| -@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 821505: Report overall duration when a build set is finished https://review.opendev.org/c/zuul/zuul/+/821505 | 02:26 | |
| -@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 821505: Report overall duration when a build set is finished https://review.opendev.org/c/zuul/zuul/+/821505 | 02:29 | |
| -@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 821505: Report overall duration when a build set is finished https://review.opendev.org/c/zuul/zuul/+/821505 | 02:32 | |
| @jim:acmegating.com | zuul-maint: i think opendev is pretty stable... how does this look for a zuul release? commit 9a2fd432f1ee87c8b9cb9fbb89e15590c544b844 (HEAD -> master, tag: 4.11.0, origin/master, refs/changes/75/821275/2) | 14:55 |
|---|---|---|
| @avass:vassast.org | What's left for a V5 release after that? | 15:12 |
| @jim:acmegating.com | i think we'll get the docs reorg and the gearmanectomy in | 15:15 |
| @ashleybullock:matrix.org | Heya, I've been looking into an odd issue with how nodepool marks an AWS ec2 as "Ready", I was wondering if anyone has encountered something similar. | 15:17 |
| The ec2 is a windows 2019 server, and I've configured winrm certificate authentication. The issue is nodepool marks the instance ready for a job | ||
| before the ec2 has fully initialised (Status checks are still not intitialising) and this returns a "Device is not ready" error. | ||
| If I set min-ready to 1 and wait a little longer before scheduling a job it works fine, I've tried playing around with boot-timeout (set in nodepool config), and | ||
| winrm settings for timeouts in the executor configs but I don't think they apply as the connection is being made yet failing. | ||
| Has anyone experienced anything similar? Thanks in advance | ||
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 821711: Add IBM Cloud VPC driver https://review.opendev.org/c/zuul/nodepool/+/821711 | 15:31 | |
| @fungicide:matrix.org | corvus: 9a2fd43 merged two days after the last scheduler restart in opendev, though it doesn't look like we logged that restart on tuesday | 15:43 |
| @fungicide:matrix.org | i have a feeling we're running on 85dc6d6, so only the readiness/liveness probe for prometheus and the upgrade release note about delete-state merged after that | 15:45 |
| @fungicide:matrix.org | seems reasonable to make the current master head 4.11.0 in that case | 15:46 |
| @jim:acmegating.com | fungi: that seems likely... and i'm not sure if we would exercise the prom change in opendev either... | 15:47 |
| @fungicide:matrix.org | the only potential problem we're currently aware of in opendev is management event backlogs from bulk branch manipulation (create and possibly delete) events | 15:47 |
| @jim:acmegating.com | yeah, and if that's a problem, it's not a recent one | 15:47 |
| @jim:acmegating.com | (ie, is almost certainly in 4.10.x) | 15:48 |
| @fungicide:matrix.org | we've seen it happen twice with bulk branch creation across repos twice last week, and i'm presently monitoring to see if a bulk branch deletion is going to cause the same here in a few minutes | 15:48 |
| @fungicide:matrix.org | openstack is about to delete some retired branches from repos, about 50 deletions in total, but going to try some small batches first and space them out | 15:49 |
| @avass:vassast.org | Ashley Bullock: i think nodepool marks the node ready as soon as it can connect to the ec2 instance on the winrm socket | 16:26 |
| @avass:vassast.org | So if you can delay the server from exposing that port until it's ready then it could work a bit better. Maybe you can block the port with the windows firewall until the server is ready could work | 16:29 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 821711: Add IBM Cloud VPC driver https://review.opendev.org/c/zuul/nodepool/+/821711 | 17:09 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 821740: Update zuul-changes for use with zuul-client https://review.opendev.org/c/zuul/zuul/+/821740 | 17:18 | |
