| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2018-09-26 | |||
| 19:10:20 | timburke | tests are going to be invaluable, but we'll still need to think a whole lot about all the possible ways things could get out of order | |
| 19:10:41 | timburke | (er, not "out of order" exactly... "interleaved") | |
| 19:11:16 | mordred | the most fun code are the classes where there are more comments and test than code :) | |
| 19:11:48 | mordred | ++ yeah - interleaved, blocked, etc | |
| 19:12:45 | timburke | fwiw, a co-worker of mine wrote https://github.com/swiftstack/python-stack-xray/blob/master/python-stack-xray that's proven invaluable when i suspect that i've hit a deadlock (particularly if eventlet is involved, too) | |
| 19:13:56 | mordred | all the eventlet/gevent stuff totally blows my mind | |
| 19:15:04 | mordred | timburke: seeing as how you have more eventlet experience than I do - my assumption with the semaphore/mutex code above is that it should work fine for eventlet callers too, right? | |
| 19:15:16 | mordred | (should probaby make an eventlet test case to test that out) | |
| 19:16:40 | timburke | probably? i'd need to think on it a bit more. i suspect it'll hurt if eventlet hasn't patched out threading and queue yet, though | |
| 19:17:54 | mordred | that's a good point | |
| 19:17:56 | timburke | (ie, one greenthread would wait on the semaphore and block the hub, so no other greenthreads in that os thread would be able to execute) | |
| 19:19:23 | mordred | https://eventlet.net/doc/modules/queue.html and https://eventlet.net/doc/modules/semaphore.html exist at least - but I don't know if that means they monkeypatch those or not | |
| 19:19:34 | openstack | Launchpad bug 1710328 in OpenStack Object Storage (swift) "object server deadlocks when a worker thread logs something" [High,Fix released] - Assigned to Samuel Merritt (torgomatic) | |
| 19:19:34 | timburke | oh... but if you've got a mix of greenthreads and os threads... you might get into a situation like https://bugs.launchpad.net/swift/+bug/1710328 ... | |
| 19:21:12 | timburke | i think the logic generally goes like this though: (1) don't mix greenthreads and os threads unless you know what you're doing. (2) no one knows what they're doing, and those that think they do are fooling themselves | |
| 19:21:29 | mordred | yes. I agree with both of those | |
| 19:21:58 | Shrews | hrm, this entire conversation leads to wonder if we should leave well enough alone with the TaskManager :) | |
| 19:22:05 | mordred | BUT - that bug does indicate that eventlet patches out threading.RLock | |
| 19:22:19 | mordred | so as long as the calling program is all threads or all mutexes life should be fine | |
| 19:23:28 | mordred | Shrews: I think we're fine - we're not spawning threads or eventlets ourselves - calling code _should_ be fine from either calling convention, and we're not adding threads to the mix ourselves | |
| 19:23:59 | mordred | Shrews: OR - we could just write the semaphore class in C++ | |
| 19:24:00 | timburke | yep, for sure eventlet can patch out threading. just gotta do something like https://github.com/openstack/swift/blob/2.19.0/swift/common/utils.py#L535 | |
| 19:29:01 | mordred | cool. I'll probably need that to do an eventlet test case | |
| 21:06:12 | openstackgerrit | Samuel de Medeiros Queiroz proposed openstack/openstacksdk master: Format URL when updating image props in Glance v1 https://review.openstack.org/605534 | |
| 21:20:05 | openstackgerrit | Samuel de Medeiros Queiroz proposed openstack/openstacksdk master: Send image fields in headers for PUT in Glance v1 https://review.openstack.org/605535 | |
| 22:28:50 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Clarify error message is from nova https://review.openstack.org/605544 | |
| #openstack-sdks - 2018-09-27 | |||
