[07:04:47] _joe_: hello any idea what's up with https://alerts.wikimedia.org/?q=alertname%3DSystemdUnitFailed&q=name%3Drequestctl-credential-refresh.service ? pinging you as it's related to requestctl [07:05:11] And for anyone else in serviceops: https://alerts.wikimedia.org/?q=alertname%3DSystemdUnitFailed&q=name%3Ddocker-reporter-kubernetes-aux_eqiad-images.service [07:05:16] <_joe_> see #wikimedia-sre-private [07:06:19] ok, should I silence the alert? it's flooding the -sre-foundations channel :( [07:26:08] <_joe_> what do you think? [08:27:01] it doesn't seem urgent and actionable for now, so better to downtime, I have no idea about the docker/k8s one though [10:20:25] <_joe_> yeah I didn't look into that, but yeah while actionable, the refresh alert is only actionable by liz when she's back :) [10:21:03] <_joe_> and, not her fault; I need to look into refreshing groups at some point [13:24:06] [dc switchover] for vis to cdanis slyngs and others, this is the status after the tests done on memcached this week: https://phabricator.wikimedia.org/T440896#12441469 [13:36:04] claime cdanis, could we make use of help from dev teams at this point? [14:04:26] matthieulec: it's a good question, I don't have anything specific for them to look at unfortunately? it's more like I want a few people who actually understand Mediawiki to spend some time hunting for other recent perf regressions [15:18:59] ack, I pinged Timo if he has any lead we're not aware of