[13:42:53] <elukey>	 hi folks!
[13:43:13] <elukey>	 I'd need to roll restart logstash and loggind-hd java daemons for https://phabricator.wikimedia.org/T371874 and https://phabricator.wikimedia.org/T371961
[13:43:36] <elukey>	 I've heard that you have a new team member, maybe he can volunteer? :D :D :D
[13:44:28] <elukey>	 (all references to tappof are purely casual)
[13:48:24] <tappof>	 sure elukey: "casual" :) here I am 
[13:49:40] <elukey>	 nice :) can you read the tasks? They are tagged as security, so we can check if you are in the right groups
[13:52:42] <tappof>	 yes elukey, I can read both
[13:55:31] <elukey>	 molto bene
[13:55:58] <elukey>	 ok so we'd need to restart the logstash nodes, and the logging-hdXXXX ones
[13:56:49] <elukey>	 we could do it manually host-by-host of via cumin, but there should be a cookbook for both
[13:56:59] <elukey>	 for logstash I think we can use sre.o11y.roll-restart-reboot-logstash-collectors
[13:58:06] <elukey>	 for logging-hdXXXX, I think sre.opensearch.roll-restart-reboot should work as well
[14:55:26] <cwhite>	 Restarts of logging-hd especially should be handled with care right now.  They're oversubscribed and restarts not taking into account shard allocation could cause an outage.
[15:01:04] <elukey>	 cwhite: o/ ah okok so I'll leave those to you if you are ok, are the logstash nodes good to be restarted?
[15:11:13] <cwhite>	 elukey: I don't see logging-hd on the list anyway - they're bookworm hosts running openjdk 17?
[15:14:34] <tappof>	 cwhite: they are in the phab task https://phabricator.wikimedia.org/T371961 (the one for jdk 17)
[15:15:08] <cwhite>	 sure enough, thanks :)
[15:17:05] <elukey>	 exactly :)
[15:17:16] <elukey>	 (Moritz sent multiple updates :D)
[15:20:37] <elukey>	 ok I'll follow up with Tiziano to run the cookbook and what to check etc..