[04:05:20] <Jayprakash12345>	 hello everyone
[04:05:35] <Jayprakash12345>	 I have a question.
[04:06:14] <Jayprakash12345>	 How I can Upload .png on on tool at Toolforge
[04:44:14] <legoktm>	 !log codesearch restarted hound-search service (T188477), also I restarted hound-extensions earlier (T188450)
[04:44:18] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Codesearch/SAL
[04:44:18] <stashbot>	 T188477: CodeSearch: Add Extension:UploadWizard  - https://phabricator.wikimedia.org/T188477
[04:44:18] <stashbot>	 T188450: CodeSearch: SocialProfile extension is not being indexed by Codesearch - https://phabricator.wikimedia.org/T188450
[04:45:20] <zhuyifei1999_>	 Jayprakash12345: from where to toolforge?
[04:45:26] <zhuyifei1999_>	 your local computer?
[04:45:57] <zhuyifei1999_>	 or a remote repository?
[04:50:16] <Jayprakash12345>	 zhuyifei1999: From Local Computer
[09:08:28] <ema>	 hello! puppet is disabled on labstore1007 since 5 days ago with the message "testing"
[09:08:36] <ema>	 just FYI in case it's been forgotten :)
[09:15:55] <Sigfrido>	 !help My public ssh key is up and the access request was accepted, but I can't log into login.tools.wmflabs.org . Any idea why? I can login via ssh with keys in other  systems, so it should not be a problem in the ssh client
[09:15:55] <wm-bot5>	 Sigfrido: If you don't get a response in 15-30 minutes, please create a phabricator task -- https://phabricator.wikimedia.org/maniphest/task/edit/form/1/?projects=wmcs-team
[10:02:54] <Ester>	 Hi all, I have a question
[10:03:08] <Ester>	 I tried "webservice --backend=gridengine stop"
[10:03:22] <Ester>	 and then "webservice --backend=kubernetes start"
[10:03:35] <Ester>	 but I get: "You should stop that webservice by issuing:
[10:03:35] <Ester>	     webservice --backend=gridengine stop
[10:03:35] <Ester>	 And then start it again with backend kubernetes by issuing:
[10:03:35] <Ester>	     webservice --backend=kubernetes start"
[10:16:24] <Sveta>	 oops
[10:18:09] <Ester>	 now I'm using  webservice --backend=kubernetes nodejs shell 
[10:18:12] <Ester>	 and it works
[11:05:42] <joaquinito01>	 Hi! Please deploy MediaWiki 1.31.0-wmf.23
[11:06:15] <joaquinito01>	 Hi! Please deploy MediaWiki 1.31.0-wmf.23
[11:08:49] <jynus>	 joaquinito01: you can see here the versions and its schedule: https://tools.wmflabs.org/versions/
[11:08:59] <jynus>	 but this is the wrong channel about that
[11:10:12] <jynus>	 more info: https://phabricator.wikimedia.org/T183962
[11:10:18] <joaquinito01>	 Hi! Please deploy MediaWiki 1.31.0-wmf.23 in group 1, see the old version: https://www.mediawiki.org/wiki/Special:Version
[11:11:36] <jynus>	 again, this channel has nothing to do with mediawiki deployments
[11:12:06] <jynus>	 but if you need help or report problems, there is #wikimedia-tech
[11:21:21] <Ester_>	 cannot open ‘/public/dumps/public/enwiktionary/20180220/enwiktionary-20180220-pages-articles.xml.bz2’ for reading: Operation not permitted
[11:21:24] <Ester_>	 why?
[11:22:53] <Ester_>	 I cannot do that from ssh to etytree-b.etytree.eqiad.wmflabs
[11:25:50] <andre__>	 joaquinito01 can be kicked, it's a bot.
[11:34:24] <Ester_>	 can anyone help me set up a nodejs app on tool labs?
