[00:01:47] it lines up with the switch to redis [00:02:00] actually the post-redis curve looks broken for recycling [00:02:23] since recycling should 1hr after failure, it shouldn't line up with pops like that [00:02:30] * robla reads backlog to see how l10nupdate deploy stuff is going, but wouldn't mind spoiler alert :) [00:02:31] it's like things recycle too fast [00:03:14] doesn't really hurt anything now, but the point is to wait longer in case jobs fail due to transient things (like mail server packet loss) [00:03:21] * Aaron|home looks at some code [00:03:43] New patchset: Andrew Bogott; "Added role::lamp::labs" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59561 [00:04:07] Change merged: Andrew Bogott; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59561 [00:07:51] binasher: hmm, maybe it's actually fine [00:08:21] recycling is every 30 min, so I'd expect 2 evenish spikes/hour [00:10:08] ahh, and there is the mt_rand() on pop() for recycleAndDeleteStaleJobs(), so it would go up some when pop() does [00:10:10] meh [00:14:11] binasher: how much ram is on mw1001-mw1016? [00:14:20] seems runJobs uses a 150M limit [00:14:35] Aaron|home: 12G [00:15:09] I think that can be bumped :) [00:16:44] root@hume:~# getent passwd l10nupdate [00:16:44] l10nupdate:x:998:10002::/home/l10nupdate:/bin/false [00:17:26] it must be puppet one one server fighting with puppet on the other server over NFS [00:18:14] ooooh. /etc is on nfs ?? [00:18:32] or just changing permissions on the file i guess [00:18:45] /home/l10nupdate/.ssh/authorized_keys is on NFS [00:19:14] puppet fight: http://www.killerspoons.com/wp-content/uploads/2012/10/78129203861113742.jpg [00:19:37] right. i first thought they were fighting over what numeric id the l10nupdate user would have for the whole system (not just that file) [00:19:39] thanks for your input, robla ;) [00:20:02] You're welcome! Glad I could contribute [00:20:03] :) [00:22:42] more to the point, /home/l10nupdate/.ssh is on NFS, so if it has the wrong permissions, l10nupdate fenari can't read /home/l10nupdate/.ssh/id_rsa which is its passwordless private key [00:23:29] Oh lovely [00:23:44] New patchset: Andrew Bogott; "Include webserver::php5 in the lamp role." [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59563 [00:24:07] Change merged: Andrew Bogott; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59563 [00:25:12] New patchset: Aaron Schulz; "Doubled the memory limits for job runners." [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59564 [00:26:32] notpeter: https://gerrit.wikimedia.org/r/#/c/59564/1 [00:27:22] RoanKattouw: this is only half the reason why it doesn't work, the other half is the fact that the shell is set to /bin/false on all apaches [00:27:44] Ouch [00:27:54] This wasn't always the case, it must've changed at some point [00:28:12] has no one noticed the non-updating? [00:28:26] well, the files will be pushed out anyway when you run scap [00:29:19] PROBLEM - Puppet freshness on cp3003 is CRITICAL: No successful Puppet run in the last 10 hours [00:30:47] !log on hume: changed UID for l10nupdate to be 10002, for consistency with puppet admins.pp [00:30:59] Logged the message, Master [00:31:22] that private key shouldn't be on NFS anyway, it's insecure [00:38:26] !log blog: enabled new eventlogging by ori, removed old logging class, converted plugins/WMBlog from svn to git and sync with gerrit [00:38:35] Logged the message, Master [00:39:08] bbl. need to get to dinner and battery dying [00:39:09] mutante: weee! [00:39:28] don't eat the battery! [00:39:32] even if it's dying. [00:39:35] there are better options. [00:39:41] :) [00:42:06] hello, please can you help me? I want to download a music score from IMSLP, but the download is not allowed in my country: http://imslp.org/wiki/Lemmink%C3%A4inen_in_Tuonela,_Op.22_No.3_(Sibelius,_Jean) . Please, can you download it and then send it to me through irc? many thanks [00:45:53] New patchset: Aaron Schulz; "Added back duplicate insert job stats since it has activity now." [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59567 [00:47:01] faLUCE: this is the wrong place to ask. IMSLP is not affiliated with the Wikimedia Foundation. [00:47:43] Jasper_Deng: I know, but I need help [00:47:59] still the wrong place to ask [00:48:17] faLUCE: try #imslpchat [00:48:38] ori-l: thnks [00:57:09] TimStarling: maybe you can look at https://gerrit.wikimedia.org/r/#/c/59564/ ? [00:57:36] or I can just bug peter tomorrow [00:58:18] New patchset: Tim Starling; "Doubled the memory limits for job runners." [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59564 [00:58:53] Change merged: Tim Starling; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59564 [00:59:45] thanks [01:28:15] greg-g: so hashar will about jenkins right? [01:28:22] (e.g. emailed already) [01:33:29] Aaron|home: is it still broken? i've seen it being more sane [01:34:07] I getting -1's for everything [01:34:19] hrmmmm [01:34:39] well the things that actually have a response [01:35:13] right [01:36:20] Aaron|home: i saw stuff was now working better so i thought maybe chad did something. i guess not? [01:36:27] Aaron|home: anyway, see chad's change. 