| opendevreview | Fiyin Peter proposed openstack/python-manilaclient master: Warn when no optional args given to OSC set commands https://review.opendev.org/c/openstack/python-manilaclient/+/982518 | 00:07 |
|---|---|---|
| opendevreview | Fiyin Peter proposed openstack/python-manilaclient master: Warn when no optional args given to OSC set commands https://review.opendev.org/c/openstack/python-manilaclient/+/982518 | 00:36 |
| opendevreview | Fiyin Peter proposed openstack/python-manilaclient master: Fix copy-paste errors in share_networks and share_network_subnets docstrings https://review.opendev.org/c/openstack/python-manilaclient/+/982605 | 02:29 |
| FiyinPeter[m] | Hello, carloss. I am still struggling with self-assigning tasks. I'd like trouble-shooting tips, if there be any | 02:34 |
| FiyinPeter[m] | s/tasks/nugs/, s/trouble-shooting/troubleshooting/ | 02:35 |
| FiyinPeter[m] | s/tasks/bugs/, s/trouble-shooting/troubleshooting/ | 02:38 |
| FiyinPeter[m] | With the problem persisting still, I'd like to have this bug assigned to me | 02:38 |
| FiyinPeter[m] | * to me https://bugs.launchpad.net/manila/+bug/2057703 | 02:41 |
| opendevreview | Fiyin Peter proposed openstack/python-manilaclient master: Fix copy-paste errors in share_networks and share_network_subnets docstrings https://review.opendev.org/c/openstack/python-manilaclient/+/982605 | 09:14 |
| carloss | > paths like missing share type already use the message API to give user facing feedback. would it be worth filing a bug and adding that here too? gouthamr | 11:55 |
| carloss | nathan_akin[m] is there a log for that though? | 11:55 |
| comfortoyewole[m] | Hello @gouthamr: @carloss: | 12:00 |
| comfortoyewole[m] | I've integrated your feedback and waiting for your review on my patch. Thank you | 12:00 |
| comfortoyewole[m] | https://review.opendev.org/c/openstack/manila/+/982140 | 12:00 |
| opendevreview | Amarachi Ogbonnaya proposed openstack/manila master: Add documentation for Shared File Systems telemetry https://review.opendev.org/c/openstack/manila/+/982149 | 12:05 |
| SilviaWachira[m] | Hello everyone. carloss, gouthamr, | 12:28 |
| SilviaWachira[m] | I have a couple of patches awaiting review (some with comments already addressed). I’d really appreciate your feedback when you have time. | 12:28 |
| SilviaWachira[m] | Thank you. | 12:28 |
| nathan_akin[m] | carloss: yes there's a LOG.error in the filter itself, but it only goes to server logs the user just gets NoValidHost with no explanation. the message API would put it in the user retrievable messages so they can actually see why scheduling failed without needing access to server logs | 12:41 |
| carloss | comfortoyewole[m] SilviaWachira[m] we'll review changes as soon as we can, thanks for the ping | 12:46 |
| carloss | nathan_akin[m]: I'm checking on how to proceed. The error has an explanation that will be in the exception itself | 12:54 |
| nathan_akin[m] | carloss: oh i understand. yeah the last_filter name is there in the exception. i was thinking more about the case where the user doesn't have access to the API error detail directly, but i see your point. should i still file a bug or leave it for now? | 12:55 |
| carloss | nathan_akin[m]: have you been able to test this? | 13:03 |
| carloss | I think this is creating a user message: https://github.com/openstack/manila/blob/master/manila/scheduler/manager.py#L222-L226 | 13:03 |
| carloss | the place you mention adds a log and raises shareinstance not found, which is catched a couple of levels above | 13:05 |
| carloss | and gets to this code patch I shared | 13:05 |
| nathan_akin[m] | oh let me check it out, because i was actually looking at it from the code itself | 13:05 |
| nathan_akin[m] | thanks for the info , let me confirm | 13:07 |
| SilviaWachira[m] | carloss: Thanks, I appreciate it! | 13:08 |
| carloss | o/ stephenfin do you remember the context of https://review.opendev.org/c/openstack/manila/+/965781? I think this fix was helpful for CI, but I feel I'm still hitting this issue on https://review.opendev.org/c/openstack/manila/+/974364 | 13:30 |
| carloss | basically, manila services are taking a lot of time to restart | 13:31 |
| carloss | I am working on this job to help us test performance changes from one branch to another | 13:32 |
| carloss | but manila services are taking a very long time to restart | 13:32 |
| opendevreview | Carlos Eduardo proposed openstack/manila master: [WIP] Add rally branch comparison job https://review.opendev.org/c/openstack/manila/+/974364 | 13:59 |
| opendevreview | Lizzy Njoroge proposed openstack/python-manilaclient master: OSC: Add missing tests for share groups and fix snapshot filter https://review.opendev.org/c/openstack/python-manilaclient/+/982663 | 14:18 |
