| opendevreview | OpenStack Proposal Bot proposed openstack/glance master: Imported Translations from Zanata https://review.opendev.org/c/openstack/glance/+/926597 | 03:39 |
|---|---|---|
| opendevreview | Tobias Urdin proposed openstack/glance master: Remove LegacyEngineFacade https://review.opendev.org/c/openstack/glance/+/926560 | 06:20 |
| opendevreview | Josephine Seifert proposed openstack/glance master: WIP migrate cinder_encryption* to os_encrypt* in db for Image Encryption https://review.opendev.org/c/openstack/glance/+/926905 | 08:32 |
| opendevreview | Stephen Finucane proposed openstack/glance-specs master: Remove use of sphinxcontrib-*diag packages https://review.opendev.org/c/openstack/glance-specs/+/927247 | 09:51 |
| tobias-urdin | im out of ideas why functional test fails for this DB changes https://review.opendev.org/c/openstack/glance/+/926560 – if anybody has any good ideas let me know | 10:02 |
| tobias-urdin | i tried setting stestr to run serially to see if it was reproducable then which it was | 10:03 |
| opendevreview | Merged openstack/glance master: Refresh Glance example configs for dalmatian m3 https://review.opendev.org/c/openstack/glance/+/927071 | 10:48 |
| frickler | tobias-urdin: "We should have deleted the task while querying for it, so make sure that our task is now marked as deleted." I didn't look closer yet, but that sounds like one of your sessions might need to become writer instead of reader | 11:12 |
| frickler | tobias-urdin: ack, I think I found it, commented on the review | 11:23 |
| frickler | and I do agree with anyone looking confused that "task_get_all()" doesn't sound like a function that one would expect to do updates | 11:46 |
| opendevreview | Jens Harbott proposed openstack/glance master: Remove LegacyEngineFacade https://review.opendev.org/c/openstack/glance/+/926560 | 11:48 |
| frickler | ^^ didn't find out how to run this locally, so let's just test my idea | 11:49 |
| tobias-urdin | frickler: nice catch! i was searching through everything but that was indeed confusing | 14:33 |
| frickler | tobias-urdin: sadly seems the CI is still failing with the same error | 14:36 |
| opendevreview | Tobias Urdin proposed openstack/glance master: Remove LegacyEngineFacade https://review.opendev.org/c/openstack/glance/+/926560 | 14:38 |
| tobias-urdin | frickler: found another one, the same! | 14:38 |
| opendevreview | Stephen Finucane proposed openstack/glance master: tests: Simplify version test https://review.opendev.org/c/openstack/glance/+/926646 | 14:51 |
| opendevreview | Stephen Finucane proposed openstack/glance master: conf: Add '[DEFAULT] allowed_hashing_algorithms' https://review.opendev.org/c/openstack/glance/+/922128 | 14:51 |
| opendevreview | Stephen Finucane proposed openstack/glance master: api: Make os_hash_algo user-configurable https://review.opendev.org/c/openstack/glance/+/922129 | 14:51 |
| opendevreview | Dan Smith proposed openstack/glance master: Use format_inspector from oslo https://review.opendev.org/c/openstack/glance/+/927291 | 15:26 |
| dansmith | abhishekk: ^ | 15:26 |
| dansmith | untested yet, but figure best to let zuul run it in parallel to my local testing | 15:26 |
| abhishekk | dansmith: ack | 15:45 |
| abhishekk | 1st glance looks good, lets wait for results | 15:49 |
| tobias-urdin | frickler: thanks! functional is finally passing now | 15:53 |
| frickler | \o/ | 16:05 |
| dansmith | abhishekk: yep, it's going to need another tempest exclusion like nova did, I just want to get a baseline | 17:09 |
| dansmith | and looks like I didn't run functionals, even though there's no format_inspector stuff in there, so I'm surprised it's failing | 17:09 |
| abhishekk | yeah, couple of tests are failing in functional | 17:15 |
| abhishekk | 2024-08-27 16:29:38.887 8375 DEBUG glance.location [-] Enabling in-flight format inspection for qcow2 set_data /home/zuul/src/opendev.org/openstack/glance/glance/location.py:588 | 17:19 |
| abhishekk | 2024-08-27 16:29:38.888 8375 DEBUG oslo_utils.imageutils.format_inspector [-] Format inspector failed, aborting: Signature KDMV not found: b'QFI\xfb' _process_chunk /home/zuul/src/opendev.org/openstack/glance/.tox/functional-py39/lib/python3.9/site-packages/oslo_utils/imageutils/format_inspector.py:1302 | 17:19 |
| abhishekk | this is failing while staging the data | 17:20 |
| abhishekk | 2024-08-27 16:29:38.892 8375 INFO glance.location [-] Image declared as qcow2 but detected as None, not updating virtual_size | 17:20 |
| opendevreview | Dan Smith proposed openstack/glance master: Use format_inspector from oslo https://review.opendev.org/c/openstack/glance/+/927291 | 17:41 |
| dansmith | abhishekk: yeah, I forgot I need to close the wrapper for it to finish any of the ones that look at the end of the stream | 17:41 |
| dansmith | I think this ^ will fix that | 17:42 |
| abhishekk | cool | 17:42 |
| dansmith | the vmdk log message is probably too verbose as that's expected to fail if you're looking at a qcow of course | 17:42 |