59556 [01:40:11] New review: MZMcBride; "Thanks for this." [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59441 [01:43:03] PROBLEM - Puppet freshness on virt3 is CRITICAL: No successful Puppet run in the last 10 hours [02:10:22] !log LocalisationUpdate completed (1.22wmf1) at Wed Apr 17 02:10:22 UTC 2013 [02:10:31] Logged the message, Master [02:19:57] !log LocalisationUpdate completed (1.22wmf2) at Wed Apr 17 02:19:57 UTC 2013 [02:20:05] Logged the message, Master [02:21:33] TimStarling: I saw your comments in scrollback about LocalisationUpdate inappropriately logging success. Is that filed as a bug? [02:25:47] no [02:27:18] Okay, I'll file a bug. [02:32:12] https://bugzilla.wikimedia.org/show_bug.cgi?id=47301 [03:13:57] !log LocalisationUpdate ResourceLoader cache refresh completed at Wed Apr 17 03:13:56 UTC 2013 [03:14:04] Logged the message, Master [03:20:07] PROBLEM - Puppet freshness on ms-fe3001 is CRITICAL: No successful Puppet run in the last 10 hours [04:29:03] bit.ly/wikisal [04:29:26] LocalisationUpdate purges RL cache now? [04:29:36] That's new isn't it? [04:35:11] New patchset: Liangent; "(bug 47305) Add /zh-mo as an alias for wikipedia.org/w/index.php" [operations/apache-config] (master) - https://gerrit.wikimedia.org/r/59580 [04:49:59] New patchset: Isarra; "(Bug 47299) Update MediaWiki.org favicon" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59589 [04:50:59] New patchset: Isarra; "(Bug 47299) Update MediaWiki.org favicon" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59589 [04:51:47] New patchset: Isarra; "(Bug 47299) Update MediaWiki.org favicon" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59589 [05:00:32] PROBLEM - SSH on cp1043 is CRITICAL: Server answer: [05:00:32] PROBLEM - SSH on gadolinium is CRITICAL: Server answer: [05:01:32] RECOVERY - SSH on gadolinium is OK: SSH OK - OpenSSH_5.9p1 Debian-5ubuntu1 (protocol 2.0) [05:02:32] RECOVERY - SSH on cp1043 is OK: SSH OK - OpenSSH_5.9p1 Debian-5ubuntu1 (protocol 2.0) [05:05:29] PROBLEM - SSH on cp1043 is CRITICAL: Server answer: [05:06:29] RECOVERY - SSH on cp1043 is OK: SSH OK - OpenSSH_5.9p1 Debian-5ubuntu1 (protocol 2.0) [06:00:43] PROBLEM - Puppet freshness on lvs1005 is CRITICAL: No successful Puppet run in the last 10 hours [06:00:43] PROBLEM - Puppet freshness on lvs1004 is CRITICAL: No successful Puppet run in the last 10 hours [06:00:43] PROBLEM - Puppet freshness on lvs1006 is CRITICAL: No successful Puppet run in the last 10 hours [06:47:38] !log LocalisationUpdate completed (1.22wmf1) at Wed Apr 17 06:47:37 UTC 2013 [06:47:46] Logged the message, Master [06:51:32] !log LocalisationUpdate completed (1.22wmf2) at Wed Apr 17 06:51:28 UTC 2013 [06:51:36] Logged the message, Master [06:57:02] !log LocalisationUpdate ResourceLoader cache refresh completed at Wed Apr 17 06:57:02 UTC 2013 [06:57:09] Logged the message, Master [07:04:28] PROBLEM - Puppet freshness on virt1005 is CRITICAL: No successful Puppet run in the last 10 hours [07:35:51] New review: MaxSem; "Since it breaks the b/c anyway, why don't we just remove these files completely? Especially since No..." [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59443 [07:53:46] New patchset: Ori.livneh; "Add initial 'eventlogging::notebook' class" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59596 [08:42:29] mark: the lame issue I was fighting with yesterday magically solved :-] [08:42:39] uh? [08:42:46] it didn't reload the manifests? [08:42:51] puppet no more attempt to mount sda3 / sdb3 for the mobile cache [08:43:06] I suspect the replication of ops/puppet on labs was broken [08:45:20] likely [08:48:15] mark: and I have deployed your mobile host rewrites patch ( https://gerrit.wikimedia.org/r/#/c/59401/ ) on beta seems to work [08:48:51] cool [08:49:14] it seemed a bit nicer that way [08:49:52] so if you feel brave enough, I think you can get them merged :-] [08:50:16] hehe [08:50:18] both of them? [08:50:22] I did a followup change on it [08:50:24] did you test that one too? [08:51:00] https://gerrit.wikimedia.org/r/#/c/59401/ only has one patchset [08:51:24] which depends on https://gerrit.wikimedia.org/r/#/c/47567/11 (latest patchset) [08:51:27] yes that's the followup [08:51:27] that includes your change [08:51:31] so I guess that is fine :-] [08:51:44] cool [08:55:20] anyone wants to look at rt 4868 / https://gerrit.wikimedia.org/r/#/c/58520/ would be great to get the new libvpx deployed soon, failed encodes are piling up because of this [08:56:49] New