[04:26:25] <icinga-wm>	 PROBLEM - Check the last execution of monitor_refine_eventlogging_analytics on an-coord1001 is CRITICAL: CRITICAL: Status of the systemd unit monitor_refine_eventlogging_analytics https://wikitech.wikimedia.org/wiki/Analytics/Systems/Managing_systemd_timers
[08:46:31] <neilpquinn>	 jkumalah: which instructions have you been following? https://wikitech.wikimedia.org/wiki/Production_shell_access#Setting_up_your_SSH_config is the best documentation I know of; if the instructions there aren't working for you, let me know and I can improve them :)
[09:02:25] <icinga-wm>	 RECOVERY - Check the last execution of monitor_refine_eventlogging_analytics on an-coord1001 is OK: OK: Status of the systemd unit monitor_refine_eventlogging_analytics https://wikitech.wikimedia.org/wiki/Analytics/Systems/Managing_systemd_timers
[17:34:33] <awight>	 nuria: hi!  well, they're currently string comparators but what I was noticing is that a bunch of reportupdater-queries require a full table scan, e.g.: https://gerrit.wikimedia.org/r/plugins/gitiles/analytics/reportupdater-queries/+/refs/heads/master/pingback/php#22
[17:35:40] <awight>	 The problem I was thinking about re. 12 < x < 1 was just a naive approach to flipping the test around, to allow efficient use of partitions.
[17:36:18] <awight>	 Here's what I did instead, not that cool however because it hardcodes the granularity: https://gerrit.wikimedia.org/r/#/c/analytics/reportupdater-queries/+/542419/1/reference-previews/baseline
[17:44:39] <nuria>	 awight
[17:45:06] <nuria>	 awight: with this data being so small joal recomendation is just leave it as is plus adding year=blah for partition
[17:54:55] <wikibugs>	 10Analytics, 10Analytics-EventLogging, 10Analytics-Kanban, 10Wikimedia-Portals: Sudden drop in WikipediaPortal events - https://phabricator.wikimedia.org/T234461 (10Nuria) hive (event)> select  webhost, count(*) from WikipediaPortal where year=2019 and month=9 and day in (13,14,15) group by webhost, day; w...
[17:55:12] <wikibugs>	 10Analytics, 10Analytics-EventLogging, 10Analytics-Kanban, 10Wikimedia-Portals: Sudden drop in WikipediaPortal events - https://phabricator.wikimedia.org/T234461 (10Nuria) Fixed now
[18:35:23] <awight>	 nuria: great, we'll try to keep a low profile, and will be done in a few months.  Still, it seems worthwhile to come up with a recommended where-clause since others will have the same issue, and may have more data than us.
[21:45:12] <wikibugs>	 (03PS2) 10Awight: New report for Reference Previews [analytics/reportupdater-queries] - 10https://gerrit.wikimedia.org/r/542419 (https://phabricator.wikimedia.org/T233108)
[21:57:24] <wikibugs>	 (03PS3) 10Awight: New report for Reference Previews [analytics/reportupdater-queries] - 10https://gerrit.wikimedia.org/r/542419 (https://phabricator.wikimedia.org/T231529)
[22:32:30] <wikibugs>	 (03PS4) 10Awight: New report for Reference Previews [analytics/reportupdater-queries] - 10https://gerrit.wikimedia.org/r/542419 (https://phabricator.wikimedia.org/T231529)
[22:57:08] <awight>	 I just found something fun, but would like to discuss it semi-privately if anyone is interested over the next few weeks: reader footnote clicks per pageview varies by at least 10x between wikis.  So many questions about why this might be!