[00:25:59] <roy649>	 On tools-sgebastion-11, when I run "webservice --backend=kubernetes python3.11 shell", I get:
[00:26:10] <roy649>	 Traceback (most recent call last):
[00:26:11] <roy649>	   File "/usr/local/bin/webservice", line 11, in <module>
[00:26:11] <roy649>	     load_entry_point('toolforge-webservice==0.1', 'console_scripts', 'webservice')()
[00:26:11] <roy649>	   File "/usr/lib/python3/dist-packages/toolsws/cli/webservice.py", line 195, in main
[00:26:11] <roy649>	     KubernetesBackend.get_types(),
[00:26:11] <roy649>	   File "/usr/lib/python3/dist-packages/toolsws/backends/kubernetes.py", line 291, in get_types
[00:26:12] <roy649>	     kubeconfig=Kubeconfig.load(), user_agent="webservice"
[00:26:12] <roy649>	   File "/usr/lib/python3/dist-packages/toolforge_weld/kubernetes_config.py", line 98, in load
[00:26:12] <roy649>	     raise KubernetesConfigFileNotFoundException(str(path.resolve()))
[00:26:13] <roy649>	 toolforge_weld.kubernetes_config.KubernetesConfigFileNotFoundException: /mnt/nfs/labstore-secondary-tools-home/roysmith/.kube/config
[00:27:12] <taavi>	 you forgot to `become` your tool account, and apparently the error message for that scenario is bugged
[01:09:21] <roy649>	 got it, thanks
[03:56:31] <roy649>	 what would it take if I wanted to run my own instance of elasticsearch on a VPS instance?
[03:56:39] <roy649>	 Is that something that needs to get installed via puppet?
[08:44:40] <wm-bb>	 <marcosmarado> Hi there, and sorry if this is the wrong channel for this. I noticed that Wikimedia Meet* is down. Can that be fixed, or should I be asking elsewhere? * https://meta.m.wikimedia.org/wiki/Wikimedia_Meet
[08:49:26] <wm-bb>	 <JeanFred> https://lists.wikimedia.org/hyperkitty/list/cloud@lists.wikimedia.org/message/SW6WVPGKOV5CRRIRCZXNKEU6QJ7ML2UX/ (re @marcosmarado: Hi there, and sorry if this is the wrong channel for this. I noticed that Wikimedia Meet* is down. Can that be fixed, or should ...)
[08:52:45] <wm-bb>	 <marcosmarado> Thanks! (re @JeanFred: https://lists.wikimedia.org/hyperkitty/list/cloud@lists.wikimedia.org/message/SW6WVPGKOV5CRRIRCZXNKEU6QJ7ML2UX/)
[12:45:29] <Guest95>	 https://w.wiki/_zNvb
[15:25:34] <wm-bb>	 <Klein> Hey everyone! While searching for my pywikibot's name on Google I randomly found an open ticket about it in Phab about needing it to migrate on Kubernetes. I am aware of the migration process happening but I had always assumed that it wouldn't require manual work from the tool maintainers (most of the time). Apparently I was wrong. Now I'm trying to migrate my tool there but I'm  <clipped message>
[15:25:35] <wm-bb>	 <Klein> a bit unsure of what I should do to consider the migration successful. Should I only change the crontab command and restart my webservice? (Also some change to the venv...) These are the three resources mentioned in the said task (T320048). I would appreciate any guidance.
[15:29:47] <wm-bb>	 <lucaswerkmeister> Klein it looks like your webservice is already running on kubernetes, so turning the crontab commands into `toolforge jobs` should be the main thing, I think
[15:31:38] <wm-bb>	 <Klein> Thank you for your fast answer! What about the recreation of the virtual environment? I'm not sure how much you can help in that because it can be different for each account I suppose but... (re @lucaswerkmeister: Klein it looks like your webservice is already running on kubernetes, so turning the crontab commands into toolforge jobs should...)