| @jim:acmegating.com | i think there's no reason not to go ahead and merge that ^ now (even prior to the gearmanectomy); it will help folks prepare. i believe opendev should be able to use that script unmodified. | 17:20 |
| @clarkb:matrix.org | If it isn't landed by this afternoon I should be able to take a look alter today after meetings | 17:38 |
| @jim:acmegating.com | fungi: i reproduced the event-merge error and found the bug... i'm going to work on a fix and then double check my assertion that it was present before 4.10. if so, i'll make the 4.11 release as planned; if not, let's regroup. | 18:03 |
| @fungicide:matrix.org | > <@jim:acmegating.com> fungi: i reproduced the event-merge error and found the bug... i'm going to work on a fix and then double check my assertion that it was present before 4.10. if so, i'll make the 4.11 release as planned; if not, let's regroup. | 18:29 |
| oh, awesome. thanks! i agree if it's a regression after 4.10 then fixing in 4.11 would be preferable. if it was already a problem in 4.10 then fixing after 4.11 is fine | ||
| @jim:acmegating.com | fungi: i was wrong. the bug is not yet released, so i think we should delay 4.11 for the fix. | 18:59 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 821771: Fix merging reconfiguration events https://review.opendev.org/c/zuul/zuul/+/821771 | 19:00 | |
| @jim:acmegating.com | zuul-maint: ^ let's hold 4.11 for that bugfix | 19:01 |
| @maheshbarai1234:matrix.org | Hi Team, I am trying to add Windows node via SSH key to the zuul executor. I am getting below error: | 19:06 |
| - Openssh server is working fine on Windows node and we are able to ssh from zuul executor into Windows machine. | ||
| Any Clue? | ||
| 2021-12-09 15:27:30,074 DEBUG zuul.AnsibleJob.output: [e: fc63d780-58fb-11ec-9cbc-6f4f0d4c3d87] [build: f652bbddd8914af5a4425d443a79dd5c] Ansible output: b'[WARNING]: Platform unknown on host windows-vm-stark is using the discovered' | ||
| 2021-12-09 15:27:30,074 DEBUG zuul.AnsibleJob.output: [e: fc63d780-58fb-11ec-9cbc-6f4f0d4c3d87] [build: f652bbddd8914af5a4425d443a79dd5c] Ansible output: b'Python interpreter at /usr/bin/python, but future installation of another' | ||
| 2021-12-09 15:27:30,074 DEBUG zuul.AnsibleJob.output: [e: fc63d780-58fb-11ec-9cbc-6f4f0d4c3d87] [build: f652bbddd8914af5a4425d443a79dd5c] Ansible output: b'Python interpreter could change this. See https://docs.ansible.com/ansible/2.9/' | ||
| 2021-12-09 15:27:30,074 DEBUG zuul.AnsibleJob.output: [e: fc63d780-58fb-11ec-9cbc-6f4f0d4c3d87] [build: f652bbddd8914af5a4425d443a79dd5c] Ansible output: b'reference_appendices/interpreter_discovery.html for more information.' | ||
| @clarkb:matrix.org | corvus: noted, and that is now high on my list of todos after the infra meeting | 19:09 |
| @avass:vassast.org | Mahesh Barai: have you configured https://zuul-ci.org/docs/nodepool/aws.html#attr-providers.[aws].cloud-images.shell-type ? | 19:15 |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 821740: Update zuul-changes for use with zuul-client https://review.opendev.org/c/zuul/zuul/+/821740 | 19:21 | |
| @tobias.henkel:matrix.org | corvus: +2 with q on 821771 | 19:24 |
| @clarkb:matrix.org | tobiash: I attempted to answer | 19:31 |
| @jim:acmegating.com | yep correct | 19:58 |
| @clarkb:matrix.org | I'ev approved that fix just now | 20:00 |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 821771: Fix merging reconfiguration events https://review.opendev.org/c/zuul/zuul/+/821771 | 22:14 | |
| @jim:acmegating.com | we just rolling-restarted opendev with that. in doing so, Clark spotted another bug relating to semaphores. i believe i understand the issue and will work on a test/fix. | 23:04 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!