[06:24:01] morning people, as fyi I just restarted the varnish backend on cp3040/3033 due to 503s [07:52:34] elukey: thank you! [13:05:09] did you folks see the cp* cronspam today? [13:05:15] /etc/cron.daily/logrotate: [13:05:15] Job for varnishkafka-webrequest.service failed. See 'systemctl status varnishkafka-webrequest.service' and 'journalctl -xn' for details. [13:05:18] error: error running non-shared postrotate script for /var/cache/varnishkafka/webrequest.stats.json of '/var/cache/varnishkafka/webrequest.stats.json ' [13:05:21] run-parts: /etc/cron.daily/logrotate exited with return code 1 [13:05:23] for a bunch of hosts [13:08:59] 10netops, 10Cloud-Services, 10Operations, 10ops-eqiad: labsdb1001's switch port negociating at 100M - https://phabricator.wikimedia.org/T177130#3650565 (10faidon) [13:52:39] sigh [13:52:54] I didn't get those emails [13:52:56] spam? [13:53:12] of course [13:55:52] ahhh these are spare::systems [13:56:01] but the cronjob is still there [13:56:15] I was really scared for a brief moment :D [13:59:15] cumin 'cp4* and R:class = role::spare::system' 'rm -f /etc/logrotate.d/varnishkafka*' [13:59:34] paravoid: --^ [14:00:01] ok [14:25:23] sorry, I caused that with the decom, I didn't re-install after re-roling them [14:25:35] tried to get away with the simpler "stop all the cp*-specific daemons" approach [14:50:02] 10netops, 10Operations, 10fundraising-tech-ops: remove fundraising firewall rules related to ganglia - https://phabricator.wikimedia.org/T176319#3650965 (10Jgreen) 05Open>03Resolved this is done [15:48:23] bblack: is T175636 waiting on you or godog? [15:48:23] T175636: prometheus -> grafana stats for per-numa-node meminfo - https://phabricator.wikimedia.org/T175636 [15:51:53] paravoid: whoever first has time? we can probably downgrade priority to Low too, since "numa_networking: isolate" apparently doesn't work out so hot even on our new assymetric cache nodes :) [15:52:15] 10Traffic, 10Operations, 10monitoring, 10Patch-For-Review: prometheus -> grafana stats for per-numa-node meminfo - https://phabricator.wikimedia.org/T175636#3651253 (10BBlack) p:05Normal>03Low [15:52:42] it'd still be a nice-to-have in the non-isolated case to stare at, and we may yet end up using isolate on the LVSes if it tests well there [15:53:50] from the sort-of-related-but-not-really department https://medium.com/netflix-techblog/serving-100-gbps-from-an-open-connect-appliance-cdb51dda3b99 [15:56:45] yeah tons of useful details in there :) [15:59:21] indeed, I wish we had a nicer way to share interesting links/reads we find over the internet [16:23:22] 10Traffic, 10Operations, 10Performance-Team (Radar): Upgrade to Varnish 5 - https://phabricator.wikimedia.org/T168529#3651406 (10BBlack) This task hasn't been updated for various IRC/Hangouts discussions since. We did decide to move forward with V5 upgrades. Arzhel has built preliminary packages, and we ha... [16:23:52] 10Traffic, 10Operations, 10Performance-Team (Radar): Upgrade cache_misc to Varnish 5 - https://phabricator.wikimedia.org/T177233#3651409 (10BBlack) [16:59:42] godog: maybe add a section at the end of the weekly meeting notes? [17:00:19] Or a wikitech page that people can subscribe to, so it's public [17:01:50] XioNoX: good ideas! yeah wikitech might work better as it is public [17:02:44] I'll get sth started and mention it at the next h/o meeting [17:47:00] 10Traffic, 10Operations, 10Reading-Admin: TEST: redirect small portion of unauthenticated desktop users to mobile web - https://phabricator.wikimedia.org/T117826#3651798 (10CKoerner_WMF) [17:59:56] 10Traffic, 10Operations, 10Performance-Team (Radar): Upgrade to Varnish 5 - https://phabricator.wikimedia.org/T168529#3651828 (10Gilles) Is the plan to use it with hitch in front of it rather than nginx? Or just an upgrade for now and we'll see about that part later? [18:56:10] 10Traffic, 10Operations, 10Performance-Team (Radar): Upgrade to Varnish 5 - https://phabricator.wikimedia.org/T168529#3651975 (10BBlack) Definitely not looking at Hitch presently. Just swapping out Varnish4 for Varnish5 in the existing software stack for both the frontend and backend cache processes. Forwa...