| opendevreview | chandan kumar proposed openstack/project-config master: Add snu-csl/nvmevirt to available repos https://review.opendev.org/c/openstack/project-config/+/977194 | 06:13 |
|---|---|---|
| *** tkajinam_ is now known as tkajinam | 10:13 | |
| jgilaber | Hi! I'm not sure if this is the right place, but I'm seeing this error in Watcher py310 unit tests and grenade jobs: "ModuleNotFoundError: No module named 'typing_extensions'" | 11:06 |
| jgilaber | it seems to be happening in other projects as well, e.g in heat https://zuul.opendev.org/t/openstack/build/ec370b503a064dba8210b0f9ccb0703e and nova https://zuul.opendev.org/t/openstack/build/509b69e561e847dc8d4a570954ee1007 | 11:07 |
| jgilaber | has this been reported? | 11:07 |
| jgilaber | it seems to be widespread https://zuul.opendev.org/t/openstack/builds?result=POST_FAILURE&skip=0 | 11:08 |
| frickler | jgilaber: yes, fix was already merged, see https://github.com/testing-cabal/testtools/pull/570 | 11:16 |
| jgilaber | ack thanks frickler that was quick! | 11:17 |
| *** croeland1 is now known as croelandt | 14:18 | |
| *** vhari_ is now known as vhari | 14:59 | |
| *** promethe- is now known as prometheanfire | 15:50 | |
| *** prometheanfire is now known as Guest3193 | 15:50 | |
| *** Guest3193 is now known as prometheanfire | 15:50 | |
| frickler | clarkb: fungi: can you please check https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/968510 this really should have gone in early in the cycle, but best don't delay even further (stumbled about that while looking at the next PTI proposal) | 15:56 |
| opendevreview | Merged openstack/openstack-zuul-jobs master: Add jobs for Python 3.14 https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/968510 | 16:09 |
| frickler | thx | 16:28 |
| sean-k-mooney | is zull ok. https://review.opendev.org/c/openstack/cyborg/+/975046 does not seam to be in check or gate env thoug its approved and i recheck | 19:33 |
| sean-k-mooney | im just wrappign up for the day so its not critial just looks like something is stuck | 19:33 |
| fungi | sean-k-mooney: it has an outdated parent change | 19:34 |
| sean-k-mooney | that shoudl not block it the parent is merged and its not in merge conlfict | 19:34 |
| sean-k-mooney | so this shoudl run agains master | 19:34 |
| sean-k-mooney | or rather it shoudl be rebased by zuul against master | 19:34 |
| fungi | its parent is change 970152 patchset 7 (3c4ed53) which does not exist in master | 19:35 |
| sean-k-mooney | yes v8 merged | 19:35 |
| sean-k-mooney | https://review.opendev.org/c/openstack/cyborg/+/970152/8 | 19:35 |
| fungi | change 970152 patchset 8 is caecf34 | 19:36 |
| sean-k-mooney | yep | 19:36 |
| fungi | so 975046 has a parent which does not exist in master, there is no way for it to be merged in git | 19:36 |
| sean-k-mooney | but i manually rechecked so zuul shoudl have said its in merge conflict or seen its partent is merged and check it agaisn master | 19:36 |
| sean-k-mooney | it shoudl able to | 19:37 |
| fungi | it can be rebased or cherry-picked onto a commit which is in the master branch history (such as caecf34) | 19:37 |
| fungi | where will git find the parent if it doesn't exist on the branch? | 19:37 |
| sean-k-mooney | zuul shoudl see the parren review is mergd and apply this directly on top of master breakign the parent relashioship | 19:38 |
| sean-k-mooney | taht what it used to do in the past | 19:38 |
| fungi | if it were to rebase or cherry-pick the change automatically that would be possible | 19:38 |
| fungi | no, zuul has never done that | 19:38 |
| sean-k-mooney | im prety sure whe have not had to manually rebase later comiit in a serise if we respon an earlier patch and htere was no merge conflict | 19:39 |
| fungi | zuul performs a git merge when it tests a change | 19:39 |
| fungi | git merge can't handle missing parents on the branch unless you also merge the missing parents | 19:39 |
| fungi | it has always been necessary to rebase changes in a series in gerrit if you alter their parent changes | 19:40 |
