[10:00:36] GitLab needs a short maintenance restart in one hour [11:08:59] GitLab maintenance finished [11:51:13] Could I get a +1 on https://gerrit.wikimedia.org/r/c/operations/docker-images/production-images/+/1065187 please, to add a LABEL to the cephadm images? I've done a test build locally already [13:03:26] sorry but I package so infrequently that I have to ask questions every time, How can I make sure cumin1002 doesn't get the new version but cumin2002 would? I'm worried if I upload it to component library (https://wikitech.wikimedia.org/wiki/Debian_packaging#Upload_to_a_component_directory) it automatically show up in cumin1002 [13:04:32] the package would show up automatically but you have to manually install it anyway. so that's one way. basically, upload the package but only install/upgrade it on cumin2002. [13:05:40] Thanks! [13:28:13] https://gerrit.wikimedia.org/r/c/operations/docker-images/production-images/+/1065218 is it Friday yet? [13:30:25] Emperor: ty <3 [16:30:09] low priority, but if anyone has time to look at a 1-line alerts patch I will truly be in your debt! https://gerrit.wikimedia.org/r/c/operations/alerts/+/1065248 [16:31:58] ryankemper ^^ [17:21:10] +1ed [18:01:23] cccccbliegdunhlgnlrlhkrndtutjijdgebvkurteikt [18:03:03] inflatador: +1. [18:03:18] (whooops disregard) [18:09:11] cwhite, have a moment to help me understand my logstash failure? [18:09:23] https://gerrit.wikimedia.org/r/c/operations/puppet/+/1065247 seems right to me but... nothing is showing up on my dashboard [18:16:04] andrewbogott: I think cwhite is OOO this week. [18:16:04] The patch LGTM, if logs are not showing up I'd suggest checking if logs are sent to the correct index and I'd also suggest checking if the logs from `magnum-api` and `magnun-conductor` are sent to the correct Kafka topic. [18:19:04] thanks denisse. I'm not sure I know how to check those things... I was expecting them to get routed to the same place as the things right above them (for instance I can see trove-* logs) [18:20:35] andrewbogott: What's the name of the dashboard you're using? [18:21:19] openstack eqiad ecs https://logstash.wikimedia.org/app/dashboards#/view/3ef008b0-c871-11eb-ad54-8bb5fcb640c0?_g=h@865c245&_a=h@0f5fc48 [18:21:45] but then I'm twiddling the 'openstack services' setting to specify magnum|heat [18:21:47] Thanks, we can also continue talking on the -observability channel if you want. :) [18:21:53] ok! [19:13:20] Thanks denisse and b-rouberol! [20:36:23] Hello cdanis ~ I was watching an old demo about your intro to tracing (super awesome & TY!) and I had some questions about my team's current board: https://trace.wikimedia.org/search?end=1724443441695000&limit=20&lookback=6h&maxDuration&minDuration&service=wikifunctions&start=1724421841695000 I love that we are able to see the API endpoints and [20:36:23] durations, etc. however, we have a couple of other services we go through past the PHP layer and I am very curious if there is a feasible way to extract more granularity onto our board somehow. Would you be able to guide me down a potential way forward? And anyone else who has some insight? TY in advance. [21:57:28] ^ answering in #distributed-tracing on Slack where ecarg also asked and is still online :) [22:06:24] {◕ ◡ ◕} [22:14:26] Wooo TY cdanis yesssss