[00:00:31] <andrewbogott>	 !log glampipe stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:00:33] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Glampipe/SAL
[00:00:43] <andrewbogott>	 !log https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:00:43] <stashbot>	 andrewbogott: Unknown project "https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge"
[00:01:04] <andrewbogott>	 !log globalcu stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:01:06] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Globalcu/SAL
[00:01:48] <andrewbogott>	 !log graphql stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:01:50] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Graphql/SAL
[00:02:34] <andrewbogott>	 !log ircwebchat stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:02:35] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Ircwebchat/SAL
[00:05:25] <andrewbogott>	 !log sentry stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:05:27] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Sentry/SAL
[00:06:06] <andrewbogott>	 !log thumbor stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:06:08] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Thumbor/SAL
[00:06:39] <andrewbogott>	 !log wikidata-autodesc stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:06:40] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Wikidata-autodesc/SAL
[00:07:18] <andrewbogott>	 !log wikidata-realtime-dumps stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:07:20] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Wikidata-realtime-dumps/SAL
[00:08:53] <andrewbogott>	 !log annotation stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:08:55] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Annotation/SAL
[00:09:07] <andrewbogott>	 !log antiharassment stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:09:09] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Antiharassment/SAL
[00:09:23] <andrewbogott>	 !log community-labs-monitoring stopping all VMs as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2021_Purge
[00:09:25] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Community-labs-monitoring/SAL
[14:44:39] <Steinsplitter>	 Hello, I have the problem that on toolsforge big imagemagick tasks are getting killed. Allocating more mem isn't resolving it, there seems to be some.... protection thing to automatically kill some jobs.
[15:26:43] <urbanecm>	 hello Steinsplitter , is that happening in the queue, or at bastion?
[17:14:13] <Steinsplitter>	 urbanecm: both.
[17:17:33] <AntiComposite>	 I believe Debian ships ImageMagick resource limits by default
[17:17:40] <AntiComposite>	 are you getting an erorr message?
[17:17:46] <urbanecm>	 Steinsplitter: so, for bastion, i think there are indeed some quotas to avoid expensive actions happening at the bastions. For job queue, the only reason I recall is not enough memory -- there might be some indication in the job's .err file?
[17:19:56] <Steinsplitter>	  urbanecm: killed. Enough mem allocated.
[17:25:59] <AntiComposite>	 which tool?
[17:26:56] <Steinsplitter>	 rotbot (Rotatebot), this is a issue wich i try to solve since months :/
[17:33:36] <urbanecm>	 Steinsplitter: did you check whether the tool indeed leaves some allocated memory unused? 
[17:39:27] <Steinsplitter>	 urbanecm: ofc. I suspekt there is some sort of .... protection which kills the convert command (via exec). Will do some other tests, sec. :)
[17:40:05] <urbanecm>	 (you can do that with `qstat -j <job id>` for running jobs and via `qacct` for historical jobs)
[17:52:40] <Steinsplitter>	 thanks.