| opendevreview | Radosław Piliszek proposed openstack/project-config master: Add new repository for NebulOuS testing data https://review.opendev.org/c/openstack/project-config/+/916876 | 10:56 |
|---|---|---|
| opendevreview | Radosław Piliszek proposed openstack/project-config master: [NebulOuS] Add repository for SAL https://review.opendev.org/c/openstack/project-config/+/917943 | 11:07 |
| clarkb | #status log Added noonedeadpunk to freezer-core, freezer-tempest-core, and freezer-release to enact https://review.opendev.org/c/openstack/governance/+/914911 | 15:40 |
| opendevstatus | clarkb: finished logging | 15:40 |
| clarkb | #status log Also added noonedeadpunk to freezer-stable-maint | 15:45 |
| opendevstatus | clarkb: finished logging | 15:45 |
| tonyb | What are the requirements for hosting code on opendev? I've been asked if opendev is an option instead of rdoproject and software-factory running theier own gerrit instances. There are many many open questions but in theory is it a thing? | 15:51 |
| clarkb | tonyb: we're happy to host OSI licensed projects. The ultimate goal is to also get the involved projects to directly contribute to opendev to help make the hosting sustainable long term | 15:53 |
| clarkb | So basically be open source and be willing to at the very least help yourselves in using the system but even better if you can help others too is the sort of appraoch we'd like to see | 15:54 |
| tonyb | Thanks. For sure they're OpenSource and I like to think that at least some of the contributors would be willing to help, there's reasonable overlap at least in the RDO team/community. | 15:56 |
| clarkb | The gerrit community meeting was helpful. Looks like the topic change limit should only apply to open changes and they don't think it will affect the upgrade if you have any over the limit. You just won't be able to push new changes to the topic until some changes merge/abandon/close | 15:57 |
| clarkb | I think I'll still try to do some lightweight testing of those assumptions but I don't anticipate problems anymore | 15:57 |
| tonyb | Okay that's cool. Also somewhat easy to test on a held node | 15:57 |
| clarkb | oh and we do need to have explicit bazel targets for building without webfonts as bazel is very explicit and they didn't think there was a way to alias targets to get a release war without webfonts | 15:58 |
| clarkb | so I'll work on that and probably try to upstream it too so that we can just build release_webfont_free or similar | 15:58 |
| tonyb | Nice | 15:58 |
| clarkb | I'm going to pop out now for breakfast and parent teacher meetings. I'll be back in a couple hours probably | 15:59 |
| opendevreview | Tony Breeds proposed opendev/system-config master: [DNM] Force etherpad failure https://review.opendev.org/c/opendev/system-config/+/917990 | 16:20 |
| fungi | tonyb: clarkb: i'm not really here (sitting in lga waiting for my bgr flight to board), but technically we already provide some services to rdo since our irc bots are joined to #rdo (so channel logging, status announcements, et cetera) | 16:50 |
| tonyb | fungi: good to know. | 16:52 |
| fungi | also it looks like their projects are currently in a gerrit they manage, based on the rdogerrit bot i see announcing changes in their channel | 16:54 |
| fungi | keep in mind that moving from review.rdoproject.org to review.opendev.org would be a straight git clone import, i don't think we'd be able to retain their old reviews since the review numbers would almost certainly conflict | 16:56 |
| tonyb | Yeah. I there would be some information loss due to that. RDO also has it's own zuul which I guess would need to be transitioned to something more like a 3rd party. | 16:58 |
| tonyb | I don't think there would be an option to add the rdo nodepool but it's all very much in the "I wonder if" stage. | 16:59 |
| *** mordred1 is now known as Guest2920 | 17:03 | |
| tonyb | clarkb: 917865: Tighten permissions on Etherpad settings file | https://review.opendev.org/c/opendev/system-config/+/917865 looks good to me. Any preference when we land it? | 17:06 |
| clarkb | ya there are complicated methods of importing gerrit changes that people seem to always run into problems with | 18:15 |
| clarkb | tonyb: I'm back around now so anytime now is fine | 18:15 |
| clarkb | er anytime after now | 18:15 |
| tonyb | clarkb: Does UTC2000 work for you? That probably conflicts with either/both of a bike ride and school run | 18:17 |
| clarkb | thats perfect. Schools out today so that parents can do the teacher meetings and I don't plan to bike ride today. Sort of annyoed that a big "winter" storm is about to roll through the next few days though | 18:18 |
