| *** mlavalle has quit IRC | 00:38 | |
| tbarron | tonyb: just following up on one thing at a time: "it's a self inflicted problem". Dumb question probably, but how? How would we avoid it? | 00:43 |
|---|---|---|
| tonyb | tbarron: Red Hat / Cannonical often uprev libvirt on a released branch (in the RHEL/CentOS case 7.x -> 7.(x+1)). Even though as an OpenStack Community we've explained this isn't supported and frankly is kinda rude. | 00:48 |
| tonyb | $distros say "please don't uprev things on stable branches" and then go ahead and do that thing | 00:49 |
| tonyb | tbarron: pragmatically I get it but it's still self inflicted from that POV | 00:49 |
| *** dave-mccowan has joined #openstack-stable | 01:15 | |
| *** mriedem has quit IRC | 01:23 | |
| tbarron | tonyb: ah, i get it now -- you and I didn't decide to do that but our employer did :) | 02:38 |
| tbarron | I don't understand yet -- practically -- how to demonstrate that the change would be safe for the other releases. | 02:42 |
| tbarron | is there a job that we can clone and change the $nodeset for each and run it with 4.8.0 or 4.10.0 and demonstrate that it's safe for each nodeset that we have? | 02:43 |
| tbarron | then rinse and repeat for each targeted stable/$branch? | 02:44 |
| tbarron | s/for the other releases/for the other distros/ | 02:45 |
| tbarron | mebbe nova lvm job, i dunno which job exercises libvirt properly | 02:46 |
| tbarron | don't see melwitt or mriedem ... w.r.t. ^^^ | 02:55 |
| tonyb | tbarron: I never said you did. | 03:12 |
| tonyb | tbarron: I'm quite aware it's a corporate decision | 03:13 |
| tonyb | tbarron: So the critera are: | 03:14 |
| tonyb | 1) That libvirt-python has to be > libvirt.so | 03:14 |
| tonyb | 2) libvirt-python listed in upper-constraints works with the libvirt.so in distros that consume this branch. Including CentOS-7, Ubuntu LTS, SLES and highly desirable Gentoo. | 03:14 |
| tonyb | 3) Validation that the new libvirt-python and libvirt.so are compatible with Nova's expectations. | 03:14 |
| tonyb | 4) From the Extended Maintenance policy, the libvirt-python on branch n-1 be >= libvirt-python on branch n, and not spot backported. | 03:15 |
| tonyb | 1) and 2) require basically the same work. Find the version of libvirt in $distro(s) that supports $branch and ensure that it is <= libvirt-python | 03:16 |
| tonyb | 3) check the libvirt driver in nova for the min (eg http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/libvirt/driver.py#n223) which I really doubt will be a problem. I *thought* there was a cap but I could be wrong | 03:18 |
| tonyb | and 4) is just doing that work on rocky, queens, pike and ocata (in order) so that we don't have to downgrade libvirt-python when upgrading a release | 03:19 |
| tonyb | tbarron: It's less important that the verion you pick match master and waay more important that it meets the distro expectations | 03:20 |
| tonyb | In terms of testing a no-op in chnage in nova that Depends-On the branch specific review and running the experimental queue would be very informative | 03:21 |
| tonyb | tbarron: https://wiki.openstack.org/wiki/LibvirtDistroSupportMatrix *may* have mich of the information you need | 03:22 |
| tonyb | tbarron: Sorry I don't have time to do more than just explain what needs to be done :( | 03:23 |
| *** dave-mccowan has quit IRC | 03:52 | |
| *** udesale has joined #openstack-stable | 04:17 | |
