[07:12:28] <gehel>	 ejoseph: I have another appointment at the hospital with Oscar this morning. I'll be late for our meeting. I'll ping you when back.
[07:13:20] <ejoseph>	 Ok sure
[09:12:39] <gehel>	 ejoseph: I'm still in the bus on the way back :/
[09:13:03] <ejoseph>	 I am on the call
[09:13:09] <ejoseph>	 meet*
[09:13:15] <gehel>	 ejoseph: did you see T293326 ?
[09:13:16] <stashbot>	 T293326: Add EJoseph to #wmf-nda - https://phabricator.wikimedia.org/T293326
[09:13:38] <gehel>	 ejoseph: still at least 30' before I'm home. I'll ping you.
[09:19:39] <ejoseph>	 ok
[09:51:09] <gehel>	 ejoseph: I'm finally back!
[09:51:44] <ejoseph>	 give me 10mins
[09:52:17] <gehel>	 ack
[11:17:18] <gehel>	 Lunch
[11:28:20] <dcausse>	 lunch 2
[14:11:01] <gehel>	 Errand, back in 20'
[16:01:14] <ebernhardson>	 meh, camera not starting. back in a sec
[17:39:55] <cbogen_>	 FYI, Fiona from the GLAm team says "There's a lot of concern in the community about the 4-day WCQS outage, particularly as it follows the recent Pattypan breakage:
[17:39:55] <cbogen_>	 https://phabricator.wikimedia.org/T297454
[17:39:55] <cbogen_>	 Is there some kind of update, or plan, that we can share?"
[17:40:08] <cbogen_>	 ebernhardson: I know you were looking at this, what do you suggest I say?
[17:40:18] <ebernhardson>	 cbogen_: it's a beta service, so finding something broken on a friday means don't fix it untill next week
[17:40:23] <cbogen_>	 totally
[17:40:28] <cbogen_>	 but it'd be great to give an update/estimate on fixing it
[17:41:19] <ebernhardson>	 i don't have any particular update, it will probably be up today but I can't really say for sure it could be some much worse thing that caused it to go into the power shutdown phase
[17:41:44] <cbogen_>	 okay, thanks
[17:41:51] <cbogen_>	 if you learn anything just update the ticket - appreciate it
[17:57:27] <ryankemper>	 Looks like it's just a matter of the underlying host running the VM lacking disk space: https://phabricator.wikimedia.org/T297454#7567079
[18:00:25] <ryankemper>	 cbogen_: just for my own curiosity, do you know what avenue fiona was getting that concern from? Like a talk page on commons wiki somewhere? I checked village pump and didn't see anything but I didn't look super hard
[18:03:51] <ryankemper>	 but yeah, echoing what erik said, we need to make sure we're managing expectations properly. not only is the service itself in beta but also the beta doesn't look anything like how production wcqs will look infra wise (ie it's running on a vm rather than physical hosts, etc)
[18:14:14] <ebernhardson>	 probably not coming up today, may require a data reload 
[19:15:25] <ebernhardson>	 i'm going to guess the flink session cluster running in yarn, started dec 8, was a test that didn't get shut down yet?
[19:18:41] <dcausse>	 probably Zbyszko, it seems empty so fine to kill
[19:19:03] <cbogen_>	 ryankemper: she said it was in "backchannels", not really sure what that menas
[19:19:20] <cbogen_>	 I asked her to emphasize that it's a beta service, etc. 
[19:19:30] <dcausse>	 I remember Zbyszko liked the idea of having a session cluster always up
[19:20:36] <ebernhardson>	 ok, thanks!