patchset: Mark Bergsma; "Use /wiki/Main_Page for the mobile URL check." [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59600 [08:58:27] Change merged: Mark Bergsma; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59600 [09:01:40] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [09:02:30] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.125 second response time [09:10:42] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [09:11:32] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.125 second response time [09:27:15] Change merged: Mark Bergsma; [operations/debs/libvpx] (master) - https://gerrit.wikimedia.org/r/58520 [09:48:53] New patchset: Hashar; "preparing package to be uploaded to the debian repo" [operations/debs/python-voluptuous] (master) - https://gerrit.wikimedia.org/r/59605 [09:49:48] paravoid: is our Debian mighty guru around ? :-D I got a debian/changelog question for you and how to support both Debian unstable and our precise-wikimedia distribution :-] the lame change is https://gerrit.wikimedia.org/r/59605 [09:50:27] !log Inserted libvpx 1.1.0-1+wmf1 packages into the precise-wikimedia APT repository [09:50:34] Logged the message, Master [09:54:37] j^: installing that package on tmh* now [10:08:27] New patchset: Mark Bergsma; "Revert "Revert "Rm special-casing for root URLs of mobile sites""" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59606 [10:08:52] Change merged: Mark Bergsma; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59606 [10:09:05] i think i'll merge the other mobile changes tomorrow [10:09:11] I'm on the road a lot this afternoon and tonight [10:12:01] mark: works for me :-] [10:29:59] PROBLEM - Puppet freshness on cp3003 is CRITICAL: No successful Puppet run in the last 10 hours [10:30:39] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [10:31:39] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 5.611 second response time [10:49:04] hashar: if you want this to be uploaded to Debian, you should drop all the changelog entries but the latest one [10:49:17] and make it -1 [10:49:43] 0.6.1-1 that is [10:49:50] (and that's why we do ~wmf1...) [10:50:13] and you should add "* Initial uploaded (Closes: #NNNNN)", with the ITP bug [10:54:28] paravoid: Apollon "apoikos" Oikonomopoulos told me so in #debian-python :-] [10:54:32] paravoid: thx for the confirmation! [10:54:43] btw I am converting the package-builder misc class to a module [10:54:54] haha [10:54:56] will add support for Debian sid chroot too [10:55:12] I guess you know Apollon :-] [10:55:17] apollon is a long-time friend of mine, we shared an office at grnet :) [10:56:19] he started being more active in Debian lately and I've been sponsoring his packages to [10:56:23] *too [11:04:44] https://integration.wikimedia.org/ci/job/operations-mw-config-tests/2261/console [11:04:53] https://integration.wikimedia.org/ci/job/operations-mw-config-tests/2262/console [11:04:56] o_0 [11:43:10] PROBLEM - Puppet freshness on virt3 is CRITICAL: No successful Puppet run in the last 10 hours [11:49:15] Change merged: Faidon; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59454 [11:57:40] PROBLEM - Puppetmaster HTTPS on stafford is CRITICAL: CRITICAL - Socket timeout after 10 seconds [11:59:08] New patchset: Hashar; "convert package-builder to a module" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59611 [11:59:23] paravoid: one more misc to module migration :-] ^^^^ [11:59:30] RECOVERY - Puppetmaster HTTPS on stafford is OK: HTTP OK: Status line output matched 400 - 336 bytes in 0.123 second response time [11:59:31] woo [11:59:44] I have no idea whether it works though [12:01:16] pbuilder is a bit confusing [12:01:32] pbuilder is just the name of one piece of software which is used for package building [12:01:51] but then most building suites are referred to as "pbuilder" [12:01:53] New patchset: Demon; "Swap string "true" for boolean true in $ganglia_aggregator" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59613 [12:02:08] why not just package-builder? [12:02:41] oh my [12:02:42] god [12:02:44] I hate puppet [12:02:53] I should migrate the lame definion to a shell script [12:03:57] I wanted our class to support creating images for debian-unstable [12:06:53] New review: Faidon; "(7 comments)" [operations/puppet] (production) C: -1; - https://gerrit.wikimedia.org/r/59611 [12:11:30] New review: Mark Bergsma; "(2 comments)" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59611 [12:12:02] see [12:12:11] also paravoid confuses it for a pbuilder module instead of what it is, a package builder role [12:12:29] New patchset: Odder; "(bug 44164) Add 'Portal' and 'Author' namespaces to iswikisource" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59614 [12:13:11] there's a role class on this change as well [12:13:24] and this is where lucid/precise should