[08:12:55] <arturo>	 !log tools depool sgeexec-0907 (T290798)
[08:13:00] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL
[08:13:00] <stashbot>	 T290798: tools-sgeexec-0907 filesystem corruption - https://phabricator.wikimedia.org/T290798
[08:13:29] <arturo>	 oh, it was already depooled :-)
[08:13:53] <majavah>	 yeah, I did that before filing the task
[08:14:16] <arturo>	 cool
[08:14:18] <arturo>	 rebooting it now
[08:14:27] <arturo>	 !log tools rebooting sgeexec-0907 (T290798)
[08:14:30] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL
[08:55:48] <arturo>	 !log tools repooling sgeexec-0907 (T290798)
[08:55:52] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL
[08:55:53] <stashbot>	 T290798: tools-sgeexec-0907 filesystem corruption - https://phabricator.wikimedia.org/T290798
[08:57:07] <arturo>	 !log tools cleared grid queues error states (T290844)
[08:57:09] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL
[08:57:10] <stashbot>	 T290844: toolforge grid master queue error cleanup - https://phabricator.wikimedia.org/T290844
[12:43:32] <addshore>	 Is there a commended best practice for if I want yarn in a node image on k8s toolforge?
[12:43:40] <addshore>	 (as it is not there out of the box it seems)
[13:08:24] <addshore>	 hmm, and doing `npm install yarn` in an itneractive containers seems to run out of memory :/
[15:44:14] <majavah>	 !log toolsbeta deploy volume-admission-controller in background; T279106
[15:44:17] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Toolsbeta/SAL
[15:44:17] <stashbot>	 T279106: Establish replacement for PodPresets in Toolforge Kubernetes - https://phabricator.wikimedia.org/T279106
[16:12:11] <wm-bb>	 <DennisPriskorn> who can I talk to about consuming Kafka events and stream them into a column store in the wmf cloud?
[16:12:11] <wm-bb>	 <DennisPriskorn> I'm planning to play with Rya there.
[16:12:13] <wm-bb>	 <DennisPriskorn> My working hypothesis is that sparql update is infeasible in the long run and just consuming the change stream is faster and easier 😃
[16:17:28] <wm-bb>	 <DennisPriskorn> IDK but loosing internet was not a problem today for me on paws terminal. 
[16:17:29] <wm-bb>	 <DennisPriskorn> I refreshing the tab brought up a working terminal 😃 (re @wmtelegram_bot: <KCVelaga> Will PAWS notebooks keep running in the background even if the browser tab is closed?)
[16:21:00] <bd808>	 @DennisPriskorn See https://wikitech.wikimedia.org/wiki/Help:Cloud_VPS_project and https://phabricator.wikimedia.org/project/view/2875/ for information on requesting a Cloud VPS project. That's what you would need I think for the compute and storage parts of your idea. A potential issue to think about would be how much data you think you will need in the near/mid/long term. Disk storage is our most constrained resource.
[16:21:49] <majavah>	 What "kafka cluster" are you referring to? I don't think the WMF has any publicly available ones
[16:24:39] <wm-bb>	 <DennisPriskorn> Yeah, that's a problem I guess. 😅
[16:24:40] <wm-bb>	 <DennisPriskorn> I looked at the code for the new updater, but I'm unsure if that gets stuff from Kafka. (re @wmtelegram_bot: <majavah> What "kafka cluster" are you referring to? I don't think the WMF has any publicly available ones)
[16:24:57] <wm-bb>	 <DennisPriskorn> Thanks 😃 (re @wmtelegram_bot: <bd808> @DennisPriskorn See https://wikitech.wikimedia.org/wiki/Help:Cloud_VPS_project and https://phabricator.wikimedia.org/project/view/2875/ for information on requesting a Cloud VPS project. That's what you would need I think for the compute and storage parts of your idea. A potential issue to think about would be how much data you think you will need in the near/mid/long term. Disk sto
[16:25:29] <wm-bb>	 <DennisPriskorn> Thanks 😃 I'm thinking about starting small with a subset of the WD dump. (re @wmtelegram_bot: <bd808> @DennisPriskorn See https://wikitech.wikimedia.org/wiki/Help:Cloud_VPS_project and https://phabricator.wikimedia.org/project/view/2875/ for information on requesting a Cloud VPS project. That's what you would need I think for the compute and storage parts of your idea. A potential issue to think about would be how much 
[16:26:19] <bd808>	 https://wikitech.wikimedia.org/wiki/Event_Platform/EventStreams is the public interface to the kafka event queue
[16:30:40] <wm-bb>	 <DennisPriskorn> I consumed that before in a small bot I wrote.
[16:30:40] <wm-bb>	 <DennisPriskorn> I'm not sure the individual triples are exposed there? 
[16:30:41] <wm-bb>	 <DennisPriskorn> I would have to resort to downloading the whole item for every change I guess.
[16:43:20] <addshore>	 If anyone can review https://gerrit.wikimedia.org/r/c/operations/docker-images/toollabs-images/+/720788 it <3 them forever :)
[16:44:43] <majavah>	 addshore: I'm going to defer that to bstorm, who's out today
[16:44:52] <addshore>	 ack, okay!
[19:37:40] <majavah>	 !log tools.ircservserv add -releng to config
[19:37:45] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.ircservserv/SAL
[21:50:12] <addshore>	 Hmmmm `AttributeError: module 'platform' has no attribute 'linux_distribution'` when trying a webservice start on k8s node
[21:52:19] <addshore>	 oh im a fool