[00:23:33] PROBLEM - RAID on searchidx1001 is CRITICAL: CHECK_NRPE: Socket timeout after 10 seconds. [00:24:32] RECOVERY - RAID on searchidx1001 is OK: OK: State is Optimal, checked 4 logical device(s) [00:26:02] PROBLEM - Puppet freshness on sq41 is CRITICAL: No successful Puppet run in the last 10 hours [00:43:18] (03PS1) 10Bsitu: Update the address in email footer [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76668 [01:01:22] @replag s3 [01:01:22] Krinkle: [s3 (DEFAULT)] db1019: 0s, db1003: 0s, db1010: 0s, db1035: 0s [01:01:32] @replage centralauth [01:01:32] Krinkle: [centralauth: s7] db1041: 0s, db1007: 0s, db1024: 0s, db1028: 0s [01:24:54] (03CR) 10Hashar: [C: 031] Disable UploadBlacklist [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76229 (owner: 10Reedy) [01:25:33] (03CR) 10Ottomata: "(1 comment)" [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 (owner: 10Ottomata) [01:36:14] TimStarling: I went to go amend the ptwiki CAPTCHA changeset to include a hard expiry and you'd already merged it. :-) [01:36:20] I think that was a good call. [01:44:47] Replied to you on your "break the internet" bug as well. :-) [01:44:53] (03PS1) 10Dzahn: rename wikivoyages to wikivoyage, fix timestamps on index page [operations/debs/wikistats] - 10https://gerrit.wikimedia.org/r/76673 [01:53:41] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [01:54:31] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.133 second response time [01:58:17] (03PS1) 10Dzahn: renable updates via file_get_contents() [operations/debs/wikistats] - 10https://gerrit.wikimedia.org/r/76674 [02:17:40] (03PS1) 10Dzahn: add 1.22wmf12 to good versions [operations/debs/wikistats] - 10https://gerrit.wikimedia.org/r/76675 [02:17:55] !log LocalisationUpdate completed (1.22wmf12) at Tue Jul 30 02:17:54 UTC 2013 [02:18:06] Logged the message, Master [02:33:31] !log LocalisationUpdate completed (1.22wmf11) at Tue Jul 30 02:33:27 UTC 2013 [02:33:41] Logged the message, Master [02:43:53] (03PS1) 10Tim Starling: Add accounts::tstarling to Locke [operations/puppet] - 10https://gerrit.wikimedia.org/r/76677 [02:45:42] (03CR) 10Tim Starling: [C: 032 V: 032] Add accounts::tstarling to Locke [operations/puppet] - 10https://gerrit.wikimedia.org/r/76677 (owner: 10Tim Starling) [02:52:52] !log LocalisationUpdate ResourceLoader cache refresh completed at Tue Jul 30 02:52:51 UTC 2013 [02:53:02] Logged the message, Master [02:57:09] !log racked and plugged back pappas [02:57:10] RECOVERY - Host pappas is UP: PING OK - Packet loss = 0%, RTA = 26.70 ms [02:57:19] Logged the message, Master [02:57:51] mutante: What's $goodversions used for in wikistats? [02:58:21] https://github.com/wikimedia/operations-debs-wikistats/search?q=goodversions [02:58:22] https://github.com/wikimedia/operations-debs-wikistats/search?q=version_color [02:58:25] can't find anything [02:59:10] <^demon> Yo Krinkle [02:59:27] hi [02:59:53] <^demon> Any chance you remember that site that tries to keep track of installed MW extensions? I know I saw it recently-ish but I can't find it in my history or anywhere else. [03:00:22] I know which you mean [03:00:26] with the bees around the mediawiki logo [03:00:53] <^demon> I can't find the dang url anywhere. [03:02:36] ^demon: http://wikiapiary.com/ [03:02:52] ^demon: found in gmail by looking for Special:ConfirmEmail, I knew I created an account htere [03:03:02] <^demon> Yes! That's it! [03:03:04] they got a new skin since I last visited [03:03:04] <^demon> Thank youuuuuu [03:03:06] yw [03:07:23] <^demon> Hahah, this site totally proved my point. [03:07:58] ^demon: context? [03:08:01] <^demon> Earlier today I heard claim about 3rd parties possibly using CodeReview. http://wikiapiary.com/wiki/Extension:CodeReview - would claim otherwise. [03:13:40] indeed, the one use of it isn't active: http://marknet.myhome.cx/science-wiki/index.php/%E7%89%B9%E5%88%A5:%E3%82%B3%E3%83%BC%E3%83%89 [03:13:43] no repos set up [03:15:13] <^demon> :) [03:32:04] (03PS1) 10Tim Starling: Puppetize /home/tstarling/.bashrc [operations/puppet] - 10https://gerrit.wikimedia.org/r/76678 [03:34:10] (03CR) 10Tim Starling: "Please review." [operations/puppet] - 10https://gerrit.wikimedia.org/r/76678 (owner: 10Tim Starling) [03:38:45] (03CR) 10MZMcBride: "(1 comment)" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76678 (owner: 10Tim Starling) [03:43:18] (03CR) 10Demon: "(2 comments)" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76678 (owner: 10Tim Starling) [03:48:18] (03PS1) 10Ryan Lane: Remediate BEAST on apache services [operations/puppet] - 10https://gerrit.wikimedia.org/r/76679 [03:49:18] (03CR) 10Ryan Lane: [C: 032] Remediate BEAST on apache services [operations/puppet] - 10https://gerrit.wikimedia.org/r/76679 (owner: 10Ryan Lane) [03:54:02] PROBLEM - Puppet freshness on lvs1005 is CRITICAL: No successful Puppet run in the last 10 hours [03:54:02] PROBLEM - Puppet freshness on lvs1004 is CRITICAL: No successful Puppet run in the last 10 hours [03:54:02] PROBLEM - Puppet freshness on lvs1006 is CRITICAL: No successful Puppet run in the last 10 hours [03:54:02] PROBLEM - Puppet freshness on virt3 is CRITICAL: No successful Puppet run in the last 10 hours [03:54:02] PROBLEM - Puppet freshness on virt1 is CRITICAL: No successful Puppet run in the last 10 hours [03:54:03] PROBLEM - Puppet freshness on virt4 is CRITICAL: No successful Puppet run in the last 10 hours [03:54:06] (03CR) 10Tim Starling: "(2 comments)" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76678 (owner: 10Tim Starling) [03:56:59] CVS! so retro. [03:57:26] Just in case it makes a comeback. [03:58:13] Heh, Tim is attempting to use Cunningham's Law. [04:03:14] (03PS1) 10Demon: More BEAST remediation [operations/puppet] - 10https://gerrit.wikimedia.org/r/76680 [04:07:20] (03CR) 10Ryan Lane: [C: 032] More BEAST remediation [operations/puppet] - 10https://gerrit.wikimedia.org/r/76680 (owner: 10Demon) [04:36:55] (03PS1) 10Demon: Further BEAST remediation [operations/puppet] - 10https://gerrit.wikimedia.org/r/76681 [04:56:25] PROBLEM - Puppet freshness on ssl1004 is CRITICAL: No successful Puppet run in the last 10 hours [05:09:09] (03CR) 10Ryan Lane: [C: 032] Further BEAST remediation [operations/puppet] - 10https://gerrit.wikimedia.org/r/76681 (owner: 10Demon) [05:21:43] PROBLEM - RAID on searchidx1001 is CRITICAL: CHECK_NRPE: Socket timeout after 10 seconds. [05:26:43] RECOVERY - RAID on searchidx1001 is OK: OK: State is Optimal, checked 4 logical device(s) [05:51:48] PROBLEM - Puppet freshness on mchenry is CRITICAL: No successful Puppet run in the last 10 hours [05:53:48] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [05:55:38] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.129 second response time [06:23:47] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [06:24:37] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 1.393 second response time [06:33:50] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [06:34:37] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.134 second response time [07:03:09] PROBLEM - Puppet freshness on holmium is CRITICAL: No successful Puppet run in the last 10 hours [07:25:30] PROBLEM - SSH on lvs6 is CRITICAL: Server answer: [07:26:30] RECOVERY - SSH on lvs6 is OK: SSH OK - OpenSSH_5.9p1 Debian-5ubuntu1.1 (protocol 2.0) [08:06:31] (03PS1) 10Nikerabbit: Switch translation memory from vanadium to zinc [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76690 [08:07:17] (03CR) 10Nikerabbit: [C: 032] Switch translation memory from vanadium to zinc [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76690 (owner: 10Nikerabbit) [08:07:25] (03Merged) 10jenkins-bot: Switch translation memory from vanadium to zinc [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76690 (owner: 10Nikerabbit) [08:11:38] !log nikerabbit synchronized wmf-config/CommonSettings.php 'Translation memory to zinc from vanadium' [08:11:48] Logged the message, Master [08:12:03] PROBLEM - Puppet freshness on erzurumi is CRITICAL: No successful Puppet run in the last 10 hours [08:29:11] why do I get errors when running maintenance script on terbium: DB connection error: No working slave server: Unknown error (10.64.32.18) ? [08:51:02] PROBLEM - Puppet freshness on manutius is CRITICAL: No successful Puppet run in the last 10 hours [08:55:26] !log Translation memory migrated from vanadium to zinc and repopulated [08:55:36] Logged the message, Master [09:21:26] (03Abandoned) 10Hashar: ** WIP ** role class for puppet agents ** WIP ** [operations/puppet] - 10https://gerrit.wikimedia.org/r/64031 (owner: 10Hashar) [09:29:21] (03PS1) 10Nikerabbit: Use application/font-woff for woff files [operations/puppet] - 10https://gerrit.wikimedia.org/r/76698 [09:29:22] (03PS6) 10Akosiaris: Introducing bacula module [operations/puppet] - 10https://gerrit.wikimedia.org/r/70840 [09:38:34] (03CR) 10KartikMistry: [C: 031] "LGTM!" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76698 (owner: 10Nikerabbit) [09:39:49] heh, debian people everywhere [09:41:08] hey paravoid :-] [09:41:15] I packaged a bunch of python modules last week [09:41:39] and I was wondering how to get them inside apt or if it worth writing a small doc to get packages in apt.wm.o [09:42:02] my concern is that if I provide you the .deb package, there is nothing preventing me from sneaking in a root kit or some nasty code :D [09:42:17] so I guess you rebuild the package and would like to document the process [09:42:32] yes, I rebuild debs before putting them into apt [09:42:45] well not always [09:43:28] ceph and cloudera's debs we don't. But as a general rule we should. It also helps spot any missing bugs... [09:43:42] missed* [09:52:55] we have some trust anchors, yes [09:53:09] at least ubuntu, debian, ceph, cloudera, jenkins [09:56:11] great [09:56:34] and yesterday I fought with cowbuilder to provide it additional dependencies [09:56:53] turns out you can pass an additional source.lists deb line when building with cowbuilder [09:57:01] it is only available when refreshing the image :-D [10:00:32] PROBLEM - SSH on pdf3 is CRITICAL: CRITICAL - Socket timeout after 10 seconds [10:01:42] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [10:03:32] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.130 second response time [10:04:12] (03PS1) 10Mark Bergsma: Don't use pass for all requests with a session cookie [operations/puppet] - 10https://gerrit.wikimedia.org/r/76700 [10:04:22] RECOVERY - SSH on pdf3 is OK: SSH OK - OpenSSH_4.7p1 Debian-8ubuntu3 (protocol 2.0) [10:05:55] (03PS2) 10Mark Bergsma: Don't use pass for all requests with a session cookie [operations/puppet] - 10https://gerrit.wikimedia.org/r/76700 [10:12:38] (03CR) 10Mark Bergsma: [C: 032] Don't use pass for all requests with a session cookie [operations/puppet] - 10https://gerrit.wikimedia.org/r/76700 (owner: 10Mark Bergsma) [10:26:21] PROBLEM - Puppet freshness on sq41 is CRITICAL: No successful Puppet run in the last 10 hours [10:32:11] ah I got a first pass of Debian packages :-D all python related! [10:32:27] akosiaris: paravoid: if one of you has anytime to get a look at my python