[03:27:04] Which is better for translators 1) [http://URL?q=$1 Google] or 2) {'url': 'http://URL?q=$1', 'label': 'Google'} [03:53:48] 6Labs, 6Operations: Manual creation of labs account - https://phabricator.wikimedia.org/T125172#2030209 (10Cobi) Any update on this? [05:48:37] PROBLEM - SSH on tools-exec-1220 is CRITICAL: CRITICAL - Socket timeout after 10 seconds [05:58:28] RECOVERY - SSH on tools-exec-1220 is OK: SSH OK - OpenSSH_6.6.1p1 Ubuntu-2ubuntu2~wmfprecise2 (protocol 2.0) [06:24:06] 6Labs, 10Tool-Labs, 10pywikibot-core: Tool Labs: shared Pywikibot code not available - https://phabricator.wikimedia.org/T127035#2030274 (10Ato_01) 3NEW [06:50:11] PROBLEM - Puppet failure on tools-exec-1410 is CRITICAL: CRITICAL: 71.43% of data above the critical threshold [0.0] [08:05:47] 6Labs, 10Tool-Labs, 10pywikibot-core: Tool Labs: shared Pywikibot code not available - https://phabricator.wikimedia.org/T127035#2030386 (10zhuyifei1999) [08:36:21] 6Labs, 10Tool-Labs, 10pywikibot-core: Tool Labs: shared Pywikibot code not available - https://phabricator.wikimedia.org/T127035#2030436 (10Alchimista) Seems the same situation of T126666. Could anyone delete the problematic files and re-run the update script again? [08:58:19] 6Labs, 10Wikimedia-Fundraising, 10Wikimedia-Fundraising-CiviCRM, 7Tracking: Create new labs project: fundraising-integration - https://phabricator.wikimedia.org/T88599#2030479 (10hashar) @danny_b Please stop spamming old bugs. Thank you. [09:21:37] hi all [11:21:11] Damn https://github.com/rohit-dua/BUB/issues/47 [11:27:15] 10MediaWiki-extensions-OpenStackManager, 10Notifications, 3Collaboration-Team-Current, 5Patch-For-Review: Update OpenStackManager notifications to new language and format - https://phabricator.wikimedia.org/T125691#2030778 (10SBisson) >>! In T125691#2029956, @Etonkovidova wrote: > In the the screenshot (p... [11:40:27] (03CR) 10Aklapper: "Multichill: What is the status of this? Should this get merged? Should this get abandoned? Thanks for your help cleaning up some rotting s" [labs/tools/multichill] - 10https://gerrit.wikimedia.org/r/223670 (owner: 10Multichill) [11:40:41] (03CR) 10Aklapper: "Multichill: What is the status of this? Should this get updated / merged? Should this get abandoned? Thanks for your help cleaning up some" [labs/tools/multichill] - 10https://gerrit.wikimedia.org/r/223671 (owner: 10Multichill) [11:56:45] 6Labs, 10Wikimedia-Fundraising, 10Wikimedia-Fundraising-CiviCRM: Create new labs project: fundraising-integration - https://phabricator.wikimedia.org/T88599#2030814 (10Danny_B) [11:57:31] 6Labs, 10Wikimedia-Fundraising, 10Wikimedia-Fundraising-CiviCRM: Create new labs project: fundraising-integration - https://phabricator.wikimedia.org/T88599#1015947 (10Danny_B) @hashar: Please stop re-adding incorrect tags. This is //not// a #tracking task. Thank you. [13:32:44] (03PS1) 10Nikerabbit: Remove ContentTranslations bugs from #mediawiki-i18n [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/270946 [14:07:45] (03CR) 10Aklapper: "How to proceed here? These 26 lines have been sitting here for five months without a review..." [labs/tools/extreg-wos] - 10https://gerrit.wikimedia.org/r/240422 (owner: 10Legoktm) [14:10:40] 6Labs: Bingbot scraping tools? - https://phabricator.wikimedia.org/T127066#2031170 (10Magnus) 3NEW [14:24:59] 6Labs: Bingbot scraping tools? - https://phabricator.wikimedia.org/T127066#2031242 (10Magnus) Now also seeing Yahoo Slurp: ``` 10.68.21.49 tools.wmflabs.org - [16/Feb/2016:14:12:55 +0000] "HEAD /glamtools/glamorous.php HTTP/1.1" 200 0 "-" "Mozilla/5.0 (compatible; Yahoo! Slurp; http://help.yahoo.com/help/us/ys... [14:32:56] Hello [14:33:52] Still trying to push tools.vocabulary-index to gerrit. [14:34:35] I can clone, fetch, pullto gerrit but i can't push [14:35:36] I breifly resume my situation there: https://wikitech.wikimedia.org/wiki/User:Youni_Verciti/Memo_cmd [14:36:26] first section is what i can do the second is what i can't and the error message [14:37:43] Change on 12wikitech.wikimedia.org a page Nova Resource:Tools/Access Request/Kharkiv07 was created, changed by Kharkiv07 link https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/Access_Request/Kharkiv07 edit summary: Created page with "{{Tools Access Request |Justification=Bot work |Completed=false |User Name=Kharkiv07 }}" [14:43:16] PROBLEM - ToolLabs Home Page on