go imho [12:14:45] yeah [12:14:48] paravoid: doing that [12:14:57] but then I have to add the $defaultdist in the role class too [12:15:14] that's fine [12:15:24] while you're at it, I'd probably parameterize the mirror as well [12:15:37] and components/suites [12:16:54] as for being a generic "package builder" or a pbuilder class, yes, I guess I'm confused too :) [12:18:07] yeah, moving such things to the role class is fine of course [12:18:54] and hashar [12:18:55] what's so bad about NRPE public? ;) [12:19:42] http://people.canonical.com/~ubuntu-security/cve/2013/CVE-2013-1362.html [12:19:58] yeah but we have that disabled don't we? [12:20:00] we do [12:20:12] * paravoid double-checks [12:20:27] so my stance in the past few years has been that NRPE is fine for a basic case of commands reporting non-private status output [12:20:27] yes [12:20:30] without arguments [12:20:38] as long as people don't do stupid things with that, it should be fine [12:20:52] of course, we have so many people now, it's hard to monitor people doing stupid things these days :/ [12:21:18] nrpe has a history of being badly maintained and not a great security track record (iirc) [12:21:30] yeah I know [12:21:34] I don't particularly like it either [12:21:57] apparently it's in main though! [12:22:04] hehe [12:22:17] paravoid: so each define should be in its own file? :D [12:22:19] nagios people suck at security [12:23:17] hashar: yeah... [12:26:52] New review: Hashar; "(6 comments)" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59611 [12:33:33] New patchset: Hashar; "convert package-builder to a module" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59611 [12:34:20] New review: Hashar; "* addresses various issues reported by Faidon on PS1" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59611 [12:34:33] mark: paravoid: I am not sure how to name the module though :( [12:36:40] "package-builder" [12:36:46] like how I named the misc class from the start? :) [12:37:18] ah [12:37:38] paravoid: what do you think about naming the module 'package-builder' ? [12:37:46] (I don't want to cause an edit war among ops) [12:38:41] what is the issue in the first place? that's what the original class was called [12:38:45] why would it be a problem now? :) [12:41:55] simply making sure Faidon is not going to request to change the name back or to something else hehe [12:46:50] will rename it [12:46:52] faidon is lost :-] [12:50:35] New patchset: Hashar; "convert package-builder to a module" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59611 [12:50:57] New review: Hashar; "renamed the module from 'pbuilder' to 'package-builder' per mark" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59611 [12:51:28] getting pages [12:51:44] yep [12:52:16] all ipv6 esams [12:52:44] fastest flapping evah [12:53:02] New review: ArielGlenn; "I was trying to figure out why, since ^((?!www).+\.|)wikidata.org includes the case =wikidata.org w..." [operations/apache-config] (master) - https://gerrit.wikimedia.org/r/49069 [12:54:03] ooof course, icinga is down [12:54:16] something broke it :( [12:54:17] right when you need it [12:55:17] wasn't me *puts screwdriver behind back* [12:58:22] jeremyb_: https://bugzilla.wikimedia.org/show_bug.cgi?id=47315 <-- it's definitely better this time :p [12:59:08] is someone interested in mobile version? I think I found a bug some days ago, though I'd like to make some tests before filling one on bugzilla [12:59:46] Vito: #wikimedia-mobile [13:00:10] Vito: event is 12-17? so really 14-19? what about buffer? [13:02:00] I'll restart it [13:02:59] !log jenkins: refreshing job mediawiki-core-phpcs-HEAD (points to a wrong git repo path) [13:03:07] Logged the message, Master [13:04:36] !Log restarted icinga on neon (the usual: Could not create external command file '/var/lib/nagios/rw/nagios.cmd' as named pipe) [13:04:39] grrr [13:04:44] Logged the message, Master [13:04:59] oh yay someone did the capital letter fix! [13:05:04] whoever they are, bless them [13:09:11] Vito: ? [13:09:58] New patchset: Jeremyb; "add account abaso and add to mortals (RT-4956)" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59453 [13:21:03] PROBLEM - Puppet freshness on ms-fe3001 is CRITICAL: No successful Puppet run in the last 10 hours [13:30:30] New review: Jeremyb; "well even if there was a reason to have both lines it wouldn't work because redirects.conf only has ..." [operations/apache-config] (master) C: -1; - https://gerrit.wikimedia.org/r/49069 [13:35:30] lol, # Uploads are offsite (except on yaseo) [13:40:20] aude: ping [13:41:10] or Silke_WMDE ? [13:41:20] here [13:41:34] aude might still be in a meeting [13:42:14] Silke_WMDE: so we're redirecting wikidata to www. what about langcodes? [13:42:24] redirect? 