| opendevreview | Noppakorn Kaewsalabnil proposed openstack/cinder-tempest-plugin master: Unskip the project reader RBAC tests https://review.opendev.org/c/openstack/cinder-tempest-plugin/+/1014269 | 02:53 |
|---|---|---|
| *** mhen_ is now known as mhen | 03:00 | |
| opendevreview | minwoo seo proposed openstack/cinder master: Add cinder-manage image-cache warm and list actions https://review.opendev.org/c/openstack/cinder/+/986740 | 05:36 |
| opendevreview | minwoo seo proposed openstack/cinder master: Add cinder-manage image-cache warm and list actions https://review.opendev.org/c/openstack/cinder/+/986740 | 05:37 |
| opendevreview | minwoo seo proposed openstack/cinder-specs master: Add spec for cinder-manage image-cache subcommand https://review.opendev.org/c/openstack/cinder-specs/+/986741 | 05:44 |
| opendevreview | Merged openstack/cinder master: NetApp - ASAr2 Driver Snapshot and Cache Acceleration Support https://review.opendev.org/c/openstack/cinder/+/971174 | 06:05 |
| opendevreview | Kubilay Kaptanoglu proposed openstack/cinder master: Record source volume type on backups https://review.opendev.org/c/openstack/cinder/+/1015013 | 07:51 |
| *** elodilles_pto is now known as elodilles | 08:52 | |
| opendevreview | Francesco Pantano proposed openstack/devstack-plugin-ceph master: Bump Ceph to Umbrella release https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/1015017 | 08:57 |
| opendevreview | Francesco Pantano proposed openstack/devstack-plugin-ceph master: Bump Ceph to Umbrella release https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/1015017 | 08:58 |
| opendevreview | Francesco Pantano proposed openstack/cinder master: DNM - Test dependency on devstack-plugin-ceph bump to Squid https://review.opendev.org/c/openstack/cinder/+/936359 | 09:12 |
| opendevreview | Francesco Pantano proposed openstack/cinder master: DNM - Test dependency on devstack-plugin-ceph bump to Squid https://review.opendev.org/c/openstack/cinder/+/936359 | 09:26 |
| *** jlarriba_ is now known as jlarriba | 10:21 | |
| opendevreview | Kubilay Kaptanoglu proposed openstack/cinder master: Record source volume type on backups https://review.opendev.org/c/openstack/cinder/+/1015013 | 11:53 |
| opendevreview | Jyotsna Lothe proposed openstack/cinder master: HPE Alletra MP: Skip sync wait for same-size clones https://review.opendev.org/c/openstack/cinder/+/1008049 | 12:13 |
| opendevreview | Jyotsna Lothe proposed openstack/cinder master: HPE Alletra MP: Skip sync wait for same-size clones https://review.opendev.org/c/openstack/cinder/+/1008049 | 12:16 |
| opendevreview | Jyotsna Lothe proposed openstack/cinder master: HPE Alletra MP: Skip sync wait for same-size clones https://review.opendev.org/c/openstack/cinder/+/1008049 | 12:21 |
| opendevreview | Eric Harney proposed openstack/cinder master: Move check_ssh_injection to ssh_utils https://review.opendev.org/c/openstack/cinder/+/859319 | 12:37 |
| rosmaita | Cinder meeting in #openstack-cinder at 1400 UTC | 14:01 |
| rosmaita | jungleboyj rosmaita smcginnis tosky whoami-rajat m5z e0ne geguileo eharney jbernard hemna fabiooliveira yuval tobias-urdin adiare happystacker dosaboy hillpd msaravan sp-bmilanov Luzi sfernand simondodsley zaubea nileshthathagar flelain wizardbit agalica lutimura kaisers aloke_dev Anoop_Shukla erlon Jyotsna rakshay | 14:01 |
| rosmaita | #startmeeting cinder | 14:01 |
| opendevmeet | Meeting started Wed Nov 11 14:01:41 2026 UTC and is due to finish in 60 minutes. The chair is rosmaita. Information about MeetBot at http://wiki.debian.org/MeetBot. | 14:01 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 14:01 |
| opendevmeet | The meeting name has been set to 'cinder' | 14:01 |
| rosmaita | #topic roll call | 14:01 |
| simondodsley | o/ | 14:01 |
| pungrumpy | o/ | 14:01 |
| aloke_dev | o/ | 14:02 |
| hvlcchao1 | o/ | 14:02 |
| erlon28 | \o | 14:02 |
| harsh | o/ | 14:03 |
| opendevreview | Merged openstack/cinder master: Fix ssh injection sanitizer utility https://review.opendev.org/c/openstack/cinder/+/1014983 | 14:04 |
| *** aloke_dev is now known as alokeC | 14:04 | |