[11:50:11] <CFisch_remote>	 Hej hej. I created a tool on toolforge and it does not let my connect to the DB with the credentials given... :-/
[11:50:42] <CFisch_remote>	 Can anyone help? ;-)
[11:51:52] <CFisch_remote>	 ( ERROR 1045 (28000): Access denied for user 's53668'@'10.68.16.44' (using password: YES) )
[12:38:14] <CFisch_remote>	 ... :-)
[12:47:45] <CFisch_remote>	 bd808: ^ *poke*
[13:31:51] <chicocvenancio>	 CFisch_remote: what is the tool? 
[13:32:20] <CFisch_remote>	 tools.wmde-inline-movedparagraphs
[13:32:21] <chicocvenancio>	 Ester: sure, if I remember correctly you're having difficulty with npm
[13:32:27] <CFisch_remote>	 chicocvenancio: \
[13:32:44] <Reedy>	 CFisch_remote: mysql doesn't want you to move the paragraphs
[13:32:49] <CFisch_remote>	 :-D
[13:32:54] <CFisch_remote>	 seems so
[13:40:30] <chicocvenancio>	 CFisch_remote: there is something wrong with the user for that tool, could you create a phab task?
[13:40:55] <CFisch_remote>	 okay, sure
[13:45:44] <CFisch_remote>	 chicocvenancio: https://phabricator.wikimedia.org/T188508 feel free to add tags / subscribers
[15:03:18] <Nudin_WMDE>	 Technical Advice IRC meeting starting in 60 minutes in channel #wikimedia-tech, hosts: @addshore & @Christoph_Jauera_(WMDE) - all questions welcome, more infos: https://www.mediawiki.org/wiki/Technical_Advice_IRC_Meeting
[15:25:50] <zhuyifei1999_>	 Ester: for that operation not permitted thing, try rebooting
[15:26:31] <zhuyifei1999_>	 I encountered it before, it’s some sort of NFS client ID collision
[15:59:31] <Ester_>	 zhuyifei1999_:thanks
[16:03:15] <chicocvenancio>	 Ester_: how is the node.js tool going? 
[16:03:21] <chicocvenancio>	 Do you need help? 
[16:03:55] <Ester_>	 Hi! thanks! I followed your instructions, now I'm trying to do npm install and I'm fixing a bug
[16:04:01] <Ester_>	 will keep you posted. Thanks a lot!
[16:04:56] <chicocvenancio>	 you're welcome
[16:06:32] <Ester_>	 I tried to install and I get two warnings :
[16:06:33] <Ester_>	 fs: re-evaluating native module sources is not supported. If you are using the graceful-fs module, please update it to a more recent version.
[16:06:38] <Ester_>	 webpack-jquery-ui@1.0.0 No repository field.
[16:07:04] <Ester_>	 besides this, how do I arrage folders?
[16:07:12] <Ester_>	 and how do I start the app?
[16:07:25] <Ester_>	 I saw on the documentation that I should create a folder www
[16:07:41] <Ester_>	 and inside www a folder js with the javascript file
[16:08:35] <Ester_>	 but my package structure is a bit different as I have a webpack.config.js file and a folder src
[16:08:53] <Ester_>	 within src I have the javascript code
[16:09:21] <chicocvenancio>	 to start the app, the best way would be to setup the folder structure as k8s expects it and run `webservice --backend=kubernettes nodejs start`
[16:09:40] <chicocvenancio>	 *`webservice --backend=kubernetes nodejs start` (single t)
[16:10:01] <Ester_>	 and webpack currently writed to  the dist foder
[16:10:26] <Ester_>	 *writes
[16:10:32] <Ester_>	 *folder (sorry)
[16:11:04] <zhuyifei1999_>	 'fs: re-evaluating native module sources is not supported. If you are using the graceful-fs module, please update it to a more recent version.' <= iirc it's some version incompatibilities
[16:12:14] <Ester_>	 in my package.json file: "start": "./node_modules/.bin/webpack-dev-server --inline --progress"
[16:12:54] <zhuyifei1999_>	 Ester_: for the folder structure as long as www/js/server.js is able to load your main script and execute the server it's okay
[16:13:17] <zhuyifei1999_>	 the webservice code do not currently read package.json
[16:13:46] <chicocvenancio>	 yes, I see you have no server.js
[16:13:52] <Ester_>	 right
[16:13:54] <Ester_>	 I don't
[16:13:56] * chicocvenancio trying to understand the code
[16:14:55] <Ester_>	 webpack creates a file at ./dist/bundle.js
[16:15:21] <Ester_>	 and copies .src/html/index.html into ./dist/index.html
[16:15:37] <chicocvenancio>	 is this actually to be served by node?