[15:32:11] <wm-bb>	 <lucaswerkmeister> ah, I see, the cronjobs use python…
[15:32:28] <wm-bb>	 <lucaswerkmeister> if you use a Python image with the same version as on the grid engine, using the same venv might work
[15:32:45] <wm-bb>	 <lucaswerkmeister> but it’s probably easier to just make a new one and reinstall the dependencies there
[15:33:31] <wm-bb>	 <lucaswerkmeister> you can use `webservice python3.11 shell` to get an interactive shell with Python 3.11, and set up the new venv in there
[15:33:41] <wm-bb>	 <lucaswerkmeister> and then use the `python3.11` image when running the jobs
[15:34:02] <wm-bb>	 <lucaswerkmeister> (I don’t remember which Python version the grid has but it’s definitely an older one, maybe 3.7 if I had to guess)
[15:34:50] <wm-bb>	 <lucaswerkmeister> alternatively, https://wikitech.wikimedia.org/wiki/Help:Toolforge/Python#Jobs has another set of instructions to set up a new venv
[15:39:05] <wm-bb>	 <Klein> Hmm... I see. Well, I'm doing the easy stuff by updating the crontab. Then I'll try to better understand how to deal with the Python version upgrade using the information you just gave to me. One last newbish question: Let's say I don't change anything beside the crontab, what happens? My script fails to activate on the given date and thus I understand that I need to deal with ve <clipped message>
[15:39:06] <wm-bb>	 <Klein> nv as well to make it work? Can I do that even beyond the deprecation date referencing the 2024 January?
[15:40:09] <wm-bb>	 <lucaswerkmeister> I think your scripts will just stop running
[15:40:17] <wm-bb>	 <lucaswerkmeister> starting the new jobs to replace them should be possible at any point
[15:40:32] <wm-bb>	 <lucaswerkmeister> I’m not sure if the old crontab will remain available… maybe run `crontab -l > crontab-backup.txt` now just in case ;)
[15:43:02] <wm-bb>	 <Klein> Okay. Thank you very much for your concise information! Sorry in advance if I come up again here soon asking for help getting unstuck in any of the steps. Haha
[15:44:19] <wm-bb>	 <lucaswerkmeister> ok, sure ^^
[15:44:45] <wm-bb>	 <lucaswerkmeister> I have a question of my own, actually… is there any size limit on toolforge build pack container images?
[15:44:58] <wm-bb>	 <lucaswerkmeister> I got “ERROR: failed to export: failed to write image to the following tags … unexpected status code 413 Request Entity Too Large” from the build command
[15:45:21] <wm-bb>	 <lucaswerkmeister> (lucaswerkmeister-test tool, but using the wd-shex-infer tool repo, k8s ref)
[15:47:04] <wm-bb>	 <lucaswerkmeister> trying again now after deleting some old builds, maybe that helps
[16:03:32] <wm-bb>	 <lucaswerkmeister> still failed :(
[16:05:36] <wm-bb>	 <lucaswerkmeister> o_O apparently the image… exists, it just couldn’t be tagged as `latest`? https://tools-harbor.wmcloud.org/harbor/projects/21347/repositories/tool-lucaswerkmeister-test/artifacts-tab/artifacts/sha256:77584ae6cb7f56db39f4bbfa1f1e91a32032a0b51b0086c95ae059b3c5d5be9f?publicAndNotLogged=yes
[16:16:54] <wm-bb>	 <lucaswerkmeister> meh, can’t get it to work with `toolforge jobs run` though
[16:58:59] <wm-bb>	 <Klein> @lucaswerkmeister, I tried updating my crontab but when saving the new info, it was automatically converted to use jsub again (as the hatnote says it would). Clearly I'm misunderstanding something.
[17:00:19] <RhinosF1>	 Toolforge jobs has its own system
[17:00:29] <RhinosF1>	 You don't need to add it through crontab
[17:03:59] <wm-bb>	 <Klein> Hmm... So do I need to do anything in regard to the kubernetes migration process? I thought I needed to upgrade my crontab as with the kubernetes equivalent as shown in the Phab task associated link. :/
[17:07:33] <RhinosF1>	 https://wikitech.wikimedia.org/wiki/Help:Toolforge/Jobs_framework#Creating_scheduled_jobs_(cron_jobs)
[17:07:46] <RhinosF1>	 Once you've added the job with a schedule
[17:07:54] <RhinosF1>	 You can delete it from your crontab
[17:10:21] <wm-bb>	 <Klein> Oh, so I create a "schedule" and that's it, right? (Also empty my crontab.)