| nathan_akin[m] | hey gouthamr carloss just wanted to give an update , i tested the fix for bug #2058064 on a devstack environment. before the patch, sending X-OpenStack-Manila-API-Version: latest returns a 400 error. after the patch it works as expected. patch is here if you get a chance to review it https://review.opendev.org/c/openstack/manila/+/982536 | 14:19 |
| opendevreview | Carlos Eduardo proposed openstack/manila master: Load manila settings during upgrade's shutdown phase https://review.opendev.org/c/openstack/manila/+/982667 | 15:17 |
| opendevreview | Carlos Eduardo proposed openstack/manila master: [WIP] Add rally branch comparison job https://review.opendev.org/c/openstack/manila/+/974364 | 15:17 |
| opendevreview | Merged openstack/manila master: Fix service listing with a wrong status https://review.opendev.org/c/openstack/manila/+/982165 | 15:26 |
| opendevreview | Silvia Wachira proposed openstack/python-manilaclient master: Mock sleep in ClientTest to reduce unit test runtime https://review.opendev.org/c/openstack/python-manilaclient/+/982695 | 17:01 |
| FiyinPeter[m] | <FiyinPeter[m]> "Hello, carloss. I am still..." <- Hi carloss and gouthamr . I'd appreciate pointers on how to proceed with. I have completed assigned Tasks and my Karma still reads 0. | 17:12 |
| carloss | FiyinPeter[m]: done | 17:37 |
| opendevreview | Olamide Ojo proposed openstack/manila master: Add missing unit test to cephfs driver for line 490. Updating test coverage to 80% https://review.opendev.org/c/openstack/manila/+/981606 | 17:53 |
| opendevreview | Olamide Ojo proposed openstack/manila master: Add missing unit test to cephfs driver for lines 633-678 and 635 https://review.opendev.org/c/openstack/manila/+/981861 | 17:53 |
| opendevreview | Olamide Ojo proposed openstack/manila master: Add missing unit test to cephfs driver for line 804 https://review.opendev.org/c/openstack/manila/+/981873 | 17:53 |
| opendevreview | Olamide Ojo proposed openstack/manila master: Add missing test to cephfs driver for line 831. Increases test coverage to 81% https://review.opendev.org/c/openstack/manila/+/982181 | 17:53 |
| gouthamr | carloss: looks like we hit a snag with https://review.opendev.org/c/openstack/manila/+/982308 | 18:12 |
| gouthamr | the tempest plugin will run with the latest microversion on the branch, and, we have some tight assertions that'll need Sai's tempest change to merge, or, for us to excise a small enough change to get this change in.. | 18:13 |
| opendevreview | Olamide Ojo proposed openstack/manila master: Add missing test to cephfs driver for line 591 https://review.opendev.org/c/openstack/manila/+/982704 | 18:13 |
| gouthamr | Sai's change meaning this one: https://review.opendev.org/c/openstack/manila-tempest-plugin/+/975489 - that's not been updated for a while, so i think we can do a surgical fix | 18:14 |
| carloss | gouthamr: sure, makes sense | 18:15 |
| gouthamr | we're running out of time to do an RC3 though.. | 18:15 |
| gouthamr | we have a day, and its too short of a heads up to the release team | 18:15 |
| carloss | yeah :| | 18:15 |
| gouthamr | so i think we can "mea culpa" this and do a release asap with the fix | 18:15 |
| nathan_akin[m] | gouthamr: do you need any help? | 18:15 |
| carloss | gouthamr: agree with it... we already plan to release asap anyways, considering the API regression | 18:16 |
| gouthamr | nathan_akin[m]: nah, thanks for asking.. this is a release/maintainer issue that needs some careful orchestration. | 18:16 |
| carloss | gah, the manila-tempest-plugin itself was already failing | 18:17 |
| nathan_akin[m] | gouthamr: oh i understand i am always here to assist | 18:17 |
| gouthamr | carloss: let's chat on #openstack-release and let them know, the release isn't "broken" by any means, an API regression that we regret atm - we have those rarely | 18:17 |
| carloss | yeah | 18:31 |
| carloss | I think we should have the fix ready anyways, just in case | 18:32 |
| * carloss looking into the fix | 18:33 | |
| carloss | I think updating the three tests in manila-tempest-plugin would suffice | 18:33 |
| carloss | In version 2.95 it is expected that the metadata field is present as part of the response, we just need to make sure that manila-tempest-plugin is accounting for that | 18:34 |
| opendevreview | Asma Mothana proposed openstack/manila master: Fix Share networks not being retrieved while showing security service https://review.opendev.org/c/openstack/manila/+/858356 | 18:36 |
| gouthamr | yeah, that'd be a good stop gap while Sai fixes up anything else on his change | 18:42 |
| gouthamr | thanks carloss | 18:42 |
| carloss | ++ thanks gouthamr | 18:42 |