| opendevreview | Merged openstack/cinder master: api-ref: remove multiattach from volume-create request sample https://review.opendev.org/c/openstack/cinder/+/983147 | 14:04 |
| agalica_ | o/ ? | 14:04 |
| rosmaita | hello everyone | 14:05 |
| rosmaita | #link https://etherpad.opendev.org/p/cinder-indri-meetings | 14:05 |
| rosmaita | so, i made the mistake of updating and rebooting my system this morning | 14:05 |
| rosmaita | which logged me out of gerrit | 14:05 |
| rosmaita | and there are issues happening with Launchpad and Ubuntu1 | 14:06 |
| rosmaita | so probably no one can get into launchpad | 14:06 |
| simondodsley | ah - that explains it... | 14:06 |
| rosmaita | and if you aren't already logged into gerrit, you can't get there either | 14:06 |
| rosmaita | i checked, and the openstack-infra team haven't changed anything on our end | 14:06 |
| simondodsley | that's not helpful | 14:07 |
| rosmaita | so we just need to be patient | 14:07 |
| rosmaita | and this may be a short meeting | 14:07 |
| rosmaita | #topic announcements | 14:07 |
| rosmaita | i guess the above counts as an announcement | 14:07 |
| rosmaita | next, spec freeze is next friday on november 20 | 14:08 |
| rosmaita | spec proposers, please keep an eye on your specs and reply to comments | 14:09 |
| rosmaita | and everyone else, please look over the specs and leave helpful comments | 14:09 |
| rosmaita | #topic public security issues | 14:09 |
| rosmaita | two public issues this week | 14:10 |
| simondodsley | there are a lot of open specs for Indri - so please look | 14:10 |
| rosmaita | "Attachment show returns connection_info of another project" | 14:10 |
| rosmaita | #link https://bugs.launchpad.net/cinder/+bug/2173158 | 14:10 |
| opendevreview | Inyong Hong proposed openstack/cinder master: NetApp ONTAP: apply space allocation to cloned LUNs https://review.opendev.org/c/openstack/cinder/+/999400 | 14:10 |
| rosmaita | this one was discovered and fixed by pungrumpy | 14:10 |
| rosmaita | and possibly the fix has merged to master, though i can't tell at the moment | 14:11 |
| simondodsley | what is the fix patch - i'll look | 14:11 |
| pungrumpy | rosmaita: not merged yet. https://review.opendev.org/c/openstack/cinder/+/1014460 has two +2 and Workflow +1 since 13:04 UTC and is in the gate now. thanks for the reviews. | 14:11 |
| erlon28 | on launchpad, we are having some issues: https://status.canonical.com/#/ | 14:11 |
| simondodsley | feels like a DDOS on Canonical | 14:12 |
| rosmaita | it's worth taking a look at the bug/fix, it identifies some 10 year old code that we have all been looking at and thought was just fine | 14:12 |
| erlon28 | rosmaita: I have reviewed/merged that one today | 14:13 |
| rosmaita | erlon28: ty! | 14:14 |
| rosmaita | we'll need to backport that one to all the stable branches, and leave a patch for unmaintained/2025.1 | 14:14 |
| rosmaita | luckily since the code is so old, there should be no conflicts all the way to yoga! | 14:14 |
| rosmaita | the other issue is "[ops][cinder] remote command execution via ssh-based volume drivers" | 14:15 |
| rosmaita | #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/GVGTCVPAK7FUP72SYT2Y3RHNX4JDFU4N/ | 14:15 |
| rosmaita | that email has links to the bug and patches | 14:16 |
| rosmaita | the master patch merged just as this meeting was starting | 14:17 |
| opendevreview | Merged openstack/os-brick master: Bump mypy to 2.3.x https://review.opendev.org/c/openstack/os-brick/+/1001541 | 14:18 |
| rosmaita | (i'll give people a few minutes to read that email in case there are questions) | 14:19 |
| rosmaita | ok, that's all the announcements | 14:23 |
| rosmaita | #topic cross-test state leaks | 14:24 |
| rosmaita | zigo has been experimenting with a new scheduler for stestr | 14:24 |
| rosmaita | #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/OZ2CUAWB5HFR7BJE3YTQVGYMQWOMH5WO/ | 14:24 |
| rosmaita | i guess the key thing for cinder is that he's got 3 patches up: | 14:25 |
| rosmaita | #link https://review.opendev.org/q/hashtag:%22zigo-test-state-leak%22+(status:open%20OR%20status:merged | 14:25 |
| simondodsley | yes - i'm looking at the Everpure one right now | 14:25 |
