[01:19:16] 10serviceops, 10Mobile-Content-Service, 10Page Content Service, 10Patch-For-Review, 10Reading-Infrastructure-Team-Backlog (Kanban): "worker died, restarting" mobileapps issue - https://phabricator.wikimedia.org/T229286 (10Mholloway) [06:14:31] <_joe_> urandom: I think the reason is that modifies a configmap and you have no reference to the change of that config map in the deployment as an annotation [06:14:40] <_joe_> see eventgate that uses [06:14:42] <_joe_> checksum/config: {{ include "config.app" . | sha256sum }} [06:15:41] <_joe_> you have the same, uhm [06:15:54] <_joe_> maybe you didn't include all the configmaps you have in this list? [06:18:39] <_joe_> no, I just looked at the chart, it looks ok, so I have no explanation for you [06:18:50] <_joe_> (also today I'm supposed to be off :P) [09:27:52] 10serviceops: Offer Loadtesting as as Service - https://phabricator.wikimedia.org/T230530 (10Tarrow) [09:28:11] 10serviceops: Offer Loadtesting as a Service - https://phabricator.wikimedia.org/T230530 (10Tarrow) [09:28:14] 10serviceops, 10Release-Engineering-Team: Offer Loadtesting as a Service - https://phabricator.wikimedia.org/T230530 (10Jdforrester-WMF) [09:30:11] _joe_: ^^made a ticket for an engineered load testing solution as requested :) Hope that's in sort of a useful place for you to work with [11:04:45] Hey! Our smoke tests yesterday we good; however we realise that the expected load has actually changed "a lot" since we requested the service. My back of the envelope calculations have gone from 1/s -> 8/s. We'd therefore like to see if everything is still green at 40/s. Is that likely to still be ok on the api app servers? [11:08:53] it looks to me like that still "a drop in the the ocean" compared to the normal 6k+ req/s average [13:24:01] _joe_: thanks :)