[16:15:38] <Ester_>	  ./dist/index.html in turn calls bundle.js
[16:15:59] <chicocvenancio>	 I think bundle.js is js to be called client-side
[16:16:27] <zhuyifei1999_>	 chicocvenancio: same question :)
[16:16:41] <chicocvenancio>	 yep, https://webpack.js.org/ confirms it
[16:17:11] <chicocvenancio>	 so, Ester_  you only need the node part to build the bundle, that is done
[16:17:52] <chicocvenancio>	 you will need to do it again if the code changes (might be good to leave that as cron job or something later)
[16:18:14] <zhuyifei1999_>	 to serve the static html and js you can use lighttpd if you want. probably easiest to setup
[16:18:59] <bd808>	 webpack is fundamentally a static site generator, but it does have a "live" mode that it typically only used for local development.
[16:19:33] <Ester_>	 can you pont me to docs on  lighttpd?
[16:19:37] <Ester_>	 thanks a lot guys!
[16:20:01] <bd808>	 https://wikitech.wikimedia.org/wiki/Help:Toolforge/Web/Lighttpd
[16:20:10] <Ester_>	 great
[16:21:02] <zhuyifei1999_>	 bd808: I'm wondering, can tools-static serve the site if it's 100% static?
[16:21:11] <bd808>	 if you put your html+js+css into $HOME/public_html and run `webservice --backend=kubernetes php5.6 start` you are pretty much done
[16:22:10] <bd808>	 zhuyifei1999_: yes. the $HOME/www/static directory is served as https://tools-static.wmflabs.org/$TOOL/
[16:22:44] <bd808>	 the webserver for this is nginx, but there is no way to do per-tool/directory configuration changes
[16:23:27] <zhuyifei1999_>	 we can symlink :)
[16:24:28] <bd808>	 yes, symlink farms should work
[16:29:51] <chicocvenancio>	 Ester_: feel free to ping me if you need help with any of that
[16:30:41] <Ester_>	 so do I put my js css and html in public html?
[16:30:54] <Ester_>	 chicocvenancio: thanks 
[16:31:22] <chicocvenancio>	 yes, an index.html that includes the bundle.js script will work
[17:01:15] <zhuyifei1999_>	 Ester_: you may also be able to symlink public_html to dist
[17:30:54] <Ester_>	 ok I have set webpack output directly to thw public_html folder
[17:31:12] <Ester_>	 I'm trying webservice --backend=kubernetes php5.6 start
[17:31:24] <Ester_>	 but I get "Looks like you already have another webservice running, with a gridengine backend
[17:31:24] <Ester_>	 You should stop that webservice by issuing:
[17:31:24] <Ester_>	     webservice --backend=gridengine stop
[17:31:24] <Ester_>	 And then start it again with backend kubernetes by issuing:
[17:31:24] <Ester_>	     webservice --backend=kubernetes start"
[17:31:31] <Ester_>	 I logged out
[17:31:35] <Ester_>	 and logged in again
[17:32:33] <Ester_>	 zhuyifei1999_: what do you mean by "try rebooting?
[17:32:36] <Ester_>	 "
[17:32:56] <chicocvenancio>	 Ester_: rebooting was for the VPS error
[17:33:13] <chicocvenancio>	 you can go to horizon and ask to soft reboot the instance
[17:33:45] <Ester_>	 hmmm... ok
[17:34:00] <zhuyifei1999_>	 Ester_: for that webservice, did you stop your currently running webservice?