[17:12:50] <RhinosF1>	 Yes
[17:19:44] <wm-bb>	 <Klein> Thank you!
[17:19:45] <wm-bb>	 <Klein> 
[17:19:47] <wm-bb>	 <Klein> In my case I have some jobs that follow this format:
[17:19:48] <wm-bb>	 <Klein> 
[17:19:50] <wm-bb>	 <Klein> 
[17:19:51] <wm-bb>	 <Klein> 0 13 1 * * /usr/bin/jsub -N Smallem-Transclusion -once ~/venv/bin/python3 ~/Smallem-Transclusion.sh sq wikipedia Aktivizimi mujor i Smallem-it për numërimin e përdorimit të stampave
[17:19:53] <wm-bb>	 <Klein> (copied from my crontab)
[17:19:54] <wm-bb>	 <Klein> 
[17:19:56] <wm-bb>	 <Klein> Would it be possible to provide me a working example how would I schedule this job in Kubernetes? I can follow with other cases myself after that I believe.
[17:23:24] <RhinosF1>	 It's cron syntax
[17:23:36] <RhinosF1>	 So you can still copy '0 13 1 * *'
[17:25:44] <wm-bb>	 <Klein> Yes but what about the path part + jsub? Should I use something like /usr/bin/Toolforge... Etc?
[17:31:34] <RhinosF1>	 You don't need jsub
[17:31:51] <RhinosF1>	 Or to put anything in a crontab
[17:35:49] <wm-bb>	 <Klein> 0 13 1 * * /usr/bin/toolforge jobs run myjob --command -N Smallem-Transclusion -once ~/venv/bin/python3 ~/Smallem-Transclusion.sh sq wikipedia Aktivizimi mujor i Smallem-it për numërimin e përdorimit të stampave --image bullseye
[17:35:50] <wm-bb>	 <Klein> 
[17:35:51] <wm-bb>	 <Klein> So is this correct? Or should I read more? :P Sorry, I'm not very good on such technicalities.
[17:36:06] <RhinosF1>	 You should read more
[17:36:59] <RhinosF1>	 "toolforge jobs run mycronjob --command ./daily.sh --image bullseye --schedule "@daily" " is the example it gives to set it up
[17:37:27] <RhinosF1>	 Blank your crontab and then run that with daily.sh changed to your command and @daily to the schedule
[17:38:47] <RhinosF1>	 So toolforge jobs run Smallem-Transclusion --command "~/venv/bin/python3 ~/Smallem-Transclusion.sh sq Wikipedia ....." --image bullseye --schedule "0 13 1 * *"
[17:39:42] <wm-bb>	 <Klein> Thank you very much! I will try it.
[17:40:09] <wm-bb>	 <Klein> I was following the example that I found in the Phab task link.
[17:40:30] <wm-bb>	 <Klein> https://wikitech.wikimedia.org/wiki/Help:Toolforge/Jobs_framework#Grid_Engine_migration
[17:41:53] <wm-bb>	 <Klein> I was made to believe from that table that there I needed to switch "jsub" to "Toolforge jobs..." and add "-- image bullseye" in the end while keeping everything unchanged.
[17:43:08] <wm-bb>	 <Klein> But apparently that is a wrong interpretation. Thought I'd say anyway because it can be interesting to know how such documentation pages may get (wrongly) interpretated by non-so-tech folks.
[17:45:07] <RhinosF1>	 It's the cron bit that makes it more complex
[18:19:58] <roy649>	 I'm trying to follow the instructions at https://wikitech.wikimedia.org/wiki/Help:Puppet, but the screenshots and instructions don't match what I'm seeing in horizon
[18:20:29] <roy649>	 There are not "project", "common" or "all" buttons
[18:48:02] <wm-bot>	 !log tools.lexeme-forms <lucaswerkmeister> deployed 7611d4e980 (l10n updates: ia, krc, sv)
[18:48:06] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.lexeme-forms/SAL