| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-08-14 | |||
| 21:02:28 | mtreinish | clarkb: hmm, that's a good point. Let's see how it turns out | |
| 21:06:21 | dims | mtreinish : please don't rebase :) | |
| 21:10:57 | mtreinish | clarkb: https://review.openstack.org/#/c/493677/ should remove the ensure logs calls | |
| 21:11:36 | clarkb | mtreinish: do you want me to go ahead nad depend on that now or wait for mine to fail on those checks to confirm they will cause problems first? | |
| 21:12:26 | mtreinish | clarkb: probably better to wait and update, just in case we don't actually need it | |
| 21:12:51 | clarkb | ok | |
| 21:17:07 | openstackgerrit | Michael Still proposed openstack/nova master: WIP: Read from console ptys using privsep. https://review.openstack.org/489486 | |
| 21:19:53 | openstackgerrit | Michael Still proposed openstack/nova master: WIP: Read from console ptys using privsep. https://review.openstack.org/489486 | |
| 21:24:13 | openstackgerrit | Merged openstack/nova master: doc: Extend nfv feature matrix with pinning/NUMA https://review.openstack.org/327126 | |
| 22:06:37 | clarkb | mtreinish: http://logs.openstack.org/73/493673/1/check/gate-grenade-dsvm-neutron-ubuntu-xenial/7e5d3b8/logs/grenade.sh.txt.gz#_2017-08-14_21_57_18_157 it failed as expected, adding the depends-on now | |
| 22:07:24 | mtreinish | clarkb: ok cool, I'll fast approve both after your respin | |
| 22:07:57 | mtreinish | unless dansmith wants to review the grenade side patch | |
| 22:12:07 | dansmith | mtreinish: review which? | |
| 22:12:59 | mtreinish | dansmith: https://review.openstack.org/493677 | |
| 22:13:54 | dansmith | mtreinish: ah okay.. you want me to +2+W that? | |
| 22:14:01 | mtreinish | dansmith: sure | |
| 22:14:02 | dansmith | I trust your logic | |
| 22:16:13 | mtreinish | dansmith: modifiy isn't a real word? :) | |
| 22:17:01 | dansmith | mtreinish: modif-i-y like d-i-y maybe.. you could make up a definition | |
| 22:23:49 | clarkb | mtreinish: dansmith hrm http://logs.openstack.org/73/493673/1/check/gate-grenade-dsvm-neutron-ubuntu-xenial/7e5d3b8/logs/old/screen-q-l3.txt.gz is not how I expected it, I expected that to go in journald | |
| 22:24:07 | clarkb | so maybe that devstack change isn't quite what we need either, but in any case we definitely need to not error on missing log files while we sort this out | |
| 22:25:34 | mtreinish | clarkb: could it be d-g doing the collection? | |
| 22:25:37 | clarkb | heh that is beacuse base branch was ocata in that change | |
| 22:25:42 | clarkb | not pike | |
| 22:25:45 | mtreinish | ah, ok | |
| 22:26:00 | clarkb | so there may be an ordering here was need to accomodate, but I think change is otherwise sound | |
| 22:26:21 | clarkb | how did that other job end up being pike to master? | |
| 22:28:00 | clarkb | mtreinish: https://review.openstack.org/#/c/493057/11/devstack-vm-gate-wrap.sh I think we want to depends on that change as well? | |
| 22:28:15 | clarkb | and just live with the weirdness until ^ is in | |
| 22:31:02 | smcginnis | clarkb: Yep, that patch is what sets up the pike to master. ^ | |
| 22:31:20 | clarkb | ok let me push a depends on that too and we'll just live with it a bit | |
| 22:31:42 | clarkb | oh wait we already did it the othe rway around | |
| 22:31:46 | clarkb | in that case onward :) | |
| 22:32:04 | smcginnis | clarkb: Or it can be rebased. Whatever makes it easier to figure out the right way here. | |
| 22:32:26 | clarkb | no thats fine they are in different repos so depends-on is what works | |
| 22:32:35 | openstackgerrit | Michael Still proposed openstack/nova master: WIP: Read from console ptys using privsep. https://review.openstack.org/489486 | |
| 22:53:29 | clarkb | mtreinish: smcginnis also is this going to be a place where we need rolling changes to grenade? | |