toollabs is CRITICAL: CRITICAL - Socket timeout after 10 seconds [14:43:53] ^ works for me [14:47:33] is toollabs home a tool itself? [14:48:10] RECOVERY - ToolLabs Home Page on toollabs is OK: HTTP OK: HTTP/1.1 200 OK - 786675 bytes in 3.122 second response time [14:48:59] 6Labs, 10Tool-Labs, 10pywikibot-core: Tool Labs: shared Pywikibot code not available - https://phabricator.wikimedia.org/T127035#2031358 (10Ato_01) It looks to be solved. [14:55:15] (03CR) 10Siebrand: [C: 031] Remove ContentTranslations bugs from #mediawiki-i18n [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/270946 (owner: 10Nikerabbit) [15:01:50] doh, is Hiera: namespace broken in wikitech? [15:01:51] https://wikitech.wikimedia.org/wiki/Hiera:Analytics [15:04:42] PHP Fatal error: Call to a member function getId() on a non-object in /srv/mediawiki/php-1.27.0-wmf.13/extensions/OpenStackManager/nova/OpenStackNovaRole.php on line 46 [15:08:24] 6Labs: Hiera: seems broken - https://phabricator.wikimedia.org/T127073#2031442 (10Ottomata) 3NEW [15:08:27] Gerrit don't let me push: [15:08:29] To push into this reference you need 'Push' rights. [15:09:12] or : ! [remote rejected] master -> master (prohibited by Gerrit) [15:09:36] or: fatal: Could not read from remote repository. [15:10:27] ottomata: let andrewbogott know, I imagine that's fallout from last week [15:11:06] thanks chasemp, CCing [15:11:22] 6Labs: Hiera: seems broken - https://phabricator.wikimedia.org/T127073#2031454 (10Ottomata) [15:12:28] ottomata: I fixed that exact bug, I don’t know why it’s back. I’ll look [15:12:38] danke! [15:12:47] it really looks like all my changes from the weekend have been rolled back somehow [15:13:06] probably a stray gravity wave [15:16:15] yeah man, all my patches are gone from the deployment host. They must’ve been dropped during something something mira something tin [15:16:26] bummer [15:27:22] my ssh connection to gerrit is ok from my station shell [15:27:55] but when i test from toolabs i have permission denied private key [15:30:44] 10Tool-Labs-tools-Other, 6TCB-Team, 7German-Community-Wishlist: "pageviews" tool link at bottom creates invalid HTML syntax (due to missing HTML encoding of characters like '&') - https://phabricator.wikimedia.org/T126975#2031521 (10MusikAnimal) Let's first try to reliably reproduce the issue. > As I said,... [15:54:53] andrewbogott: can I help in any way? most of the work I was going to do today requires the Hiera: thing [15:55:07] ottomata: it’ll all get fixed during scap [15:55:20] I just don’t want to poke around on the deploy servers right before a window [15:55:29] (and scap starts in 5 minutes) [15:56:15] oh that's upcoming cool [15:56:18] great thanks [16:01:09] I resolved my ssh access from tools.vocabulary-index but the git push continue failing on: [16:01:12] f8rx1vtM2OO5 [16:02:02] sorry, fails on : You are not allowed to perform this operation. [16:02:19] To push into this reference you need 'Push' rights. [16:02:39] User: y-verciti [16:02:59] ! [remote rejected] master -> master (prohibited by Gerrit) [16:02:59] hello :) [16:05:17] Youni: i am not sure what you are trying to do, but usually we don't do normal pushes to gerrit, we 'git review' (which is a push...but special) [16:05:18] https://www.mediawiki.org/wiki/Gerrit/Tutorial#Push_your_change_set_to_Gerrit [16:06:55] i'm trying to fill the gerrit repo with the files i stored on tool.vocabulary-index [16:07:26] Youni: typically one-time repo population like that still goes through git-review or if too large or impractical perms to allow direct push [16:07:41] you may need to ask in -releng as they usually handle perms for gerrit [16:13:14] ottomata: better? [16:13:33] git review asked my username i told "y-verciti" [16:13:55] than returned: fatal: bad revision 'remotes/gerrit/ssh://y-verciti@gerrit.wikimedia.org:29418/labs/tools/vocabulary-index' [16:16:12] chasemp: yes better thank you! [16:16:16] i mean [16:16:18] andrewbogott: thank you! [16:24:18] (03CR) 10Legoktm: [C: 032] Remove ContentTranslations bugs from #mediawiki-i18n [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/270946 (owner: 10Nikerabbit) [16:29:48] (03Merged) 10jenkins-bot: Remove ContentTranslations bugs from #mediawiki-i18n [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/270946 (owner: 10Nikerabbit) [16:30:34] !log tools.wikibugs Updated channels.yaml to: f4951fed10578ac63d80a6ec95e0927aaa90411e Remove ContentTranslations bugs from #mediawiki-i18n [16:30:37] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.wikibugs/SAL, Master [16:37:22] Youni: something is funky with that remote, its a bad revision/sha and is also a bad remote [16:37:24] not sure how you got that [16:45:49] 6Labs: Hiera: seems broken - https://phabricator.wikimedia.org/T127073#2031848 (10Andrew) 5Open>3Resolved a:3Andrew This was broken by a deployment screwup that was entirely my fault. It's working again now. [16:51:10] * yuvipanda waves at andrewbogott and chasemp [16:51:20] * andrewbogott waves back [16:52:10] glibc vulnerability! [16:52:53] (03PS1) 10BryanDavis: Ignore comments posted to Phabricator by Stashbot [labs/tools/wikibugs2] - 10https://gerrit.wikimedia.org/r/270989 [17:00:33] mornin' yuvipanda [17:00:56] hello chasemp [17:10:12] RECOVERY - Puppet failure on tools-exec-1410 is OK: OK: Less than 1.00% above the threshold [0.0] [17:11:06] hi guys, hi yuvipanda. i'd like to explain a problem to see if anyone could show me a different way or info i'm not taking into account. [17:11:06] [17:11:07] I'm trying to do some queries to the enwiki database to obtain data from revision table. after having tried several strategies using python mysqldb module, i am now retrieving rows into a file directly using a mysql command line output to a file with tsv. [17:11:08] my goal is to obtain revision by user and I do several queries in batches. so i use rev_user (see the queries: http://pastebin.com/saMKfW3v). first i take all the anon (rev_user=0) and since they are a lot, i divide them by the number of ip number they start with (or other address). then I segment other rev_id into different batches. [17:11:10] however, at the very first query for the anon users, mysql dies... do you think I can segment the batches better or do sth else to allocate more memory? [17:26:29] yuvipanda, I'm getting another PHP bug again on labs. [17:37:50] Cyberpower678: yuvipanda is out of office this week I think. You probably should file a bug in phabricator with the details of your issue to get help. [17:42:08] It kind of needs an immediate fix though [17:42:22] I'm getting a syntax error where non should exist. [17:42:26] bd808, ^ [17:43:05] that sounds unlikely actually. You mean that you have found a bug in the PHP interpreter? [17:43:13] Yes [17:43:17] Cyberpower678: pastebin your code? [17:43:21] Hey marmick [17:43:28] Because the exact code works on my machine [17:43:50] If you are doing session computation, it turns out that you don't have to partition the data by user. [17:43:57] PHP 5.3 vs PHP 5.something else? [17:44:06] But when I load it on labs, I get a syntax error [17:44:10] Since few users are actively editing at the same time, you can just have revisions sorted in chronological order. [17:44:18] So I recommend time-bounding your queries. [17:44:32] I was able to get a query for each year to complete for enwiki. [17:44:48] bd808, My computer uses 5.4 [17:44:48] E.g. 2001-2005, 2006, 2007, 2008, 2009, 2010, 2011, ... [17:44:54] I think labs uses 5.6 [17:45:03] halfak: yes, dividing by years may work, but i need data by users [17:45:10] marmick, how come? [17:45:11] Cyberpower678: You can check that easily enough with php --version... [17:45:35] i need to examine each of the edits for every user [17:46:05] Trusty is 5.5.9; precise is 5.3.3 [17:46:12] 5.5.9 [17:46:54] Cyberpower678: you still haven't revealed the actual syntax error [17:47:00] marmick, gotcha. So are you not performing the session computation -- or are you hoping to look at the results of session computation on a per-user basis? [17:47:17] yeh, it's a session computation on a per-user basis [17:47:27] bd808, I tend to forget to reveal the important details. [17:47:28] maybe it's too expensive computationally [17:47:30] :p [17:47:45] bd808, [17:47:46] PHP Parse error: syntax error, unexpected '(' in /data/project/cyberbot/bots/cyberbot-ii/IABot/Core.php on line 150 [17:47:46] PHP Stack trace: [17:47:46] PHP 1. {main}() /data/project/cyberbot/bots/cyberbot-ii/IABot/deadlink.php:0 [17:47:47] PHP 2. require_once() /data/project/cyberbot/bots/cyberbot-ii/IABot/deadlink.php:20 [17:47:48] halfak: http://pastebin.com/saMKfW3v this is what i generated as .sql [17:48:00] marmick, actually, that's exactly what I'm talking about. If you want