301 or 302? [13:42:50] ööhhh [13:43:10] * Silke_WMDE is not part of the Wikidata team any more. [13:43:19] yeah, i can never remember [13:43:22] Lydia_WMDE: ^ :) [13:43:33] Silke_WMDE: TS and what else? [13:43:51] jeremyb_: Internal Office IT at WMDE [13:44:09] (but TS mainly these days) [13:44:12] ahhh, you're like yossie kinda. i think [13:44:20] yossie? [13:47:16] err, having some trouble finding him [13:48:33] So I'm establishing IT-related office processes. [13:49:40] jeremyb_: (and I have to admit that I miss my labs instances a bit ;) ) [13:49:51] Silke_WMDE: you should try out ganeit [13:49:54] ganeti* [13:50:36] and depending on what you're doing could still use labs a bit [13:50:37] Will we meet at the Amsterdam Hackathon? [13:51:00] Silke_WMDE: i haven't made any plans to be there yet so probably not :) [13:51:16] * jeremyb_ waves to the south african [14:05:58] New patchset: Odder; "(bug 47315) Add $wmfThrottlingExceptions for it.wiki GLAM event" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59625 [14:07:37] New patchset: Jeremyb; "(bug 45005) Redirect wikidata.org to www.wikidata.org" [operations/apache-config] (master) - https://gerrit.wikimedia.org/r/49069 [14:08:22] aude: apergos: see last push ^^ [14:09:46] looking [14:10:18] ah ha [14:10:19] ok thanks [14:11:45] ok well that's pretty hilarious after 10 other patchsets [14:11:49] anyways let's see what folks say [14:12:17] apergos: we discussed it some in #wikimedia-wikidata. i think it's good to go but of course Denny_WMDE, etc. could give a +1 [14:12:38] New review: Faidon; "(5 comments)" [operations/puppet] (production) C: -1; - https://gerrit.wikimedia.org/r/59611 [14:12:57] errr [14:13:02] huh [14:13:11] oh, i didn't go far enough [14:13:17] jeremyb_, did you see https://bugzilla.wikimedia.org/show_bug.cgi?id=47276 [14:13:34] (nvm, all is good) [14:13:40] New review: Denny Vrandecic; "I cannot comment on the implementation, but I give a +1 on the intent. We want wikidata.org to be al..." [operations/apache-config] (master) C: 1; - https://gerrit.wikimedia.org/r/49069 [14:14:11] 10 patchsets for a redirect? seriously? :) [14:14:20] jeremyb_: gave it a +1 [14:14:50] Thehelpfulone: just a policy issue... whatever people want [14:15:01] redirects are hard paravoid ;) [14:15:12] Thehelpfulone: https://rt.wikimedia.org/Ticket/Display.html?id=4830 [14:15:18] New patchset: Hashar; "package builder now supports Debian.org unstable" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59626 [14:17:04] thanks, didn't know it was already in gerrit, I'll poke a couple of people and see what they've got to say [14:17:54] hashar: shouldn't dists include precise too? [14:18:07] Thehelpfulone: well that's a little different. but i guess could be simultaneously decided [14:18:35] hashar: also, if we're going that road with non-wikimedia dists, then you should use "lucid-wikimedia" as a dist (you might have precise later) [14:19:46] paravoid: defaultdist does provide precise :D [14:19:53] paravoid: though I have to test that out in labs [14:21:15] apergos or paravoid, can you merge https://gerrit.wikimedia.org/r/#/c/57647/ please? [14:23:28] boy E: Release signed by unknown key (key id AED4B06F473041FA) [14:23:44] New patchset: Demon; "Don't link to draft patches" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59628 [14:24:23] hashar: old box? [14:24:39] jeremyb_: na an ubuntu box attempting to install debian packages :D [14:24:49] hashar: haha [14:25:12] I guess the debian-archive-keyring in Ubuntu is not up to date [14:25:35] well are you using sid packages? :) [14:25:52] jeremyb_: yeah I am trying to build a Debian/sid chroot under a Ubuntu/Precise host [14:26:14] hrmmm, debootstrap? [14:26:18] debian-archive-keyring *** 2010.08.28 0 [14:26:23] I am using cowbuilder [14:26:32] which indeed calls debootstreap [14:26:39] i guess if it's a throwaway host then security doesn't matter so much? [14:26:50] you could just download the new package and install it [14:26:52] Iwill get the package back ported :-] [14:26:59] t says needs rebase or has dependency and won't let me submit MaxSem [14:27:08] owchie [14:28:27] apergos: err? that's a lie [14:28:39] it's not a lie for me [14:28:46] ## production...origin/production [ahead 1] [14:28:48] gah [14:28:50] maybe you can do it [14:29:19] anyway, it's a fastforward! [14:29:26] New patchset: MaxSem; "Send Zero notifications to #wikimedia-mobile" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/57647 [14:29:34] rebased^^^ [14:29:50] yes and that one doesn't whine [14:30:32] ohhhh [14:30:46] waiting for jenkinsbot to verify... come one you slow thing [14:30:52] you know it helps so much if we're all looking at the same changeset??? [14:30:55] :-( [14:31:15] hahaha [14:31:18] nice :-) [14:31:21] * jeremyb_ clicked the link for 59628 above by accident (2 lines below where MaxSem said it) [14:31:26] woops [14:31:27] it's a perfect fast forward! [14:31:34] :-D