| rosmaita | that link is probably bad, missing a right paren at the end | 14:25 |
| rosmaita | i need to take some time to understand why he found these issues that we haven't seen | 14:27 |
| rosmaita | because i'm pretty sure we have --random in tox for the unit tests | 14:27 |
| zigo | rosmaita: I just had another idea for the stestr scheduler though: explicitely add some fuzzing in it. | 14:27 |
| zigo | I'm kind of convinced that this will uncover some more failures everywhere. | 14:28 |
| opendevreview | Merged openstack/cinder master: IBM SVC: downgrade SSH error log to warning in _run_ssh https://review.opendev.org/c/openstack/cinder/+/983111 | 14:28 |
| opendevreview | Merged openstack/cinder master: Scope attachments by their own volume's project https://review.opendev.org/c/openstack/cinder/+/1014460 | 14:28 |
| rosmaita | you are probably correct | 14:28 |
| rosmaita | zigo: can you explain why the dynamic scheduler works better than --random for forcing these situations? | 14:29 |
| zigo | Does stestr already has a --random option? | 14:29 |
| zigo | ahve | 14:29 |
| zigo | have | 14:29 |
| rosmaita | pretty sure | 14:29 |
| zigo | I didn't know about it. Well then, maybe everyone in OpenStack should be using it then. | 14:30 |
| rosmaita | well, what's interesting is that we do use it, and you uncovered some stuff we didn't see | 14:30 |
| rosmaita | or maybe it's stuff we did see and just ignored | 14:31 |
| rosmaita | (which may be the real answer) | 14:31 |
| eharney_ | yeah, i'd like to figure out why --random didn't expose the same problems -- i think we put it in there because we expected it to | 14:31 |
| rosmaita | exactly | 14:31 |
| zigo | rosmaita: If one uses --dynamic with also --random, this mode sacrifices the class-affinity grouping, so tests relying on per-class fixtures set up once per worker would re-run them per test. | 14:32 |
| rosmaita | and zigo, is that dynamic scheduler a general stestr thing, or unique to your distro? | 14:32 |
| zigo | That's not what we want, we want to uncover the setUp ran once, and then destroyed by poinsoner. | 14:32 |
| simondodsley | they solve different problems | 14:32 |
| simondodsley | --random only shuffles the order of tests inside each worker’s fixed share; it doesn’t change how much work each worker gets. A dynamic scheduler hands tests to workers as they free up, so no worker sits idle while another grinds through a badly estimated partition. Different problems: --random finds order-dependent tests, a dynamic scheduler shortens wall-clock time. | 14:33 |
| zigo | rosmaita: I'd like for it to be upstreamed, though Mathew doesn't seem to have worked on it: https://github.com/mtreinish/stestr/pull/400 | 14:33 |
| zigo | Actually, I can shoot another email to him to see if he's got bandwidth for review. | 14:35 |
| rosmaita | zigo: ok, that's what i was wondering ... so we can't just use --dynamic yet | 14:35 |
| zigo | Not in the CI yet, indeed. | 14:35 |
| rosmaita | ok, well thanks for working on this, and thank you for the patches | 14:36 |
| zigo | :) | 14:37 |
| rosmaita | before moving on, just want to leave a note for driver developers to take a look | 14:37 |
| rosmaita | simondodsley already knows about Pure | 14:38 |
| zigo | I hope to do more. Also, AI helped a lot. | 14:38 |
| simondodsley | Everpure please :) | 14:38 |
| simondodsley | Yep - just +1'ed it | 14:38 |
| rosmaita | we need to drink a shot every time you say that | 14:38 |
| rosmaita | agalica: there's a hitachi patch | 14:39 |
| agalica_ | I see mine too, but can’t login | 14:39 |
| rosmaita | #link https://review.opendev.org/c/openstack/cinder/+/1014694 | 14:39 |
| agalica_ | Yeah, I see it | 14:39 |
| erlon28 | I +2 them | 14:39 |
| simondodsley | and one for Vast | 14:39 |
| rosmaita | #link https://review.opendev.org/c/openstack/cinder/+/1014702 | 14:39 |
| simondodsley | anyone from Vast ever turn up here? | 14:39 |
| rosmaita | apparently not today | 14:41 |
| rosmaita | #topic cinder-manage image-cache warm/list | 14:43 |
| rosmaita | #link https://review.opendev.org/c/openstack/cinder-specs/+/986741 | 14:43 |