session computation for all users, just run it across the whole chronological history of Wikipedia. [17:48:13] Not very computationally intensive once you get the revisions out of the DB in sorted order. [17:49:19] Cyberpower678: I asked you to pastebin your code earlier...we can't help unless you share the code. [17:49:24] Cyberpower678: ok. "unexpected '('" is one of those crappy errors that just tells you that PHP's parser got confused somewhere. Probably going to need a paste of the file or a link to version control to help [17:49:27] halfak: once the revisions are out in a file in sorted order, i need to put them in memory to do the sorting though. isn't that a lot of data in memory? [17:49:45] marmick, once they are in sorted order, you don't need to sort again. [17:50:04] legoktm, bd808: give me sec... Uploading new code to GitHub [17:50:10] Also, mysql as well as the "sort" utility will use file-based merge sort. [17:50:25] But I recommend mysql over the "sort" utility generally. [17:50:40] ** not really generally, so much as when you are pulling things out of the DB :) [17:51:53] i'm not sure if i get you, but i see sorting them by years and exporting out in a file as a middle step. my goal is to obtain the data sorted by user, so i have a line for each editor with the rev, timestamp, etc. [17:52:16] one line per editor? [17:52:43] nope, one per edit [17:53:39] Why do you need the data sorted by user? [17:53:43] legoktm, bd808: https://github.com/cyberpower678/Cyberbot_II/tree/test-code/IABot [17:53:49] halfak 2352 page_title1 timestampx, halfak 2352 page_title2, timestampx+1,...until today [17:53:50] All of IABot's code. [17:53:58] 10Tool-Labs-tools-Other, 6TCB-Team, 7German-Community-Wishlist: "pageviews" tool link at bottom creates invalid HTML syntax (due to missing HTML encoding of characters like '&') - https://phabricator.wikimedia.org/T126975#2032210 (10PatoLogic) you got me wrong: the link "works" but its syntax in the html cod... [17:54:34] because i want to examine the 'editing lifetime' of each editor (number of edits in periods, types of edits, etc.) [17:54:57] bd808, andrewbogott yuvipanda any clue about https://phabricator.wikimedia.org/T126580 ? should I keep retrying? [17:55:17] halfak: for instance, i want the number of edits in the first week, which is sth u were doing [17:55:41] Cyberpower678: https://3v4l.org/2vRv1 it runs fine in 5.5+, but not <=5.4. So run it on trusty, not precise. [17:55:47] subbu: I haven’t seen that, I don’t think — try again and see if it’s still happening? [17:56:05] marmick, will you be looking to have the "sessions" dataset sorted by user or the "session_revisions" dataset sorted by user? [17:56:40] legoktm, and how do I do that? [17:56:49] I can't control cyberbot-exec [17:56:57] At least I don't know how. [17:57:09] -l release=trusty, but idk about your current setup [17:57:18] you can probably re-write your code to be 5.3 compliant [17:57:59] * Cyberpower678 tries [17:58:54] (03CR) 10Legoktm: "I just haven't gotten around to deploying it yet." [labs/tools/extreg-wos] - 10https://gerrit.wikimedia.org/r/240422 (owner: 10Legoktm) [17:59:26] I just ssh'd into the exec node [17:59:43] That node does indeed run 5.3.10 [18:00:28] 6Labs, 10whatcanidoforwikimedia.org: Project wcidfwm (What can I do for wikimedia) - https://phabricator.wikimedia.org/T115092#2032261 (10Dzahn) >>>! In T115092#2023416, @Dzahn wrote: >> is this project also for http://whatcanidoforwikimediacommons.org/ ? > The project is hosted on wmpl toolserver and the sou... [18:07:01] Cyberpower678: restarting your job with "-l release=trusty" in the command will move you to an exec node with 5.5 [18:08:54] andrewbogott, ok .. let me see if it is a 3rd time lucky scenario. [18:12:59] andrewbogott, [subbu@earth ~] ssh ssastry@nlwiki.wikitextexp.eqiad.wmflabs [18:12:59] Permission denied (publickey). [18:13:17] ok, looking... [18:26:47] subbu: I think I see how to fix this, but I’ve no idea why it’s happening in your project in particular. [18:26:56] It will take a while to roll out the fix; I’ll let you know when it’s ready [18:27:32] andrewbogott: need help? [18:27:52] chasemp: nope, no big deal. The bug is in the base images which I need to rebuild to pick up security fixes anyway [18:27:55] andrewbogott, ok. thanks. [18:28:02] but I’m going to delay until those security fixes actually