[17:34:00] <bd808>	 Ester_: what's your tool name? I've seen this problem before, but I'd like to take a look at the current state before I tell you how to fix it
[17:34:13] <chicocvenancio>	 bd808: tools.etytree
[17:34:15] <Ester_>	 etytree is the name of the tool
[17:34:26] <Ester_>	 yes I did
[17:34:31] <Ester_>	 I think
[17:34:39] <Ester_>	 webservice --backend=kubernetes stop
[17:35:15] <bd808>	 yeah. this is the goofy state that can happen.
[17:35:16] <zhuyifei1999_>	 I guess bd808 will debug it :) /me is on mobile rn
[17:35:46] <bd808>	 `qstat` shows no grid jobs. `kubctl get pods` shows no running pods.
[17:35:54] <Ester_>	 one more question: do I have to put the package in the www/js folder?
[17:36:00] <Ester_>	 and run nom install from there?
[17:36:13] <Ester_>	 in kubernetes?
[17:36:18] <Ester_>	 *npm
[17:36:24] <chicocvenancio>	 bd808: `webservice status` says it is not running as well
[17:36:56] <bd808>	 !lot tools.etytree Deleted $HOME/service.manifest
[17:37:00] <bd808>	 !log tools.etytree Deleted $HOME/service.manifest
[17:37:03] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.etytree/SAL
[17:37:16] <bd808>	 !log tools.etytree Started with `webservice --backend=kubernetes php5.6 start`
[17:37:17] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.etytree/SAL
[17:37:29] <bd808>	 Ester_: it should be running now
[17:37:50] <Ester_>	 I can see that
[17:37:55] <Ester_>	 it is almost working
[17:38:04] <Ester_>	 there areother things that do not work :)
[17:38:06] <bd808>	 the "trick" is deleting the service.manifest file. That seems to clear up the confusion that the webservice watchdog can have
[17:38:09] <Ester_>	 anyways I have a meeting now
[17:38:20] <Ester_>	 can I do this myself?
[17:38:30] <bd808>	 Ester_: yes
[17:38:31] <Ester_>	 (delete the manifest file)
[17:38:35] <chicocvenancio>	 sure, but hopefully you don't see that error again
[17:38:40] <Ester_>	 oh ok
[17:38:47] <Ester_>	 ok guys thanks a lot
[17:38:52] <bd808>	 but you should only do it when webservice says that something is running but it is really not
[17:38:54] <Ester_>	 I'll ping you later or tomorrow
[17:39:26] <bd808>	 chicocvenancio: we need to start those docs about node things. :)
[17:39:57] <bd808>	 especially this use case of using node tools to generate content rather than running an active node webservice
[17:43:50] <Hauskatze>	 bd808: on stewardbots we have those service.{log,maniphest,access} files; if we get rid of them, do they recreate themselves or would the tool break?
[17:44:19] <bd808>	 Hauskatze: if you get rid of service.manifest the web interface will break
[17:44:45] <Hauskatze>	 ah oh
[17:44:45] <bd808>	 removing service.log while the webservice is running probably won't end well either
[17:44:57] <Hauskatze>	 so better to webservice stop first
[17:45:00] <bd808>	 yeah
[17:45:15] <Hauskatze>	 webservice restar alone would suffice?
[17:45:19] <Hauskatze>	 *restart
[17:45:39] <Hauskatze>	 I think last time I restarted it was using 'trusty'
[17:45:54] <bd808>	 no, sadly. To clean up old logs you really need to stop the webservice, move or delete the logs, and then start it back up
[17:45:54] <Hauskatze>	 anyway, just a question, I don't want to do nothing for now
[19:21:34] <mmecor>	 hi guys, i'm having a little problem with utf-8 encoding with my python script in tool forge
[19:21:49] <mmecor>	 does anyone know about it?
[19:22:20] <mmecor>	 the thing is: when I run my code in my toolforge account
[19:22:27] <mmecor>	 tool account
[19:22:44] <mmecor>	 "UnicodeEncodeError: 'ascii' codec can't encode character '\xe0' in position 41: ordinal not in range(128)"
[19:23:00] <mmecor>	 when i try to print a dictionary that contains arrays of strings
[19:23:04] <chicocvenancio>	 mmecor: what is tool? where are you running it?