backports http://noc.wikimedia.org/~hashar/debs/ :-] [10:32:49] (03PS1) 10Mark Bergsma: Ignore busy objects on requests with session cookies [operations/puppet] - 10https://gerrit.wikimedia.org/r/76702 [10:33:41] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [10:33:50] (03CR) 10Mark Bergsma: [C: 032] Ignore busy objects on requests with session cookies [operations/puppet] - 10https://gerrit.wikimedia.org/r/76702 (owner: 10Mark Bergsma) [10:34:13] why do we need those? [10:34:23] ? [10:34:27] sorry [10:34:30] hashar: why do we need those? [10:35:23] that is a good question :-] [10:35:31] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.128 second response time [10:35:37] d2to1 and python-setuptools-git are used by openstack python modules [10:35:45] they end up being build dependencies for some other packages [10:35:59] for example, to have Zuul packaged [10:36:58] the pep8 one is a python linting tool. We currently have 1.3.3 which is lagging behind, fails to detects some style errors or report wrong one and is lacking some feature that would improve our python lint jobs [10:37:09] sorry should have give context first or provided a read me maybe :D [10:44:15] meh [10:44:21] now I need to figure out how debian packaging works [10:47:58] (03PS3) 10Hashar: set some paths to use $wmfHostnames 'bits' and 'test' [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/71774 [10:49:02] (03CR) 10Hashar: "Finally came back to this change. I have added a new 'test' entry in $wmfHostnames, that makes the resulting code even simpler :-]" [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/71774 (owner: 10Hashar) [10:49:16] mark: read the docs? :D https://wikitech.wikimedia.org/wiki/Category:Package_management [10:49:48] i wrote most those docs [10:58:58] [11:27:43] Nikerabbit> why do I get errors when running maintenance script on terbium: DB connection error: No working slave server: Unknown error (10.64.32.18) ? [11:01:39] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [11:02:29] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.124 second response time [11:04:39] (03PS6) 10Mark Bergsma: Mobile Cookie Vary caching optimizations [operations/puppet] - 10https://gerrit.wikimedia.org/r/75316 [11:15:05] (03CR) 10MaxSem: "(1 comment)" [operations/puppet] - 10https://gerrit.wikimedia.org/r/75316 (owner: 10Mark Bergsma) [11:15:41] MaxSem: thanks [11:20:34] !log Pooled cp1052 as text varnish box in PyBal group 'text-varnish' with weight 1, to serve a tiny fraction of wikidata traffic [11:20:45] Logged the message, Master [11:20:56] wee [11:24:00] wee indeed [11:24:30] (03CR) 10Nemo bis: "MZ, of course I don't like it: nobody does as far as I know, at best it's considered a necessary evil (by those who voted for it)." [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/69982 (owner: 10Alex Monk) [11:24:34] so when that api bot hits the varnish box there's lots of requests in that one session [11:24:37] and then it goes quiet again [11:26:39] (03CR) 10Nikerabbit: "(1 comment)" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76556 (owner: 10Ryan Lane) [11:40:29] (03CR) 10Nikerabbit: "(2 comments)" [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/75507 (owner: 10Manybubbles) [11:53:42] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [11:54:33] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.125 second response time [12:00:40] aude: so I have a tiny bit of wikidata traffic (in the US) handled by Varnish now [12:00:51] mark: nice :) [12:01:03] of course it's all bots and such [12:01:16] heh [12:03:29] Hi, who is on RT duty now? [12:04:07] I know it's not Ryan since that was last week, but https://wikitech.wikimedia.org/wiki/Interrupts_Rotation doesn't mention this week's person and the topic hasn't been updated [12:08:09] Week of July 22 - Ryan [12:08:10] Week of Aug 5 (Wikimania) - Rob [12:08:11] hah [12:08:15] looks like we missed a week [12:08:16] how nice [12:09:12] what do you need sumanah? [12:10:01] someone on RT duty, clearly [12:10:07] paravoid: http://lists.wikimedia.org/pipermail/wikitech-l/2013-July/070946.html Silke has a db replication issue; The db replication of s5 and s6 stopped on the toolserver. She's in #wikimedia-tech [12:10:51] but yeah, mark is right, I'd also like to know who's on duty so I don't interrupt random people :) [12:11:18] springle/notpeter/binasher would be the people to ask for that [12:11:52] so an RT ticket that would be assigned on one of those three :) [12:12:04] let me file one for you [12:12:15] * sumanah curtsies with thanks :) [12:12:28] PROBLEM - SSH on pdf3 is CRITICAL: CRITICAL - Socket timeout after 10 seconds [12:12:30] * mark offers a brownie [12:14:18] RECOVERY - SSH on pdf3 is OK: SSH OK - OpenSSH_4.7p1 Debian-8ubuntu3 (protocol 2.0) [12:20:33] !log Put all non-European wikidata traffic on cp1052 and cp1065 [12:20:45] Logged the message, Master [12:44:49] (03PS7) 10Mark Bergsma: Mobile Cookie Vary caching optimizations [operations/puppet] - 10https://gerrit.wikimedia.org/r/75316 [12:55:38] (03CR) 10MaxSem: [C: 031] Mobile Cookie Vary caching optimizations [operations/puppet] - 10https://gerrit.wikimedia.org/r/75316 (owner: 10Mark Bergsma) [13:01:42] (03PS8) 10Mark Bergsma: Mobile Cookie Vary caching optimizations [operations/puppet] - 10https://gerrit.wikimedia.org/r/75316 [13:28:08] (03CR) 10Santhosh: [C: 031] Use application/font-woff for woff files [operations/puppet] - 10https://gerrit.wikimedia.org/r/76698 (owner: 10Nikerabbit) [13:43:49] (03CR) 10MaxSem: [C: 031] Mobile Cookie Vary caching optimizations [operations/puppet] - 10https://gerrit.wikimedia.org/r/75316 (owner: 10Mark Bergsma) [13:54:13] PROBLEM - Puppet freshness on lvs1004 is CRITICAL: No successful Puppet run in the last 10 hours [13:54:13] PROBLEM - Puppet freshness on lvs1005 is CRITICAL: No successful Puppet run in the last 10 hours [13:54:13] PROBLEM - Puppet freshness on lvs1006 is CRITICAL: No successful Puppet run in the last 10 hours [13:54:13] PROBLEM - Puppet freshness on virt1 is CRITICAL: No successful Puppet run in the last 10 hours [13:54:13] PROBLEM - Puppet freshness on virt3 is CRITICAL: No successful Puppet run in the last 10 hours [13:54:14] PROBLEM - Puppet freshness on virt4 is CRITICAL: No successful Puppet run in the last 10 hours [14:04:44] (03PS8) 10Hashar: adapt `foreachwiki` for labs [operations/puppet] - 10https://gerrit.wikimedia.org/r/55059 [14:07:28] (03PS4) 10Hashar: beta: symlink /a/common [operations/puppet] - 10https://gerrit.wikimedia.org/r/65254 [14:07:52] (03PS10) 10Ottomata: Puppetizing HA NameNode via Quorum Based JournalNode. [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 [14:08:14] (03CR) 10Andrew Bogott: "I have the feeling this should have a more specific name than 'git'. Can you give an example of where this will be used?" [operations/puppet] - 10https://gerrit.wikimedia.org/r/75855 (owner: 10Hashar) [14:09:38] (03Abandoned) 10Hashar: mobile $wgLoadScript now uses $wmfHostnames['bits'] [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/71779 (owner: 10Hashar) [14:10:12] andrewbogott: my first ever working rspec test in puppet! [14:11:04] hashar: That's more than I've written :) [14:11:40] I need to find out why the "git" modules specs are not invoked by the rake file :/ [14:12:03] ahh oh [14:12:56] [11:27:43] Nikerabbit> why do I get errors when running maintenance script on terbium: DB connection error: No working slave server: Unknown error (10.64.32.18) ? [14:14:39] (03PS2) 10Hashar: git::userconfig to easily craft .gitconfig files [operations/puppet] - 10https://gerrit.wikimedia.org/r/75855 [14:17:49] (03CR) 10Hashar: "I used `puppet-rspec init` to create the basic spec structure. The Rakefile generated did not rely on puppetlabs_spec_helper/rake_tasks. " [operations/puppet] - 10https://gerrit.wikimedia.org/r/75855 (owner: 10Hashar) [14:18:16] andrewbogott: good catch on the rspec not being run :-] [14:18:42] andrewbogott: I used puppet-rspec init which does not provide all the target we need and does not rely on the puppet labs spec helper hehe [14:19:18] (03CR) 10Hashar: "And I missed commenting about using the 'git' name. I simply could not find a better idea." [operations/puppet] - 10https://gerrit.wikimedia.org/r/75855 (owner: 10Hashar) [14:20:36] hashar: wouldn't it look better to split that commit in two? [14:20:55] one generic to setup the spec framework, and one specific for the current patch? [14:21:33] (03PS2) 10Hashar: deployment: abstract out MW_RSYNC_HOST [operations/puppet] - 10https://gerrit.wikimedia.org/r/72491 [14:21:37] code + test, that is an atomic change imho [14:22:19] modules/git/spec/spec_helper.rb doesn't seems to be atomic to .gitconfig [14:22:32] (03PS1) 10Ottomata: Puppetizing Hadoop JournalNode. [operations/puppet] - 10https://gerrit.wikimedia.org/r/76722 [14:24:16] !g b4b96fe [14:24:16] https://gerrit.wikimedia.org/r/#q,b4b96fe,n,z [14:24:17] !g cb3baa9 [14:24:18] https://gerrit.wikimedia.org/r/#q,cb3baa9,n,z [14:24:20] !g 22db649 [14:24:20] https://gerrit.wikimedia.org/r/#q,22db649,n,z [14:24:45] (03CR) 10Hashar: "See also bug 34552 and RT 2517" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76679 (owner: 10Ryan Lane) [14:24:49] (03CR) 10Hashar: "See also bug 34552 and RT 2517" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76680 (owner: 10Demon) [14:24:54] (03CR) 10Hashar: "See also bug 34552 and RT 2517" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76681 (owner: 10Demon) [14:25:39] wtf? [14:25:42] "You are not authorized to access bug #34552." [14:26:20] what's so secret with that bug? [14:27:02] something to do with ssl I assume [14:27:12] security hole? [14:27:40] that is related to the Gerrit changes above [14:27:44] and yes that is an internal bug. [14:27:55] Nikerabbit: how about !log ging that? [14:28:46] hashar: internal as in "only need to know"? [14:30:05] Nemo_bis: that's not a place to get help [14:30:08] I assume it's related to CVE-2011-3389 [14:31:08] which is pretty old [14:31:23] don't know why it would need to be secret on bugzilla [14:31:34] Nikerabbit: apparently this channel isn't useful for that either :) at least today [14:33:05] hashar: has there been an recently undisclosed update on BEAST? [14:33:20] * AzaToth is clueless [14:37:04] can't comment there sorry. [14:39:28] hashar, that module just creates .gitconfig right? So maybe name it gitconfig? [14:39:35] or gituser [14:55:09] andrewbogott: yup could do that [14:55:25] I thought we might want to reuse the 'git' module base to add more commands later on [14:55:32] like the evil git::clone() we have in generic-definitions [14:56:13] hashar, that might get done with https://gerrit.wikimedia.org/r/#/c/74099/ [14:57:06] PROBLEM - Puppet freshness on ssl1004 is CRITICAL: No successful Puppet run in the last 10 hours [14:58:10] andrewbogott: :]] [14:58:42] so git to gitconfig ? [14:58:42] then I would stick the define in init.pp and name it define gitconfig() [14:59:12] if you think you'll put other config things in there, then gitconfig. If you think you'll put other generic git stuff in there, maybe gitutils? [14:59:40] * andrewbogott