Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-03-28
15:36:20 mriedem jaypipes-away: dansmith: alex_xu_: efried: edleafe: did i miss the point about https://review.openstack.org/#/c/555413/ being ok to change the member_of behavior in the 1.21 microversion?
15:36:52 dansmith mriedem: it's not going to change 1.21
15:37:01 efried mriedem: You're right, we weren't being particularly precise, but it was understood that we were going to cut a new microversion.
15:37:28 mriedem ok https://review.openstack.org/#/c/555413/3/specs/rocky/approved/alloc-candidates-member-of.rst@172
15:37:31 mriedem i see that now
15:37:33 openstackgerrit Artom Lifshitz proposed openstack/nova-specs master: NUMA-aware live migration https://review.openstack.org/552722
15:37:33 mriedem whew
15:38:28 melwitt dansmith: can you hit these queens backports pls https://review.openstack.org/#/c/555914
15:39:02 dansmith melwitt: when I'm done with downstream bug paperwork :/
15:39:13 melwitt :(
15:40:14 mriedem so i guess we should re-open https://blueprints.launchpad.net/nova/+spec/alloc-candidates-member-of
15:40:17 mriedem right?
15:42:11 dansmith mriedem: I don't care what we do for the process
15:43:15 openstackgerrit Claudiu Belu proposed openstack/nova master: vmware: Fixes _detach_instance_volumes method https://review.openstack.org/557377
15:44:28 openstackgerrit Eric Fried proposed openstack/nova master: Slugification utilities for placement names https://review.openstack.org/556628
15:44:46 claudiub who's working on the vmware driver?
15:45:00 claudiub so i can add them as reviewers to the patch
15:45:34 mriedem claudiub: cdent is the gateway
15:45:40 mriedem b/c the others aren't ever in channel
15:45:46 claudiub ah, cool.
15:45:58 claudiub thanks. :)
15:46:32 claudiub added him to the vmware driver patch, he doesn't seem to be here though.
15:46:48 openstackgerrit Merged openstack/nova master: Add more functional test for placement.usage https://review.openstack.org/513264
15:46:56 openstackgerrit Merged openstack/nova master: Fix spelling mistake of HTTPNotFound exception https://review.openstack.org/557187
15:47:04 openstackgerrit Merged openstack/nova-specs master: Amend the member_of spec for multiple query sets https://review.openstack.org/555413
15:49:21 melwitt claudiub: thanks for the patch, I'll take a look at it too
15:50:34 claudiub np. :)
15:51:10 claudiub IMO, we should merge the autospec stuff as soon as possible, so we can avoid simple stuff like this. :)
15:51:28 mriedem so uh,
15:51:38 mriedem this means you basically can't attach/detach volumes with the vmware driver,
15:51:43 mriedem since...who knows how long, right?
15:51:48 mriedem and their CI never caught this?
15:51:54 claudiub yeah, that's pretty much what I get from it as well.
15:52:08 mriedem should be a -2 on all vmware driver specs until they can fix their CI then
15:52:24 openstackgerrit Eric Fried proposed openstack/nova master: Slugification utilities for placement names https://review.openstack.org/556628
15:52:29 melwitt yeah, I'm trying to find what's up with their CI. maybe it only runs on vmware driver patches?
15:52:40 claudiub that method is called in destroy
15:52:43 melwitt I noticed the xen driver CI works like that. it only runs on xen driver patches
15:53:51 claudiub i think that they can still attach / detach volumes, but it will fail when they're destroying instances with volumes attached.
15:54:08 claudiub still should have been caught by their CI in boot-from-volume cases
15:54:16 mriedem so maybe just got lucky by tempest always detaching volumes before they were deleted
15:54:58 claudiub can you detach a volume which is used as boot-from-volume?