| rosmaita | minwooseo says that the spec had been updated to address comments, so please re-review | 14:44 |
| simondodsley | already on it | 14:44 |
| rosmaita | there's also a reference implementation | 14:44 |
| rosmaita | #link https://review.opendev.org/c/openstack/cinder/+/986740 | 14:44 |
| rosmaita | #topic spec for all_tenants read policies | 14:45 |
| pungrumpy | hi, thanks. last week you asked for a spec for https://review.opendev.org/c/openstack/cinder/+/1012421, it is up as https://review.opendev.org/c/openstack/cinder-specs/+/1014452 and passes Zuul. | 14:45 |
| pungrumpy | short version: one <resource>:all_tenants policy per list, each defaulting to is_admin:True, the check the API makes today. so who can read across projects does not change until a deployer grants one. no new microversion proposed. | 14:45 |
| pungrumpy | two asks: could someone review it before the Nov 20 freeze? and there is one open question for reviewers in the REST API impact section, on whether two with_count edge cases need a microversion. | 14:45 |
| erlon28 | seems a pretty good idea | 14:45 |
| erlon28 | the image-cache ^ | 14:46 |
| simondodsley | pungrumpy: i'll look at it | 14:46 |
| pungrumpy | simondodsley: thank you | 14:46 |
| rosmaita | #topic open discussion | 14:47 |
| rakshay | @erlon28 @jbernard if you guys could please help me with this patchset: https://review.opendev.org/c/openstack/cinder/+/997962. all the comments have been addressed on it and it already has a +2. | 14:47 |
| erlon28 | pungrumpy: the spec was in the commit message before :P | 14:47 |
| pungrumpy | a CI note: openstack-tox-py314 timed out on 8 cinder check runs since Nov 6. in each one a single stestr worker stops early, 7 of 8 inside test_backup_ceph.BackupCephTestCase. looks like LP #2168492, and https://review.opendev.org/c/openstack/cinder/+/1007324 is still open for it. could it get a review? | 14:47 |
| rosmaita | pungrumpy: thanks for pointing that out, unfortunately i cannot review it myself | 14:49 |
| erlon28 | rakshay: I was trying to understand your comments and patch there | 14:49 |
| erlon28 | I'll get to it again | 14:49 |
| rakshay | sure. thanks! | 14:49 |
| erlon28 | What is your timezone? It would be nice to get you around to answer questions | 14:49 |
| harsh | Could someone please help on a second +2 for https://review.opendev.org/c/openstack/cinder/+/982617 | 14:52 |
| rakshay | @erlon28 I'm in IST (UTC+5:30), usually online around 09:30 to 18:30 IST | 14:52 |
| rosmaita | was just going to mention that, there's been a +2 on it for a month | 14:52 |
| rosmaita | also, the two short glance-related patches each have a +2, would be nice to close those out | 14:53 |
| erlon28 | rosmaita, it was my -1. | 14:53 |
| hong-p[m] | Hi, could some cores take a look at https://review.opendev.org/c/openstack/cinder/+/999400 ? NetApp cloned-LUN space allocation fix, with one open question at block_cmode.py L1322 for cores. Thanks! | 14:53 |
| alokeC | simondodsley, I have addressed the review comment on #link https://review.opendev.org/c/openstack/os-brick/+/986070. Please take a look when you get a chance. Thanks! | 14:53 |
| erlon28 | on the image-volume-cache-management spec, is it way to long and technical or it's just me that am getting too lazzy? | 14:54 |
| rosmaita | erlon28: all the specs are getting too long | 14:54 |
| simondodsley | alokeC: will try to get to it today | 14:54 |
| rosmaita | i suspect claude/gemini | 14:54 |
| simondodsley | the longer the spec, the better the implementation will be | 14:55 |
| alokeC | eharney_, simondodsley, erlon28: Also, I have uploaded PS8 on #link https://review.opendev.org/c/openstack/cinder/+/986072. Please take a look at the changes done when you get a moment. | 14:55 |
| rosmaita | simondodsley: not sure | 14:55 |
| rosmaita | really long spec hides a lot of issues | 14:56 |
| fqi | Thank you very much for your cooperation in reviewing the cinder spec. I would like to ask for your continued support in reviewing the patches. https://review.opendev.org/q/topic:%22retype-volume-including-snapshots%22 | 14:56 |