is not helpful [15:01:19] hashar, https://plus.google.com/hangouts/_/32bd58dfec1f8fbc583410aebb70559fa9e47c10 [15:02:51] <^demon|zzz> hashar: Hi. I didn't know about RT 2517, so I had filed 5541 last night for the issue. [15:10:59] akosiaris1: andrewbogott : quick meetings are the best ones :] [15:11:04] yep! [15:11:29] :-) [15:12:00] I'm cleaning up the calendar for the next few weeks [15:12:11] cool. I was about to do it [15:12:26] Oh wait, I can't, you'll have to [15:12:29] I guess you own the entries [15:12:39] I'm out until 8/26 same as hashar [15:12:45] \O/ [15:13:15] ok ... I will schedule next on 3/9 (or 9/3 depending on where you live) [15:13:40] and will try to give you full access on that google cal entry [15:13:57] paravoid: off-topic: I had to unwatch lumail's bug mail, got really busy/lots of churn this week. Looks like a good amount of infra is setup for things like gpg and addressbooks now [15:14:46] ^demon|zzz: the more ticket we have the best it is :-] [15:14:58] ^demon|zzz: mine went unnoticed for the last 18 months or so hehe [15:15:16] andrewbogott: it does say that guests can modify event. You are sure you can not edit it ? [15:15:45] Ah, I can, just the interface is different from events that I own [15:16:10] cool [15:16:30] andrewbogott: regarding the git module, how should I rename it ? gitconfig ? [15:16:39] yeah, let's go gitconfig for now [15:20:06] <^demon|zzz> hashar: Well, Ryan and I should've taken care of almost everything last night. [15:20:38] <^demon|zzz> :) [15:21:13] :D [15:21:19] !log Purging old SecurePoll private data from all wikis [15:21:34] Logged the message, Master [15:22:27] greg-g: yep, I'm waiting for the vmime switch :) [15:22:47] ah puppet I hate you :D [15:23:00] <^demon|zzz> We should put that on a t-shirt or something. [15:23:09] given a module named 'gitconfig', how could I have it create a define named 'gitconfig' ? :-D [15:23:22] I thought about sticking it in init.pp but that does not seem to work [15:25:57] (03CR) 10Reedy: "http://docs.puppetlabs.com/references/latest/type.html" [operations/puppet] - 10https://gerrit.wikimedia.org/r/74592 (owner: 10Reedy) [15:28:20] hashar: file named gitconfig.pp with a define gitconfig::gitconfig (params) {} [15:28:45] which looks ugly btw :P [15:28:49] ahhh [15:29:02] and how one would invoke that define? gitconfig::gitconfig ? [15:29:10] exactly [15:31:19] yeah that is ugly [15:31:31] I have renamed it gitconfig::user :-∆ [15:32:16] (03PS3) 10Hashar: gitconfig::user to easily craft .gitconfig user files [operations/puppet] - 10https://gerrit.wikimedia.org/r/75855 [15:33:38] (03CR) 10Hashar: "renamed module from 'git' to 'gitconfig'" [operations/puppet] - 10https://gerrit.wikimedia.org/r/75855 (owner: 10Hashar) [15:36:46] I am out, got to catch might daughter :-} [15:36:59] (03CR) 10Andrew Bogott: [C: 04-1] "(5 comments)" [operations/puppet] - 10https://gerrit.wikimedia.org/r/75087 (owner: 10Ori.livneh) [15:50:30] (03PS1) 10Petr Onderka: writing different kinds of dumps (stub/pages, current/history) [operations/dumps/incremental] (gsoc) - 10https://gerrit.wikimedia.org/r/76736 [15:52:47] PROBLEM - Puppet freshness on mchenry is CRITICAL: No successful Puppet run in the last 10 hours [16:11:57] (03CR) 10Faidon: [C: 04-1] "(7 comments)" [operations/puppet] - 10https://gerrit.wikimedia.org/r/75087 (owner: 10Ori.livneh) [16:14:04] (03CR) 10Andrew Bogott: [C: 032] gitconfig::user to easily craft .gitconfig user files [operations/puppet] - 10https://gerrit.wikimedia.org/r/75855 (owner: 10Hashar) [16:14:34] module gitconfig? ew :) [16:14:38] I guess I'm too late though [16:15:33] (03CR) 10Andrew Bogott: [C: 032] style for beta.pp role [operations/puppet] - 10https://gerrit.wikimedia.org/r/76283 (owner: 10Hashar) [16:16:13] andrewbogott: I'm wondering if ori-l is right and we should just make a git module, move git::clone and gitconfig under it and be done with it [16:16:58] sure, vcsrepo compatibility is nice, but if we need to have our own gitconfig module, things are starting to get a bit hairy [16:17:12] (03CR) 10Andrew Bogott: "Could we get some class headers while you're at it?" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76284 (owner: 10Hashar) [16:17:45] paravoid: maybe hashar can explain about that module… it seems sort of orthogonal to vcsrepo. [16:18:02] I think he just wants it as a convenience to tune user settings. [16:18:27] (03CR) 10Faidon: [C: 04-1] "Also, if we're going style guide on this, ditch the nested subclasses." [operations/puppet] - 10https://gerrit.wikimedia.org/r/76284 (owner: 10Hashar) [16:20:08] (03PS2) 10Petr Onderka: writing different kinds of dumps (stub/pages, current/history) [operations/dumps/incremental] (gsoc) - 10https://gerrit.wikimedia.org/r/76736 [16:53:52] (03PS4) 10Reedy: (bug 49001) Restrict editing the Query namespace on Wikidata [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/66192 (owner: 10Hazard-SJ) [16:53:59] (03CR) 10Reedy: [C: 032] (bug 49001) Restrict editing the Query namespace on Wikidata [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/66192 (owner: 10Hazard-SJ) [16:54:19] (03Merged) 10jenkins-bot: (bug 49001) Restrict editing the Query namespace on Wikidata [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/66192 (owner: 10Hazard-SJ) [16:55:03] (03PS2) 10Reedy: Make FlaggedRev rights available to global groups [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/75070 (owner: 10Jalexander) [16:55:16] (03CR) 10Reedy: [C: 032] Make FlaggedRev rights available to global groups [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/75070 (owner: 10Jalexander) [16:55:37] (03Merged) 10jenkins-bot: Make FlaggedRev rights available to global groups [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/75070 (owner: 10Jalexander) [16:56:31] (03PS2) 10Reedy: Exclude *.wmflabs.org from the need to enter captchas [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/67953 (owner: 10Raimond Spekking) [16:56:37] (03CR) 10Reedy: [C: 032] Exclude *.wmflabs.org from the need to enter captchas [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/67953 (owner: 10Raimond Spekking) [16:59:34] (03CR) 10Anomie: [C: 031] set some paths to use $wmfHostnames 'bits' and 'test' [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/71774 (owner: 10Hashar) [16:59:35] (03Merged) 10jenkins-bot: Exclude *.wmflabs.org from the need to enter captchas [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/67953 (owner: 10Raimond Spekking) [16:59:36] (03PS2) 10Reedy: Enable Secure Login everywhere [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/74268 (owner: 10Ryan Lane) [16:59:37] (03CR) 10Reedy: [C: 032] Enable Secure Login everywhere [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/74268 (owner: 10Ryan Lane) [16:59:38] (03PS2) 10Ottomata: Puppetizing Hadoop JournalNode and Standby HA NameNode [operations/puppet] - 10https://gerrit.wikimedia.org/r/76722 [16:59:39] (03Merged) 10jenkins-bot: Enable Secure Login everywhere [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/74268 (owner: 10Ryan Lane) [16:59:40] (03PS2) 10Reedy: EventLogging: Add schema for mobile watchlist interactions [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76026 (owner: 10Jdlrobson) [16:59:41] (03PS4) 10Reedy: Update protection configs for core change I6bf650a3 [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/71538 (owner: 10Anomie) [16:59:42] (03CR) 10Reedy: [C: 032] Update protection configs for core change I6bf650a3 [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/71538 (owner: 10Anomie) [16:59:43] (03Merged) 10jenkins-bot: Update protection configs for core change I6bf650a3 [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/71538 (owner: 10Anomie) [16:59:47] (03PS2) 10Reedy: (bug 52076) set wgUploadNavigationUrl to a local page for kowiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76235 (owner: 10TTO) [16:59:54] (03CR) 10Reedy: [C: 032] (bug 52076) set wgUploadNavigationUrl to a local page for kowiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76235 (owner: 10TTO) [17:00:10] (03Merged) 10jenkins-bot: (bug 52076) set wgUploadNavigationUrl to a local page for kowiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76235 (owner: 10TTO) [17:00:32] (03PS2) 10Reedy: (bug 52101) add Index and Page namespaces to FlaggedRevs for ruwikisource [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76240 (owner: 10TTO) [17:00:36] secure login? secure login?! /me grabs popcorn [17:00:38] (03CR) 10Reedy: [C: 032] (bug 52101) add Index and Page namespaces to FlaggedRevs for ruwikisource [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76240 (owner: 10TTO) [17:00:58] (03Merged) 10jenkins-bot: (bug 52101) add Index and Page namespaces to FlaggedRevs for ruwikisource [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76240 (owner: 10TTO) [17:01:06] (03PS3) 10Reedy: (bug 52149) Correct AFTv5 e-mail for fr.wp oversighters [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76311 (owner: 10Odder) [17:01:13] (03CR) 10Reedy: [C: 032] (bug 52149) Correct AFTv5 e-mail for fr.wp oversighters [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76311 (owner: 10Odder) [17:01:39] (03Merged) 10jenkins-bot: (bug 52149) Correct AFTv5 e-mail for fr.wp oversighters [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76311 (owner: 10Odder) [17:02:18] (03PS3) 10Reedy: Set wgUploadNavigationUrl to Commons for ckbwiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76237 (owner: 10TTO) [17:02:30] (03CR) 10Reedy: [C: 032] Set wgUploadNavigationUrl to Commons for ckbwiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76237 (owner: 10TTO) [17:02:40] (03Merged) 10jenkins-bot: Set wgUploadNavigationUrl to Commons for ckbwiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76237 (owner: 10TTO) [17:03:28] (03PS2) 10Reedy: (bug 52145) Set wmgBabelCategoryNames for wikidatawiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76309 (owner: 10Odder) [17:03:46] (03CR) 10Reedy: [C: 032] (bug 52145) Set wmgBabelCategoryNames for wikidatawiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76309 (owner: 10Odder) [17:03:56] (03Merged) 10jenkins-bot: (bug 52145) Set wmgBabelCategoryNames for wikidatawiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76309 (owner: 10Odder) [17:04:01] PROBLEM - Puppet freshness on holmium is CRITICAL: No successful Puppet run in the last 10 hours [17:05:23] <^d> manybubbles: What e-mail address you using for BZ? [17:05:28] (03PS2) 10Reedy: Update the address in email footer [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76668 (owner: 10Bsitu) [17:05:39] <^d> Creating a new component for Cirrus and was gonna put you on default cc list. [17:05:43] (03CR) 10Reedy: [C: 032] Update the address in email footer [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76668 (owner: 10Bsitu) [17:05:52] (03Merged) 10jenkins-bot: Update the address in email footer [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76668 (owner: 10Bsitu) [17:06:06] ^d: [17:06:08] neverett+bugzilla@wikimedia.org [17:06:18] <^d> Ah, found it. Missing the +bugzilla :) [17:06:35] <^d> Ok, MediaWiki