15:55:12 melwitt I didn't think so
15:55:18 mriedem no
15:55:39 mriedem their CI doesn't usually report at all
15:55:45 mriedem https://review.openstack.org/#/c/553411/ was posted on march 15
15:55:48 mriedem no CI response
15:56:13 melwitt ah, bummer. I checked on the xen one recently and it has recent runs but only on xen patches
15:56:43 mriedem i'm personally fine with 3rd party CI running on restricted sets of the tree
15:56:48 mriedem for the most part
15:57:01 mriedem but to not run on changes for that driver at all, is unacceptable
15:57:05 melwitt yup, same. the only downside is it's harder to notice when it goes MIA
15:57:14 mriedem i used to have to post a quarterly thing to the ML about how the vmware CI wasn't running or was broken
15:57:15 melwitt but that's a lot more minor thing
15:57:50 melwitt yeah, I remember
16:00:00 melwitt I'll mail the list about the vmware CI, see if anyone responds this time
16:01:48 openstackgerrit Claudiu Belu proposed openstack/nova master: vmware: Fixes _detach_instance_volumes method https://review.openstack.org/557377
16:02:26 openstackgerrit Matt Riedemann proposed openstack/nova master: vmware: log quality warning on startup https://review.openstack.org/557398
16:04:27 openstackgerrit Eric Fried proposed openstack/nova master: Make generation optional in ProviderTree https://review.openstack.org/539324
16:12:32 mriedem edleafe: nvm, the docs thing for placement and 1.21 can be fixed in https://review.openstack.org/#/c/554357/
16:13:05 edleafe mriedem: heh, I was just about to push a fix
16:13:33 edleafe I'll just combine them
16:18:33 lyarwood dansmith: https://review.openstack.org/#/c/557348/ - would you mind looking at this retry_on_request removal backport to stable/pike that came up internally earlier?
16:18:58 dansmith lyarwood: I assume we don't need to do that in upstream stable because of g-r right?
16:19:12 dansmith (if it's not already broken I mean)
16:19:54 lyarwood dansmith: well upstream outside of TripleO no we don't need to do this
16:20:04 lyarwood dansmith: but even with g-r I think this is broken for TripleO
16:20:16 dansmith I would defer to mriedem on that
16:20:36 dansmith if it's capped by u-c I think we're fine as-is, and if not, we broke ourselves
16:21:02 dansmith I would suspect maybe upstream tripleo isn't honoring u-c?
16:21:15 lyarwood very likely
16:21:37 lyarwood it's all RPMs there and I don't recall seeing the same u-c rules in the specs
16:23:04 mriedem -
16:23:06 mriedem -1
16:23:08 mriedem -5
16:23:14 mriedem b/c of what dan said
16:23:25 mriedem the decorator was removed in oslo.db 4.30,
16:23:32 mriedem stable/pike is capped at 4.25.1 https://github.com/openstack/requirements/blob/stable/pike/upper-constraints.txt#L433
16:23:41 mriedem so tripleo should be relying on u-c
16:23:52 lyarwood mriedem: nice thanks
16:24:08 edleafe ugh, heavy lightning around here - power drops have kicked me off the internet 3 times already.
16:25:30 dansmith mriedem: oh they're special alright
16:28:13 cdent mriedem, melwitt: all I'm able to update on the vmware sitiuation is that you've got their attention, so reasonable strategy
16:29:29 melwitt cdent: ack
16:29:45 openstackgerrit Merged openstack/nova stable/queens: Preserve multiattach flag when refreshing connection_info https://review.openstack.org/555029
16:30:53 mriedem dansmith: tssurya: melwitt: i assume we're having a cells meeting today?
16:30:59 mriedem there is at least one new bug i wanted to talk about
16:31:25 dansmith mriedem: I was going to ask if we needed to
16:31:28 dansmith sounds like we do
16:31:40 melwitt yeah, if we have topics, we can meet
16:32:12 mriedem unless we just want to skip and talk about the specific bug here
16:32:40 melwitt either way wfm
16:33:42 dansmith I'm good to skip if it's just one bug
16:33:45 dansmith I got nothing else
16:37:04 mriedem wait to see what tssurya says, she's the busy one
16:37:57 tssurya I am good to skip too
16:38:02 mriedem hot dog
16:38:19 mriedem ok so just wanted to talk about options for https://bugs.launchpad.net/nova/+bug/1759316
16:38:19 openstack Launchpad bug 1759316 in OpenStack Compute (nova) queens "pre-cells_v2 nova-osapi_compute service in database breaks instance lookup" [Medium,Confirmed]
16:38:39 dansmith okay

Earlier   Later