| opendevreview | Merged opendev/base-jobs master: Update fedora-latest to Fedora 36 https://review.opendev.org/c/opendev/base-jobs/+/854331 | 01:24 |
|---|---|---|
| opendevreview | Merged opendev/base-jobs master: Add Rocky Linux 9 nodeset https://review.opendev.org/c/opendev/base-jobs/+/854332 | 01:24 |
| opendevreview | Merged opendev/infra-manual master: Stop recommending PyPI project name squatting https://review.opendev.org/c/opendev/infra-manual/+/852584 | 01:30 |
| opendevreview | Merged opendev/git-review master: Clarify that test rebases are not kept https://review.opendev.org/c/opendev/git-review/+/850054 | 01:39 |
| *** rlandy is now known as rlandy|out | 02:47 | |
| *** rcastillo|rover_ is now known as rcastillo | 03:56 | |
| *** chandankumar is now known as chkumar|ruck | 05:02 | |
| *** ysandeep|out is now known as ysandeep | 05:22 | |
| opendevreview | Ian Wienand proposed openstack/diskimage-builder master: Start running dib-lint again https://review.opendev.org/c/openstack/diskimage-builder/+/855589 | 06:18 |
| *** arxcruz is now known as arxcruz|rover | 07:12 | |
| opendevreview | Merged openstack/diskimage-builder master: Add subscription-manager repo names for RHEL-9 https://review.opendev.org/c/openstack/diskimage-builder/+/851734 | 07:35 |
| *** jpena|off is now known as jpena | 07:36 | |
| opendevreview | Rafal Lewandowski proposed openstack/diskimage-builder master: changed release check logic in lvm element https://review.opendev.org/c/openstack/diskimage-builder/+/855616 | 08:06 |
| opendevreview | Rafal Lewandowski proposed openstack/diskimage-builder master: changed release check logic in lvm element enabling lvm for all ubuntu releases except for trusty Change-Id: I77184990270be74afa6542fd16b8e05bd1a27e04 https://review.opendev.org/c/openstack/diskimage-builder/+/855616 | 08:14 |
| opendevreview | Rafal Lewandowski proposed openstack/diskimage-builder master: changed release check logic in lvm element https://review.opendev.org/c/openstack/diskimage-builder/+/855616 | 08:14 |
| *** ysandeep is now known as ysandeep|afk | 09:49 | |
| *** rlandy|out is now known as rlandy | 10:37 | |
| opendevreview | Rafal Lewandowski proposed openstack/diskimage-builder master: changed release check logic in lvm element https://review.opendev.org/c/openstack/diskimage-builder/+/854566 | 10:40 |
| *** ysandeep|afk is now known as ysandeep | 11:08 | |
| opendevreview | Rafal Lewandowski proposed openstack/diskimage-builder master: - added elrepo element https://review.opendev.org/c/openstack/diskimage-builder/+/853817 | 11:49 |
| chkumar|ruck | Hello #opendev, can we get this review https://review.opendev.org/c/openstack/tripleo-quickstart/+/855587/ force merge to clear tripleo gates | 12:27 |
| chkumar|ruck | we have got the vote from slagle on the patch | 12:30 |
| chkumar|ruck | on this patch https://zuul.opendev.org/t/openstack/status#855587, tripleo-ci-centos-9-undercloud-containers - https://zuul.opendev.org/t/openstack/stream/52ffc5d44ebd456e94bbbab7b740a1c0?logfile=console.log is stuck from long time | 12:31 |
| chkumar|ruck | if we update it then it going to take more than 6+ hr to kand | 12:31 |
| chkumar|ruck | *land | 12:31 |
| chkumar|ruck | that's why we need to force merge it | 12:31 |
| opendevreview | Rafal Lewandowski proposed openstack/diskimage-builder master: added elrepo element https://review.opendev.org/c/openstack/diskimage-builder/+/853817 | 12:38 |
| fungi | chkumar|ruck: how about i move it to the front of the gate queue so it's the next change to merge? | 12:52 |
| chkumar|ruck | fungi: just updated the patch | 12:53 |
| chkumar|ruck | fungi: sounds good | 12:53 |
| fungi | all the other changes in the gate will get their testing restarted behind it either way | 12:53 |
