| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Update test-prepare-workspace-git to use a module https://review.opendev.org/c/zuul/zuul-jobs/+/925539 | 00:00 |
|---|---|---|
| *** ykarel__ is now known as ykarel | 13:30 | |
| *** ykarel__ is now known as ykarel | 13:57 | |
| *** darmach7 is now known as darmach | 14:42 | |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Update test-prepare-workspace-git to use a module https://review.opendev.org/c/zuul/zuul-jobs/+/925539 | 14:52 |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Synchronize test-prepare-workspace-git to prepare-workspace-git https://review.opendev.org/c/zuul/zuul-jobs/+/925540 | 14:52 |
| *** darmach3 is now known as darmach | 15:08 | |
| clarkb | https://zuul.opendev.org/t/openstack/builds?project=openstack%2Frequirements&pipeline=periodic&skip=0 I think this shows the new wheel keytab is working | 15:18 |
| fungi | oh, yep i meant to point tat out earlier but got sidetracked by e-mail | 15:23 |
| clarkb | apparently you can't use failed when on a block level in ansible | 16:02 |
| * clarkb will fix that change up shortly | 16:02 | |
| opendevreview | Clark Boylan proposed opendev/base-jobs master: More improvement to base-test cleanup and post playbooks https://review.opendev.org/c/opendev/base-jobs/+/925549 | 16:05 |
| opendevreview | Clark Boylan proposed opendev/base-jobs master: Update cleanups for base and base-minimal https://review.opendev.org/c/opendev/base-jobs/+/924787 | 16:05 |
| clarkb | infra-prod-letsencrypt appears to have last run on july 28, this is why the gitea certs haven't renewed yet | 16:11 |
| clarkb | infra-prod-base has been failing | 16:12 |
| clarkb | oh its because the linaro mirror is unreachable | 16:13 |
| * clarkb writes a change | 16:13 | |
| opendevreview | Clark Boylan proposed opendev/system-config master: Remove most linaro cloud resources https://review.opendev.org/c/opendev/system-config/+/925611 | 16:23 |
| clarkb | that ended up being a fair bit more involved than I expected. Please review carefully | 16:24 |
| clarkb | I can also push up a change that just modifies the inventory list for now if we prefer then do broader claenup as a followup, but I think this is relatively safe just need to make sure we didn't miss anything important whcih will result in more failures | 16:24 |
| fungi | oh, hah | 16:25 |
| clarkb | https://review.opendev.org/c/opendev/system-config/+/925611/1/inventory/service/groups.yaml this is maybe the most interesting bit of the change | 16:26 |
| fungi | yeah, i think that should be fine | 16:28 |
| clarkb | I don't know if we've tested empty lists in groups with teh yamlgroup thing, I suspect it will work fine though | 16:28 |
| clarkb | actually we have a test that should cover functionality of that its why I have to change https://review.opendev.org/c/opendev/system-config/+/925611/1/playbooks/roles/install-ansible/files/inventory_plugins/test-fixtures/results.yaml | 16:28 |
| clarkb | so if that test passes I think we're good | 16:28 |
| clarkb | https://review.opendev.org/c/opendev/system-config/+/925611 passes testing. I think the jobs that ran are robust enough to cover any major issues so this should be fine | 17:26 |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Synchronize test-prepare-workspace-git to prepare-workspace-git https://review.opendev.org/c/zuul/zuul-jobs/+/925540 | 17:32 |
| clarkb | changes like ^ reall make me want gerrit to allow pulling up arbitrary files side by side. ThoughI guess in this case diffing them is less useful /me will just do it locally | 17:32 |
| fungi | went ahead and approved 925611 so our certs will get updating again | 17:43 |
| clarkb | looks like one of the gate jobs failed on the ansible rc -13 issue so that will need to be rechecked | 18:31 |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Update test-prepare-workspace-git to use a module https://review.opendev.org/c/zuul/zuul-jobs/+/925539 | 18:46 |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Synchronize test-prepare-workspace-git to prepare-workspace-git https://review.opendev.org/c/zuul/zuul-jobs/+/925540 | 18:46 |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Update test-prepare-workspace-git to use a module https://review.opendev.org/c/zuul/zuul-jobs/+/925539 | 18:47 |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Synchronize test-prepare-workspace-git to prepare-workspace-git https://review.opendev.org/c/zuul/zuul-jobs/+/925540 | 18:47 |
| clarkb | fungi: I have rechecked that change. But I'm about to pop out for lunch now. Feel free to -W if that is prudent | 18:56 |
| fungi | thanks, i was waiting for the failure notification, you beat me to it | 18:56 |
| opendevreview | Julia Kreger proposed openstack/diskimage-builder master: Change centos DIB_RELEASE default to 9-stream https://review.opendev.org/c/openstack/diskimage-builder/+/925629 | 19:53 |
| opendevreview | Merged opendev/system-config master: Remove most linaro cloud resources https://review.opendev.org/c/opendev/system-config/+/925611 | 21:31 |
| *** clarkb is now known as Guest1974 | 21:33 | |
| Clark[m] | Jobs for ^ are running and will also renew certs since that job got enqueued due to the vars updates | 21:51 |
| Clark[m] | Hrm the let's encrypt job is failing | 22:16 |
| Clark[m] | Progress that it is running I guess. But not sure why it is sad | 22:17 |
| Guest1974 | the job failed because nb04 filled /opt and the job couldn't manage the acme.sh script there | 22:37 |
| Guest1974 | I'll stop the service there then clear out /opt/dib_tmp and hopefully the daily LE run will be happy again | 22:38 |
| Guest1974 | also I'm a guest now apparently. Will fix that too I guess | 22:38 |
| fungi | yeah, seems there was a netsplit or ircd restart. i also got disconnected briefly | 22:40 |
| Guest1974 | ok I've got deletion srunning in a screen because I'm mobile and iirc that can be slow on these arm nodes | 22:41 |
| Guest1974 | as far as I can tell that was the only node that was sad with the LE run | 22:42 |
| *** Guest1974 is now known as clarkb | 22:44 | |
| clarkb | much better | 22:44 |
| clarkb | related to being mobile and having a freshly installed laptop it is slightly annoying that sshfp records are hex and the fingerprint ssh spits out when you try to connect the first time is in base64 (without proper padding even). Anyway a short script later and some brute forced padding appending and now I have a script to make this easier | 23:10 |
| clarkb | fungi: one thing I've noticed looking at the acme.sh repo is that way we checkout the working release then cherry pick commits results in a very busy reflog (we're basically always making edits every day to the same state we were in before) | 23:29 |
| clarkb | I think you wrote the change to patch that stuff. Anyway its got me wondering if we should use another approach (just apply patches in a working dir or maybe check if the state is good first or fork and the ncheckout a branch/tag that we control?) | 23:29 |
| clarkb | I don't think its a huge impact the repo is only 24M but something to consider I guess | 23:30 |
| clarkb | or maybe upstream finally fixed things and we can go back to using latest | 23:30 |
| fungi | i'll try to remember to take another look at it next week | 23:38 |
| clarkb | ya I don't think it is urgent its been working this way for a while | 23:38 |
| clarkb | but I think that is one reason the acme stuff is so sensitive to the disk filling | 23:39 |
| fungi | maybe they've fixed whatever we were patching it for | 23:39 |
| clarkb | ya that would be the best outcome | 23:39 |
| fungi | i don' remember at the moment why we did it, but expect it was meant to be temporary | 23:41 |
| clarkb | good news is the deletins are outpacing new image builds so we should be good, and as suspected this is not fast so I'm glad I put it in screen | 23:46 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!