[01:31:50] <wikibugs>	 10Traffic, 10SRE, 10Patch-For-Review, 10Performance-Team (Radar): 8-10% response start regression (Varnish 5.1.3-1wm15 -> 6.0.6-1wm1) - https://phabricator.wikimedia.org/T264398 (10dpifke) Alternatively, can we get identical results just by incrementing `grace` by `keep`?  (And possibly setting `keep` to 0...
[11:38:22] <wikibugs>	 10Traffic, 10SRE, 10Patch-For-Review, 10Performance-Team (Radar): 8-10% response start regression (Varnish 5.1.3-1wm15 -> 6.0.6-1wm1) - https://phabricator.wikimedia.org/T264398 (10BBlack)  >>! In T264398#7040458, @dpifke wrote: > Alternatively, can we get identical results just by incrementing `grace` by...
[11:42:49] <wikibugs>	 10Traffic, 10SRE, 10Patch-For-Review, 10Performance-Team (Radar): 8-10% response start regression (Varnish 5.1.3-1wm15 -> 6.0.6-1wm1) - https://phabricator.wikimedia.org/T264398 (10BBlack) Continuing the thought above: varnishlog data may infer that most of the perf impact could be restored just by extendi...
[11:50:54] <bblack>	 FYI I might (or might not) be unusualy-late to our traffic meeting.  My vax appt got moved to this morning, and in theory I should be out of there in time to be on time, but we'll see!
[11:53:30] <sukhe>	 bblack: good luck for the vax appointment!
[12:08:28] <jayme>	 I need to do a rolling restart of pybals (for https://gerrit.wikimedia.org/r/c/operations/puppet/+/683248) in codfw, eqsin and ulsfo. Are there any specifics steps to follow?
[12:08:37] <jayme>	 If not I would just go from the secondary ones starting with low-traffic -> high-traffic1 and the primaries afterward in the same order
[12:08:51] <jayme>	 if you ack that :)
[12:19:09] <jayme>	 vgutierrez maybe or ema?
[12:21:54] * vgutierrez checking
[12:22:35] <vgutierrez>	 jayme: your suggested order seems perfect
[12:24:00] <jayme>	 vgutierrez: nice. Thanks for taking a look
[13:21:08] <wikibugs>	 10Traffic, 10SRE, 10Patch-For-Review, 10Performance-Team (Radar): 8-10% response start regression (Varnish 5.1.3-1wm15 -> 6.0.6-1wm1) - https://phabricator.wikimedia.org/T264398 (10ema) >>! In T264398#7037012, @Gilles wrote: > The change's claimed behaviour is definitely consistant with the change we obser...
[14:05:01] <jbond42>	 hi all any objections to adding pybal logs to ELK https://gerrit.wikimedia.org/r/c/operations/puppet/+/682566