| fungi | chkumar|ruck: it'll need cr+2,w+1 votes added | 12:57 |
| chkumar|ruck | fungi: getting the votes | 12:57 |
| *** dasm|off is now known as dasm | 13:00 | |
| opendevreview | Rafal Lewandowski proposed openstack/diskimage-builder master: added elrepo element https://review.opendev.org/c/openstack/diskimage-builder/+/853817 | 13:09 |
| fungi | chkumar|ruck: i have to step away for a few minutes, but let me know once the latest revision has approval votes and i'll enqueue/promote it in the gate | 13:10 |
| chkumar|ruck | fungi: done, thank you :-) | 13:14 |
| *** jpena is now known as jpena|off | 14:07 | |
| *** ysandeep is now known as ysandeep|out | 14:34 | |
| clarkb | ok looks like that promotion happened | 15:08 |
| *** dviroel is now known as dviroel|lunch | 15:13 | |
| fungi | yeah | 15:15 |
| fungi | i'm heading out to the biergarten for lunch, but once i'm back i'll be available for meetpad upgrade testing | 15:16 |
| fungi | probably back by 16:30 utc, 17:00 at the latest | 15:17 |
| clarkb | fungi: sounds good. Should I approve the change in about 45 minuets? | 15:17 |
| fungi | yeah, that should be plenty safe since it'll probably take a while deploying anyway | 15:17 |
| clarkb | ok I'll do that | 15:17 |
| fungi | anyway, headed out, bbiaw | 15:17 |
| clarkb | enjoy the bier garten | 15:17 |
| fungi | thanks! | 15:17 |
| clarkb | tumbleweed must've udpated gcc as I now have a bunch of packages to upgrade | 15:21 |
| clarkb | approving the jitsi meet udpate now | 16:12 |
| *** johnthetubaguy[m] is now known as johnthetubaguy | 16:14 | |
| opendevreview | Clark Boylan proposed opendev/project-config-example master: Update deprecated zuul syntax https://review.opendev.org/c/opendev/project-config-example/+/855684 | 16:19 |
| opendevreview | Clark Boylan proposed openstack/project-config master: Update deprecated merge-failure zuul syntax https://review.opendev.org/c/openstack/project-config/+/855686 | 16:21 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Update example zuul config to remove deprecated syntax https://review.opendev.org/c/opendev/system-config/+/855687 | 16:23 |
| clarkb | corvus: ^ fyi opendev cleanups for the merge-failure removal | 16:24 |
| clarkb | 855686 is the only oen that represents a real breaking issue. The others are docs updates | 16:24 |
| fungi | okay, back and catching up | 16:35 |
| clarkb | the jitsi meet change should merge in a few minutes right around when the hourly deploys finish | 16:40 |
| clarkb | probably in 20-30 minutes we should plan to do testing | 16:40 |
| *** dviroel|lunch is now known as dviroel | 16:41 | |
| opendevreview | Merged opendev/system-config master: Update Jitsi configs to latest upstream samples https://review.opendev.org/c/opendev/system-config/+/855388 | 16:42 |
| fungi | awesome, thanks for approving! | 16:44 |
| corvus | clarkb: lgtm -- those are all safe to merge now | 16:48 |
| clarkb | fungi: https://meetpad.opendev.org/isitbroken | 16:49 |
| clarkb | I was able to pull that up and the etehrpad is there | 16:49 |
| clarkb | I'm not muted but I think it may mute me when someone else joins | 16:49 |
| fungi | that was quick. i'll make sure i delete all my cookies/permissions before i test | 16:51 |
| opendevreview | Jeremy Stanley proposed opendev/system-config master: Move our jitsi-meet interface config to defaults https://review.opendev.org/c/opendev/system-config/+/855693 | 17:30 |
| opendevreview | Jeremy Stanley proposed opendev/system-config master: Move our jitsi-meet interface config to defaults https://review.opendev.org/c/opendev/system-config/+/855693 | 17:32 |
| clarkb | fungi: if you get a chance https://review.opendev.org/q/topic:merge-failure-cleanup is straightfowrard updates for zuul deprecations that will be removed soon | 17:54 |