| 22:53:42 | clarkb | I guess not becuse we only check logs on the target size | |
| 22:53:44 | clarkb | *side | |
| 22:53:55 | smcginnis | clarkb: Seems like we should. | |
| 22:54:28 | clarkb | smcginnis: I think we may get away with it because only the new/target side is being checkedwhich means running master of grenade anyways | |
| 22:56:37 | openstackgerrit | Merged openstack/nova master: Always use application/json accept header in report client https://review.openstack.org/489772 | |
| 23:17:29 | openstackgerrit | Eric Fried proposed openstack/nova master: [Trivial] docstrings, typos, minor refactoring https://review.openstack.org/493701 | |
| 23:29:19 | openstackgerrit | Eric Fried proposed openstack/nova master: nova.utils.get_ksa_adapter() https://review.openstack.org/488137 | |
| 23:29:19 | openstackgerrit | Eric Fried proposed openstack/nova master: Use ksa adapter for placement conf & requests https://review.openstack.org/492247 | |
| #openstack-nova - 2017-08-16 | |||
| 01:46:24 | openstackgerrit | Naichuan Sun proposed openstack/nova master: xenapi: cached images should be cleaned up by time https://review.openstack.org/465954 | |
| 02:07:28 | dims | dansmith : melwitt : around? i am chasing a problem with grenade pike->master, nova is ending up creating 2 ports and tempest ends with SSH timeout (base smoke test) | |
| 02:08:04 | dims | dansmith : melwitt : kevinbenton helped go through the logs and saw 2 ports getting created - detail/chat is here - http://paste.openstack.org/show/618473/ | |
| 02:08:38 | dims | dansmith : melwitt : short version, when the instance gets rescheduled things go wrong | |
| 02:09:51 | dims | smcginnis : clarkb : here's the latest ^ | |
| 02:15:24 | clarkb | and the test fails because the vm only brings up one interface and the floating IP is attached to the other | |
| 02:16:11 | melwitt | dims: I don't have time to look into it right now but fyi this has come up in the past if it might help https://bugs.launchpad.net/nova/+bug/1609526 | |
| 02:16:13 | openstack | Launchpad bug 1639230 in nova (Ubuntu Xenial) "duplicate for #1609526 reschedule fails with ip already allocated error" [Undecided,New] - Assigned to Felipe Reyes (freyes) | |
| 02:16:48 | melwitt | or this. I don't know if it's the same thing you're seeing https://bugs.launchpad.net/nova/+bug/1639230 | |
| 02:16:49 | openstack | Launchpad bug 1639230 in nova (Ubuntu Xenial) "reschedule fails with ip already allocated error" [Undecided,New] - Assigned to Felipe Reyes (freyes) | |
| 02:17:25 | gaoyuan | Hi, everyone. I met a problem. Please help if you can. The desc is simple. When I shut down a vm, it sometimes got start up automatically by nova-compute | |
| 02:17:36 | gaoyuan | Anyone met the same problem before? | |
| 02:17:49 | gaoyuan | I was using liberty, when met this problem. | |
| 02:47:12 | openstackgerrit | jichenjc proposed openstack/nova master: update comment for dropping support https://review.openstack.org/494061 | |
| 03:35:48 | openstackgerrit | Kevin Zhao proposed openstack/nova master: Make uefi as the default boot perproties for AArch64 https://review.openstack.org/489951 | |
| 06:35:38 | openstackgerrit | Maciej Jozefczyk proposed openstack/nova master: Remove host filter for _cleanup_running_deleted_instances periodic task https://review.openstack.org/491808 | |
| 07:26:14 | jaypipes | gibi: good morning sir. any new bugs today? | |
| 07:26:32 | gibi | jaypipes: good morning | |
| 07:27:11 | gibi | jaypipes: no new bug today. yet. I have the evacuate and the shelve offload fix proposed if you have time to look at some reviews | |
| 07:27:18 | openstackgerrit | zhangbailin proposed openstack/nova-specs master: Modify spelling error in nova-spec document https://review.openstack.org/494094 | |
