[05:34:12] <_joe_> James_F: yes [06:05:56] Excellent. [07:56:45] I checked the kafka logging consumer lag, and in the udp_localhost-info topic there seem to be a lot of [07:57:02] <_joe_> a lot of? [07:57:05] "Use of ResourceLoaderSkinModule::getAvailableLogos with $wgLogoHD set instead of $wgLogos was deprecated in MediaWiki 1.35. [Called from ResourceLoaderSkinModule::getAvailableLogos" [07:57:15] I was copy/pasting gimme 1 min :D [07:57:26] <_joe_> ahahah it seemed it was truncated [07:57:28] <_joe_> :P [07:58:37] ah ok in the 24h view there was a horrible lag (that recovered) from 20:00 UTC yesterday [07:58:41] when mediawiki was deployed [07:58:58] (https://grafana.wikimedia.org/d/000000484/kafka-consumer-lag?from=now-24h&to=now&orgId=1&v&var-datasource=eqiad%20prometheus%2Fops&var-cluster=logging-eqiad&var-topic=All&var-consumer_group=All) [07:59:59] <_joe_> 🤔 [08:03:55] I commented here https://phabricator.wikimedia.org/T233868#5906174 [08:04:28] ah snap https://phabricator.wikimedia.org/T245778 [12:30:57] is this known? Seems deployment breakage? https://grafana.wikimedia.org/d/000000561/logstash?orgId=1&refresh=5m&from=1582115428259&to=1582288228259&fullscreen&panelId=9&var-datasource=eqiad%20prometheus%2Fops&var-input=gelf%2F12201 [12:32:02] couldn't find a task [13:05:21] https://phabricator.wikimedia.org/T245778 [13:05:25] found it [18:18:05] o/ [18:18:26] hello [22:59:41] uhm.. new servers in .wikimedia.org .. did not have a role yet. can't ssh to them. but also can't use install-console (anymore) .. odd [23:00:29] IMHO we should not install without a role, I would even make puppet fail in those case if it was for me :) [23:00:32] uh oh.. something must be wrong with DNS or VLAN.. 100% packet loss with ICMP [23:00:57] ah.. well.. maybe they are just not installed yet :) [23:01:11] i think i just mixed them up with other servers [23:01:43] ack, i just added one.. installing or reinstalling [23:07:28] haha, ignore me. the Ganeti VMs were not even created :) [23:07:55] lol [23:10:37] wasn't sure we have 180GB left to spare for virtual disks right now [23:27:24] we do. creating 2 new VMs for APT repos (without TFTP/DHCP) [23:40:44] bbl..off for now