[19:23:20] <mmecor>	 wcdo account
[19:23:54] <mmecor>	 it must be there: /data/project/wcdo/
[19:24:12] <chicocvenancio>	 where are you seeing the error?
[19:24:26] <jynus>	 mmecor: the errors says it, there is a mix of ascii and utf8 there- probably python2, which was horrible to handle those
[19:24:39] <mmecor>	 when i run it in my terminal just for quick feedback and when i sned the task in the output .err
[19:24:46] <mmecor>	 i'm using python3
[19:24:49] <jynus>	 oh
[19:24:55] <chicocvenancio>	 jynus: since its in print it can be just the locale set for the terminal
[19:24:56] <mmecor>	 and i put utf-8 at the header
[19:25:01] <mmecor>	 but the very weird thing is
[19:25:05] <mmecor>	 that in my laptop, when i run the script
[19:25:07] <mmecor>	 there is no problem
[19:25:15] <mmecor>	 in both cases i am using python3
[19:25:22] <jynus>	 echo $LANG ?
[19:25:25] <chicocvenancio>	 hence my question of where you are running this, grid/bastion/k8s have different locale setup
[19:25:38] <mmecor>	 chicocvenancio: how can i check?
[19:25:51] <chicocvenancio>	 what did you do to get the error?
[19:25:51] <mmecor>	 echo $LANG
[19:25:51] <mmecor>	 en_US.UTF-8
[19:26:59] <mmecor>	 i load a dataframe from .csv file, i check some rows, and i put them into arrays in a dictionary
[19:27:02] <mmecor>	 then i print hte dictionary
[19:27:10] <mmecor>	 this is what it is in my .py file
[19:27:32] <mmecor>	 https://pastebin.com/s3ANePca
[19:27:38] <chicocvenancio>	 mmecor: I see no k8s pods, no grid jobs and no python terminal  or processes opened for that tool in tools-bastion-03
[19:28:04] <mmecor>	 nope, wait a sec, i'll generate a job
[19:28:09] <mmecor>	 and you'll see the .err
[19:28:23] <chicocvenancio>	 wait
[19:28:37] <chicocvenancio>	 so you're sending it to the grid?
[19:28:47] <mmecor>	 i'm doing it right now
[19:28:59] <mmecor>	 1794535 0.30000 prova      tools.wcdo   r     02/28/2018 19:28:38 task@tools-exec-1432.tools.eqi     1 
[19:29:22] <chicocvenancio>	 this is a known bug, give me a minute to get the docs 
[19:30:01] <chicocvenancio>	 https://wikitech.wikimedia.org/wiki/Help:Toolforge/Grid#An_error_with_"ascii"_codepage,_"file_not_found",_or_UnicodeEncodeError
[19:30:39] <chicocvenancio>	 simplest workarround is to add `-v LC_ALL=en_US.UTF-8` to jsub
[19:31:47] <mmecor>	 the task is still in the grid
[19:31:56] <mmecor>	 with no output, that's strange
[19:32:11] <mmecor>	 qstat tells it is running
[19:33:11] <mmecor>	 chicocvenancio: i'll try to add that in the command, but in any case, it is also failing when i run the script in the termianl (not sending in the grid... that i know i sholdn't do normally, but i just do it for quick stuff, nto really heavy)
[19:35:30] <mmecor>	 ok, the prova.err came out from the grid job
[19:35:31] <mmecor>	 with the error
[19:36:10] <chicocvenancio>	 I can run the script in the bastion without error
[19:36:17] <chicocvenancio>	 how are you connecting to bastion?
[19:36:22] <chicocvenancio>	 mmecor: ^
[19:36:31] <mmecor>	 first i login with my username
[19:36:40] <mmecor>	 to @tools-login.wmflabs.org
[19:36:44] <mmecor>	 then i become the tool
[19:36:47] <chicocvenancio>	 if your terminal has ascii set this will be a problem with bastion login as well
[19:36:50] <mmecor>	 then i activate the venv
[19:36:58] <chicocvenancio>	 mmecor: what program do you use to login?
