[00:12:06] <paladox>	 bd808 hi, i guess we can remove the old labs vagrant class, it seems no instances use it now
[00:12:07] <paladox>	 https://tools.wmflabs.org/openstack-browser/puppetclass/role::labs_vagrant
[00:12:53] <paladox>	 oh, there's a deprecated role too
[00:12:55] <paladox>	 https://tools.wmflabs.org/openstack-browser/puppetclass/role::deprecated::labsvagrant
[00:14:32] <bd808>	 I'm pretty sure we moved role::labs_vagrant to role::deprecated::labsvagrant years ago
[00:15:21] <bd808>	 role::labs::vagrant was the old name
[00:16:46] <paladox>	 ah thanks bd808, i guess it should be updated here https://github.com/wikimedia/puppet/blob/e959321aa620b77403cc9379db2e86080323c6e8/modules/role/manifests/labs/mediawiki_vagrant.pp#L18 ?
[00:17:26] <bd808>	 paladox: if you want to help with this, you could update T121477 to have the current list of VMs and gently nudge owners of live vms
[00:17:26] <stashbot>	 T121477: Migrate projects using ::role::deprecated::labsvagrant to ::role::labs::mediawiki_vagrant - https://phabricator.wikimedia.org/T121477
[00:17:35] <paladox>	 ok
[00:17:41] <paladox>	 will update the list now
[00:18:22] * bd808 is reminded to finish replacing grantreview-dev.grantreview.eqiad.wmflabs
[00:18:45] <Sagan>	 bd808: apropos reminder: guess the 30 minutes are over now :D
[00:19:51] <bd808>	 Sagan: looks like it ran :)
[00:20:07] <Sagan>	 bd808: nice, thank you very much :)
[00:21:58] <paladox>	 i think i updated the list correctly :)
[00:22:53] <paladox>	 bd808 https://tools.wmflabs.org/openstack-browser/server/grantreview-dev.grantreview.eqiad.wmflabs has an unkown image heh
[00:23:10] <bd808>	 that just means it is really old
[00:23:18] <paladox>	 heh :)
[00:23:44] <bd808>	 I built a new host there last week but haven't moved traffic over to it yet
[00:24:01] <paladox>	 ah ok
[00:24:19] <Sagan>	 what does that host do? the name sounds interesting
[00:24:40] <paladox>	 tgr hi, im wondering would you be able to migrate sentry-testwiki.sentry.eqiad.wmflabs from ::role::deprecated::labsvagrant to ::role::labs::mediawiki_vagrant please? :)
[00:24:48] <bd808>	 it's the demo server for https://iegreview.wikimedia.org/
[00:25:33] <paladox>	 legoktm hi, im wondering would you be able to migrate gadgets.social-tools.eqiad.wmflabs from ::role::deprecated::labsvagrant to ::role::labs::mediawiki_vagrant please? :)
[00:25:49] <bd808>	 Sagan: the app is https://www.mediawiki.org/wiki/Grants_Review_app
[00:25:55] <Sagan>	 bd808: ah, thx
[00:26:02] <paladox>	 this has the same puppet class applied twice https://tools.wmflabs.org/openstack-browser/server/drmf.math.eqiad.wmflabs
[00:26:20] <paladox>	 and it's using the new and old puppet class
[00:26:21] <paladox>	 role::deprecated::labsvagrant
[00:26:21] <paladox>	 role::labs::lvm::srv
[00:26:21] <paladox>	 role::labs::lvm::srv
[00:26:21] <paladox>	 role::labs::mediawiki_vagrant
[00:26:31] <bd808>	 probably some bug int he reporting tool
[00:29:04] <paladox>	 bd808 sorry for another ping, i've done https://gerrit.wikimedia.org/r/#/c/389295/ (i wonder should the fail check be moved to the top?)
[00:30:19] <bd808>	 order doesn't matter in puppet manifests
[00:30:37] <paladox>	 ah ok thanks.
[00:32:51] <paladox>	 it seems that labs_vagrant/maniphest/init.pp exist
[00:32:58] <paladox>	 modules/labs_vagrant/manifests/init.pp
[00:33:14] <bd808>	 yes. that's the module that the deprecated role applies
[00:35:00] <bd808>	 https://github.com/wikimedia/puppet/blob/production/modules/role/manifests/deprecated/labsvagrant.pp
[00:35:24] <paladox>	 ah thanks
[00:35:33] <bd808>	 the role applies the class which provisions mediawiki-vagrant in the old weird way
[00:36:57] <paladox>	 ok.
[00:40:09] <paladox>	 Reedy hi, im wondering would you be able to migrate xsstest.security-tools.eqiad.wmflabs from ::role::deprecated::labsvagrant to ::role::labs::mediawiki_vagrant please? :)
[00:40:53] <paladox>	 leogktm also im wondering could you do the same for butterfly-m4m.butterfly.eqiad.wmflabs please? :)
[00:41:03] <paladox>	 woops misplet legoktm :)
[03:23:50] <legoktm>	 if I'm going to (for fun) write a tool in PHP, are there any recommended examples/libraries/frameworks in use on toolforge?
