[13:09:55] <Reedy>	 legoktm: https://phabricator.wikimedia.org/T140459
[15:37:50] <anomie>	 robla: Busy?
[15:40:03] <robla>	 hi anomie ... I'll PM
[19:21:17] <bd808>	 anomie: hmmm.. I don't think that new job type can ride the train can it?
[19:22:00] <bd808>	 since it runs on the local wiki that will mean that group0 could submit a job that tries to run on group2 and goes boom
[19:22:22] <anomie>	 bd808: Good question. Does it go boom, or does the job just get dropped?
[19:22:32] <bd808>	 I ... have no idea
[19:23:09] <anomie>	 https://phabricator.wikimedia.org/diffusion/MW/browse/master/includes/jobqueue/JobRunner.php;ee71c2658eb8388039248010bfc99b0ca90d5c34$106-110 ?
[19:23:48] <bd808>	 nice. so it will just be a no-op
[19:24:07] <anomie>	 And in this case, something else will just resubmit it again the next time it's hit after the train runs.
[19:24:18] <bd808>	 yeah
[19:25:09] <bd808>	 My random guess is that most of the dangling wikis are in group1
[19:25:18] <bd808>	 commons and wikidata
[19:30:01] <anomie>	 I'm not so sure, the issue from T119736#2458851 seems reasonably common (although not as common as the Echo thing) and can hit any wiki. I'm hoping Aaron will want to look into that one more next week.
[19:30:01] <stashbot>	 T119736: Could not find local user data for {Username}@{wiki} - https://phabricator.wikimedia.org/T119736
[19:32:25] <bd808>	 *nod* and that seems to be still related to echo right? because they are running lots of api requests in the hook?
[19:32:36] <bd808>	 it could happen for any reason, sure
[19:33:53] * bd808 -> lunch
[19:39:20] <anomie>	 Maybe, if Echo is making processes hold the transaction open and that's blocking other processes.