[19:37:11] <mmecor>	 chicocvenancio: but when i run it locally in my mac-python3 there is no problem
[19:37:17] <mmecor>	 so i guess it is not my terminal
[19:37:27] <chicocvenancio>	 oh, yes, its a mac
[19:37:42] <mmecor>	 mmm
[19:38:08] <mmecor>	 but in the mac works and not in the linux from the servers, shouldn't it be the other way around?
[19:38:20] <mmecor>	 when i "echo $LANG"
[19:38:33] <mmecor>	 in my mac, the response is nothing an empty line
[19:38:36] <mmecor>	 could this be the problem?
[19:38:46] <mmecor>	 i mean in a terminal i just opened
[19:39:14] <chicocvenancio>	 $LANG is the system locale, it might not match a particular terminal encoding
[19:39:22] <mmecor>	 with the tool account i'm logged in bastion at "(venv)tools.wcdo@tools-bastion-03"
[19:39:54] <mmecor>	 i have english as my primary language
[19:40:08] <chicocvenancio>	 have you tried sending it to the grid with `-v LC_ALL=en_US.UTF-8`
[19:40:49] <mmecor>	 i just did now
[19:41:02] <mmecor>	 it says running,.... let's see
[19:42:52] <chicocvenancio>	 I think https://stackoverflow.com/a/4606636/3930971 might be the answer for the mac terminal, but I'm not sure
[19:43:33] <mmecor>	 i have utf-8 enabled
[19:43:37] <mmecor>	 in the mac terminal
[19:44:11] <chicocvenancio>	 does a simple `echo à` work ?
[19:44:24] <mmecor>	 yup
[19:44:36] <mmecor>	 i run the script in my mac and it works
[19:44:40] <chicocvenancio>	 when logged in to toolforge as well?
[19:44:57] <mmecor>	 ahhh
[19:44:59] <mmecor>	 right
[19:45:01] <mmecor>	 not when logged in
[19:45:32] <chicocvenancio>	 does it work with your user, before the become wcdo?
[19:45:37] <mmecor>	 in any other terminal tab
[19:45:51] <mmecor>	 nope, not with my user 
[19:46:27] <mmecor>	 this quite weird, all this used to work one year and half ago. it was the same laptop and operating system
[19:49:42] <mmecor>	 oh
[19:49:52] <mmecor>	 chicocvenancio: that's bad, adding that line in the jsub didn't work
[19:50:02] <mmecor>	 the same error appeared in the .err
[19:57:52] <Hauskatze>	 !log tools.stewardbots New maintainer: rxy
[19:57:55] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.stewardbots/SAL
[20:00:03] <mmecor>	 chicocvenancio: the more i check things i can do about setting the terminal into utf-8, the less i understand. everything seems ok. in the different configurations there are the right lines
[20:00:47] <chicocvenancio>	 the problem is the terminal enconding
[20:00:57] <chicocvenancio>	 let me see what happens if I run this in k8s
[20:01:42] <mmecor>	 https://pastebin.com/JbiR6F3k
[20:01:49] <mmecor>	 this is what the locale says when i'm in toolforge
[20:02:58] <mmecor>	 so, apparently, there is utf-8, but then i cannot echo à
[20:03:27] <chicocvenancio>	 I can, in the same env
[20:04:10] <chicocvenancio>	 which is why I'm assuming the "not working in bastion" part to be related to the ssh in mac os
[20:05:40] <mmecor>	 chicocvenancio: i now found the way to print 'echo à'
[20:05:46] <mmecor>	 https://blog.remibergsma.com/2012/07/10/setting-locales-correctly-on-mac-osx-terminal-application/
[20:05:53] <mmecor>	 "Escape non-ASCII input"
[20:05:59] <mmecor>	 i checked this checkbox
[20:06:11] <mmecor>	 and it seems that i can print in the toolforge in bastion
[20:06:18] <mmecor>	 but the script gives the same error
[20:06:23] <chicocvenancio>	 in bastion?
[20:06:28] <mmecor>	 yes
[20:06:46] <chicocvenancio>	 `python3 ccc_selection.py`?