extensions -> CirrusSearch now exists. [17:08:08] !log reedy synchronized wmf-config/ [17:08:08] <^d> manybubbles: So yeah, if we want to toss bugs in there ourselves (or that other people find as they play with beta), we can now :) [17:08:21] Logged the message, Master [17:08:58] ^d: thanks! I've been getting the regression tests working against beta this morning. it is almost done but they beta seemed to start flaking out about an hour ago [17:09:39] !log reedy synchronized wmf-config/InitialiseSettings.php [17:09:54] Logged the message, Master [17:10:38] (03PS2) 10Reedy: (bug 52015) set uca-fa category collation for ckbwiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76242 (owner: 10TTO) [17:10:39] (03CR) 10Reedy: [C: 032] (bug 52015) set uca-fa category collation for ckbwiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76242 (owner: 10TTO) [17:10:41] (03Merged) 10jenkins-bot: (bug 52015) set uca-fa category collation for ckbwiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76242 (owner: 10TTO) [17:12:29] (03PS5) 10Reedy: Disable UploadBlacklist [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76229 [17:12:34] (03CR) 10Reedy: [C: 032] Disable UploadBlacklist [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76229 (owner: 10Reedy) [17:12:46] (03Merged) 10jenkins-bot: Disable UploadBlacklist [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76229 (owner: 10Reedy) [17:13:39] !log reedy synchronized wmf-config/ 'Disable UploadBlacklist' [17:13:54] Logged the message, Master [17:16:06] (03CR) 10Andrew Bogott: "I'm generally pro-upstream-module... is fixing the upstream version and getting that change merged upstream an option?" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76059 (owner: 10Ori.livneh) [17:16:26] (03PS2) 10Reedy: Add DataTypes extension [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76481 (owner: 10Aude) [17:27:11] (03PS6) 10Andrew Bogott: Fixed the path to icinga cmd file all around +file [operations/puppet] - 10https://gerrit.wikimedia.org/r/52726 (owner: 10Matanya) [17:27:56] (03CR) 10Andrew Bogott: "rebased and resolved a conflict" [operations/puppet] - 10https://gerrit.wikimedia.org/r/52726 (owner: 10Matanya) [17:28:18] LeslieCarr, 10-second review: https://gerrit.wikimedia.org/r/#/c/52726/ [17:28:47] 10…. 9….. 8..... [17:29:25] (03PS1) 10Jalexander: enable translate extension on collabwiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76746 [17:33:00] (03PS7) 10Lcarr: Fixed the path to icinga cmd file all around +file [operations/puppet] - 10https://gerrit.wikimedia.org/r/52726 (owner: 10Matanya) [17:33:26] andrewbogott: can you counter review? ;) [17:34:23] (03CR) 10Andrew Bogott: [C: 032] Fixed the path to icinga cmd file all around +file [operations/puppet] - 10https://gerrit.wikimedia.org/r/52726 (owner: 10Matanya) [17:36:05] !log Created Translate tables on collabwiki [17:36:18] * Reedy pokes morebots [17:36:19] Logged the message, Master [17:36:32] (03PS11) 10Ottomata: Puppetizing HA NameNode via Quorum Based JournalNode. [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 [17:37:14] (03CR) 10Andrew Bogott: "This one could use a class header as well." [operations/puppet] - 10https://gerrit.wikimedia.org/r/76285 (owner: 10Hashar) [17:39:07] (03PS2) 10Reedy: enable translate extension on collabwiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76746 (owner: 10Jalexander) [17:40:04] (03CR) 10Reedy: [C: 032] enable translate extension on collabwiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76746 (owner: 10Jalexander) [17:40:16] (03Merged) 10jenkins-bot: enable translate extension on collabwiki [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76746 (owner: 10Jalexander) [17:40:31] (03CR) 10Ryan Lane: "(1 comment)" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76556 (owner: 10Ryan Lane) [17:40:31] LeslieCarr, thanks [17:42:10] !log reedy synchronized wmf-config/InitialiseSettings.php 'Enable Translate on collabwiki' [17:42:24] Logged the message, Master [17:43:58] PROBLEM - RAID on virt6 is CRITICAL: CHECK_NRPE: Socket timeout after 10 seconds. [17:44:58] RECOVERY - RAID on virt6 is OK: OK: Active: 16, Working: 16, Failed: 0, Spare: 0 [17:46:57] (03PS1) 10Cmjohnson: rt5527 remove everything with $svnkey [operations/puppet] - 10https://gerrit.wikimedia.org/r/76750 [17:47:13] (03CR) 10jenkins-bot: [V: 04-1] rt5527 remove everything with $svnkey [operations/puppet] - 10https://gerrit.wikimedia.org/r/76750 (owner: 10Cmjohnson) [17:50:14] [11:27:43] Nikerabbit> why do I get errors when running a maintenance script on terbium: DB connection error: No working slave server: Unknown error (10.64.32.18) ? [17:54:48] (03PS12) 10Ottomata: Puppetizing HA NameNode via Quorum Based JournalNode. [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 [17:57:24] It's external storage [17:59:08] (03Abandoned) 10Cmjohnson: rt5527 remove everything with $svnkey [operations/puppet] - 10https://gerrit.wikimedia.org/r/76750 (owner: 10Cmjohnson) [18:00:24] Reedy: hmm? I don't quite understand... why are we encouraged to run maintenance scripts on terbium if they don't work there? [18:01:01] Ryan_Lane: topic says you're on RT duty, could you look into why we're getting raw HTML pages on beta labs as of a few minutes ago? [18:01:33] Jul 22-26 [18:02:06] <^d> chrismcmahon: I was getting that off and on yesterday. I think the bits setup in beta is flakey. [18:02:58] Nikerabbit: Do they work elsewhere? [18:03:17] Reedy: yeah it worked on tin (ttmserver-export) [18:03:34] chrismcmahon: http://en.wikipedia.beta.wmflabs.org/wiki/Main_Page <-- I'm getting CSS [18:03:46] ^d: I think it was OK until about 40 minutes ago when I saw beta labs 503 for a few minutes. when it came back, it was pure HTML [18:03:52] Ryan_Lane: try to login [18:03:55] I'm also not getting pure HTML [18:04:00] Ryan_Lane: when you hit a random page or login? [18:04:18] ah. I see [18:04:35] <^d> Ah yes, on SSL. [18:04:45] <^d> Forcing secure login for beta broke something ;-) [18:05:00] I see [18:05:04] it's because of the ssl cert [18:05:05] Yeah, that will break beta tests [18:05:17] CSS comes from bits [18:05:20] and will throw an error [18:05:23] <^d> We could set secure login to false for beta. [18:05:27] but won't present the error [18:05:37] Ryan_Lane: you suck [18:05:46] greg-g: oh? in what way? :) [18:05:49] * greg-g approves all requests [18:06:14] *confetti* [18:06:21] war is over! [18:06:21] I'm more than happy to have a real cert for beta, assuming every project admin has an NDA [18:06:23] since the cert for beta is bogus anyway, secure login is going to wreak havoc on automated browsers [18:06:34] * greg-g +2's all of ori's changes [18:07:08] greg-g: could you take a look at pdf servers? [18:07:16] (03PS3) 10Ottomata: Puppetizing Hadoop JournalNode and Standby HA NameNode [operations/puppet] - 10https://gerrit.wikimedia.org/r/76722 [18:07:17] :( [18:07:31] I'm already tripping over the lack of a real cert for beta, let's not require secure login without that? [18:07:52] greg-g: hahaha. I just noticed you changed the topic :) [18:08:00] AaronSchulz: ping? [18:08:03] Ryan_Lane: I uh, didn't [18:08:07] oh. ori did [18:08:08] :D [18:08:09] 4 days idle, I guess not many chances there [18:08:10] <^d> paravoid: He's on vacation this week. [18:08:14] oh, okay [18:08:14] I thought you did for some reason.... [18:08:14] thanks [18:08:26] Ryan_Lane: you = you ;) [18:08:47] my backscroll shows ori changing the topic to make you on duty ;) [18:09:12] <^d> We should find an anonymous user and put them on RT duty. [18:09:27] thanks paravoid [18:10:02] Ryan_Lane: yeah, I misread my backscroll. I thought it was in retaliation to my snark in _security [18:10:04] * paravoid -1's all of ori's changes [18:10:08] :D [18:10:09] heh [18:10:09] * ori-l quietly sweeps confetti. [18:10:22] (03CR) 10Dzahn: [C: 032] rename wikivoyages to wikivoyage, fix timestamps on index page [operations/debs/wikistats] - 10https://gerrit.wikimedia.org/r/76673 (owner: 10Dzahn) [18:10:23] (03CR) 10Dzahn: [V: 032] rename wikivoyages to wikivoyage, fix timestamps on index page [operations/debs/wikistats] - 10https://gerrit.wikimedia.org/r/76673 (owner: 10Dzahn) [18:12:56] PROBLEM - Puppet freshness on erzurumi is CRITICAL: No successful Puppet run in the last 10 hours [18:14:59] (03CR) 10Dzahn: [C: 032 V: 032] renable updates via file_get_contents() [operations/debs/wikistats] - 10https://gerrit.wikimedia.org/r/76674 (owner: 10Dzahn) [18:15:01] (03CR) 10Dzahn: [C: 032 V: 032] add 1.22wmf12 to good versions [operations/debs/wikistats] - 10https://gerrit.wikimedia.org/r/76675 (owner: 10Dzahn) [18:17:27] (03CR) 10Dzahn: [C: 032] "http://www.w3.org/TR/WOFF/#appendix-b" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76698 (owner: 10Nikerabbit) [18:18:24] Ryan_Lane: so it was this that took out beta labs? https://gerrit.wikimedia.org/r/#/c/74268 [18:18:53] yes [18:19:30] !log deploying fix for Bug 51593 - Correct the mime type for WOFF to application/font-woff [18:19:50] Logged the message, Master [18:19:50] so, for changes like that, we'd really love to have a deploy window, btw [18:20:07] I'd be kicked by opsen for doing something like that without a window :) [18:20:23] oh, the one above? [18:20:25] Ryan_Lane: can we revert that or amend it to get the test environment backi? [18:20:39] (03PS3) 10Petr Onderka: writing different kinds of dumps (stub/pages, current/history) [operations/dumps/incremental] (gsoc) - 10https://gerrit.wikimedia.org/r/76736 [18:20:47] mutante: yeah [18:20:47] greg-g: which one? the mime type as well? [18:20:54] mutante: I meant the secure longin [18:20:57] -n [18:21:02] greg-g: ok [18:21:53] chrismcmahon: just change the config for beta to disable it [18:21:59] absolutely don't just revert it [18:22:36] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [18:22:52] Ryan_Lane: OK, hang on I'll have a config change here shortly [18:23:16] chrismcmahon: so, yeah, this is a dev thing. ops isn't needed [18:23:31] this is purely mediawiki config [18:24:17] (03PS1) 10Cmjohnson: removing $svnkey entries [operations/puppet] - 10https://gerrit.wikimedia.org/r/76755 [18:25:26] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.127 second response time [18:26:23] (03PS13) 10Ottomata: Puppetizing HA NameNode via Quorum Based JournalNode. [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 [18:26:36] (03CR) 10Petr Onderka: [C: 032 V: 032] writing different kinds of dumps (stub/pages, current/history) [operations/dumps/incremental] (gsoc) - 10https://gerrit.wikimedia.org/r/76736 (owner: 10Petr Onderka) [18:27:22] scapping in a second [18:28:22] (03CR) 10Ottomata: "Ok, this is tested and applied in labs. Went pretty smoothly!" [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 (owner: 10Ottomata) [18:29:12] hi [18:29:25] i wonder if it's just me or related to the new varnish caching... [18:29:39] when i request https://www.wikidata.org/w/api.php?action=wbgetclaims&entity=q3457234&format=json from the browser (from europe?) [18:29:49] i get 4 items in the list [18:30:00] when i request from php (from north america), i get