exist... [18:29:10] bd808, I use the cyberbot exec node [18:29:18] It's node specifically for me. [18:41:37] andrewbogott, chasemp: no sign of Ubuntu updates yet [18:41:56] moritzm: ok. I’m going to rebuild the debian base image, will continue to ignore trusty [18:42:15] ok, I'll start poking ubuntu security people after dinner [18:42:29] moritzm: mention the cloud archive while you’re at it :) [18:42:51] I'm afraid that's a different department at Canonical :-) [18:45:12] yet another case of Debian being faster than Ubuntu :) [18:45:45] Change on 12wikitech.wikimedia.org a page Nova Resource:Tools/Access Request/Kharkiv07 was modified, changed by Tim Landscheidt link https://wikitech.wikimedia.org/w/index.php?diff=310271 edit summary: [18:48:51] bd808, there we go [18:48:59] Bot is up again [18:51:30] moritzm: the ‘security’ people don’t care about security if it’s in a different repo? [19:14:08] 10Tool-Labs-tools-Other, 6TCB-Team, 7German-Community-Wishlist: "pageviews" tool link at bottom creates invalid HTML syntax (due to missing HTML encoding of characters like '&') - https://phabricator.wikimedia.org/T126975#2032579 (10MusikAnimal) I'm sorry I do not follow... I inspected the source of https://... [19:30:46] 6Labs, 10Tool-Labs: Copy user database to c1 - https://phabricator.wikimedia.org/T127107#2032628 (10Superyetkin) 3NEW [19:47:58] subbu: I’m still not sure why it was breaking for you, but try again now? I updated the jessie base image [19:48:05] ok .. [19:48:43] you mean spin up a new instance right? not try to login. [19:50:54] right, a new instance [19:51:39] andrewbogott, a new error this time around [19:51:40] [subbu@earth ~] ssh ssastry@nlwiki.wikitextexp.eqiad.wmflabs [19:51:41] channel 0: open failed: administratively prohibited: open failed [19:51:41] ssh_exchange_identification: Connection closed by remote host [19:52:08] ah, looks like stale keys. [19:52:10] subbu: that means the domain name is incorrect [19:52:17] (or some DNS issue) [19:52:22] i purged it and retried. [19:52:27] and back to [subbu@earth ~] ssh ssastry@nlwiki.wikitextexp.eqiad.wmflabs [19:52:27] Permission denied (publickey). [19:54:11] subbu: has it finished its first puppet run? (see the instance console) [19:56:25] https://gist.githubusercontent.com/subbuss/f2878e945e0f6575a780/raw/1b64891699d1b6786c8199b2331ad637b83522bd/gistfile1.txt [19:58:06] subbu: Ah, there's an issue with the activated puppet classes [19:59:00] subbu: it's better to wait with adding classes to an instance until after it has fully booted and finished its first puppet run [19:59:11] i haven't added any classes. [19:59:18] huh. [19:59:29] i was waiting for it to boot up before adding them. [19:59:39] ok [19:59:54] subbu: did you create a new instance? If so where is it? [20:00:32] ok, so the hiera config for the project has this single line "---" so .. maybe that is the issue? [20:00:37] https://wikitech.wikimedia.org/w/index.php?title=Hiera:Wikitextexp&action=edit [20:00:55] and andrewbogott https://wikitech.wikimedia.org/wiki/Nova_Resource:Nlwiki.wikitextexp.eqiad.wmflabs is the instance [20:01:05] * subbu is hungry [20:01:06] not sure, that might just mean 'empty' (I'm never quite sure with yaml) [20:01:49] * valhallasw`cloud is at a loss [20:02:31] * subbu is going to eat lunch [20:03:35] subbu|lunch: I don’t know, but I’m going to remove that hiera page [20:04:24] --- is just the yaml document starter, so this is indeed empty [20:12:21] subbu|lunch: I’ve created two new instances in that project without trouble. I don’t know why — would break things but it must’ve been; that would explain the behavior you got before as well. [20:18:43] 6Labs, 10Labs-Infrastructure: Make instance creation failures more verbose - https://phabricator.wikimedia.org/T47483#2032855 (10scfc) [20:18:45] 6Labs, 10Labs-Infrastructure, 10MediaWiki-extensions-OpenStackManager: On create instance failure, give link to quotas page and explain the error - https://phabricator.wikimedia.org/T121368#2032854 (10scfc) [20:24:36] andrewbogott, great. i'm going to remove your instance and create a new one with the name i want. [20:24:51] ok [20:26:07] 10MediaWiki-extensions-OpenStackManager: openstackmanager-configureproject-serviceuserinfo overly escaped - https://phabricator.wikimedia.org/T60519#2032895 (10scfc) Perhaps due to the Keystone backend move, I as a project admin no longer see the "Configure project" (?) action at https://wikitech.wikimedia.org/w... [20:29:00] andrewbogott, oh no ... maybe it hates me .. or maybe it doesn't like the nlwiki name .. or maybe it wants you to create instances :-) but same permission denied problem again. :( [20:29:10] i could login to the thisworks.