[08:57:21] <hashar>	 moritzm: hi :]  when you have rebooted deployment-mediawiki-07 we had some some 503 issued 
[08:57:38] <hashar>	 only thing I found is in the php fpm error log:
[08:57:39] <hashar>	 [WARNING] [pool www] seems busy (you may need to increase pm.start_servers, or pm.min/max_spare_servers), spawning 8 children, there are 0
[08:57:49] <hashar>	 not sure whether that is related
[08:58:03] <hashar>	 or maybe the 503 was issued by varnish itself since it could not reach the backend
[08:58:24] <moritzm>	 seeem so, yes
[08:59:36] <hashar>	 out of curiosity, do we have a monitoring for the php-fpm pool business?
[09:05:05] <moritzm>	 yeah if too many get depooled in prod, there's an alert
[09:07:50] <_joe_>	 moritzm: that's not what hashar was referring to
[09:07:55] <_joe_>	 hashar: we have monitoring yes
[09:12:24] <hashar>	 _joe_: good :-]
[09:18:42] <_joe_>	 we collect data and they're shown in grafana
[09:19:44] <_joe_>	 hashar: our busiest server https://grafana.wikimedia.org/d/000000550/mediawiki-application-servers?panelId=66&fullscreen&orgId=1&var-source=eqiad%20prometheus%2Fops&var-cluster=api_appserver&var-node=mw1348
[09:20:59] <hashar>	 _joe_: that sounds perfect :-]