| opendevreview | Atsushi Kawai proposed openstack/cinder master: Hitachi: support extending volume with snapshot https://review.opendev.org/c/openstack/cinder/+/850830 | 03:39 |
|---|---|---|
| *** shoffmann is now known as Guest3972 | 07:39 | |
| *** shoffmann1 is now known as shoffmann | 07:39 | |
| opendevreview | Atsushi Kawai proposed openstack/cinder master: Hitachi: Support retype to another pool for VSP driver https://review.opendev.org/c/openstack/cinder/+/842905 | 08:10 |
| Kumar_Kanishka | Hi all, please review my patch | 08:36 |
| Kumar_Kanishka | https://review.opendev.org/c/openstack/cinder/+/869305 | 08:36 |
| geguileo | whoami-rajat: fyi Cinder's DB unit tests are horribly broken... The MySQL tests are actually run on SQLite | 10:48 |
| whoami-rajat | geguileo, oh, that's bad news ... | 10:50 |
| whoami-rajat | can we see it run on mysql somewhere? or we need to run it locally? | 10:50 |
| geguileo | whoami-rajat: I'm running it locally to reproduce this gate failure: https://zuul.opendev.org/t/openstack/build/a217b74a44f14ab6b77ccde6525c2cd1 | 10:51 |
| geguileo | whoami-rajat: I couldn't tell looking at upstream logs that that was the problem | 10:52 |
| geguileo | whoami-rajat: sfinucan is on PTO :-( | 10:55 |
| whoami-rajat | geguileo, I'm also not able to understand why the groups table is missing there but i didn't dig enough | 10:59 |
| whoami-rajat | stephenfin, when you're around, please take a look at ^ | 10:59 |
| whoami-rajat | hopefully he will see the message tomorrow or we can ping him again | 10:59 |
| geguileo | whoami-rajat: he's on PTO until the 13th | 10:59 |
| whoami-rajat | oh ... | 10:59 |
| geguileo | whoami-rajat: oh, that's the misleading thing, the problem is not the groups table! | 11:00 |
| geguileo | whoami-rajat: so the migration thinks it's going to mysql, but only the DB creation is going there | 11:00 |
| geguileo | whoami-rajat: all the migration commands are going to sqlite | 11:00 |
| geguileo | whoami-rajat: and because the migration thinks it's going to mysql, it runs a command that only works on mysql but fails on sqlite | 11:01 |
| geguileo | whoami-rajat: and that's the failure, running a non supported feature in sqlite | 11:01 |
| geguileo | it took me a while to get there... | 11:01 |
| whoami-rajat | geguileo, oh ok, I'm still thinking about what changed that triggered this issue in our gates | 11:06 |
| whoami-rajat | i see oslo.db was recently released https://pypi.org/project/oslo.db/#history | 11:06 |
| geguileo | whoami-rajat: oh, this is NOT something "new" | 11:06 |
| whoami-rajat | and that is used in our gate | 11:06 |
| whoami-rajat | 2023-02-06 18:28:00.591945 | ubuntu-jammy | Downloading https://mirror.bhs1.ovh.opendev.org/pypifiles/packages/b4/45/eaabc90bfc62ef41af24b869dee0e556624bfe6211da1e5653015c8a9cc4/oslo.db-12.3.1-py3-none-any.whl (177 kB) | 11:06 |
| geguileo | whoami-rajat: this is most likely to do with the alembic migration changes, but I'm not 100% sure | 11:07 |
| geguileo | whoami-rajat: if this was a problem with packages, then Ironic whould be failing | 11:07 |
| whoami-rajat | geguileo, i understand it could be an old issue but we didn't face it in the gate before, maybe a random behavior? | 11:08 |
| geguileo | whoami-rajat: no, no, the behavios is only triggering now because I'm using in that patch a new function | 11:08 |
| geguileo | whoami-rajat: https://review.opendev.org/c/openstack/cinder/+/819669/4/cinder/db/migrations/versions/daa98075b90d_add_resource_indexes.py | 11:09 |
| geguileo | whoami-rajat: L44 | 11:09 |
| whoami-rajat | geguileo, oh got it | 11:10 |