* vm you had created. [20:30:23] i could login now .. maybe this time around .. i had to wait. [20:30:24] so, \o/ [20:30:29] subbu: works for me [20:30:41] yes, i am in too. [20:31:15] maybe i was just impatient and had to wait for the puppet run to finish .. anyway, i am still learning all the rituals .. i should check console output to veirfy puppet run is complete before trying to login. [20:31:36] andrewbogott: the ubuntu security team is only responsible for the main archive, from what I can tell the cloudarchive is maintained by their openstack people [20:31:59] 10MediaWiki-extensions-OpenStackManager: openstackmanager-configureproject-serviceuserinfo overly escaped - https://phabricator.wikimedia.org/T60519#2032921 (10Andrew) Yeah, I've removed the 'configure project' gui entirely. That means this bug is moot, right? [20:33:21] moritzm: and I take it the ubuntu security team hasn’t done their job yet either? [20:34:55] so, after i use the configure link and select 2 puppet groups for the vm, i suppose i need to run the puppet agent command on the vm? [20:34:59] doesn't seem so, they focused on a libreoffice update for now. priorities :-) [20:35:12] http://www.ubuntu.com/usn/usn-2899-1/ [20:35:27] subbu: correct, or wait for puppet to run itself [20:35:47] ok. [20:40:35] bd808, andrewbogott should i have spun up an ubuntu vm for vagrant to work? because i see this error from puppet .. [20:40:47] Error: Could not retrieve catalog from remote server: Error 400 on SERVER: OS ubuntu >= trusty required. at /etc/puppet/modules/vagrant/manifests/init.pp:13 on node nlwiki.wikitextexp.eqiad.wmflabs [20:41:16] I don’t really know the state of labs+vagrant [20:41:39] ok. i'll wait for bd808 to be around. [20:46:36] subbu: vagrant doesn't work on jessie, needs trusty. [20:46:52] sorry :( [20:47:04] alright. not a problem. [21:01:38] andrewbogott: filed T127113 for the cloud archive problem [21:01:38] T127113: Some labvirt systems use qemu from "cloud archive" (which doesn't get security support) - https://phabricator.wikimedia.org/T127113 [21:05:53] 6Labs, 10Tool-Labs, 10pywikibot-core: Tool Labs: shared Pywikibot code not available - https://phabricator.wikimedia.org/T127035#2033034 (10Ato_01) 5Open>3Resolved [21:15:34] 10Tool-Labs-tools-Other, 6TCB-Team, 7German-Community-Wishlist: "pageviews" tool link at bottom creates invalid HTML syntax (due to missing HTML encoding of characters like '&') - https://phabricator.wikimedia.org/T126975#2033075 (10Raymond) Fixed with https://de.wikipedia.org/w/index.php?title=MediaWiki:Wik... [21:19:55] 10Tool-Labs-tools-Other, 6TCB-Team, 7German-Community-Wishlist: "pageviews" tool link at bottom creates invalid HTML syntax (due to missing HTML encoding of characters like '&') - https://phabricator.wikimedia.org/T126975#2033110 (10Raymond) >>! In T126975#2028686, @Aklapper wrote: > (And personally I'm puz... [21:21:57] moritzm: did you log a but with the cloud archive? Or just with me? [21:22:01] *bug [21:23:02] tools-bastion1002 having an issue is me, sorry addresing it now [21:33:29] !log tools reboot of bastion-1002 [21:33:33] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL, Master [21:44:33] bd808: I just created a new project on Tool Labs, but it doesn't have a replica.my.cnf file. Any idea why that might be? [21:45:32] yuvipanda: ^ [21:45:41] kaldari: I think it can take until the next puppet run [21:45:52] bd808: how long does that take? [21:45:54] so ~20 minutes or so [21:46:13] cool, I'll update the docs at https://wikitech.wikimedia.org/wiki/Help:Tool_Labs/Database#Creating_new_databases to mention that [21:47:18] is that the mysql perms file dropped off per tool? [21:48:31] chasemp: yes [21:48:55] I have a suspicion labdb1002 being down may be affecting it [21:49:02] 6Labs, 5Patch-For-Review: Instances broken on initial provision with dns setup issues - https://phabricator.wikimedia.org/T126580#2033180 (10ssastry) [21:49:12] kaldari: if you don't see it in the next bit log a ticket and toss it my way [21:49:14] please [21:49:48] ok [22:13:41] chasemp: kaldari bd808 it should take about 5mins [22:13:48] it's on a daemon that waits 5mins, not puppet [22:14:01] yuvipanda: I'm pretty sure the service is crashing on lack of reaching labsdb1002 [22:14:16] but I'm tied up talk though this fr stuff [22:14:18] ah [22:14:23] chasemp: yeah I can take care of that [22:15:14] andrewbogott: labsdb1002 is out of commision for the foreseeable future, right? [22:15:31] yuvipanda: until disk replacement at least [22:15:35] yuvipanda: until we get it a new disk. That’s probably a Chris question [22:16:23] chasemp, yuvipanda, bd808: replica.my.cnf file never did show up [22:16:57] kaldari: yeah I think it's fallout from other issues, may be a bit but we'll sort it out and ping you [22:17:56] I'm going to take labstore1002 out of rotation [22:17:59] err [22:18:01] labsdb [22:18:53] right :) [22:27:52] bd808, please set https://phabricator.wikimedia.org/V7 to public [22:28:25] Krenair: it isn't public? [22:28:33] nope, all users [22:28:58] Krenair: fixed [22:29:09] thanks! [22:29:15] wish upstream would give that option a more helpful name [22:29:20] like 'logged-in users only' [22:31:51] registered users [22:31:56] idk [22:33:36] Krenair: I think I was trying to ensure that anons couldn't vote when I set that perm [22:33:46] but maybe anons can't vote anyhow? [22:33:47] I don't think phab would permit that [22:33:56] they can't do stuff like comment anywhere [22:34:55] anons never can take a action at phab [22:34:58] they can only view [22:35:21] andrewbogott: what do you mean? whether I raised this with Ubuntu? will do that [22:35:31] updates for Ubuntu are out now, BTW [22:35:43] kaldari: bd808 I just merged a change, should be back up shortly. I'll poke you when deplyoed [22:35:46] http://www.ubuntu.com/usn/usn-2900-1/ [22:35:53] moritzm: oh, great! I’ll build new trusty and precise images [22:36:59] actually... despite it being marked as public now, I can't view it while logged out? [22:37:00] weird [22:37:05] kaldari: bd808 done, should be there now [22:37:06] oh well, upstream's problem [22:38:24] 6Labs, 10Tool-Labs: Copy user database to c1 - https://phabricator.wikimedia.org/T127107#2033392 (10scfc) a:3jcrespo [22:39:49] chasemp: just a fyi, the labsdb user creation stuff is sorted now [22:40:02] thanks man [22:40:29] yuvipanda: fyi back at you :) we have a failed disk in labsdb1002 and on labstore1001 [22:40:34] tickets for both out to chris [22:41:20] chasemp: woah, labstore1001? [22:41:30] is it one of the internal disks or is it the NFS disk? [22:41:37] nfs [22:41:43] ouch [22:41:54] so we've degraded NFS performance now, I take it? [22:41:58] for now it's no biggie, obv further failures mean issues [22:42:27] no just reduced redundancy I believe as one of the RAID0's under the RAID10 is now inadequate [22:42:38] iiuc [22:44:08] right [22:44:13] ok [22:44:27] yuvipanda: it's there now. Thanks! [22:44:43] 6Labs, 10Tool-Labs, 10DBA: Copy user database to c1 - https://phabricator.wikimedia.org/T127107#2033426 (10scfc) [22:44:54] kaldari: np! thanks for reporting [22:46:18] 6Labs, 10Tool-Labs, 10Wikimedia-Site-Requests, 10wikitech.wikimedia.org, and 2 others: Create a wiki documentation page for each tool - https://phabricator.wikimedia.org/T122865#2033435 (10Legoktm) I supported the poll since I agree with adding the namespaces, but don't think it sets a good precedent of us... [23:04:05] I really should have timed how long it takes to set up a new mw-v instance on labs [23:04:25] > ==> default: Notice: Finished catalog run in 691.27 seconds [23:04:29] heh [23:05:18] bd808: did we ever figure out a solution for mwv not starting up after restarts? [23:05:24] because it's gonna happen thursday :) [23:07:52] yuvipanda: ugh. No I never did anything about that [23:08:09] let me file a bug and we can see if there's an easy way out [23:09:04] yuvipanda: we could make an upstart script for it pretty easily [23:09:05] Hmm, I set up http://bf-wmpageview.wmflabs.org/ as a proxy, curl works fine while logged into the server, but proxy 504s... [23:09:16] legoktm: project firewall [23:09:20] blegh [23:09:35] * legoktm wonders how the other mw instance in that project works then [23:09:44] oh, I have to whitelist 8080 [23:09:49] yup [23:09:59] it's in the instructions ;) [23:10:35] ...and now it works! [23:13:07] thanks bd808 :) [23:13:22] legoktm: yw. We've all done that one [23:15:48] 