| *** dhellmann has quit IRC | 05:19 | |
| *** rtjure has quit IRC | 05:19 | |
| *** dhellmann has joined #openstack-stable | 05:20 | |
| *** rtjure has joined #openstack-stable | 05:22 | |
| *** e0ne has joined #openstack-stable | 07:13 | |
| *** rcernin has quit IRC | 07:30 | |
| *** alexchadin has joined #openstack-stable | 07:30 | |
| *** e0ne has quit IRC | 08:31 | |
| *** jpich has joined #openstack-stable | 09:01 | |
| *** mtreinish has quit IRC | 09:11 | |
| *** dtantsur|afk is now known as dtantsur | 09:19 | |
| *** derekh has joined #openstack-stable | 09:44 | |
| *** mtreinish has joined #openstack-stable | 09:57 | |
| *** e0ne has joined #openstack-stable | 10:08 | |
| *** udesale has quit IRC | 11:02 | |
| *** udesale has joined #openstack-stable | 11:03 | |
| *** e0ne has quit IRC | 11:30 | |
| *** dtantsur is now known as dtantsur|brb | 11:56 | |
| *** dave-mccowan has joined #openstack-stable | 12:40 | |
| *** dtantsur|brb is now known as dtantsur | 13:20 | |
| *** eharney has joined #openstack-stable | 13:40 | |
| *** dtantsur is now known as dtantsur|brb | 13:56 | |
| *** mriedem has joined #openstack-stable | 14:13 | |
| *** irclogbot_2 has quit IRC | 14:32 | |
| *** e0ne has joined #openstack-stable | 14:38 | |
| *** irclogbot_2 has joined #openstack-stable | 14:43 | |
| *** irclogbot_2 has quit IRC | 14:52 | |
| *** jlibosva has joined #openstack-stable | 14:55 | |
| *** mlavalle has joined #openstack-stable | 15:02 | |
| *** irclogbot_2 has joined #openstack-stable | 15:04 | |
| *** alexchadin has quit IRC | 15:22 | |
| *** dtantsur|brb is now known as dtantsur | 15:25 | |
| *** udesale has quit IRC | 16:19 | |
| *** e0ne has quit IRC | 17:21 | |
| *** jpich has quit IRC | 17:26 | |
| *** dtantsur is now known as dtantsur|afk | 17:26 | |
| *** derekh has quit IRC | 18:03 | |
| *** mriedem is now known as mriedem_lunch | 18:21 | |
| *** mriedem_lunch is now known as mriedem | 19:11 | |
| *** e0ne has joined #openstack-stable | 19:20 | |
| *** e0ne_ has joined #openstack-stable | 19:27 | |
| *** e0ne has quit IRC | 19:27 | |
| *** e0ne_ has quit IRC | 19:30 | |
| *** e0ne has joined #openstack-stable | 19:34 | |
| *** e0ne_ has joined #openstack-stable | 19:39 | |
| *** e0ne has quit IRC | 19:41 | |
| *** e0ne has joined #openstack-stable | 19:44 | |
| *** e0ne_ has quit IRC | 19:45 | |
| *** e0ne has quit IRC | 19:45 | |
| *** e0ne has joined #openstack-stable | 19:46 | |
| *** e0ne has quit IRC | 19:54 | |
| *** jlibosva has quit IRC | 20:01 | |
| *** kmalloc is now known as notmorgan | 20:10 | |
| *** notmorgan is now known as morgan | 20:10 | |
| *** e0ne has joined #openstack-stable | 20:15 | |
| *** e0ne has quit IRC | 20:16 | |
| tbarron | tonyb: prometheanfire: sorry for the async back-conversations in the meeting but thanks for the help | 21:02 |
| prometheanfire | tbarron: it's all good :D | 21:02 |
| *** rcernin has joined #openstack-stable | 21:06 | |
| tonyb | tbarron: all good. It was helpful | 21:29 |
| *** obre_ has joined #openstack-stable | 22:25 | |
| *** andreykurilin has quit IRC | 22:29 | |
| *** obre has quit IRC | 22:29 | |
| *** mlavalle has quit IRC | 22:29 | |
| *** smcginnis has quit IRC | 22:29 | |
| *** mlavalle has joined #openstack-stable | 22:30 | |
| *** smcginnis has joined #openstack-stable | 23:08 | |
| *** mlavalle has quit IRC | 23:13 | |
| *** dave-mccowan has quit IRC | 23:19 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!