[10:01:50] lunxh [12:40:28] dcausse: could you be ready for our ITC for our 1:1 tomorrow morning? [12:50:06] gehel: sure, I'll prep something [12:50:11] o/ [12:50:21] dcausse: or do you want to do it live like last time? [12:50:41] sure, live is nice [12:55:50] then let's do it live! [13:24:42] https://gerrit.wikimedia.org/r/c/operations/puppet/+/946547 small patch to get the codfw flink-zk cluster going [13:43:59] dcausse re: your comment. I thought we needed a zk cluster for codfw as well? [13:48:08] inflatador: my bad, my comment was not very explanatory, meant to name the nodes with *2*00[123] instead of *1*00[123] [13:48:09] codfw nodes generatlly start with a 2 [13:48:09] dcausse Oh yeah, thanks. (Can you tell it's Monday morning for me? ;P ) [13:48:09] :P [13:48:41] OK, fixed [14:22:59] Just checking that someone is aware of this: https://phabricator.wikimedia.org/T343319 PuppetCertificateAboutToExpire: for `search.svc.${site}.wmnet` [14:23:25] ryankemper, inflatador: ^ [14:23:30] I'm happy to look at it, but I don't want to tread on anyone's toes if it's already under control. [14:44:55] btullis ACK, I'll check it out [14:52:17] looks like this is how we did it in 2018: https://gerrit.wikimedia.org/r/c/operations/puppet/+/454788 [14:53:02] inflatador: Awesome! Feel free to add me for review/cc etc. [15:01:39] ebernhardson, inflatador, pfischer: triage meeting: https://meet.google.com/eki-rafx-cxi [15:58:50] * ebernhardson is still surprised when docker containers come back up after a reboot [17:15:54] dinner [17:35:55] lunch, back in ~1h [18:48:45] sorry, been back [20:01:40] ryankemper heads-up, reimaged wcqs2001-02 to Bullseye on Friday. Didn't see any issues, so I'm starting wcqs2003 now [20:02:11] inflatador: ack, so 2001/2002 are back in svc? [20:03:02] ryankemper confirmed [20:03:30] SGTM [20:04:47] {◕ ◡ ◕} [20:22:28] meh...gitlab is annoying :P I was trying to review an MR that has a dozen patches attached so I opened each patch in a new tab. Now i have a dozens tabs with all the same name :P [20:23:21] it really feels like gitlab tries hard to not care what a patch is, even though that is the critical unit of work in git [20:23:51] s/patch/commit/ [21:41:15] * ebernhardson has now renamed this metric that reports Special:Search load time at least 5 times :P [23:37:14] ebernhardson: ^ is my big gripe with github/gitlab in general. I feel like they both treat the pull request as the fundamental "unit of work" rather than the commit