| carloss | gouthamr: wondering if I should go ahead and bump the api version in manila-tempest-plugin altogether at this point | 18:55 |
| carloss | I do not expect any other issues though | 18:56 |
| gouthamr | true.. no point leaving that out | 18:56 |
| opendevreview | Carlos Eduardo proposed openstack/manila-tempest-plugin master: Bump API testing version to 2.95 and fix replica tests https://review.opendev.org/c/openstack/manila-tempest-plugin/+/982714 | 19:06 |
| carloss | gouthamr: ^ worked on my env, let's see if CI likes it too | 19:07 |
| opendevreview | Carlos Eduardo proposed openstack/manila master: Add rally branch comparison job https://review.opendev.org/c/openstack/manila/+/974364 | 19:26 |
| opendevreview | Carlos Eduardo proposed openstack/manila master: Add rally branch comparison job https://review.opendev.org/c/openstack/manila/+/974364 | 19:28 |
| opendevreview | Oluwasola Akintewe proposed openstack/manila master: Validate 'ensuring' filter value in services list API https://review.opendev.org/c/openstack/manila/+/982716 | 19:49 |
| nathan_akin[m] | hey gouthamr carloss so i noticed that the 'ensuring' filter in the services list API silently ignores invalid values ?ensuring=maybe returns all services as if no filter was passed at all, while the ?status=invalid correctly returns an empty list, so i tested it on devstack. and filed bug #2146840 and pushed a fix https://review.opendev.org/c/openstack/manila/+/982716 , would appreciate if you could give it a look | 19:52 |
| nathan_akin[m] | Please i also want to ask a question. so i saw that the rally job patch is tagged under eventlet-removal and sets OS_MANILA_DISABLE_EVENTLET_PATCHING for all the services. is the plan to fully remove eventlet from Manila? i'm curious how big of a change that is and whether there are any contribution opportunities around it for someone who wants to get more involved in the project.... | 19:57 |
| opendevreview | Goutham Pacha Ravi proposed openstack/manila stable/2026.1: Fix jsonschema enum (list instead of tuple) https://review.opendev.org/c/openstack/manila/+/982720 | 19:59 |
| gouthamr | nathan_akin[m]: i'd add a release note to https://review.opendev.org/c/openstack/manila/+/982716 | 20:00 |
| gouthamr | ty for checking about the eventlet removal, yes - it's an ongoing project and our plan is to drop eventlet completely in this release cycle ('hibiscus', due October 2026) | 20:01 |
| gouthamr | we've done a bunch of work already: https://review.opendev.org/q/project:openstack/manila+topic:eventlet-removal | 20:02 |
| nathan_akin[m] | gouthamr: oh wow | 20:02 |
| gouthamr | so using "OS_MANILA_DISABLE_EVENTLET_PATCHING", you can stop using eventlet with the release we are shipping in two days; but we're asking folks to wait for the next release because: | 20:05 |
| gouthamr | 1) we're chasing down potential perf/scale regressions | 20:05 |
| gouthamr | 2) we're looking for popular web servers to support/test/document for anyone still using the native eventlet-based web service we had | 20:05 |
| gouthamr | 3) we'll switch all default testing to be run without eventlet, and who knows, we may find some bugs | 20:05 |
| nathan_akin[m] | gouthamr: ok ill do that, thanks | 20:06 |
| opendevreview | Oluwasola Akintewe proposed openstack/manila master: Validate 'ensuring' filter value in services list API https://review.opendev.org/c/openstack/manila/+/982721 | 20:08 |
| nathan_akin[m] | gouthamr: ok so i just pushed patchset 2 on 982716 with the release note but actually it got a new change id, it's now at https://review.opendev.org/c/openstack/manila/+/982721. | 20:10 |
| gouthamr | nathan_akin[m]: use "git commit --amend" when you update an existing commit | 20:11 |
| nathan_akin[m] | gouthamr: ok thank you | 20:11 |
| nathan_akin[m] | gouthamr: that's really cool, so when you say perf or scale regressions , what kind of things are you actually watching for? like is it request latency, throughput, something with how threads are handled and the popular web servers you mentioned, are you looking at something like gunicorn or uvicorn to replace the eventlet-based one? | 20:13 |
| opendevreview | Oluwasola Akintewe proposed openstack/manila master: Validate 'ensuring' filter value in services list API https://review.opendev.org/c/openstack/manila/+/982716 | 20:14 |
| nathan_akin[m] | i used the git commit --amend, this is the original one, the release note is now here. https://review.opendev.org/c/openstack/manila/+/982716. let me abandon the other one | 20:14 |
| gouthamr | nathan_akin[m]: request latency, it's hard to quantify it based on simple API calls, so we're focussing on the APIs doing "work" through the stack.. carloss has been working on using the rally benchmarking tool to get us some results. | 20:15 |