| 03:07:51 | openstackgerrit | Merged openstack/openstacksdk master: Normalize security groups when using Neutron https://review.openstack.org/602147 | |
| 03:26:18 | openstackgerrit | Merged openstack/openstacksdk master: Allow JMESPath on searching networking resources https://review.openstack.org/599078 | |
| 10:38:31 | openstackgerrit | Merged openstack/openstacksdk master: Change the method of role update https://review.openstack.org/600671 | |
| 12:26:44 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Remove all the deprecated stuff https://review.openstack.org/605508 | |
| 12:27:07 | dtantsur | s/links/likes/ | |
| 12:27:08 | mordred | \o/ | |
| 12:27:19 | mordred | dtantsur: I love deleting code | |
| 12:28:59 | mordred | dtroyer: if you have any spare time amongst all of the starlingxing ... https://review.openstack.org/#/c/600292 is the last main big-boy before 1.0 and is ready for review. it's GIANT, so I don't blame you if you don't want to review it - but a lot of the patch is just removing unneeded lines from unit tests | |
| 12:39:22 | tosky | mordred: hi! Do you remember us discussing about https://review.openstack.org/#/c/589465/ ? Apparently openstackclient-check-plugins is still failing on older stable branches | |
| 12:39:25 | tosky | for example http://logs.openstack.org/36/588336/4/check/openstackclient-check-plugins/348b548/ | |
| 12:47:52 | mordred | tosky: oy. that's ugly | |
| 12:48:16 | mordred | tosky: I remember talking about it - I don't remember what we said or if we had come up with a solution that just needed implementing | |
| 12:53:36 | tosky | mordred: I think that the last discussion (before the PTG) was more or less "oh, ok, something is wrong" | |
| 12:58:29 | mordred | tosky: darn. | |
| 12:58:40 | mordred | tosky: well, I agree with pre-ptg us | |
| 13:10:08 | tosky | mordred: could it be that openstackclient runs the tests from its master branch, thus requirement master requirements.txt, which conflicts with the requirements from a stable branch? | |
| 13:10:40 | tosky | that would explain the issue with python-keystonclient 3.15.0 installed, but conflicting with the requirement of python-keystoneclient 3.17.0 | |
| 13:19:37 | mordred | tosky: it should be stable/queens of osc -- http://logs.openstack.org/36/588336/4/check/openstackclient-check-plugins/348b548/job-output.txt.gz#_2018-09-27_12_18_28_095085 | |
| 13:20:02 | mordred | oh - but openstack/openstackclient is master | |
| 13:20:03 | tosky | mordred: that's python-openstackclient, not openstackclient | |
| 13:20:05 | tosky | yep | |
| 13:20:08 | mordred | http://logs.openstack.org/36/588336/4/check/openstackclient-check-plugins/348b548/job-output.txt.gz#_2018-09-27_12_18_31_882086 | |
| 13:20:11 | mordred | yeah | |
| 13:21:04 | mordred | but requirements is also on stable/queens http://logs.openstack.org/36/588336/4/check/openstackclient-check-plugins/348b548/job-output.txt.gz#_2018-09-27_12_18_34_622015 | |
| 13:22:19 | tosky | mordred: but this requirements.txt comes from openstackclient, so from master http://git.openstack.org/cgit/openstack/openstackclient/tree/playbooks/openstackclient-check-plugins/run.yaml#n11 | |
| 13:22:29 | mordred | yah | |
| 13:22:39 | mordred | I wonder if we need stable branch versions of that repo | |
| 13:23:01 | mordred | dtroyer, dhellmann: ^^ I'm not 100% sure of how that hangs together | |
| 13:23:45 | tosky | it used to have stable branches - was everything originally under openstack/openstackclient and then the main code was moved to openstack/python-openstackclient? | |
| 13:30:35 | dhellmann | tosky : no, the code was always in openstack/python-openstackclient. The openstackclient repo is a meta-package for pulling in all of the extra dependencies to make it easier to install | |
| 13:30:53 | dhellmann | tosky , mordred ,dtroyer : if we're testing against things that branch, we need to branch | |