6Labs, 10MediaWiki-Vagrant: mediawiki-vagrant on labs requires manual action after instance restart - https://phabricator.wikimedia.org/T127129#2033512 (10yuvipanda) 3NEW [23:15:59] bd808: ^ [23:16:23] bd808: yeah, upstart script sounds nice :D but then would it start it as soon as you apply the role? [23:16:25] yuvipanda: thanks. [23:16:50] I'm trying to figure out a guard right now to only start if there is an existing container [23:17:50] * yuvipanda nods [23:19:08] 6Labs, 10Beta-Cluster-Infrastructure: Completely remove Beta Cluster dependency on NFS - https://phabricator.wikimedia.org/T102953#2033534 (10yuvipanda) [23:19:10] 6Labs, 10Beta-Cluster-Infrastructure: Disable /data/project for instances in deployment-prep that do not need it - https://phabricator.wikimedia.org/T125624#2033533 (10yuvipanda) [23:19:17] 6Labs, 10Beta-Cluster-Infrastructure: Disable /data/project for instances in deployment-prep that do not need it - https://phabricator.wikimedia.org/T125624#1992944 (10yuvipanda) This is unrelated to swift, since this is only removing the mounts from instances that do not use them at all and is effectively a n... [23:21:51] 6Labs, 10MediaWiki-Vagrant: mediawiki-vagrant on labs requires manual action after instance restart - https://phabricator.wikimedia.org/T127129#2033544 (10bd808) Something like [[http://stackoverflow.com/a/24592073/8171|this upstart script from Stack Overflow]] is probably the right way to go. We'd want to add... [23:24:31] 6Labs, 10Beta-Cluster-Infrastructure: Disable /data/project for instances in deployment-prep that do not need it - https://phabricator.wikimedia.org/T125624#2033562 (10yuvipanda) https://wikitech.wikimedia.org/w/index.php?title=Hiera%3ADeployment-prep&type=revision&diff=311783&oldid=311781 turns off NFS for ne... [23:26:56] 6Labs, 10MediaWiki-Vagrant: mediawiki-vagrant on labs requires manual action after instance restart - https://phabricator.wikimedia.org/T127129#2033580 (10bd808) p:5Triage>3High [23:27:06] 6Labs, 10MediaWiki-Vagrant, 15User-bd808: mediawiki-vagrant on labs requires manual action after instance restart - https://phabricator.wikimedia.org/T127129#2033583 (10bd808) a:3bd808 [23:27:49] 6Labs, 10Beta-Cluster-Infrastructure: Disable /data/project for instances in deployment-prep that do not need it - https://phabricator.wikimedia.org/T125624#2033587 (10yuvipanda) I've set `mount_nfs: true` on all the instances I listed earlier. These are all still no-ops, since I actually need to unmount them... [23:28:12] chasemp: andrewbogott fyi, I'm slowly getting rid of NFS from about 20+ instances on deployment-prep now [23:30:10] 52 nodes, to be exact [23:32:07] sweet [23:32:27] yuvipanda: what's involved in that exactly. editing teh yaml file and mount -a? [23:32:53] change hiera to exclude nfs, puppet run, mount -a, edit yaml [23:34:59] 6Labs, 10Beta-Cluster-Infrastructure: Disable /data/project for instances in deployment-prep that do not need it - https://phabricator.wikimedia.org/T125624#2033610 (10yuvipanda) I've unmounted /data/project on deployment-poolcounter01 and verified that puppet doesn't bring it back. I'll unmount it on all the... [23:37:09] 10MediaWiki-extensions-OpenStackManager: openstackmanager-configureproject-serviceuserinfo overly escaped - https://phabricator.wikimedia.org/T60519#2033624 (10Krenair) 5Open>3Resolved a:3Krenair Well, the message was deleted from en.json... It should've been deleted from the other i18n files too, will ask... [23:37:16] 10MediaWiki-extensions-OpenStackManager: openstackmanager-configureproject-serviceuserinfo overly escaped - https://phabricator.wikimedia.org/T60519#2033629 (10Krenair) a:5Krenair>3Andrew [23:41:48] yuvipanda: welcome back [23:42:08] I had a question while you were gone [23:58:03] 10MediaWiki-extensions-OpenStackManager, 10Notifications, 3Collaboration-Team-Current, 5Patch-For-Review, 5WMF-deploy-2016-02-23_(1.27.0-wmf.15): Update OpenStackManager notifications to new language and format - https://phabricator.wikimedia.org/T125691#2033706 (10jmatazzoni) @SBisson is right. No bold... [23:58:10] 10MediaWiki-extensions-OpenStackManager, 10Notifications, 3Collaboration-Team-Current, 5Patch-For-Review, 5WMF-deploy-2016-02-23_(1.27.0-wmf.15): Update OpenStackManager notifications to new language and format - https://phabricator.wikimedia.org/T125691#2033707 (10jmatazzoni) 5Open>3Resolved