[20:06:51] <mmecor>	 yes
[20:07:56] <chicocvenancio>	 https://www.irccloud.com/pastebin/ScqU6TDk/expected_output
[20:08:06] <mmecor>	 ok, now i unchecked 'set locale environment variables on startup'
[20:08:27] <mmecor>	 chicocvenancio: that's what's supposed to be printed, yes
[20:08:39] <mmecor>	 i unchecked that and checked the 'escape non-ascii input'
[20:08:48] <mmecor>	 and i don't know exactly what is going on but works!
[20:09:05] <mmecor>	 thanks! :)
[20:09:09] <mmecor>	 now let's try the jsub
[20:09:09] <chicocvenancio>	 lol, thats the best kind of debugging
[20:10:06] <mmecor>	 thanks for help :)
[20:10:48] <mmecor>	 pointing the mac out has been key haha
[20:11:24] <chicocvenancio>	 where are you placing the `-v LC_ALL=en_US.UTF-8`?
[20:11:37] <mmecor>	 many problems are related to that... i guess that if i using ubuntu i would come across very similar problems. been in mac for 5 years and i can't remember how it was with ubuntu :)
[20:11:41] <mmecor>	 nope, chicocvenancio 
[20:11:45] <mmecor>	 i'm trying first without that line
[20:12:16] <chicocvenancio>	 that won't work, unless you set locale in the prova.sh script
[20:12:21] <mmecor>	 what i'm seeing is that although it says running, it usually takes longer now to start running
[20:12:45] <mmecor>	 chicocvenancio: how can i set the locale in the script?
[20:12:56] <chicocvenancio>	 but one thing I was doing wrong is to place -v LC_ALL=en_US.UTF-8 at the end of the jsub command
[20:13:08] <chicocvenancio>	 it must come before the prova.sh part
[20:13:14] <mmecor>	 ah, before
[20:13:17] <mmecor>	 i was putting it after too
[20:17:05] <chicocvenancio>	 the bash script prova.sh could have a `export LC_ALL=en_US.UTF-8` that would solve things for the grid as well
[20:17:20] <mmecor>	 chicocvenancio: inside the prova.sh?
[20:18:16] <chicocvenancio>	 yes, it has to be before the python line, ideally on the second line
[20:19:25] <chicocvenancio>	 ohhh, I forgot, but `export PYTHONIOENCODING=UTF-8`probably solves it for bastion as well
[20:19:36] <mmecor>	 export PYTHON... where?
[20:19:46] <mmecor>	 we have 3 options here i see
[20:19:57] <mmecor>	 first the param. in the jsub
[20:20:00] <chicocvenancio>	 although you might get weird characters in the terminal if you had not solved the macos config
[20:20:01] <mmecor>	 second in the bash script
[20:20:04] <mmecor>	 third...?
[20:21:04] <chicocvenancio>	 those are the options for the grid
[20:21:19] <bd808>	 there is a phabricator task somewhere with a really detailed analysis that zhuyifei1999_ did on python locale issues and the grid...
[20:21:31] <chicocvenancio>	 for the bastion problem you already solved it with the macos config
[20:21:56] <chicocvenancio>	 but export PYTHONIOENCODING=UTF-8 will let the script run even without that fix
[20:21:56] <mmecor>	 yes, bastion's solved
[20:22:09] <mmecor>	 chicocvenancio: where should i put this line?
[20:22:57] <zhuyifei1999_>	 this one or one of the merged tasks of https://phabricator.wikimedia.org/T60784 
[20:23:18] * zhuyifei1999_ goes back to doing Commons POTY stuff
[20:23:25] * chicocvenancio is again too slow
[20:23:54] <chicocvenancio>	 mmecor: never mind that line, as long as bastion doesn't give you unicodedecode errors again
[20:24:14] <mmecor>	 chicocvenancio: bastion is ok
[20:24:24] <chicocvenancio>	 then that line is not needed
[20:24:33] <mmecor>	 now i was looking for a fix in the script for the jobs
[20:25:04] <mmecor>	 although it seems that adding -v LC_ALL=en_US.UTF-8 before the .py works
[20:25:20] <chicocvenancio>	 either `-v LC_ALL=en_US.UTF-8` in the jsub command or `export LC_ALL=en_US.UTF-8`in the bash script
[20:25:56] <mmecor>	 export LC_ALL=en_US.UTF-8 in the bash? 