| fungi | oh, yeah thanks i saw you mention that while i was out | 17:55 |
| clarkb | also once we're happy with jitsi meet we can land the gravatar change to gitea if we think that is safe enough (it should be and easy to revert if necessary too) | 18:13 |
| fungi | oh, yep now's a great time for that as well. shall i approve it now, or wait? | 18:14 |
| clarkb | I'm happy either way. I think the main thing would be if we have two broken things at the asme time we'd have to pick one to fix. But gitea is proibably higher priority so no confusion over what to focus on | 18:15 |
| fungi | i went ahead and approved it | 18:16 |
| clarkb | thanks | 18:18 |
| clarkb | It is possible that we would need to manually restart the services for that gitea change too now that I think of it | 18:19 |
| clarkb | The process for that is a bit involved and has me thinking maybe we should split it out into a separate playbook/task file | 18:19 |
| fungi | oh, yeah since it only touches a config template | 18:19 |
| clarkb | we have to start the web service before starting the sshd otherwise replication doesn't work properly | 18:19 |
| fungi | i don't suppose there's a minor image update we could push through too? | 18:20 |
| clarkb | fungi: not that I'm aware of. I can just manually do it. It isn't too bad just have to remember to follow the process in he larger role | 18:20 |
| fungi | sounds fine | 18:20 |
| opendevreview | Merged opendev/system-config master: Update example zuul config to remove deprecated syntax https://review.opendev.org/c/opendev/system-config/+/855687 | 18:21 |
| opendevreview | Merged openstack/project-config master: Update deprecated merge-failure zuul syntax https://review.opendev.org/c/openstack/project-config/+/855686 | 18:22 |
| opendevreview | Merged opendev/project-config-example master: Update deprecated zuul syntax https://review.opendev.org/c/opendev/project-config-example/+/855684 | 18:24 |
| clarkb | fungi: we may also need to manually restart meetpad containers. | 18:48 |
| fungi | oh, yes that too | 18:49 |
| clarkb | I'm still trying to download all my tumbleweed updates. I think when they update gcc and rebuild the world it creates quite the thundering herd of installs trying to fetch all the packages | 18:50 |
| fungi | that's not surprising | 18:50 |
| opendevreview | Merged opendev/system-config master: Disable Gravatar in Gitea https://review.opendev.org/c/opendev/system-config/+/854678 | 19:16 |
| opendevreview | Merged opendev/system-config master: Move our jitsi-meet interface config to defaults https://review.opendev.org/c/opendev/system-config/+/855693 | 19:16 |
| clarkb | Those will deploy in that order once the hourly jobs are done | 19:22 |
| clarkb | fungi: I need to eat lunch but then can help test things afterwards | 19:22 |
| clarkb | and do restarts as necessary | 19:22 |
| fungi | same, i'm popping out in a sec to pick up takeout, but won't be long | 19:37 |
| clarkb | confirming that gitea appears to need a manual restart | 19:57 |
| clarkb | I'll wait for the job to complete before doing that just to be sure I don't interact with the deploy job unexpecedly | 19:57 |
| clarkb | gitea01 has been restarted. Seems fine so I'll do the others in sequence | 20:04 |
| clarkb | and all done. I don't see gravatar requests in my browser | 20:11 |
| clarkb | And the service seems to function otherwise | 20:11 |
| clarkb | The meetpad deploy got stuck behind the next hourly run :/ | 20:11 |
| fungi | back now and gitea lgtm | 20:15 |
| clarkb | meetpad should update shortly | 20:42 |
| fungi | i'm ready to restart it. just the containers on meetpad.o.o i guess, or should i do the jvb servers too? | 20:45 |
| fungi | ready to restart it once deployment completes of course, i mean | 20:45 |