| abhishekk | yes | 17:45 |
| dansmith | lol, I suck | 17:55 |
| opendevreview | Dan Smith proposed openstack/glance master: Use format_inspector from oslo https://review.opendev.org/c/openstack/glance/+/927291 | 18:00 |
| abhishekk | looks like functional is fixed now, tempest run will take around 2-3 hours | 18:05 |
| dansmith | tempest should just fail the footer test until I exclude it like I did for. nova, but I want to record that before I do | 18:06 |
| opendevreview | Dan Smith proposed openstack/glance master: Use format_inspector from oslo https://review.opendev.org/c/openstack/glance/+/927291 | 18:10 |
| dansmith | abhishekk: can we get someone to approve the iso patch below? | 18:11 |
| dansmith | perhaps croelandt | 18:11 |
| abhishekk | I think so, but he might have left for the day | 18:12 |
| abhishekk | rosmaita: around ^^ | 18:13 |
| rosmaita | i'll take a look in a bit | 18:16 |
| abhishekk | thank you! | 18:30 |
| dansmith | abhishekk: I'm seeing functional failures like this locally: https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_6fa/927291/4/check/openstack-tox-functional-py310/6fa2fbf/testr_results.html | 18:32 |
| dansmith | although for me all those new tests fail 100% of the time | 18:32 |
| abhishekk | looking | 18:33 |
| abhishekk | :/ | 18:37 |
| abhishekk | Not failing for me locally | 18:37 |
| dansmith | lemme re-run and get you a list | 18:38 |
| abhishekk | which platform you are running it on?, I have ubuntu + py310 | 18:38 |
| abhishekk | ack | 18:38 |
| dansmith | fedora something 3.11 | 18:38 |
| abhishekk | ack | 18:39 |
| dansmith | abhishekk: these six fail 100% for me: https://termbin.com/jd08 | 18:41 |
| abhishekk | looking | 18:41 |
| abhishekk | let me spin vm with fedora and run it locally, mostly it will take some time | 18:42 |
| dansmith | I doubt it's fedora specific, maybe py311 ish, but that job ^ was 3.10 | 18:42 |
| abhishekk | right | 18:43 |
| abhishekk | could you please also provide one full failure strace? | 18:43 |
| abhishekk | tomorrow my time I will install py311 on ubuntu and try | 18:46 |
| dansmith | abhishekk: v | 18:53 |
| dansmith | https://termbin.com/i9uv | 18:53 |
| dansmith | abhishekk: https://termbin.com/i9uv | 18:56 |
| dansmith | abhishek_: ^ | 18:56 |
| abhishek_ | thank you! | 18:56 |
| abhishek_ | strange, it is working for me fedora + py311 | 18:56 |
| dansmith | *shrug*.. fails 100% of the time for me :/ | 18:57 |
| abhishek_ | I will provide this to pdeore and will have a look together | 18:57 |
| dansmith | ++ thanks | 18:57 |
| abhishek_ | cat /etc/redhat-release | 18:58 |
| abhishek_ | Fedora release 37 (Thirty Seven) | 18:58 |
| dansmith | same | 18:58 |
| abhishek_ | np! thank you for the logs | 18:58 |
| abhishek_ | just last question, is this on master ? | 19:00 |
| dansmith | abhishek_: it's on top of your iso patch | 19:00 |
| dansmith | https://review.opendev.org/c/openstack/glance/+/927291/4 | 19:00 |
| abhishek_ | OK and I can see 2024-08-27 11:39:17.410 2188323 WARNING stevedore.named [-] Could not load location_import | 19:01 |
| dansmith | unless your iso patch is breaking it, I can re-test, but I very much doubt it because it's only testing locations | 19:01 |
| abhishek_ | is this the reason? | 19:01 |
| dansmith | idk | 19:01 |
| abhishek_ | ack, applied above patch and running the tests | 19:03 |
| dansmith | same tests fail on master | 19:03 |
| dansmith | just replicated there | 19:03 |
| abhishek_ | ack | 19:03 |
| abhishek_ | for me it is passing with above patch as well | 19:05 |
| dansmith | ack, well, it's clearly not just me because it failed in that CI run | 19:06 |
| dansmith | and since it was just one, it's probably a race of some sort | 19:06 |
| abhishek_ | hmm, between do you have location_import in glance.egg-info/entry_points.txt | 19:08 |
| dansmith | no | 19:09 |
| dansmith | wait, in .tox you mean? | 19:09 |
| abhishek_ | can you add below line under glance.flows | 19:09 |
| abhishek_ | location_import = glance.async_.flows.location_import:get_flow | 19:09 |
| dansmith | oh tox links to . I guess | 19:10 |
| abhishek_ | in /opt/staack/glance/glance.egg-info/entry_points.txt | 19:10 |
| dansmith | I'm running locally not in a devstack | 19:10 |
| abhishek_ | ok | 19:10 |
| abhishek_ | Now I have removed that line from my above file and 6 tests are failing | 19:10 |
| dansmith | I cleaned my glance.egg-info and am rebuilding | 19:11 |
| dansmith | but, again, it failed the same way in the CI job | 19:11 |
| abhishek_ | ack | 19:11 |
| dansmith | yep, rm -Rf glance.egg-info and now it passes | 19:11 |
| abhishek_ | yeah 1 with some kind of race | 19:11 |
| abhishek_ | I will highlight that to pranali to have a look | 19:11 |
| dansmith | it looked like the same failure behavior but okay | 19:11 |
| dansmith | ack | 19:12 |
| abhishek_ | thank you | 19:12 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!