| abhishekk | tow short glance-related patches, https://review.opendev.org/c/openstack/cinder/+/1011471 and https://review.opendev.org/c/openstack/cinder/+/934261 | 14:56 |
| rosmaita | the claude text sounds very reasonable all the time and seems to cover issues | 14:56 |
| erlon28 | @all, we should target specs commit messages and comments for conciseness and clarity, and for human readers. Otherwise, we will have to copy and paste every review and ask for summaries. | 14:56 |
| simondodsley | more detail allows spec reviewers to better understand what is being suggested, and also allow a deeper analysis of the impact it will have to the existing code base | 14:56 |
| opendevreview | Anoop Kumar Shukla proposed openstack/cinder master: NetApp ONTAP: Fix migration from/to FlexGroup pool https://review.opendev.org/c/openstack/cinder/+/1011433 | 14:56 |
| simondodsley | in the long run it will save CI sysles | 14:57 |
| erlon28 | simondodsley: for the AI, yes, which is hard for us to follow especially giving the number of specs and patches | 14:57 |
| rosmaita | there's a difference between detail and verbosity | 14:57 |
| rosmaita | ok 2 minutes left ... anything else? | 14:58 |
| pungrumpy | thanks, I'll watch the spec review. | 14:59 |
| erlon28 | for example: Notifications impact | 14:59 |
| erlon28 | No new notifications. The transient volume sends the existing | 14:59 |
| erlon28 | ``volume.create.*`` / ``volume.delete.*``........ | 14:59 |
| simondodsley | do we want existing verbose specs to be redone? | 15:00 |
| rosmaita | not right now, but we need to keep this kind of thing in mind for next time | 15:00 |
| erlon28 | Not the ones that have already received enough feedback | 15:00 |
| rosmaita | erlon28: ++ | 15:00 |
| rosmaita | pre-AI, it was a PITA to type all this stuff out ... you would stop at "No new notifications" and not say anything else! | 15:01 |
| erlon28 | It's important (at least to me to understand the high level architecture) and design decisions | 15:01 |
| rosmaita | ok, thanks everyone for attending | 15:02 |
| rosmaita | #endmeeting | 15:02 |
| opendevmeet | Meeting ended Wed Nov 11 15:02:16 2026 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:02 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/cinder/2026/cinder.2026-11-11-14.01.html | 15:02 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/cinder/2026/cinder.2026-11-11-14.01.txt | 15:02 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/cinder/2026/cinder.2026-11-11-14.01.log.html | 15:02 |
| abhishekk | thank you!! | 15:04 |
| erlon28 | \o | 15:07 |
| *** auniyal4 is now known as auniyal | 15:21 | |
| zigo | rosmaita: I know why we're not using --random with stestr. It's simply silently dropped if using the --parallel mode ! | 16:14 |
| rosmaita | i did not know that | 16:18 |
| opendevreview | Saikumar Pulluri proposed openstack/cinder master: NetApp ONTAP: fix SVM-scoped REST SSC permission probe https://review.opendev.org/c/openstack/cinder/+/1014888 | 16:22 |
| opendevreview | Datta Srivathsava Gollapinni proposed openstack/cinder master: NetApp ONTAP: added unit tests for REST metrics-based pool utilization https://review.opendev.org/c/openstack/cinder/+/1010451 | 16:25 |
| zigo | rosmaita: It looks like a bug to me. | 16:31 |
| zigo | (I just checked...) | 16:31 |
| zigo | rosmaita: Added --random support here: https://github.com/mtreinish/stestr/pull/400/changes/9e440adf7b3de404b8a6dc20330267dce20d1ecb (part of my #400 PR). | 16:48 |
| zigo | This adds support for --random in the --dynamic mode, and also repairs it for --parallel. | 16:49 |
| simondodsley | zigo: do you want me to review your #400 PR? | 18:21 |
| zigo | That would be great, thanks ! | 18:22 |
| opendevreview | Merged openstack/cinder master: Simplify fields status constant usage https://review.opendev.org/c/openstack/cinder/+/1012322 | 18:22 |
| *** eharney_ is now known as eharney | 18:58 | |
| opendevreview | Eric Harney proposed openstack/cinder master: Remove unused get_volumes_image_metadata code https://review.opendev.org/c/openstack/cinder/+/1015206 | 19:43 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!