just one [18:30:06] like old data [18:30:14] can anyone help confirm? [18:30:55] could it be related? [18:31:12] heya LeslieCarr, i know that HA NameNode commit is pretty huge [18:31:19] got some time to go over it with me today? [18:31:51] it means my bot can't operate and would affect others [18:32:17] mark: ^ around? [18:32:28] aude: I get the same output from both India and USA (toollabs) [18:32:29] 4 [18:32:33] (if that helps [18:32:34] oh, really? [18:32:46] how can i be getting old data? [18:32:47] let me confirm that [18:32:48] moment [18:32:49] Ryan_Lane: heya, since you were the owner of that gerrit change, do you know the associated bug report and/or discussion for the UI change to login? [18:32:58] i am getting the same data [18:33:03] wow [18:33:42] Ryan_Lane: also, why can't it be reverted? [18:33:54] aude: yeah, I get 4 items. India (local connection) and toollabs [18:34:01] greg-g: why should it be? [18:34:04] hmmm [18:34:11] fixing beta is a matter of confguration [18:34:21] for beta [18:34:25] p625, p17, p131, p107 [18:34:27] Ryan_Lane: because the UI change was reviewed by Design, it doesn't seem like it was ready for deploy. [18:34:33] s/was/wasn't/ [18:34:58] why can't we follow up with the design fix? [18:35:00] Ryan_Lane: so, you said not to revert it, why not? [18:35:03] aude, mark's email said no https [18:35:04] YuviPanda: i'm trying on my laptop with the same bot code [18:35:10] why was it so need to be deployed this morning? [18:35:11] because it provides a major UI improvement [18:35:16] MaxSem: ? [18:35:21] it was accidentally deployed, I'd imagine [18:35:31] aude: I just tried on one of my other servers (North America, but not toollabs). Getting 4 items [18:35:37] <^d> I'm removing the "Stick to HTTPS" crap. [18:35:38] aude, that is, it shouldn't be coming from varnish [18:35:44] <^d> That checkbox has no reason to exist. [18:35:50] ok [18:35:52] <^d> (Or at least working on a patch to) [18:35:53] gah, git/gerrit is frozen now ? [18:35:54] my bot is using http [18:35:57] that's fine. I asked for that originally ;) [18:36:10] I also hate the checkbox [18:36:11] give me a minute.... [18:36:12] yep, gerrit slow, but coming back [18:36:14] <^d> chrismcmahon: No. [18:36:17] <^d> Works just fine. [18:36:19] (03PS1) 10Cmcmahon: override https://gerrit.wikimedia.org/r/#/c/74268 for beta labs only [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76756 [18:36:25] ^d ok, it just unstuck ^^ [18:36:42] so, without that change, if someone logs in with https, then they visit using http [18:36:46] they are no longer logged in [18:36:55] aude, confirmed - headers say it's not coming from varnish servers [18:36:56] which is *much* worse than some lock icons [18:37:10] wait, login was broken without the change? I don't remember seeing any reports of that [18:37:13] <^d> If someone logs in with HTTPS, why would we bother sending them back to HTTP? [18:37:21] <^d> That makes no sense to me. [18:37:24] ^d: that's different [18:37:24] PROBLEM - SSH on pdf3 is CRITICAL: CRITICAL - Socket timeout after 10 seconds [18:37:31] that checkbox should die in a fire :) [18:37:37] <^d> Ok, working on that fire [18:37:40] (03CR) 10Ottomata: "Which, ahem, means totally ready for review!" [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 (owner: 10Ottomata) [18:37:42] * ^d rubs some sticks together [18:37:45] I'm talking about the behavior this enforces [18:37:59] so, with this change enabled, all users are forced to login via https [18:38:05] but it also adds a new cookie [18:38:21] if you are logged in via https, if you visit an http page you get redirected back to https [18:38:34] I think keeping that without the checkbox to disable is sane. [18:38:38] without this change in, you don't get redirected and you aren't logged in [18:39:01] Right, which sucks and is confusing unless you understand the technical difference between HTTP/S [18:39:02] greg-g Ryan_Lane ^d not sure who should review https://gerrit.wikimedia.org/r/#/c/76756 , this is not my usual beat. [18:39:10] StevenW: indeed [18:39:28] (03CR) 10Demon: [C: 032] override https://gerrit.wikimedia.org/r/#/c/74268 for beta labs only [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76756 (owner: 10Cmcmahon) [18:39:32] <^d> chrismcmahon: Done [18:39:37] (03Merged) 10jenkins-bot: override https://gerrit.wikimedia.org/r/#/c/74268 for beta labs only [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76756 (owner: 10Cmcmahon) [18:39:39] thansk ^d [18:39:44] !log bsitu Started syncing Wikimedia installation... : Update Echo, Thanks, PageTriage to master [18:39:53] ottomata: yeah i should be able to after lunch ? [18:39:53] I'd prefer we keep this change in and follow up. the current UI nastiness is not as bad as it would be if it was reverted [18:39:54] i wonder if i can merge stuff that includes #The Initial Developer of the Original Code is Netscape Communications . Copyright (C) 1998 Netscape Communications Corporation. .. but also # Mozilla Public License Version 1.1 [18:39:58] Logged the message, Master [18:40:14] RECOVERY - SSH on pdf3 is OK: SSH OK - OpenSSH_4.7p1 Debian-8ubuntu3 (protocol 2.0) [18:40:20] Ryan_Lane: the Ui would be different after reverting than it was before deploying? [18:40:26] yes [18:40:35] how so? [18:40:41] * greg-g doesn't know this feature [18:40:41] on my macbook, with the same bot code, i get 4 properties [18:40:44] I just explained why :) [18:40:50] on my linode and labs i get one [18:40:59] uhhh, no [18:41:06] greg-g: say you log in with https [18:41:17] then you google something and click a wikipedia link [18:41:19] right, the "then visit via http" thing yeah [18:41:29] so, that's different now than it was yesterday? [18:41:32] LeslieCarr: cool, sounds good, i'm here for another couple of hours [18:41:37] this feature adds a cookie [18:41:47] i want a cookie Ryan_Lane [18:41:56] LeslieCarr: what kind? oatmeal? [18:42:13] greg-g: if the cookie is set, if you visit an http page it redirects you to https [18:42:13] only if it has chocolate chips [18:42:20] LeslieCarr: would you like to remain connected with HTTPS after your cookie? [18:42:21] raisins in cookies are evil tricksy things [18:42:27] LeslieCarr: white chocolote chocolate chips? [18:42:28] * ori-l trolls. [18:42:28] Ryan_Lane: right, I get that [18:42:39] but, that behavior is set by....? [18:42:44] the preference [18:42:49] in the change [18:43:05] so, I'm still confused why we just can't go back to what we had working on MOnday [18:43:16] white chocolate is fine :) [18:43:24] ewww [18:43:29] greg-g: what did we have working on monday? [18:43:33] oh man, chocolate cookie with some cayenne + white chocolate chips is probably my favorite cookie [18:43:34] the site :) [18:43:38] login [18:43:40] etc [18:44:00] might not have been what we wanted in a perfect world, but this was a premature/unplanned/accidental deploy [18:44:01] because this cookie wouldn't be set? [18:44:39] omg, Talk:Chocolate_chip_cookie is in Category:Low-importance Food and drink articles .. who did that [18:44:41] the issue right now is whether the interaction with the site is better or worse with this change enabled [18:44:42] so, would the site break if we reverted this change? [18:44:46] and it's much, much better [18:44:58] I consider the site broken without this change [18:45:35] sure, whatever, but that's not the point. I just wanted an answer if the site would break if we reverted this change, not if we liked where we were before. [18:46:01] ooo: [18:46:02] http://tox.im/ [18:46:06] well, the issue is that we don't like things to be deployed accidentally that make changes to UI in confusing ways [18:46:09] it wouldn't break, but it would go back to working in a state that's functionally broken for https users [18:47:03] mutante: be bold my friend [18:47:10] sorry, it's just frustrating that a change that was accidentally deployed and does unwanted things is being pushed through [18:47:12] * aude confused [18:47:45] unplanned, even, for a relatively major change (that would have really been good to announce before hand) [18:47:57] this was actually pushed out months ago [18:48:02] and was announced back then [18:48:02] not the config change [18:48:06] yes [18:48:07] it was [18:48:15] ? ok, now I'm more confused :) [18:48:18] it was reverted due to a bug in the feature [18:48:42] it was supposed to be pushed out with the new login stuff, but wasn't. so, I pushed in the config change [18:49:17] ok, it seems that csteipp was surprised it was deployed this morning, and he was out all last week when the only comments on it were made (so he probably missed it in his backlog) [18:49:28] and he pretty much owns this feature [18:49:29] it wasn't scheduled for today [18:49:35] right, which is annoying :) [18:49:41] it "just happened" [18:49:51] I was surprised too, but I'm happy it was deployed nonetheless [18:50:22] ok, in the meantime, is there something to be bounced or whatever to get bits back for beta labs? [18:50:37] * Ryan_Lane sighs [18:50:42] fine, just revert it [18:50:46] I lost internet connectin during scap, I believe i18n update is done, should I sync-dir to make sure everything go through? [18:51:00] <^d> chrismcmahon: Bits is fine. It's secure login that was messy on beta. [18:51:00] <^d> Which you disabled :) [18:51:04] PROBLEM - Puppet freshness on manutius is CRITICAL: No successful Puppet run in the last 10 hours [18:51:08] <^d> ...I thought [18:51:11] So, just like we get pissed when wikidata make changes to configs that break beta, we also get pissed when things like this happen that break beta, and whether or not it is a wanted change, we want to make sure that it doesn't happen the same way again [18:51:38] Ryan_Lane: I just wanted understanding, not a fight. [18:51:47] ^d: still seeing raw HTML on beta random pages [18:52:00] just a way forward [18:52:01] <^d> meh [18:52:09] greg-g: to be fair I earlier said absolutely don't revert a production feature just because beta was broken [18:52:30] bsitu1: you should probably re-scap, but check with greg-g and Ryan_Lane, they may want to piggyback a config change [18:52:32] s/said/meant/ [18:52:48] <^d> I do tend to agree with Ryan_Lane on that point. 