| 07:28:12 | gibi | jaypipes: shelve offload regression + bugfix https://review.openstack.org/#/c/493062/ | |
| 07:28:55 | gibi | jaypipes: I've just noticed that the evac bugfix needs a rebase https://review.openstack.org/#/c/493037/ | |
| 07:29:01 | gibi | doing that right now | |
| 07:30:58 | gibi | jaypipes: yesterday I started looking at the soft delete and periodic reclaim case but no definit result yet if there is a bug or not | |
| 07:32:47 | jaypipes | gibi: k, reviewing. even though I hate shelving and all that it entails :) | |
| 07:35:56 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: delete allocation of evacuated instance https://review.openstack.org/493037 | |
| 07:36:17 | gibi | jaypipes: thanks. I share your feelings about shelve. | |
| 07:46:26 | jaypipes | gibi: k, https://review.openstack.org/#/c/493062/ reviewed. | |
| 07:46:36 | jaypipes | moving on to https://review.openstack.org/#/c/493834/ | |
| 07:47:56 | openstackgerrit | zhangbailin proposed openstack/nova-specs master: Modify spelling error in nova-specs document https://review.openstack.org/494094 | |
| 07:48:22 | gibi | jaypipes: thanks, I'm going to check your comments | |
| 07:49:49 | jaypipes | gibi: ok, so let's chat about Ia93168b1560267178059284186fb2b7096c7e81f | |
| 07:50:11 | gibi | jaypipes: sure | |
| 07:50:21 | jaypipes | gibi: it makes me uneasy (and probably dansmith too) to add code to the compute manager to directly call the scheduler report client from the compute manager | |
| 07:51:25 | gibi | jaypipes: should that code moved in to resource_tracker? | |
| 07:51:27 | jaypipes | gibi: I understand why you'd want to do this that way, though (to isolate when the delete_allocation_for_instance() method is called to only be during shelve_offloading() instead of trying to reconcile these things in the perioidic interval task of the RT | |
| 07:53:14 | gibi | jaypipes: do you suggest to try to integrate it to _update_usage_from_instance ? https://github.com/openstack/nova/blob/master/nova/compute/resource_tracker.py#L1012 | |
| 07:54:19 | jaypipes | gibi: no, that method is already trying to do too much :) | |
| 07:54:32 | gibi | jaypipes: agree | |
| 07:54:48 | jaypipes | gibi: my guess is that for shelve_offloaded instances, the periodic interval is going to hit this code block here: https://github.com/openstack/nova/blob/master/nova/compute/resource_tracker.py#L1183 | |
| 07:55:12 | jaypipes | gibi: because instance.host has not yet been un-set and the allocation for the instance is going to be still in the placmeent API | |
| 07:55:41 | jaypipes | gibi: instead of modifying that code block, I'd prefer to add a new remove_offloaded_instance_allocation() method to the *resource tracker* and call that from the compute manager. | |
| 07:57:19 | jaypipes | gibi: I'll add a comment saying so | |
| 07:57:26 | jaypipes | gibi: does that make sense? | |
| 07:57:41 | gibi | jaypipes: that seems like an easy refactoring on the current patch | |
| 07:57:47 | gibi | jaypipes: so I can do that | |
| 07:57:48 | jaypipes | gibi: ya | |
| 07:58:03 | jaypipes | gibi: basically just hiding the reportclient interaction behind an RT call | |
| 07:58:13 | gibi | jaypipes: make sense | |
| 07:58:28 | gibi | jaypipes: especially as the current patch depends on the rt internals anyhow | |
| 08:01:20 | jaypipes | gibi: ya | |
| 08:05:06 | maciejjozefczyk | jaypipes: Hello, about https://review.openstack.org/#/c/491808/ ; I've removed those unit-tests are not testing anything. Tests for this case i think are already done with ./nova/tests/functional/db/test_build_request.py:test_get_by_filters* | |
| 08:05:46 | jaypipes | maciejjozefczyk: lol, I was just reviewing that :) | |
| 08:06:04 | maciejjozefczyk | jaypipes: :D | |
| 08:09:29 | jaypipes | maciejjozefczyk: k, reviewed. problem found. see review comments :) | |