[04:06:08] <bd808>	 legoktm: I can pitch things I use I guess. SAL and bash are both using https://github.com/wikimedia/wikimedia-slimapp as a base
[04:06:59] <bd808>	 I started to write a PHP version of your toolforge python library at some point but I don't think I ever finished it
[04:08:19] <bd808>	 legoktm: oh here it is -- https://github.com/toolforge/toolforge-php-mysql -- and yeah not really finished at all
[04:09:09] <bd808>	 there are several different libs for talking to the MW API
[04:09:48] <bd808>	 https://www.mediawiki.org/wiki/API:Client_code#PHP
[04:10:14] <bd808>	 I haven't really used any of them though
[04:10:32] <bd808>	 It turns out almost none of my tools actually talk to MediaWiki.
[04:11:06] <bd808>	 stashbot maybe the only thing that does and it's python
[04:11:07] <stashbot>	 See https://wikitech.wikimedia.org/wiki/Tool:Stashbot for help.
[04:11:22] * bd808 pats stashbot on the head
[05:21:23] <legoktm>	 thanks bd808
[05:22:53] <legoktm>	 bd808: also, https://wikitech.wikimedia.org/wiki/Help:Toolforge/Kubernetes#php5.6_.28Lighttpd_.2B_PHP.29 wasn't really clear if there's a specific directory structure I need to follow like for python?
[05:23:49] <bd808>	 $HOME/public_html
[05:24:54] <bd808>	 https://wikitech.wikimedia.org/wiki/Help:Toolforge/Web/Lighttpd#PHP
[05:25:40] * bd808 -> sleep
[05:27:32] <legoktm>	 thanks :) and good night
[18:20:28] <musikanimal>	 hello, what command do I use to check my k8s continuous jobs, and kill them? also, where is stdout logged?
[18:23:42] <valhallasw`cloud>	 musikanimal: something with kubectl
[18:24:28] <valhallasw`cloud>	 https://kubernetes.io/docs/user-guide/kubectl-cheatsheet/
[18:24:53] <valhallasw`cloud>	 so `kubectl get pods` and `kubectl logs <podname>`
[18:25:31] <musikanimal>	 there we go
[18:27:16] <musikanimal>	 and I use `kubectl delete` to kill it?
[18:29:52] <musikanimal>	 looks like `kubectl delete pod` brings it back up
[18:31:39] <musikanimal>	 I had to delete yaml file
[18:31:54] <musikanimal>	 going to add all of this to Wikitech!
[18:38:57] <madhuvishy>	 musikanimal: kubectl delete deployment might have done it
[18:39:38] <musikanimal>	 yeah I did `kubectl delete -f deployment.yml`
[18:39:57] <musikanimal>	 I'll try what you said next, because ideally you wouldn't have to delete the YAML file entirely
[18:40:39] <madhuvishy>	 yeah! just `kubectl get deployment` and `kubectl delete deployment <name>`
[18:41:17] <musikanimal>	 cool, and actually it doesn't delete the YAML file, it turns out
[18:43:04] <madhuvishy>	 as long as you have a deployment active, k8s will always keep trying to make whatever you specified in the config and the current state match, so if you delete a pod, and the spec is for a pod to be running, it will bring a pod back up even if you delete the pod (reconciliation loop). so you delete the deployment in this case.
[18:43:06] <madhuvishy>	 cool
[18:43:30] <madhuvishy>	 $ kubectl delete -f ./pod.json                                              # Delete a pod using the type and name specified in pod.json
[18:43:59] <musikanimal>	 got it, thanks!
[18:43:59] <madhuvishy>	 yeah looks like that would not delete the spec, just seems confusing I think
[18:44:05] <musikanimal>	 yeah it is
[18:46:35] <musikanimal>	 I'm trying to get a continuous Node script running
[18:47:11] <musikanimal>	 I was using `nohup` but it will mysteriously die after a few hours or so
[23:34:50] <bd808>	 !log tools.wsexport Tool is leaking files in /tmp; 955 files leaked on tools-webgrid-lighttpd-1428.tools.eqiad.wmflabs
[23:34:53] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.wsexport/SAL
[23:37:07] <bd808>	 !log tools Cleaned up 955 files owned by tools.wsexport on tools-webgrid-lighttpd-1428 with `sudo find /tmp -user tools.wsexport -exec rm {} \+`
[23:37:10] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL
[23:43:35] <bd808>	 !log tools.rezabot Tool is leaking files in /tmp; 109 files leaked on tools-webgrid-lighttpd-1428.tools.eqiad.wmflabs
[23:43:37] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.rezabot/SAL
[23:44:13] <bd808>	 !log tools Cleaned up 109 files owned by tools.rezabot on tools-webgrid-lighttpd-1428 with `sudo find /tmp -user tools.rezabot -exec rm {} \+`
[23:44:16] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL
[23:48:41] <bd808>	 !log tools Cleaned up 2 huge /tmp files left by tools.croptool (~6.5G)
[23:48:44] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL
[23:57:04] <bd808>	 !log testlabs Running stress-ng test on labvirt1015stresstest* vms for T171473
[23:57:07] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Testlabs/SAL
[23:57:07] <stashbot>	 T171473: labvirt1015 crashes - https://phabricator.wikimedia.org/T171473