9 times out of 10 it's because beta's misconfigured and needs fixing *anyway*, not a problem with the actual feature/bugfix/etc. [18:53:03] right, I was looking at it as "reverting a production feature becuase it was deployed accidentally and introduced confusing UI that wasn't reviewed on a peice of UI that is getting a lot of UI review and churn" [18:53:31] bsitu1: also, it's probably a good idea to run scap in a 'screen' session so that it doesn't halt if your connection is momentarily interrupted [18:53:34] +1 greg-g [18:53:35] ie: something we want to be careful with (the login experience right now) [18:53:51] ori-l: thx, I will check with them, that's a good idea [18:54:21] it's also not great to go out so quickly with a feature that we don't even discover it's broken in beta prior to deploying to production [18:54:27] the problem is, we're not fixing an incredibly massive HTTPS issue so that we don't introduce a fairly minor UI issue [18:54:46] this is the same conversation that we need to have regarding cherry-picking versus just pulling from master during this kinds of deploys [18:54:49] yeah, the fact that it just appeared is bad [18:54:49] Krinkle: delayed answer from yesterday night. it uses in display.php and that is function version_color($version) in functions.php [18:54:57] it shouldn't have gone out without a window [18:55:07] * greg-g nods [18:55:56] bsitu1 has to patch up a tricky deployment, so should he just scap? [18:56:08] the revert (if we agree on it) can wait 5-10 mins, right? [18:56:10] * csteipp was thinking this would go out with a lot of fanfair and in a carefully scheduled window [18:56:37] gre-g, Ryan_Lane: I will need to re-scap,do you want to piggyback a config change? [18:56:51] MaxSem: you say the varnish text cache is http only? [18:56:53] (03PS1) 10Ryan Lane: Revert "Enable Secure Login everywhere" [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76757 [18:56:58] bsitu1: chrismcmahon has a beta change for this, but that doesn't implicate you [18:57:04] aude, yes [18:57:07] huh [18:57:12] bsitu1: that change I just reverted [18:57:28] on different linodes, i get different results [18:57:39] on labs, i get all data with https, old data with http [18:57:55] aude, quoting the original email: "I've just put non-European, non-IPv6 and non-HTTPS wikidata traffic on two text Varnish boxes in eqiad, cp1052 and cp1065." [18:58:04] ok, I have to run to lunch, back later [18:58:16] Ryan_Lane: I'll +2 to unblock bsitu, OK? [18:58:18] ok [18:58:20] * Ryan_Lane nods [18:58:31] something is funky.... [18:58:41] (03CR) 10Ori.livneh: [C: 032] "Don't shoot! I'm unarmed!" [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76757 (owner: 10Ryan Lane) [18:58:43] seems if i try now, fresh request then i get new data [18:58:48] therefore, the headers should explicitly say that it's coming from these boxes:) [18:58:52] (03Merged) 10jenkins-bot: Revert "Enable Secure Login everywhere" [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76757 (owner: 10Ryan Lane) [18:58:52] ok [18:59:12] bsitu1: OK, can you pull that change onto tin and scap? [18:59:48] doesn't need scapping [19:00:15] Reedy: bsitu1 needs to scap for independent reasons [19:00:29] sure [19:00:33] sync-file and carry on as normal [19:01:09] if i add a bogus extra param to my json request, then i get new data1 [19:01:51] with my bot, i do the getclaims request to see if the properties are missing and then add them [19:02:08] when an item appears again in my list, i check and it should see the properties are there [19:02:31] it always worked until now, but now my bot thinks the properties are still missing and adds them again! [19:03:13] i can wait to talk to mark tomorrow, see if the problem goes away, if anyone else has this, etc. [19:04:06] I need to re-scap again, I can wait if you want to sync the file or I can have the config change piggy back on the scap, either way is fine with me [19:09:55] sync coming [19:13:18] (03PS2) 10Cmjohnson: removing $svnkey entries [operations/puppet] - 10https://gerrit.wikimedia.org/r/76755 [19:14:07] (03CR) 10Siebrand: "Is there a bug report that requests this configuration change? I'm interested what the background of the request is," [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76746 (owner: 10Jalexander) [19:15:33] !log bsitu synchronized wmf-config/InitialiseSettings-labs.php 'Revert Enable Secure Login everywhere' [19:15:48] Logged the message, Master [19:16:05] !log bsitu synchronized wmf-config/InitialiseSettings.php 'Revert Enable Secure Login everywhere' [19:16:19] Logged the message, Master [19:17:05] hi, is XFF header appended or prepended when going through the SSL proxy? [19:18:35] !log bsitu Started syncing Wikimedia installation... : Update Echo, Thanks, PageTriage to master [19:22:37] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [19:23:05] !log bsitu Finished syncing Wikimedia installation... : Update Echo, Thanks, PageTriage to master [19:23:18] Logged the message, Master [19:23:27] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.123 second response time [19:24:26] Nikerabbit: ping [19:24:38] mark around? [19:24:58] yurik: there was an XFF handling change deployed today [19:25:14] yurik: what's the problem? [19:25:19] addshore: can I help perhaps? [19:26:10] paravoid, yes, that's why i'm asking about it -- i was reading that patch and noticed that Zero's XFF handling assumes that XFF gets appended each time it goes through a proxy (like opera servers or our SSL proxy) [19:26:10] mhhm, well I spent half the day debuging some of my code that for some reason kept getting bad token errors using the wikidata api :P it started happening within 10 mins of (12:21 : Put all non-European wikidata traffic on cp1052 and cp1065) [19:26:27] any idea or should I just leave it until tomorrow? :) [19:26:35] addshore: aude was also having (perhaps unrelated) trouble [19:26:41] addshore: could you be a bit more specific about the trouble? [19:26:41] paravoid, but for some reason i was pretty sure it gets prepended. Hence - checking :) [19:26:46] YuviPanda: yee, she just told me [19:26:57] addshore, aybe cookie weirdness [19:27:01] addshor, aude: what's the isssue specifically [19:27:14] I can revert that change now [19:27:17] paravoid: well, when making trying to do an edit on wikidata with correct token I always get a lovely bad token message [19:27:25] addshore, what cookies are you sending (names)? [19:27:26] but let's collect all information first [19:27:46] MaxSem: you want my tokens? :> [19:27:55] no, just cookie names [19:28:12] I have cleared everything, logged in, logged out, tried to get new tokens etc [19:28:18] with my shell powahs, I can tell you all your tokens;) [19:28:44] MaxSem: trying to look now :P [19:29:02] its been a long time since i touched this bit of code ;p [19:29:13] heh curl_setopt( $this->ch, CURLOPT_COOKIEFILE, '/tmp/addframe.cookies.' . $this->uid . '.dat' ); [19:29:30] so you just go to a random item, click add or edit, type something, hit save and then get a token error? [19:30:05] paravoid: this code is running from labs, as far as I know it should work fine from where I am currently [19:30:26] I could probably knock together a little test script [19:30:32] I'm confused [19:32:56] could you outline the steps needed to reproduce? [19:32:57] paravoid, bots might rely on slightly different cookies than human users, and mark's change filters cookies very strictly [19:33:21] that's why I asked what exactly gets sent [19:33:27] right [19:36:59] (03PS3) 10Cmjohnson: removing $svnkey entries and fixing style errors [operations/puppet] - 10https://gerrit.wikimedia.org/r/76755 [19:38:07] right the code works fine from here, just not on labs (ie fine in europe, just not fine in the us) :/ [19:38:35] coookies! [19:38:47] I wanna cookies! [19:39:27] mhhm, the questions is how am I going to find them [19:39:28] xD [19:40:36] addshore: btw, labs is also internal network, so that might have something to do with it [19:40:48] indeed [19:41:50] internal or not, it gets served by varnish (just as planned) [19:44:05] (03PS4) 10Cmjohnson: removing $svnkey entries and fixing style errors [operations/puppet] - 10https://gerrit.wikimedia.org/r/76755 [19:45:38] * addshore gives MaxSem a cookie from his cookiejar [19:49:09] hmm, these cookies contain both "Token" and "Session" so they shouldn't get munged [19:49:30] (03CR) 10Cmjohnson: [C: 032 V: 032] removing $svnkey entries and fixing style errors [operations/puppet] - 10https://gerrit.wikimedia.org/r/76755 (owner: 10Cmjohnson) [19:49:57] MaxSem: I imagine they are all the same as well :> [19:53:05] just looked at another one of the exec nodes and they all appear to contain the same sorts of dat [19:56:06] MaxSem: any inital thoughts then? :/ [19:56:37] (03PS14) 10Ottomata: Puppetizing HA NameNode via Quorum Based JournalNode. [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 [19:57:01] nope [19:57:12] paravoid, any ideas? [19:57:15] :D [19:59:53] (03PS15) 10Ottomata: Puppetizing HA NameNode via Quorum Based JournalNode. [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 [20:02:07] (03PS16) 10Ottomata: Puppetizing HA NameNode via Quorum Based JournalNode. [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 [20:02:29] what did I miss? [20:02:45] paravoid: pong? [20:03:13] Nikerabbit: hi [20:03:14] paravoid, I got cookies from addshore, they contain both Token and Session, so shouldn't be munged [20:03:37] (03CR) 10MaxSem: [C: 032] EventLogging: Add schema for mobile watchlist interactions [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76026 (owner: 10Jdlrobson) [20:04:02] Nikerabbit: we have an RT ticket open since Jun 17th, about ULS & IE8. the ticket's last action was that someone should contact microsoft about a blacklist of theirs. would you happen to know the status or should I ping everyone via the ticket? [20:04:29] MaxSem: can I see them? [20:05:04] PM'd [20:05:11] +1 xD [20:06:00] paravoid: I'm not sure about status, but I doubt that anyone has pinged Microsoft. Iirc we were pondering whether we could send a http header to force standards mode [20:06:11] (03PS4) 10Ottomata: Puppetizing Hadoop JournalNode and Standby HA NameNode [operations/puppet] - 10https://gerrit.wikimedia.org/r/76722 [20:07:13] addshore: so what is it that you run? [20:07:29] hmm, code/framework? [20:07:45] i see some libcurl cookies [20:07:50] how do you use that? [20:08:10] (03PS5) 10Ottomata: Puppetizing Hadoop JournalNode and Standby HA NameNode [operations/puppet] - 10https://gerrit.wikimedia.org/r/76722 [20:08:45] curl_setopt( $this->ch, CURLOPT_COOKIE, $cookies ); [20:08:59] the cookiejar gets attached to the curl opts [20:09:09] urm *looks further* [20:09:15] sorry i havnt really ever touched this code ;p [20:09:47] https://github.com/addshore/addwiki/blob/master/includes/Http.php [20:10:33] okay [20:10:49] (03PS6) 10Ottomata: Puppetizing Hadoop JournalNode and Standby HA NameNode [operations/puppet] - 10https://gerrit.wikimedia.org/r/76722 [20:14:24] addshore: can you see now? [20:14:35] see if its working? [20:14:48] yes please [20:14:56] LeslieCarr: if you are back and wanna, I have another hourish to review [20:14:59] will I need to get new cookies? ;p [20:15:02] no [20:15:24] heh, well its constantly trying to edit [20:15:38] what does that mean? [20:16:36] addshore: does it work now? [20:17:15] youll have to give me a second :> [20:17:47] !log switching wikidata back to squid; reports of "bad token" messages [20:18:01] Logged the message, Master [20:18:34] 10:17 PM [[Q1174923]] B http://www.wikidata.org/w/index.php?diff=62546654&oldid=61949583&rcid=62608253 * Addbot * (+24) /* wbeditentity-update:0| */ [20:18:41] (03CR) 10Jalexander: "There isn't a bug report, it's a request from myself and the legal team team because we wanted a place where wikifying (including translat" [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76746 (owner: 10Jalexander) [20:18:42] seems to be editing again [20:18:55] (03PS1) 10Dzahn: add voyagewiki.org & .com, change some other wikivoyage redirects to go www. directly [operations/apache-config] - 10https://gerrit.wikimedia.org/r/76819 [20:18:57] k [20:19:27] we might ping you tomorrow european hours :) [20:20:30] thats fine :) [20:20:36] (03PS11) 10Dzahn: (bug 15434) Periodical run of currently disabled special pages [operations/puppet] - 10https://gerrit.wikimedia.org/r/33713 (owner: 10Nemo bis) [20:23:25] roankattouw: can i take down a parsoid server...any wtp10xx will do? Want to see if I can stop the cpu core power limit warning [20:26:12] cmjohnson1: What's the deal with wtp1008? That one's still depooled AFAIK (let me check real quick) and available for you to grab [20:26:23] I have wtp1008 down as "Chris owns this, don't touch" still [20:26:30] If you need another box I can set that up for you too [20:26:31] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [20:26:34] oh..that's fixed..but cool I will use that one [20:26:41] since it's already depooled [20:26:51] PROBLEM - Puppet freshness on sq41 is CRITICAL: No successful Puppet run in the last 10 hours [20:27:00] Yeah if it's equally convenient for you to use 1008 just go ahead and use that one [20:27:23] it is..i am looking into rt5271 [20:27:31] Everyone involved knows that 1008 is a box you're still messing it so no one will (should) touch it without talking to you first [20:27:53] Oh, that one, yeah [20:28:21] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.139 second response time [20:28:29] (03CR) 10Dzahn: [C: 032] add voyagewiki.org & .com, change some other wikivoyage redirects to go www. directly [operations/apache-config] - 10https://gerrit.wikimedia.org/r/76819 (owner: 10Dzahn) [20:28:35] cool..yeah so on wtp1008 I had to replace the mother board...all done [20:28:42] Oh nice [20:28:59] those are just kernel msgs not errors but want to make them stop [20:29:32] AFAICT they're more common when the box isn't idle [20:29:51] Anecdotally I believe they happen much less on wtp1008 than elsewhere but that only started once we depooled it. So it seems to be vaguely related to load [20:30:55] okay...if the comment on bugzilla is correct that would make sense [20:32:11] (03CR) 10Dzahn: "testing 7 urls on 1 servers, totalling 7 requests" [operations/apache-config] - 10https://gerrit.wikimedia.org/r/76819 (owner: 10Dzahn) [20:32:31] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [20:32:51] PROBLEM - Host wtp1008 is DOWN: PING CRITICAL - Packet loss = 100% [20:33:21] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.122 second response time [20:33:36] !log gracefull'ing apaches, RT-5539 & bug 51593 [20:33:52] Logged the message, Master [20:38:05] (03PS17) 10Ottomata: Puppetizing HA NameNode via Quorum Based JournalNode. [operations/puppet/cdh4] - 10https://gerrit.wikimedia.org/r/76018 [20:38:07] (03PS2) 10Hashar: style for zuul.pp [operations/puppet] - 10https://gerrit.wikimedia.org/r/76284 [20:38:08] (03CR) 10Hashar: "* got rid of nested subclasses" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76284 (owner: 10Hashar) [20:38:20] (03CR) 10Andrew Bogott: [C: 031] "Much better, thanks!" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76284 (owner: 10Hashar) [20:38:30] yay for no nested classes hashar! [20:39:22] heh [20:39:28] mark will disagree with you ;) [20:39:35] * Ryan_Lane prefers no nested [20:39:39] *nesting [20:41:48] * AzaToth nestes Ryan_Lane [20:43:21] The English Wikipedia has a long history of "trials" (started with or without consensus) which became permanent due to the sclerotization of the consensus-gaining process, the most prominent example being the "experiment" of restricting article creation to [[Newly registered user|confirmed users]]. [20:43:33] sclerotization <# [20:43:53] * AzaToth looks up wtf sclerotization means [20:43:55] <3 hmm [20:44:12] "The process by which the cuticle of an arthropod is hardened by substances other than chitin." [20:44:28] See, it's as simple as that. [20:44:29] * AzaToth is even more confused [20:44:30] (03PS2) 10Hashar: style for ipv6relay.pp [operations/puppet] - 10https://gerrit.wikimedia.org/r/76285 [20:45:04] (03CR) 10Hashar: "And the class description now has a single sentence:" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76285 (owner: 10Hashar) [20:45:12] twkozlowski: I can't dechipher what you are tryingt osay [20:45:28] I assume it's in the wrong channel anyway [20:46:51] Why would it be in the wrong channel? [20:47:47] because it's not #wikiphilosophy? [20:48:52] RECOVERY - Host wtp1008 is UP: PING WARNING - Packet loss = 54%, RTA = 50.07 ms [20:49:27] twkozlowski: mostly because it isn't on-topic here right now, until after some decision has been made on what to do, code deployment-wise [20:50:03] (03CR) 10Dzahn: [C: 032] "rebased once again and now merging per Asher's comment " Any problems shouldn't effect the site and could be easily backed out." because i" [operations/puppet] - 10https://gerrit.wikimedia.org/r/33713 (owner: 10Nemo bis) [20:50:19] MaxSem: so what. [20:52:08] that's https://meta.wikimedia.org/w/index.php?diff=5690893&oldid=5690033 BTW; Nemo_bis ^^ :-) [20:53:52] MaxSem: we could talk about the philosophy of puppet [20:54:10] -> #puppet [20:54:26] :-P [20:55:25] twkozlowski: sorry, it's a common word in Italian [20:57:31] come on, Monbiot uses it a lot according to Wiktionary, it's a must have then :) [21:01:32] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [21:02:22] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.127 second response time [21:03:30] (03PS1) 10Demon: New memcached hosts for beta [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76828 [21:03:52] (03CR) 10Demon: [C: 032] New memcached hosts for beta [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76828 (owner: 10Demon) [21:04:09] (03Merged) 10jenkins-bot: New memcached hosts for beta [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76828 (owner: 10Demon) [21:04:45] (03CR) 10Faidon: [C: 032] style for zuul.pp [operations/puppet] - 10https://gerrit.wikimedia.org/r/76284 (owner: 10Hashar) [21:06:25] (03PS1) 10Dzahn: let misc::maintenance::updatequerypages use $enabled like the other maintenance jobs, avoid puppet breakage due to Invalid parameter enabled when trying to use $enabled in site.pp [operations/puppet] - 10https://gerrit.wikimedia.org/r/76831 [21:07:51] (03CR) 10Dzahn: [C: 032] "this should now make maintenance::updatequerypages work" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76831 (owner: 10Dzahn) [21:08:35] scapping... [21:10:32] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [21:11:27] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.122 second response time [21:12:15] (03CR) 10Dzahn: "recheck" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76831 (owner: 10Dzahn) [21:12:17] (03CR) 10jenkins-bot: [V: 04-1] let misc::maintenance::updatequerypages use $enabled like the other maintenance jobs, avoid puppet breakage due to Invalid parameter enabled when trying to use $enabled in site.pp [operations/puppet] - 10https://gerrit.wikimedia.org/r/76831 (owner: 10Dzahn) [21:13:03] why is it a syntax error in 421, but not in 392 .. hmm [21:17:28] !log maxsem Started syncing Wikimedia installation... : Weekly mobile deployment [21:17:43] Logged the message, Master [21:19:02] (03PS2) 10Dzahn: let misc::maintenance::updatequerypages use $enabled like the other maintenance jobs, avoid puppet breakage due to Invalid parameter enabled when trying to use $enabled in site.pp [operations/puppet] - 10https://gerrit.wikimedia.org/r/76831 [21:20:36] bsitu: heya, see beta cluster error, something ya'll could fix so we can get beta back? [21:21:56] greg-g: what error? [21:22:20] bsitu: go to http://en.wikipedia.beta.wmflabs.org/ and look at the error at the bottom :) [21:22:54] it's loading [21:23:07] hrmm? [21:24:02] bsitu: this was the error: http://paste.debian.net/20402/ [21:24:50] bsitu: now I'm getting a cache error [21:25:03] I mean the page is blank and keeps loading [21:25:42] now I'm getting that first error again [21:25:48] (that I pastebin'd) [21:26:32] (03CR) 10Dzahn: [C: 032] let misc::maintenance::updatequerypages use $enabled like the other maintenance jobs, avoid puppet breakage due to Invalid parameter enabled [operations/puppet] - 10https://gerrit.wikimedia.org/r/76831 (owner: 10Dzahn) [21:26:33] bsitu: I'm getting that inclusion error pretty consistently now, as is ^d [21:27:15] chrismcmahon: thoughts on that pastebin above from me? [21:27:29] does this file exist? /data/project/apache/common-local/php-master/extensions/Echo/Hooks.php [21:27:58] !log maxsem Finished syncing Wikimedia installation... : Weekly mobile deployment [21:28:06] bsitu: ^d says no.... [21:28:12] Logged the message, Master [21:28:37] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [21:29:04] <^d> bsitu: It's not in /data/project/... or in /a/common/... [21:29:09] bsitu: so, how was that code deployed on production? it should just work on beta cluster... [21:29:13] <^d> (So, not on deployment-bastion *at all*) [21:29:37] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 9.862 second response time [21:29:39] a glitch in updates on beta? [21:29:46] *the matrix [21:29:49] I think I encountered this before [21:30:18] greg-g: I reverted this: https://gerrit.wikimedia.org/r/#/c/74666/ [21:30:25] greg-g: then updated Echo to master [21:30:43] greg-g: it rides on mobile's scap [21:32:27] (03PS1) 10Dzahn: disable updatequerypages cron on hume, instead enable on new maintenance host terbium [operations/puppet] - 10https://gerrit.wikimedia.org/r/76835 [21:32:53] bsitu: so echo's master is what is on production, so taht is what should be on beta cluster as well.... [21:33:03] * greg-g doesn't know why it is different [21:33:49] Nemo_bis, nice to see https://gerrit.wikimedia.org/r/#/c/33713/ got merged. Does this mean the special pages that got disabled by misermode are going to start getting updated? [21:34:17] Krenair: yes, very slowly for now [21:34:31] not the 6 pages disabled only on en.wiki though [21:35:00] greg-g: is beta running on 1.22wmf12? [21:35:05] Which ones are they? [21:35:16] ask Elsie's bug [21:35:17] bsitu: beta cluster runs master [21:35:25] bsitu: updates master every 3 minutes [21:35:33] Hello Elsie. [21:36:10] (03CR) 10Dzahn: [C: 032] disable updatequerypages cron on hume, instead enable on new maintenance host terbium [operations/puppet] - 10https://gerrit.wikimedia.org/r/76835 (owner: 10Dzahn) [21:36:19] bsitu: so, you said that echo's master is deployed on production, which is what beta cluster is running, can you see if there is a difference between the two, please? [21:36:44] * Nemo_bis hopes mutante is not going to hate him [21:38:00] heh, nah, they are just supposed to appear on both nodes, one enabled and one disabled [21:38:05] Nemo_bis, mostlinked, mostrevisions, fewestrevisions, wantedpages, mostlinkedtemplates, ancientpages based on comment 59? [21:39:04] hmm, it did create them on both ... [21:39:21] ah, no, it didnt [21:39:34] removed on one, created on the other as it should [21:39:56] bsitu: you might be off the hook :) [21:39:57] (03CR) 10Dzahn: "needed