| whoami-rajat | so cinder gate isn't broken but our UTs are that we need to fix | 11:11 |
| geguileo | whoami-rajat: correct, I think I figured out how to fix it | 11:25 |
| whoami-rajat | wow, that was quick! | 11:27 |
| geguileo | whoami-rajat: not quick, I've been at it a while... And it doesn't solve all the problems... | 11:29 |
| geguileo | so I have to keep digging | 11:29 |
| whoami-rajat | geguileo, sure, let me know if i can help anywhere | 11:31 |
| whoami-rajat | though I'm not very familiar with how alembic really works | 11:31 |
| geguileo | whoami-rajat: I'm not familiar with it either, that's why I'm stumbling a lot | 11:33 |
| geguileo | whoami-rajat: from what I can gather our current UT setup is a CF | 11:33 |
| geguileo | whoami-rajat: we have parts that read from an INI file that has hardcoded the sqlalchemy.url location | 11:34 |
| geguileo | our main context manager doesn't get properly configured... | 11:35 |
| whoami-rajat | geguileo, i see similar hardcoded url in nova as well https://github.com/openstack/cinder/blob/a803c275c177fe005f3963a85b8688663101be85/cinder/db/alembic.ini#L38 | 11:41 |
| geguileo | whoami-rajat: probably ok depending on how other things are handled... | 11:42 |
| whoami-rajat | sorry wrong url : https://github.com/openstack/nova/blob/master/nova/db/api/alembic.ini#L42 | 11:43 |
| whoami-rajat | yep, just trying to eliminate places where our code is similar to nova | 11:43 |
| geguileo | whoami-rajat: ok, I have found a way to solve it | 11:56 |
| geguileo | whoami-rajat: at least for the migration tests... | 11:56 |
| whoami-rajat | so there are more places to be fixed? | 11:59 |
| geguileo | whoami-rajat: Ooooooooh, I think we've had this bug since FOREVER!!!! | 12:00 |
| geguileo | rofl rofl rofl | 12:00 |
| geguileo | and here I was about to complain to stephenfin when it had nothing to do will his work... | 12:00 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Improve resource listing efficiency https://review.opendev.org/c/openstack/cinder/+/819669 | 12:18 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Fix Migrations UTs using wrong DB https://review.opendev.org/c/openstack/cinder/+/872907 | 12:18 |
| geguileo | whoami-rajat: ^ | 12:18 |
| geguileo | 3 hours to write 3 LOC... not very efficient | 12:19 |
| whoami-rajat | geguileo, but the learning involved to get to those 3 LOC might be worth it :D | 13:51 |
| opendevreview | Sofia Enriquez proposed openstack/cinder master: WIP Create clone from luks inside qcow2 volume https://review.opendev.org/c/openstack/cinder/+/872052 | 13:59 |
| whoami-rajat | rosmaita, enriquetaso jungleboyj and other cores, can you take a look at this change? it's blocking a feature I've done on glance side https://review.opendev.org/c/openstack/cinder/+/869051 | 14:15 |
| rosmaita | whoami-rajat: ack, may be a little while though | 14:26 |
| whoami-rajat | sure, there's time but wanted to prioritize it (else i might miss FF) | 14:35 |
| *** shoffmann is now known as Guest4022 | 15:56 | |
| *** shoffmann1 is now known as shoffmann | 15:56 | |
| *** chuanm1 is now known as chuanm | 16:42 | |
| opendevreview | Evelina Shames proposed openstack/cinder-tempest-plugin master: WIP - Test srbac on snapshots https://review.opendev.org/c/openstack/cinder-tempest-plugin/+/872986 | 17:35 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Improve resource listing efficiency https://review.opendev.org/c/openstack/cinder/+/819669 | 18:11 |
| opendevreview | Merged openstack/os-brick master: Support separate privsep logging levels https://review.opendev.org/c/openstack/os-brick/+/871835 | 18:53 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!