| clarkb | it is May and driving over the mountain passes is going to be sketchy this weekend. I'm ready for spring :) | 18:19 |
| tonyb | Okay lets call that a plan. | 18:19 |
| tonyb | we can land 917865 at least ... did you want to also do ertherpad 2.0.3 ? | 18:20 |
| clarkb | tonyb: we did 2.0.3 already | 18:20 |
| clarkb | this is the followup to that from some fungi review comments | 18:20 |
| clarkb | and thank you for double checking the container image produced the expected permissions | 18:21 |
| tonyb | Ahh okay. I was confused by https://review.opendev.org/c/opendev/system-config/+/840972 (the hold doe the 2.0.3 upgrade) still being there. I thought you'd done the real upgarde | 18:26 |
| tonyb | I was very confused by the docker output. | 18:26 |
| opendevreview | Tony Breeds proposed opendev/system-config master: Allow overriding of meetpad PUBLIC_URL https://review.opendev.org/c/opendev/system-config/+/905510 | 18:46 |
| opendevreview | Tony Breeds proposed opendev/system-config master: Use group var to define letsencrypt_certs https://review.opendev.org/c/opendev/system-config/+/905561 | 18:46 |
| opendevreview | Tony Breeds proposed opendev/system-config master: Switch meetpad test servers to jammy https://review.opendev.org/c/opendev/system-config/+/905182 | 18:46 |
| opendevreview | Tony Breeds proposed opendev/system-config master: DNM: force jitsi_meet node failure for testing https://review.opendev.org/c/opendev/system-config/+/905183 | 18:46 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Patch Gerrit's build system for webfont free war builds https://review.opendev.org/c/opendev/system-config/+/918008 | 19:10 |
| clarkb | This is a half testing chagne. I want to see if I can make that generally work before proposing a change like that upstream. Ideally we wouldn't need to carry the patch at all in merged system-config git state, but we'll see | 19:11 |
| clarkb | now that I've done that I half wonder if I should've just pushed it upstream first and then used a depends on. But that gets weird with wanting it to apply to both stable-3.8 and 3.9 before I'm ready to backport or ofrward port anything upstream. This is probably good for now to figure out if this is viable | 19:15 |
| tonyb | clarkb: fair enough. Good luck. | 19:49 |
| tonyb | clarkb: You still okay for ~10mins to land fungi's patch? | 19:49 |
| tonyb | I'm happy to monitor it as it lands etc so I expect you're on the hook for any problems I can't grok. | 19:50 |
| clarkb | yup I've finished eating lunch and am around | 19:56 |
| clarkb | tonyb: it should auto restart the service for us since the container image is updating | 19:58 |
| clarkb | then its just a amtter of checking that it works generally | 19:58 |
| tonyb | clarkb: Yup that was more or less my expectation. | 19:59 |
| tonyb | clarkb: It feels low risk | 19:59 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Patch Gerrit's build system for webfont free war builds https://review.opendev.org/c/opendev/system-config/+/918008 | 20:00 |
| clarkb | yup I agree | 20:00 |
| tonyb | # status log There will be a short etherpad outage while the service restarts to accommodate new configuration. | 20:03 |
| tonyb | clarkb: Does ^^ looks okay? | 20:03 |
| clarkb | ya that seems fine. Note that `log` won't report to every channel like `notice` does but I expect this to eb small enough that we don't need to spam for that | 20:04 |
| tonyb | clarkb: Noted. | 20:04 |
| tonyb | #status log There will be a short etherpad outage while the service restarts to accommodate new configuration. | 20:04 |
| opendevstatus | tonyb: finished logging | 20:05 |
| tonyb | clarkb: I noticed that nodepool hourly prod run failed on nb02 with: ERROR: for nodepool-builder Head "https://quay.io/v2/zuul-ci/nodepool-builder/manifests/latest": received unexpected HTTP status: 502 Bad Gateway | 20:23 |
| tonyb | I assume that's okay to ignore until the nest run? | 20:23 |
| clarkb | yes I think so. Thats it trying to fetch the container image and a 500 error implies it isn't on our end | 20:31 |
| opendevreview | Merged opendev/system-config master: Tighten permissions on Etherpad settings file https://review.opendev.org/c/opendev/system-config/+/917865 | 20:31 |
| tonyb | clarkb: cool that was my realing | 20:31 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Patch Gerrit's build system for webfont free war builds https://review.opendev.org/c/opendev/system-config/+/918008 | 20:31 |
