[09:10:09] !log tools.stashbot restarting to remove wikimedia-cloud-daily from the list of channels [09:10:12] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.stashbot/SAL [09:10:37] xd [09:11:41] !log tools.stashbot restart done, making sure it logs to sal [09:11:44] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.stashbot/SAL [09:11:47] awesome :) [09:54:06] !log toolforge deploy toolforge-jobs-framework-cli 16.0.1 [09:54:07] dcaro: Unknown project "toolforge" [09:54:08] dcaro: Did you mean to say "tools.toolforge" instead? [09:54:16] !log tools deploy toolforge-jobs-framework-cli 16.0.1 [09:54:18] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL [12:56:06] !log paws Remove 123-11 cluster T355785 [12:56:11] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Paws/SAL [12:56:11] T355785: Remove paws-123-11 cluster - https://phabricator.wikimedia.org/T355785 [14:13:37] Hi all, I am trying to read a MediaWiki History Dump dataset into a dask dataframe (in Python) and I am getting a ValueError that I do not know how to debug. From what I understand from the error it is trying to convert a string (a username) to a number, for some reason. the point is that I don't even understand which is the column or row that is [14:13:38] creating the issue. [14:15:28] Does anybody have any pointer? I don't have much experiendce with Dask (and I am just starting to regret using it) [14:27:46] If you want to see the code, it's here: https://gist.github.com/CristianCantoro/7f4aeae4f25ba57ce1df51a4156a32aa [14:49:12] CristianCantoro: you can try checking if the CSV files have inconsistent data (feels like one of the columns with data type Int64, has an entry that is a string `"Paucabot"`, it only happens on compute because it's lazy loaded I think) [15:28:16] dcaro: thanks, I will look into that. And yes, data in dask are lazy loaded. [16:31:36] for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as possible and to free skiyomi and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as possible and to free skiyomi and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as possible and to free skiyomi [16:31:36] and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as possible and to free skiyomi and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as possible and to free skiyomi and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as possible and to [16:31:37] free skiyomi and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as possible and to free skiyomi and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as possible and to free skiyomi and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as [16:31:37] possible and to free skiyomi and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as possible and to free skiyomi and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and die as soon as possible and to free skiyomi and other ltas for yamlafuck pooyo and deepfuckfuck to jump off a bridge and [16:31:38] die as soon as possible and to free skiyomi and other ltas [16:31:50] ** Warning: if there is any bot in #wikimedia-cloud which should be exempted from ozone, voice it or contact staffers before it gets caught ** [18:09:24] If I create a new security group in horizon, does that take effect immediately, or do I need to wait for puppet to run? [18:09:36] immediately [18:11:25] and I assume 172.16.0.0/21 covers the k8s hosts in addition to the bastions? [18:12:20] that would cover toolforge k8s as seen by the rest of cloud vps, yes [18:12:40] ok, cool. Thanks [18:20:31] roy649: if you have a specific rule that should be working but is not we're happy to have a look [18:22:56] no, it's working now, I just needed to actually apply the rule to the host after creating it :-) [18:23:13] that would do it :-P [19:33:59] Is there a mechanism to increase resource allocations to kubernetes pods ? [19:36:17] I'm trying to use multithreading to speed up https://link-dispenser.toolforge.org, however, I keep hitting a brick wall when I go over 10 workers (it always errors out as a BrokenPool) [19:46:26] `kubectl describe resourcequotas` might help with figuring out what quota you're running up against [19:46:41] https://wikitech.wikimedia.org/wiki/Help:Toolforge/Web#Requesting_additional_tool_memory if it is cpu or memory [19:47:01] https://wikitech.wikimedia.org/wiki/Help:Toolforge/Kubernetes#Quotas_and_Resources if you need more quota [19:48:35] as a guess it might be replica limits [19:49:41] https://k8s-status.toolforge.org/namespaces/tool-link-dispenser/ and the linked Grafana dashboard might be of help too