[08:38:01] !log testlabs created 'puppetmaster' security group and add it to the arturo-puppetmaster VM [08:38:03] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Testlabs/SAL [09:31:50] !log icinga downtime cloud* servers for 30 mins (T261866) [09:31:54] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Icinga/SAL [09:31:55] T261866: Route cloud-hosts1-b-eqiad vlan through cloudsw - https://phabricator.wikimedia.org/T261866 [09:31:57] !log admin icinga downtime cloud* servers for 30 mins (T261866) [09:32:01] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Admin/SAL [09:32:09] !log icinga please ignore previous log entry [09:32:11] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Icinga/SAL [15:12:20] bd808 hello! :) We talked a few days ago about a problem with pywikibot. As you mentioned, the bug with auto-generating user script page was solved. But now I'm back to where I started: I get this problem every time I set up a continuous job: [15:12:53] (just a sec while I copy it :P ) [15:13:56] Traceback (most recent call last): [15:14:15] It gets stuck here over and over and over again... What am I doing wrong? :/ [15:14:28] !log tools.heritage Deploy latest from Git master: f43b6aa, a379e25, b6b5e92, 5108541 [15:14:32] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.heritage/SAL [15:15:03] Bigem: all that came through was the "Traceback" header. Try putting your stack trace into https://paste.toolforge.org/ and pasting the link you get back here. [15:15:46] https://paste.toolforge.org/view/a5b0aa78 [15:15:52] Yes, write. Sorry, I forgot again. [15:15:56] right* [15:22:42] Bigem: hmmm... that looks like your environment is missing PYTHONPATH=/data/project/shared/pywikibot/stable:/data/project/shared/pywikibot/stable/scripts in the environment [15:23:02] Bigem: did that happen when trying to run on the job grid? [15:23:08] Yes. [15:23:45] The job gets accepted but it does nothing. And I went to see the error file and it gets filled with those traces. [15:24:06] Over and over again, even for days if I don't stop the job with qdel. It's an infinite loop. [15:25:16] Let me take a quick look at your wrapper script, but my guess is that your $HOME/.bash_profile is not run on the job grid. [15:25:31] Oh... Maybe I have forgot to follow this step: [15:25:40] -Import the path settings into your current session: - [15:25:53] Do I need to do that everytime I ssh? [15:26:11] I'm sure I haven't done that before running the job. [15:26:58] I did that just on the beginning, some days ago, when I wrote the $HOME/.bash_profile file. [15:27:39] Bigem: the `export PYTHONPATH=...` in $HOME/.bash_profile will do that for your interactive shells, but I'm pretty sure that jobs running on the job grid do not source $HOME/.bash_profile before running [15:28:24] I think a fix would be adding `source $HOME/.bash_profile` to your $HOME/Smallem.sh jsut before you run the script. [15:28:58] * bd808 realizes that was worded badly [15:29:44] add the line `source $HOME/.bash_profile` to $HOME/Smallem.sh between the `#!/bin/bash' line and the `python3 ...` line [15:30:27] I tried now to source from it and it worked. It is running and not showing any errors yet. (running on simulate). [15:30:58] I guess that's what was missing. Do you think I still need to add that line in the wrapper code? [15:31:24] Bigem: is it running on the job grid now, or just in your ssh session? [15:31:30] On the job grid. [15:31:44] And still no errors. <3 XD [15:32:12] oh, magical! I wonder if that means the submitted job inherits the environment it was started from? [15:32:33] I guess... That was the only step I have never followed. [15:32:43] Because I thought I only needed to do that once. [15:32:57] But now I read it carefully and... [15:32:57] if so you may still need the `source $HOME/.bash_profile` if you decide to start the grid job from your crontab [15:33:51] Oh well, just to be safe, I'm adding it on the wrapper script then. I guess it can't do any harm? [15:34:08] I should not :) [15:37:12] Yes, I added it and again the job runs correctly, without any errors. I have just one more question that isn't entirely clear to me from