[20:25:57] <mmecor>	 trying that
[20:31:21] <mmecor>	 chicocvenancio: yay, it woked with the 'export LC_ALL' in the second line of the bash script
[20:31:28] <mmecor>	 so that's the easiest part
[20:31:33] <mmecor>	 the easiest way i mean :)
[20:31:43] <mmecor>	 thanks!
[20:31:59] <chicocvenancio>	 cool, you're welcom
[20:32:07] <mmecor>	 do you know how come it takes longer now to run the tasks?
[20:32:11] <mmecor>	 or was it just my perception?
[20:32:40] <mmecor>	 a silly script that in terminal runs in a sec, when i send it.. it takes around a min although it says running
[20:35:30] <chicocvenancio>	 grid jobs times are quite unpredictable, tbh I don't understand it fully
[20:35:52] <mmecor>	 ok :)
[20:43:01] <bd808>	 The grid scheduler has to do a lot of work to setup for running a job and may need to wait for resources (RAM, CPU) to be available. It is a system designed for consistency and efficient resource utilization rather than real-time responsiveness 
[20:46:24] <mmecor>	 i see
[22:06:02] <wikibugs>	 (03Draft1) 10Paladox: Send all notifications for refs/meta/config to #wikimedia-releng [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/415461
[22:06:04] <wikibugs>	 (03PS2) 10Paladox: Send all notifications for refs/meta/config to #wikimedia-releng [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/415461
[22:06:49] <wikibugs>	 (03CR) 10Chad: Send all notifications for refs/meta/config to #wikimedia-releng (031 comment) [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/415461 (owner: 10Paladox)
[22:07:31] <wikibugs>	 (03PS3) 10Paladox: Send all notifications for refs/meta/config to #wikimedia-releng [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/415461
[22:07:41] <wikibugs>	 (03CR) 10Paladox: Send all notifications for refs/meta/config to #wikimedia-releng (031 comment) [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/415461 (owner: 10Paladox)
[22:57:02] <zhuyifei1999_>	 !log quarry killed two IO-intensive query saves
[22:57:04] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Quarry/SAL
[22:57:17] <chicocvenancio>	 thanks zhuyifei1999_ 
[22:57:51] <zhuyifei1999_>	 yeah, when people run queries like https://quarry.wmflabs.org/query/25152 :(
[22:58:23] <Reedy>	 SELECT *;
[22:58:24] <Reedy>	 wat
[22:58:29] <Reedy>	 also, lol
[22:58:42] <wikibugs>	 (03CR) 10Chad: [C: 032] Send all notifications for refs/meta/config to #wikimedia-releng [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/415461 (owner: 10Paladox)
[22:58:54] <paladox>	 thanks no_justification :)
[22:59:09] <wikibugs>	 (03Merged) 10jenkins-bot: Send all notifications for refs/meta/config to #wikimedia-releng [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/415461 (owner: 10Paladox)
[22:59:19] <wikibugs>	 (03CR) 10jenkins-bot: Send all notifications for refs/meta/config to #wikimedia-releng [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/415461 (owner: 10Paladox)
[23:00:16] <paladox>	 Does it auto deploy?
[23:02:39] <no_justification>	 Docs say yes
[23:02:48] <paladox>	 Ok, thanks.
[23:02:50] * paladox tests
[23:02:59] <paladox>	 by editing some random config file (as a test)
[23:03:29] <wikibugs>	 (03Draft1) 10Paladox: Test: DO NOT MERGE [labs/tools/wikibugs2] (refs/meta/config) - 10https://gerrit.wikimedia.org/r/415476
[23:03:49] <wikibugs>	 (03Abandoned) 10Paladox: Test: DO NOT MERGE [labs/tools/wikibugs2] (refs/meta/config) - 10https://gerrit.wikimedia.org/r/415476 (owner: 10Paladox)