| gouthamr | nathan_akin[m]: on web servers, i am leaning on allowing choice and not restricting one.. choice would mean we add testing/documentation for multiple popular options | 20:16 |
| nathan_akin[m] | gouthamr: oh i understand now, so rally is basically hammering the heavier operations like share create, migrate, etc and measuring how long they take with and without eventlet? so is the plan to run it against both the current release and master so you can catch regressions before they land? or maybe another thing | 20:17 |
| gouthamr | nathan_akin[m]: yes most of our CI is preventive. this will run on each commit: https://review.opendev.org/c/openstack/manila/+/974364 | 20:19 |
| nathan_akin[m] | gouthamr: ok thats a very cool way , i thk so users can pick what works best for their deployment. | 20:19 |
| nathan_akin[m] | gouthamr: oh nicee, let me check it out so i can understand it | 20:21 |
| nathan_akin[m] | gouthamr: So are there web servers already in mind? | 20:34 |
| opendevreview | Amarachi Ogbonnaya proposed openstack/manila master: api-ref: fix scheduler_hints min_version to 2.67 https://review.opendev.org/c/openstack/manila/+/982727 | 20:45 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: api-ref: Add missing availability_zone in replica response https://review.opendev.org/c/openstack/manila/+/982586 | 20:48 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: api-ref: Add missing availability_zone in replica response https://review.opendev.org/c/openstack/manila/+/982586 | 20:49 |
| opendevreview | Amarachi Ogbonnaya proposed openstack/manila master: api-ref: fix scheduler_hints min_version to 2.67 https://review.opendev.org/c/openstack/manila/+/982727 | 21:00 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: api-ref: Add missing versioned fields to share server docs https://review.opendev.org/c/openstack/manila/+/982588 | 21:14 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: api-ref: Add missing versioned fields to share server docs https://review.opendev.org/c/openstack/manila/+/982588 | 21:18 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: Delete deprecated contrib/ci/ scripts https://review.opendev.org/c/openstack/manila/+/982590 | 21:22 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: Delete deprecated contrib/ci/ scripts https://review.opendev.org/c/openstack/manila/+/982590 | 21:22 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: api-ref: Add metadata param to manage snapshot docs https://review.opendev.org/c/openstack/manila/+/982728 | 21:29 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: Remove stale Keystone v2 reference in client_auth.py https://review.opendev.org/c/openstack/manila/+/982589 | 21:50 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: api-ref: Fix share_server_id path param definition Closes-Bug: #2146262 Signed-off-by: Meer Hashaam Khan <hashaamkhan975@gmail.com> https://review.opendev.org/c/openstack/manila/+/982729 | 21:50 |
| hashaaam | Hi gouthamr carloss, submitted another patch fixing share_server_id path param definition. Review: https://review.opendev.org/c/openstack/manila/+/982729 | 21:51 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: api-ref: Fix share_server_id path param definition Closes-Bug: #2146262 Signed-off-by: Meer Hashaam Khan <hashaamkhan975@gmail.com> https://review.opendev.org/c/openstack/manila/+/982729 | 21:58 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: api-ref: Fix share_server_id path param definition Closes-Bug: #2146262 Signed-off-by: Meer Hashaam Khan <hashaamkhan975@gmail.com> https://review.opendev.org/c/openstack/manila/+/982729 | 21:59 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: api-ref: Add metadata param to manage snapshot docs https://review.opendev.org/c/openstack/manila/+/982728 | 22:02 |
| rogerkorantenng | Hi gouthamr, — just a friendly follow-up on my patch for the Hitachi HSP TLS certificate verification fix: https://review.opendev.org/c/openstack/manila/+/982292. Would appreciate a review when you get a chance. Thanks! | 22:07 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: Remove stale Keystone v2 reference in client_auth.py https://review.opendev.org/c/openstack/manila/+/982589 | 22:10 |
| opendevreview | Meer Hashaam Khan proposed openstack/manila master: api-ref: Fix share_group_type_id path param definition Closes-Bug: #2146263 Signed-off-by: Meer Hashaam Khan <hashaamkhan975@gmail.com> https://review.opendev.org/c/openstack/manila/+/982730 | 22:10 |
| hashaaam | Hi gouthamr carloss, submitted another patch fixing share_group_type_id path param definition in share-group-types.inc. Review: https://review.opendev.org/c/openstack/manila/+/982730 | 22:11 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!