the documentation page: [15:38:07] I can run the bot with the command jstart and theoretically (?) I wouldn't need to deal with it anymore because theoretically (?) it will be running forever, no? [15:38:34] Why would I need to set up my crontab? [15:38:48] Or is that a problem for the grid itself? To run it like that? [15:40:04] Why would one need cronjobs when you can start continuous jobs on the grid? Aren't cronjobs only needed on 1 time jobs? [15:51:42] bd808, sorry for the ping, just making sure you've read what I sent since I have to attend something in 10 minutes or so. [15:51:42] Bigem: if your script really runs forever then yes the jstart should be all you need. I have a feeling that this is actually a process that completes though, which probably means you will want to run it every (period of time) like once a day or once an hour or something [15:53:04] No, it runs forever. At least, it should. It continuously checks for errors with citations on SqWiki that can be solved with a bot and fixes what it can. [15:53:47] It just runs through every entry on the mainspace over and over again. At least, it should. [15:59:49] Well, that solves it then. Thank you a lot for everything! I really hope I have no more need to come here and ask for help in the near future. :P Thank you a lot again! You've been really helpful! [16:00:10] Have a good day! :)) [17:10:14] !log tools.bridgebot Add #wikipedia-it-sysop irc<->telegram bridge. (T261954) [17:10:17] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.bridgebot/SAL [17:14:09] !log quarry `framawiki@quarry-web-01:/tmp$ find /tmp/* -mtime +360 -user www-data -exec sudo rm -v {} \;` 775 files deleted for 10G. Again. T261909 [17:14:12] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Quarry/SAL [17:14:12] T261909: In the Quary tool the download as excel creates empty files - https://phabricator.wikimedia.org/T261909 [17:14:25] hi there. guys, can somebody do this task? https://phabricator.wikimedia.org/T261567 [17:51:21] balloons: ^ have you handled a 2FA removal yet? [17:51:38] bd808, no time like the present! [17:52:05] balloons: https://wikitech.wikimedia.org/wiki/Password_reset#Reset_two_factor_authentication [17:52:53] and also https://wikitech.wikimedia.org/wiki/Password_reset#For_users for rluts to prove identity if that has not been done yet [18:38:03] rluts, it's been removed. Thanks for your patience! [18:38:49] balloons: thanks! [20:14:32] !log wmde-dashboards increased cores to 24 and ram to 49152 [20:14:34] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Wmde-dashboards/SAL [20:23:15] balloons: handy tip about !logs, Stashbot will comment on the ticket too if you but the task's T.... id in the message [20:23:44] oO.. nice! I normally like linking tickets everywhere, but I failed this time [21:50:38] !log library-upgrader added libraryupgrader2.wmcloud.org DNS proxy and remove wmflabs.org one for automatic redirect (T261995) [21:50:42] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Library-upgrader/SAL [21:50:42] T261995: Migrate libraryupgrader2.wmflabs.org domain to wmcloud.org - https://phabricator.wikimedia.org/T261995 [22:08:19] andrewbogott: would it be possible for the "Edit proxy" option in horizon to allow switching from wmflabs.org to wmcloud.org ? [22:08:33] right now that field is disabled/read only [22:10:01] !log packagist-mirror switch domain to wmcloud.org [22:10:03] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Packagist-mirror/SAL [22:18:38] !log packagist-mirror manually kicking mirror script, it apparently got stuck on 2020-07-01 [22:18:40] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Packagist-mirror/SAL [22:24:58] legoktm: mind if I +2 https://gerrit.wikimedia.org/r/c/labs/libraryupgrader/+/624277 ? [22:25:32] hauskatze: please! [22:26:36] it's in zuul already [22:26:57] also, posted a hint as to why zuul complains on-task [22:27:08] might be that, or now [22:27:10] *not [22:30:37] hauskatze: maybe you just have the magic touch [22:31:00] * hauskatze can do things [22:33:44] zuul seems happy with https://gerrit.wikimedia.org/r/c/labs/libraryupgrader/+/623968 now as well [22:36:01] * hauskatze zZzZ