this fix, didn't have $enabled yet and that temp. broke puppet. https://gerrit.wikimedia.org/r/#/c/76831/ and https://gerrit.wiki" [operations/puppet] - 10https://gerrit.wikimedia.org/r/33713 (owner: 10Nemo bis) [21:40:52] bsitu: ^d saw that there were some uncommited changes on beta, looks like someone was doing live-hack testing and deleted that hooks.php file [21:40:57] ^d gets a cookie [21:41:06] greg-g: the file on tin is the same as current master [21:41:26] Nemo_bis: http://paste.debian.net/20408/ [21:41:39] bsitu: you don't happen to know if someone was live testing on beta from your team just now, do you? [21:42:16] greg-g: I don't think so [21:42:22] (03CR) 10Dzahn: "# Puppet Name: cron-updatequerypages-mostrevisions-s5@15" [operations/puppet] - 10https://gerrit.wikimedia.org/r/33713 (owner: 10Nemo bis) [21:42:40] bsitu: now I get to send a fun email to engineering then ;) [21:43:05] mutante: log filenames are ugly [21:43:12] greg-g: beta seems ok for me now [21:43:25] chrismcmahon: yeah, ^d fixed it (undeleted the files) [21:43:35] it seems gerrit has to learn newlines [21:43:47] chrismcmahon: you weren't deleting files on beta were you? :) [21:43:48] ^d rawks SO HARD [21:43:54] greg-g: I was not [21:43:58] good [21:43:58] :P [21:44:31] greg-g: Our team doesn't use beta lab, I don't even know how to log into betalab [21:45:04] (03CR) 10Dzahn: "http://paste.debian.net/20408/" [operations/puppet] - 10https://gerrit.wikimedia.org/r/33713 (owner: 10Nemo bis) [21:46:06] bsitu: you shouldn't need to login to it's servers, really. But you should use the beta cluster more, though :) [21:46:18] it's the best testing environment we have :) [21:46:29] greg-g: yeah, I definitely should [21:49:45] ori-l: feature request: a scap simulator [21:50:26] ori-l: ie: for those who really really want to feel the power of scap during their own deploys, but we've moved them to riding the train, they can still get a high from typing scap and seeing errors scroll by (^d's idea) [21:50:44] (03PS8) 10Dzahn: Simplify wikimania apache conf [operations/apache-config] - 10https://gerrit.wikimedia.org/r/62566 (owner: 10Reedy) [21:50:46] (03PS5) 10Reedy: Fixup docroot code to work for wikimanias all from one docroot folder [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/67341 [21:51:21] <^d> greg-g: We'll call it the "MethodOne Scapping Clinic." [21:51:22] Reedy: was that a reaction or coincidence [21:51:50] mutante: reaction. It "needed" rebasing ;) [21:52:23] heh, ok, let's do that before wikimania [21:52:47] when most people go to the site :) [21:53:30] greg-g: http://hackertyper.com/ [21:53:59] ++ "poor man / hacky routing engine " :) [21:54:13] mutante: I think those 2 commits need pushing live sort of simultaneously [21:55:11] i guess .. MWMultiVersion.php first [21:55:23] but yea, can do [21:56:24] ori-l: yes! [21:57:15] Reedy: $site = "wikipedia" and $lang === 'wikimania' .. it seems confusing [21:57:17] ori-l: blugh, I take it back, they link to a macbook air [21:57:24] but i know you'll know [21:57:57] Dividing by site/lang or family/lang is stupid and should never be done. [21:58:02] Way too many edge cases. [21:58:08] mutante: Yup.. [21:58:09] reedy@tin:~$ mwscript eval.php wikimania2013wiki [21:58:09] > var_dump( $site, $lang ); [21:58:09] string(9) "wikipedia" [21:58:09] string(13) "wikimania2013" [21:58:18] I thought it was wiki, not wikipedia. [21:58:31] no [21:58:45] wikimania2013 is an awfully silly language. [21:59:07] Elsie, are you coming btw? [21:59:17] So is mediawiki [21:59:20] MaxSem: I wasn't invited! [21:59:22] (No.) [21:59:45] Maybe next year. [22:00:36] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [22:01:38] so, can we hold off with the fires for a little bit, I want to get some work done now [22:02:26] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.128 second response time [22:03:41] <^d> greg-g: I think I'm gonna run scap a half dozen times. [22:03:43] <^d> Just for shits. [22:04:36] * greg-g giggles [22:08:40] <^d> manybubbles|away: The new memcached is live for beta. Re-running indexing so I can try to populate the parser cache a bit. [22:10:22] (03PS2) 10Reedy: Delete all the superfluous wikimania docroots [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/62565 [22:10:22] (03PS3) 10Reedy: Delete all the superfluous wikimania docroots [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/62565 [22:12:01] mutante: When do you want to do it? [22:17:11] That sounded weird. [22:22:32] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [22:23:22] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.134 second response time [22:44:09] did you just change something? [22:44:22] I have a "I got new messages" stuck [22:44:57] ^d: memcached you say? [22:45:03] bsitu, ^ [22:45:17] AzaToth: hmm, we thought we found the issue and reverted it :( [22:45:26] Related to https://gerrit.wikimedia.org/r/#/c/76824/ ? [22:45:36] ebernhardson: now it's gone again though [22:45:48] ebernhardson: haven't had a edit to my talk page since 24:th [22:46:58] AzaToth: bah, now its a ghost bug [22:47:05] AzaToth: https://gerrit.wikimedia.org/r/#/c/74666/ was the only change to the orange bar, but it has been reverted [22:47:28] bsitu: i see the orange bar on en.wo right now, too [22:47:30] I see [22:47:33] on https://en.wikipedia.org/wiki/Bulbophyllum_dusenii, fwiw [22:47:50] I don't have the orange bar any more though [22:47:56] (i didn't force-purge my cache) [22:48:13] I wonder if ctrl+shift+r fixes it AzaToth [22:48:30] bsitu: if there's some debugging you want me to do while i'mseeing it, i'mall yours [22:48:37] Krenair: can't tell as it's gone now [22:49:01] was probably as ebernhardson said a ghost bug [22:49:46] * MatmaRex is seeing ghosts [22:50:06] MatMaRex: Thanks [22:50:56] i'm still getting the version of the echo alert script which was reverted [22:50:58] if (!mw.config.get('echoNewMsgAlertDisplayed')) { [22:51:10] maybe you guys need to touch the files or something [22:52:12] I would assume it could be some apaches or something that takes time to sync or something [22:52:19] Hmm, this is what I see: if(mw.config.get('wgUserNewMsgRevisionId')&&!mw.config.get('echoNewMsgAlertDisplayed')) [22:52:31] i'll do a soft refresh now [22:52:58] i went to a random page, https://en.wikipedia.org/wiki/Ambodimahabibo , and it's still there [22:53:08] What's there? [22:53:17] phantom orange bar [22:53:19] got it again now [22:53:37] bsitu: mw1034 [22:53:44] Is there a bug about this? [22:53:47] And a screenshot? [22:54:27] Elsie: http://i.imgur.com/FibiQ0Z.png [22:54:45] Well, at least it's the new orange bar. :-) [22:54:55] It'd be more interesting if you were getting the old orange bar. [22:55:25] I can touch and re-sync the files [22:55:33] Elsie: :-P [22:56:29] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [22:57:54] bsitu: ping me when you want me to recheck [22:58:09] MatmaRex: thx [22:58:11] will do [22:58:19] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.123 second response time [23:03:22] <^d> AzaToth: Hm? [23:04:23] ^d: was some ghosts of bsitu [23:04:39] AzaToth: :) [23:04:42] indeed [23:05:20] greg-g: can I use the LD window just to touch a file? [23:05:49] bsitu: sure, RoanKattouw is doing something officially in it [23:05:51] Go ahead [23:05:53] I'm not ready yet [23:05:58] RoanKattouw: thx [23:09:31] !log bsitu synchronized php-1.22wmf11/extensions/Echo/modules/alert/ext.echo.alert.js 'Refresh js file' [23:09:41] Logged the message, Master [23:10:17] !log bsitu synchronized php-1.22wmf12/extensions/Echo/modules/alert/ext.echo.alert.js 'Refresh js file' [23:10:28] Logged the message, Master [23:10:31] bsitu: https://en.wikipedia.org/wiki/Wikipedia:Village_pump_%28technical%29#Random_.22Talk:_you_have_new_messages.22_messages.3F [23:10:32] fyi [23:11:00] oops, forget to touch the file, I am just not myself today [23:11:04] :( [23:11:21] hahahahaha [23:11:53] bsitu: I hope you reply to all of them [23:13:06] !log bsitu synchronized php-1.22wmf11/extensions/Echo/modules/alert/ext.echo.alert.js 'Refresh js file' [23:13:40] !log bsitu synchronized php-1.22wmf12/extensions/Echo/modules/alert/ext.echo.alert.js 'Refresh js file' [23:16:25] (03PS1) 10Dzahn: re-enable (non-indexed) full-text search on RT [operations/puppet] - 10https://gerrit.wikimedia.org/r/76854 [23:16:59] AzaToth: it will take me a while, :( [23:17:10] hehe [23:17:13] MatmaRex: do you still see the outdated js file [23:17:27] checking [23:17:41] still visible right now, let's give it five minutes to purge the caches? [23:17:47] (03CR) 10Dzahn: [C: 032] "http://blog.bestpractical.com/2011/06/full-text-searching.html" [operations/puppet] - 10https://gerrit.wikimedia.org/r/76854 (owner: 10Dzahn) [23:17:52] also, this has just gotten reported, gotta be closed later… https://bugzilla.wikimedia.org/show_bug.cgi?id=52301 [23:17:54] bsitu: https://bugzilla.wikimedia.org/show_bug.cgi?id=52301 [23:18:05] Though that's probably a dupe at this point... [23:18:14] (03PS1) 10Tim Starling: Send rel=canonical pointing from HTTPS to HTTP [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76856 [23:22:37] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [23:22:39] (03CR) 10Ryan Lane: [C: 031] Send rel=canonical pointing from HTTPS to HTTP [operations/mediawiki-config] - 10https://gerrit.wikimedia.org/r/76856 (owner: 10Tim Starling) [23:23:27] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.130 second response time [23:23:34] bsitu: yay, seems to be fixed now [23:23:49] or at least i don't see any more notifications (no forceful cache purging) [23:25:42] MatmaRex: thanks! I will see if others still experience this ghost bug [23:31:32] bsitu: You done now? [23:31:43] RoanKattouw: yes [23:31:52] Awesome, then I'll go [23:34:20] !log catrope synchronized php-1.22wmf11/extensions/VisualEditor 'Update VE to master' [23:34:32] Logged the message, Master [23:34:45] !log catrope synchronized php-1.22wmf12/extensions/VisualEditor 'Update VE to master' [23:34:55] Logged the message, Master [23:45:52] https://bugzilla.wikimedia.org/show_bug.cgi?id=36497 you bitch [23:46:13] This bug drives me crazy so much, grrr :-( [23:51:16] PROBLEM - SSH on pdf3 is CRITICAL: CRITICAL - Socket timeout after 10 seconds [23:52:36] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [23:53:26] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.126 second response time [23:54:56] PROBLEM - Puppet freshness on lvs1004 is CRITICAL: No successful Puppet run in the last 10 hours [23:54:56] PROBLEM - Puppet freshness on lvs1005 is CRITICAL: No successful Puppet run in the last 10 hours [23:54:56] PROBLEM - Puppet freshness on lvs1006 is CRITICAL: No successful Puppet run in the last 10 hours [23:54:56] PROBLEM - Puppet freshness on virt1 is CRITICAL: No successful Puppet run in the last 10 hours [23:54:56] PROBLEM - Puppet freshness on virt3 is CRITICAL: No successful Puppet run in the last 10 hours [23:54:57] PROBLEM - Puppet freshness on virt4 is CRITICAL: No successful Puppet run in the last 10 hours [23:58:16] RECOVERY - SSH on pdf3 is OK: SSH OK - OpenSSH_4.7p1 Debian-8ubuntu3 (protocol 2.0)