[00:02:33] <AaronSchulz>	 Krinkle: lots of wbmi_mediasearch_entities_* clutter in wancache grafana dashboard
[00:06:30] <AaronSchulz>	 ugh
[00:46:41] <Krinkle>	 AaronSchulz: results look good, data/image for posterity at https://phabricator.wikimedia.org/T302623#8404657
[00:47:07] <Krinkle>	 perhaps we'll see another drop next week without the serialize call
[00:49:11] <AaronSchulz>	 nice
[01:30:12] <Krinkle>	 AaronSchulz: something to sleep well over tonight: https://grafana.wikimedia.org/d/000000429/backend-save-timing-breakdown?orgId=1&from=now-5y&to=now&viewPanel=34 we're back around 500ms for p75 backend save timing
[01:30:18] <Krinkle>	 This is last 5 years summarised
[01:32:39] <Krinkle>	 and to assure confidence, if you navigate up to the dashboard as a whole, "rate" under "by entrypoint" shows a steady 1K/minute so it hasn't e.g. lost scope or somthing like that
[01:32:57] <Krinkle>	 (unlike stashing)
[10:07:17] <phuedx>	 Krinkle: o/ I too am struggling to find a dedicated section that details the differences between the deployment processes for primary and secondary. Regardless, when you merge a change in schemas/event/secondary.git, it will be live on schemas.wikimedia.org soon afterwards
[10:07:54] <phuedx>	 There is a brief mention at the end of this section: https://wikitech.wikimedia.org/wiki/Event_Platform/EventLogging_legacy#How_do_I_edit_schemas?
[13:58:14] <Krinkle>	 https://gerrit.wikimedia.org/r/plugins/gitiles/operations/puppet/+/0491839ae4a72850b7211afad32c47e382d5f83f/modules/eventschemas/manifests/repository.pp
[13:58:42] <Krinkle>	 So it's puppet ensuring latest git clone
[13:59:00] <Krinkle>	 https://gerrit.wikimedia.org/g/operations/puppet/+/0491839ae4a72850b7211afad32c47e382d5f83f/modules/profile/manifests/eventschemas/repositories.pp#16
[13:59:24] <Krinkle>	 in other words, after  ~30min
[14:00:03] <Krinkle>	 And any checking or polling on top from EventGate