| clarkb | for whatever reason typing webfont is really difficult and it always comes out webfront | 20:32 |
| tonyb | LOL, I read the webfont in that was webfront and was like .... "What is he talking about?" | 20:33 |
| tonyb | I guess that means I can understand the typing problem | 20:33 |
| tonyb | clarkb: the deploy run didn't get a new container, but the playbooks did change the ownership on the settings file | 20:41 |
| clarkb | hrm | 20:41 |
| clarkb | oh i know | 20:41 |
| tonyb | root@etherpad02:~# ls -l /etc/etherpad/settings.json | 20:41 |
| tonyb | -r--r----- 1 5001 5001 23450 May 1 16:20 /etc/etherpad/settings.json | 20:41 |
| tonyb | root@etherpad02:~# docker exec -it --user root etherpad-docker_etherpad_1 ls -l /opt/etherpad-lite/settings.json | 20:41 |
| tonyb | -r--r----- 1 etherpad 5001 23450 May 1 16:20 /opt/etherpad-lite/settings.json | 20:41 |
| clarkb | actually no it did run the build in check so the file matchers should be fine | 20:42 |
| clarkb | I think it has to do with only changing the job definition and not the dockerfiel though | 20:43 |
| clarkb | tonyb: maybe we should go ahead and add a rebuild comment to the dockerfile just to make sure the image rebuilds and deploys? | 20:43 |
| tonyb | Okay. I can see that | 20:43 |
| clarkb | do you want ot push that up or should I? | 20:45 |
| tonyb | I'll do it. | 20:50 |
| opendevreview | Tony Breeds proposed opendev/system-config master: Force new build of the etherpad container https://review.opendev.org/c/opendev/system-config/+/918011 | 20:52 |
| tonyb | So the container did build in system-config-upload-image-etherpad but it only published a change_917865_latest tag without a latest tag | 20:53 |
| clarkb | ya because the gate jobs didn't run the test. I think maybe when we trigger jobs because they are updated that only applies to check and not gate? | 20:54 |
| clarkb | it is a confusing behavior if that is the case and maybe zuul should run them in the gate too | 20:54 |
| clarkb | cc corvus | 20:54 |
| tonyb | clarkb: I think I understand. I'll watch what 918011 does to verify. | 20:59 |
| tonyb | clarkb: FWIW: the nodepool hourly run was fine this time: https://zuul.opendev.org/t/openstack/builds?job_name=infra-prod-service-nodepool&pipeline=opendev-prod-hourly&skip=0&limit=3 | 21:11 |
| clarkb | cool just a transient problem in quay then | 21:13 |
| tonyb | Yup. | 21:22 |
| tonyb | clarkb: I don't know if this is what you meant but, it looks like we didn't trigger the system-config-promote-image-etherpad job in the deploy pipeline because 917865 didn't touch a file that matches the files jobvar. | 21:54 |
| clarkb | oh yup that would explain it | 21:55 |
| clarkb | as you mentioend the image did end up in docker hub but since promotion didn't run we never updated the latest tag | 21:55 |
| tonyb | Yup that's my read of it. I thought system-config-upload-image-etherpad would take care of it but I understand better now. | 21:57 |
| tonyb | so we'll have another leaked chnage_xxx tag up there that we can cleanup at some stage. | 21:57 |
| opendevreview | Merged opendev/system-config master: Force new build of the etherpad container https://review.opendev.org/c/opendev/system-config/+/918011 | 22:59 |
| clarkb | arg my ssh keys have expired out of the agent | 23:01 |
| clarkb | etherpad restarted a minute ago on the new image and says it is healthy | 23:02 |
| clarkb | spot checking a few etherpads they all load for me | 23:03 |
| clarkb | tonyb: I think this is looking happy to me | 23:03 |
| tonyb | Yup just checking it now but it's looking good | 23:03 |
| tonyb | Yup permissions are as expected \o/ | 23:04 |
| clarkb | excellent | 23:06 |
| tonyb | https://paste.opendev.org/show/bZgfWtYFYXgixq2pIY3n/ | 23:06 |
| tonyb | I do notice that we didn't keep the old etherpad image around. I thought we set things up so that we would for 72hours just in case | 23:07 |
| opendevreview | Tony Breeds proposed opendev/system-config master: Add a helper script for doing the LVM setup on mirror nodes. https://review.opendev.org/c/opendev/system-config/+/901504 | 23:21 |
| opendevreview | Tony Breeds proposed opendev/system-config master: Remove some old CentOS compatibility and address FIXMEs https://review.opendev.org/c/opendev/system-config/+/918016 | 23:21 |
| opendevreview | Merged opendev/system-config master: Add a helper script for doing the LVM setup on mirror nodes. https://review.opendev.org/c/opendev/system-config/+/901504 | 23:34 |
| tonyb | clarkb: I think I'm done for the day. | 23:35 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!