| clarkb | I think those chagnes only affect meetpad since it is only the webd stuff? | 20:46 |
| clarkb | fungi: looks like it may have restarted itself | 20:47 |
| fungi | oh marvellous | 20:51 |
| *** dviroel is now known as dviroel|out | 20:56 | |
| *** dasm is now known as dasm|off | 21:32 | |
| opendevreview | Jeremy Stanley proposed opendev/system-config master: Pass PUBLIC_URL to jitsi-meet jvb containers https://review.opendev.org/c/opendev/system-config/+/855700 | 21:35 |
| clarkb | https://github.com/jitsi/jitsi-videobridge/blob/master/doc/web-sockets.md | 21:51 |
| clarkb | https://jitsi.github.io/handbook/docs/devops-guide/faq/ | 21:59 |
| fungi | https://jitsi.github.io/handbook/docs/devops-guide/turn/ "One-to-one calls should avoid going through the JVB for optimal performance and for optimal resource usage. This is why we've added the peer-to-peer mode where the two participants connect directly to each other. Unfortunately, a direct connection is not always possible between the participants. In those cases you can use a TURN server | 22:19 |
| fungi | to relay the traffic (n.b. the JVB does much more than just relay the traffic, so this is not the same as using the JVB to "relay" the traffic)." | 22:19 |
| clarkb | infra-root fungi's jitsi meet config updates exposed that our extra JVB servers have likely not been working for some time (and possibly no video was working at all on meetpad for some time). The reason for this is JVB uses "colibri websockets" to negotiate video transfer (maybe nto the video itself but the details?) and we weren't configuring this properly for the extra jvb servers | 22:41 |
| clarkb | 855700 above fixes fungi's config updates to work with jvb on the meetpad server itself but not jvb01 and jvb02. For this reason we've put jvb0X in the emergency file and manaully shut down their services | 22:42 |
| clarkb | To make them work we need to issue them ssl certs (likely via LE), then update their configs so that they listen on port 9090 with an https server. Then we update the meet.conf nginx config for meetpad.o.o to proxy to the jvb backends over port 9090 appropriately | 22:42 |
| clarkb | And likely we want firewall rules to restrict all this 9090 communication to within the cluster (since we aren't sure what sort of control that exposes if opened to the internet) | 22:43 |
| clarkb | We don't need to go through this with the all in one meetpad.o.o server bceause the existing configs work for http (not https) connectivity to the localhost jvb | 22:43 |
| clarkb | Anyway a very long way of saying our extra JVBs need some work, but meetpad.o.o is currently functional based on our testing as an all in one server. | 22:44 |
| clarkb | This also illustrates that we probably don't need two jvbs since one all in one server has been getting us by | 22:45 |
| clarkb | We'll likely update things to support this colibri websocket stuff but then shrink to a single jvb at the same time | 22:45 |
| fungi | ideally before the ptg, and then if we need to re-add more jvb servers we can do that quickly and easily | 22:46 |
| opendevreview | Merged opendev/system-config master: Pass PUBLIC_URL to jitsi-meet jvb containers https://review.opendev.org/c/opendev/system-config/+/855700 | 23:31 |
| clarkb | that change has deployed but it did not restart jitsi meet services. I'm going to manually restart them and then test | 23:49 |
| clarkb | and it seems to work between my desktop and laptop with video happening too | 23:51 |
| clarkb | I think we're all set with the all in one install. Next week we can look into fixing the separate jvbs | 23:52 |
| fungi | cool, thanks for testing and for all the help! | 23:58 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!