[08:50:48] <_joe_>	 jayme: so for our OKRs regarding envoy
[08:51:42] <_joe_>	 I think I'm going to add the service proxy to scb and maps (the rest of the stuff that is outside of k8s and we need to migrate), then we can think of what to do on k8s
[12:43:52] <moritzm>	 akosiaris: the fixed chromium 83 is now available
[14:33:13] <akosiaris>	 moritzm: oh, nice. thanks!
[15:46:40] <jayme>	 akosiaris: Gathering kubernetes stats is stats like in "N services running there" or new metrics?
[16:34:38] <wikibugs>	 10serviceops, 10Prod-Kubernetes: "Failed to fork" errors on kubernetes100[1,3,4] - https://phabricator.wikimedia.org/T257679 (10JMeybohm)
[17:49:33] <wikibugs>	 10serviceops, 10Prod-Kubernetes: "Failed to fork" errors on kubernetes100[1,3,4] - https://phabricator.wikimedia.org/T257679 (10akosiaris) Looking at logs for kubernetes1001 I notice the following  ` ul  9 22:11:44 kubernetes1001 kubelet[22510]: E0709 22:11:44.718926   22510 fsHandler.go:121] failed to collect...
[17:49:59] <rzl>	 _joe_ effie wkandek: invited you to a Thursday 1600Z meeting as discussed, but right now it's just after effie's and my 1:1s, it might make sense to swap it to 1500Z so that it's before them instead
[17:56:10] <effie>	 +1
[18:11:50] <wkandek>	 2 hours before works for me 7:00 PST/10:00 EST/ 14:00 UTC
[18:20:26] <rzl>	 oh, sure -- I wouldn't have guessed it was okay to start that early in your TZ :) done
[18:41:02] <wikibugs>	 10serviceops, 10Prod-Kubernetes: "Failed to fork" errors on kubernetes100[1,3,4] - https://phabricator.wikimedia.org/T257679 (10colewhite) >>! In T257679#6302180, @akosiaris wrote: > `/proc/sys/kernel/threads-max` and `/proc/sys/kernel/pid_max` don't look particularly probable as they are well below the limits...