a huge win particularly for consistency of test results. It avoids pollution of site packages that can happen differently across test machines. This particular type of inconsistency has been the cause of the previously mentioned headaches. I agree this is a huge win, but I am just concerned we don't have any deprecation cycle and just roll out a new requirement without a heads up. Clark ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev ___ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