| 13:31:10 | dhellmann | even if we never do anything with the branches, that's just how the test machinery works | |
| 13:34:05 | tosky | I guess it would make sense to keep the job definition in master only | |
| 13:34:56 | tosky | so that the only thing to take care is the check_osc_commands.py | |
| 13:35:07 | tosky | though in openstackclient has code in stable/queens | |
| 13:38:12 | tosky | uh, or there is something weird in my local checkout | |
| 13:40:22 | tosky | for some reason I needed `git remote prune origin` in my local checkout of openstackclient | |
| 13:46:45 | dtroyer | it is totally possible that the openstackclient repo is not properly branched… I had wanted to move the plugin tests there but don't think that was ever completed, so it needs to be handled like the main repo | |
| 13:47:55 | dtroyer | mordred: I am release-branching stx today if zuul ever catches up, but that may give me an excuse to slip into a dark corner and refrsh myself with sdk :) | |
| 13:48:15 | dtroyer | I see how you pull the strings behind the scenes :) | |
| 13:49:32 | mordred | dtroyer: I've gotta do _something_ :) | |
| 16:41:18 | elmiko | edleafe dtantsur let me know how this strikes you https://etherpad.openstack.org/p/api-sig-newsletter | |
| 16:42:03 | edleafe | elmiko: that's not a python docstring :) | |
| 16:42:31 | elmiko | haha ;) | |
| 16:42:46 | elmiko | oh you... always making things so /grammatical/ | |
| 16:43:43 | edleafe | I removed the "meeting agenda" from the bottom | |
| 16:43:52 | elmiko | ++ thanks | |
| 16:44:00 | edleafe | I guess I need to update the meeting in governance | |
| 16:44:26 | elmiko | i took an action for that, but if you want to no worries =) | |
| 16:44:51 | edleafe | oh, go ahead. I didn't know you took that action | |
| 16:45:01 | elmiko | i snuck it in the meeting XD | |
| 16:45:22 | elmiko | i was gonna follow the format for the other office hours entries, i'll make the pr shortly | |
| 16:45:37 | elmiko | am i good to publish on the newsletter? | |
| 16:46:37 | edleafe | send it | |
| 16:46:52 | elmiko | thanks! | |
| 16:49:52 | dtantsur | elmiko: LGTM | |
| 16:50:26 | elmiko | thanks | |
| 16:56:37 | edleafe | elmiko: dtantsur|afk: I updated the meetings page: https://wiki.openstack.org/wiki/Meetings/API-SIG | |
| 16:57:16 | edleafe | elmiko: you're going to take care of the changes in openstack-infra/irc-meetings, right? | |
| 16:58:56 | elmiko | edleafe: https://review.openstack.org/#/c/605808/ | |
| 16:59:01 | elmiko | thanks ed ! | |
| 17:32:56 | mnaser | https://review.openstack.org/#/c/589012 -- can we have eyes on this? | |
| 17:33:15 | mnaser | it's impossible to do live migrations without force specifying a host which isnt ideal | |
| 19:05:10 | mordred | mnaser: how about nargs='?' to make the argument to --live be optional | |
| 19:05:27 | mnaser | mordred: i like that, it doesn't break the backwards compatibility | |
| 19:06:24 | mnaser | mordred: i dunno.. should i respin that patch? push another one and this one can stay as an alternate when we want to further refactor things? | |
| 19:06:27 | mordred | mnaser: might need to also add const='<nonce>' - which will tell argparse to put '<nonce>' in as the argument to --live if none is otherwise given | |
| 19:06:39 | mnaser | could we use default=None? | |
| 19:06:45 | mordred | well, I don't think the patch in its current state is viable bdcause it's a total breaking change | |
| 19:06:46 | mnaser | i think that is the most preferable thing on the nova api side of things | |
| 19:06:54 | mordred | mnaser: maybe? it's worth a try | |
| 19:07:51 | mordred | mnaser: so - default is the value that will be put into the variable if --live isn't given. const is the value that is put in if --live is given with no argument | |
| 19:08:05 | mnaser | mordred: ah okay, so const is what we want then | |
| 19:08:10 | mordred | mnaser: so maybe nargs='?', const=False, default=None | |