| sean-k-mooney | maybe i normlay do but i didnt think that was the zuul behavior i tought hat was only need on github | 19:41 |
| sean-k-mooney | it probaly shoudl reprot somting back to gerrit in this case but ill just rebase it and let it run | 19:41 |
| fungi | the crux of the problem isn't really zuul, it's a git problem and gerrit itself will be unable to merge the change with its built-in merge method (jgit's really) | 19:42 |
| sean-k-mooney | it depend on the sumbiut stragies to some degre | 19:42 |
| sean-k-mooney | we are usign merge if nessisary | 19:42 |
| sean-k-mooney | it has other that can do this | 19:43 |
| fungi | gerrit *can* submit changes in that condition when using its rebase or cherry-pick methods, but not merge (which is what we've used since ~2012) | 19:43 |
| fungi | precisely because changing the parent commit changes the commit id of the change | 19:44 |
| sean-k-mooney | https://gerrit.cloudera.org/Documentation/project-configuration.html#merge_if_necessary i could be confusign it with downstream were i configured https://gerrit.cloudera.org/Documentation/project-configuration.html#rebase_if_necessary | 19:44 |
| sean-k-mooney | actuly no downstrema i used https://gerrit.cloudera.org/Documentation/project-configuration.html#rebase_always because we waned the reviewd on linek and appvoe by ectra | 19:45 |
| fungi | yes, rebase will handle it, but the result is that the commit that ends up in the branch may not match the commit in the tested patchset of the change | 19:45 |
| fungi | i think the result, though it's been a while since i played with it, is that gerrit auto-creates a final patchset you can see in the ui when it rebases | 19:46 |
| sean-k-mooney | its more the lack of feedback that i was rasiign then anythign else to be fair | 19:46 |
| fungi | yes, the lack of ui feedback in these situations is a problem, we've talked about it extensively in the past | 19:46 |
| fungi | gerrit does have indicators for this, but they're far too subtle | 19:47 |
| sean-k-mooney | well as i said im used to it workign diffently but that baised by downstream as well where we configured it different | 19:48 |
| fungi | in the relation chain list, the parent "(Merged)" is orange-colored instead of grey | 19:48 |
| sean-k-mooney | yep | 19:48 |
| sean-k-mooney | typically i expect it to end up in merge conflict like this https://review.opendev.org/c/openstack/cyborg/+/977020/1 as well but i guess waht zuul and gerrit will reslvoe in thsi regard differ enough for it to soemtiems be an issue | 19:49 |
| sean-k-mooney | ill try to rememebr that in the future | 19:50 |
| opendevreview | Dr. Jens Harbott proposed openstack/openstack-zuul-jobs master: Add openstack-tox-py314 jobs to python templates https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/977360 | 20:17 |
| frickler | gmaan: fungi: ^^ I wanted to check the results of some py314 jobs after 968510 merged and the noticed that we didn't actually add them to the python templates | 20:20 |
| fungi | sorry i didn't call that out on the previous change, i wrongly assumed you had planned to phase it in with a separate change for adding those jobs to the central templates after trying them in a project-local template or similar | 20:21 |
| fungi | er, project pipeline i mean | 20:22 |
| gmaan | yeah, we missed that. lgtm | 20:22 |
| opendevreview | Merged openstack/openstack-zuul-jobs master: Add openstack-tox-py314 jobs to python templates https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/977360 | 20:35 |
| opendevreview | Brian Haley proposed openstack/openstack-zuul-jobs master: Add openstack-tox-py314-arm64 jobs to python template https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/977374 | 21:53 |
| opendevreview | Jeremy Stanley proposed openstack/project-config master: Replace 2026.1/Gazpacho key with 2026.2/Hibiscus https://review.opendev.org/c/openstack/project-config/+/977380 | 22:33 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!