imagine that [14:32:00] whyyyy is jenkins not doing its thing [14:32:02] so we're just in alternate universes is all... [14:32:19] do you have the evil spock beard? enquiring minds want to know [14:32:37] no [14:32:50] bummer.. it was a cute beard [14:33:36] I'm gonna verify this hing in a sec if jenkins doesn't get its *ss in gear [14:33:48] yayyy [14:33:55] Change merged: ArielGlenn; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/57647 [14:34:29] and done [14:34:46] (not going to run it though, so patience young grasshopper til puppet goes around) [14:35:51] New patchset: Hashar; "package builder now supports Debian.org unstable" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59626 [14:36:13] thanks apergos [14:36:26] yw [14:37:23] jeremyb_: looks good to me [14:37:27] ahhh, that's why my paste failed before, 17 14:28:46 -!- Irssi: Unknown command: wmf-puppet$ [14:37:48] aude: +1 then? :) [14:37:58] yeah [14:38:06] it handles just wikidata.org -> www.wikidata.org [14:38:09] seems good [14:38:31] New review: Aude; "thanks jeremyb :) looks good" [operations/apache-config] (master) C: 1; - https://gerrit.wikimedia.org/r/49069 [14:39:01] * jeremyb_ runs away [14:39:18] for what is in main.conf, does not need redirect at this point [14:39:22] so looks good [14:42:06] New review: Demon; "I'm not opposed to that." [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59443 [14:44:24] New review: Hashar; "PS2 added --mirror to set the mirror properly to debian." [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59626 [14:45:18] paravoid: any idea how I could get the debian/sid release key on an ubuntu precise ?;) [14:45:58] paravoid: the Ubuntu/raring packages debian-archive-keyring_2012.4_all.deb does not provide any key for sid just for squeeze / wheezy [15:00:40] New review: Hashar; "The key seems to be:" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59626 [15:09:05] New patchset: MaxSem; "Delete redirect.(php|phtml)" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59443 [15:09:16] ^demon|away, ^^ [15:15:54] New review: Hashar; "Way to solve it:" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59626 [15:16:06] wave [15:39:26] New review: ArielGlenn; "What happens when gerrit is full up and apache starts to backlog with connections it can't pass on?" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/50591 [15:43:31] <^demon> MaxSem: Hrm? [15:54:28] New review: Brion VIBBER; "Icon has both 16x16 and 32x32 versions, looks good to me." [operations/mediawiki-config] (master); V: 2 C: 1; - https://gerrit.wikimedia.org/r/59589 [16:01:25] PROBLEM - Puppet freshness on lvs1005 is CRITICAL: No successful Puppet run in the last 10 hours [16:01:25] PROBLEM - Puppet freshness on lvs1004 is CRITICAL: No successful Puppet run in the last 10 hours [16:01:25] PROBLEM - Puppet freshness on lvs1006 is CRITICAL: No successful Puppet run in the last 10 hours [16:01:38] ^demon, mhm! I rewrote your patch:) [16:05:17] Reedy: reminder: today is the day that Asher is taking the 18:00-20:00 UTC window for the mariadb migration [16:05:18] <^demon> MaxSem: I saw, I was only half paying attention (was on the phone) when I said hrm. [16:05:27] <^demon> I gave it +1 cuz it's awesome. [16:12:09] New patchset: Demon; "Move connection limiting from gerrit's Jetty to Apache" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/50591 [16:12:56] New review: ArielGlenn; "after chat with demon in irc, we'll try this for awhile and see how apache does with it." [operations/puppet] (production) C: 2; - https://gerrit.wikimedia.org/r/50591 [16:18:13] Change merged: ArielGlenn; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/50591 [16:19:51] <^demon> !log running puppet on manganese (will restart gerrit) [16:19:59] Logged the message, Master [16:21:30] New patchset: MaxSem; "pngcrush everything" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59638 [17:05:08] PROBLEM - Puppet freshness on virt1005 is CRITICAL: No successful Puppet run in the last 10 hours [17:05:52] New patchset: Hashar; "package builder now supports Debian.org unstable" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59626 [17:06:42] New review: Hashar; "PS3 adds a debootstrap variable. Set to " [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59626 [17:33:30] ^demon: is jenkins still broken? [17:36:19] AaronSchulz: i think it works now [17:36:20] <^demon> I thought Antoine got it fixed earlier. [17:36:31] <^demon> I didn't do anything other than possibly break it further last night. [17:36:50] rebasing and new patchsets seems to not trigger a score [17:36:58] unless it's just taking an age and year [17:37:06] ^demon: btw, https://www.mediawiki.org/wiki/Git/New_repositories/Requests is a bit backlogged, i think [17:37:12] seems like yesterday, where it would take a year and then -1 [17:37:40] ^demon: I may start adding +2V for stuff... [17:38:16] ok, I just got one +2V from jenkins [17:38:21] hopefully that means the rest will finish [17:38:43] * AaronSchulz wonders why it's still unmerged [17:39:47] <^demon> ori-l: I know :( [17:40:13] ^demon: what are your thoughts re: opening it up more broadly, now that repo deletion is possible? [17:40:22] iirc that was one of the reasons it had to be so tightly restricted [17:40:56] <^demon> One of, but I think it's still too confusing for people. [17:41:30] <^demon> Not too confusing to allow more (which I'm cool with), but not for self-service. [17:43:03] i'd be interested in learning, fwiw. i don't think anyone in editor engagement has those powers currently which kind of puts us between the cracks, what with integration having integration/, analytics having analytics/, etc. etc. [17:46:59] <^demon> I've got it documented on-wiki. [17:47:10] <^demon> https://www.mediawiki.org/wiki/Git/Creating_new_repositories [17:47:25] i know :) but i don't have the permissions [17:49:37] !log aaron synchronized php-1.22wmf2/includes/User.php 'deployed 6a66776fd7afdf883ad6f82270453fc76833508b' [17:49:38] <^demon> I can grant it to you. You seem like a reasonable fellow :) [17:49:45] Logged the message, Master [17:50:22] <^demon> ori-l: You're now in the "Project and Group Creators" group, so you can create groups & projects. [17:53:13] ^demon, so well named! [17:53:31] <^demon> I try ;-) [17:55:10] ^demon: weee, thanks [17:55:14] ! [17:55:33] <^demon> yw. [17:56:16] ^demon|busy: is help with the backlog welcome, or should i leave those for you to filter / adjudicate / whatever? [17:56:53] oops, you're busy; i'll ping you some other time. [17:59:05] <^demon|busy> For the obvious cases--no weird ACL, or a standard extension setup with just the one group--those are fair game. [17:59:30] <^demon|busy> New hierarchies, weird ACLs, etc...I'll keep on those for now. [18:03:34] k [18:15:27] New review: JanZerebecki; "Good, I found no obvious errors in this version." [operations/apache-config] (master) C: 1; - https://gerrit.wikimedia.org/r/49069 [18:16:25] !log asher synchronized wmf-config/db-eqiad.php 'pulling db1050' [18:16:32] Logged the message, Master [18:17:41] !log installing new linecard in cr2-eqiad , .1% chance of issues [18:17:49] Logged the message, Mistress of the network gear. [18:20:16] New patchset: Asher; "preparing to switch s1/s5 masters" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59657 [18:22:03] New patchset: RobH; "lanthanum in wrong autopart section" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59659 [18:27:24] !log asher synchronized wmf-config/db-eqiad.php 'returning db1050' [18:27:31] Logged the message, Master [18:28:47] New review: RobH; "im tired of waiting on you zuul" [operations/puppet] (production); V: 2 C: 2; - https://gerrit.wikimedia.org/r/59659 [18:28:48] Change merged: RobH; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59659 [18:29:31] New patchset: Asher; "new s[15] masters" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59661 [18:30:38] New patchset: Asher; "new s[15] masters" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59661 [18:31:23] Change merged: Asher; [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59661 [18:32:06] Change merged: Asher; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59657 [18:33:39] PROBLEM - MySQL Slave Delay on db1025 is CRITICAL: CRIT replication delay 198 seconds [18:34:48] New patchset: Jgreen; "remove civicrm stock config, no reason to puppetize it" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59663 [18:35:17] Change merged: Jgreen; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59663 [18:36:44] RECOVERY - MySQL Slave Delay on db1025 is OK: OK replication delay 8 seconds [18:44:32] New review: JanZerebecki; "I think there are more vhosts that need this. Search for /zh-cn in main.conf to find what I mean." [operations/apache-config] (master) - https://gerrit.wikimedia.org/r/59580 [18:58:27] New patchset: Asher; "for master swaps" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59665 [19:00:11] !log aaron synchronized php-1.22wmf2/maintenance/doMaintenance.php 'deployed 1306792ad24be5b8d17355c58da655eb88c2fcc6' [19:00:18] Logged the message, Master [19:02:34] New patchset: Asher; "for master swaps" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59665 [19:03:22] !log aaron synchronized php-1.22wmf2/maintenance/doMaintenance.php 'deployed 1437b57bd475e5ecb06800f9f64b14ccffd0dedf' [19:03:29] Logged the message, Master [19:04:05] Change merged: Asher; [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59665 [19:06:30] ottomata: is reportcard.wm still in use? looks broken [19:08:12] its hasn't been deployed yet [19:08:13] its ready [19:08:15] but not deployed [19:08:17] the old reportcard is here [19:08:18] http://stats.wikimedia.org/reportcard/ [19:08:21] the new one is still in labs [19:11:11] ottomata: internal server error [19:11:29] eh, not on the labs one, but the one describe in statistics.pp [19:11:43] i was just looking where the metrics vhost was [19:11:56] !log asher synchronized wmf-config/db-pmtpa.php 'preparing for master swaps' [19:12:04] Logged the message, Master [19:12:59] !log asher synchronized wmf-config/db-eqiad.php 's1/5 read-only' [19:13:06] Logged the message, Master [19:17:20] !log asher synchronized wmf-config/db-eqiad.php 's1/5 writeable' [19:17:27] Logged the message, Master [19:19:05] !log asher synchronized wmf-config/db-pmtpa.php 's1/s5 writeable from pmtpa' [19:19:12] Logged the message, Master [19:23:20] PROBLEM - MySQL Slave Delay on db1025 is CRITICAL: CRIT replication delay 190 seconds [19:25:04] mwalker: Jeff_Green ^^ [19:25:20] RECOVERY - MySQL Slave Delay on db1025 is OK: OK replication delay 10 seconds [19:25:41] could be awight blowing up 1008 [19:25:45] nevermind [19:25:51] pgehres: yeah, we're doing a drupal upgrade [19:26:12] New review: Ottomata; "(14 comments)" [operations/debs/kafka] (master) - https://gerrit.wikimedia.org/r/53170 [19:26:22] New patchset: Ottomata; "Initial debian packaging using git-buildpackage" [operations/debs/kafka] (master) - https://gerrit.wikimedia.org/r/53170 [19:28:03] New patchset: Ottomata; "Initial debian packaging using git-buildpackage" [operations/debs/kafka] (master) - https://gerrit.wikimedia.org/r/53170 [19:31:11] !log jenkins / gallium has a huge queue of jobs. Investigating. [19:31:18] morebots: ping [19:31:18] Logged the message, Master [19:31:19] I am a logbot running on wikitech-static. [19:31:19] Messages are logged to wikitech.wikimedia.org/wiki/Server_Admin_Log. [19:31:19] To log a message, type !log . [19:31:38] !log jenkins gallium has a huge queue of jobs. investigating. [19:31:46] Logged the message, Master [19:31:47] seriously [19:33:24] !log jenkins: reducing number of executors from 10 to 6. [19:33:31] Logged the message, Master [19:37:07] ok paravoid, i think the kafka stuff is looking good, i'm ready for your next comments! [19:40:25] binasher: https://gerrit.wikimedia.org/r/#/c/59567/ [19:40:47] ottomata: pay it forward :P https://gerrit.wikimedia.org/r/#/c/59596/ [19:42:03] ori-l are all those packages available? [19:42:13] AaronSchulz: ooh, data.. ok, just going to merge [19:42:19] ottomata: yep [19:42:20] Change merged: Asher; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59567 [19:43:27] !log enwiki, dewiki, wikidata are fully on mariadb 5.5.30 [19:43:34] Logged the message, Master [19:43:46] binasher: is that the end of the master swaps today? [19:44:00] yeah [19:44:11] awesome [19:44:15] congrats [19:44:29] :) [19:44:53] ori-l, that's fine with me, i'm not sure if someone else needs to approve that or not [19:45:03] i don't see why [19:45:04] i think its fine [19:45:21] ottomata: yeah, the overall plan statement in the commit is one that i've rehearsed elsewhere; it isn't introduced there [19:49:41] gallium is broken :-( [19:51:26] Change merged: Ottomata; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59596 [19:51:49] ori-l^ merged [19:52:01] ottomata: yay, thanks! [19:56:32] !log gallium slowness is partly due to the ssd install. The git repositories are on the slow raid where as the workspaces are on the ssd. That means `git clone` can't do hardlink anymore :( [19:56:38] Logged the message, Master [19:57:41] <^demon|busy> hashar: We can change where we replicate to if that'd help. [19:58:01] na it clones from the Zuul repo [19:58:10] which I still have to migrate to the ssd :( [19:58:15] I got the puppet change around though [19:58:41] but yeah I will have to rethink how we clone [19:58:45] probably should replicate to the SSD [19:58:53] and uses that as references if at all possible [19:58:58] New patchset: Odder; "(bug 47325) Add 'autopatrol' right to two groups on eswikivoyage" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59712 [19:59:17] <^demon|busy> Can we clone with -s? [19:59:37] <^demon|busy> -s + being on the same disk would make it pretty damn fast. [19:59:49] that is what is done when cloning on the same disk [20:00:32] if the repo is a local path, it uses --local which clone by making hardline for .git/objects [20:00:43] hardlinks [20:00:57] I focused on getting Zuul upgraded [20:01:04] should have worked on migrating its git dir [20:07:51] so at least we have a reproduction of the Zuul issue noticed yesterday :D [20:08:12] New patchset: Jgreen; "tweak fundraising banner log rotation script" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59714 [20:09:08] Change merged: Jgreen; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59714 [20:12:04] Ok I got the fix [20:14:10] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: fishbowl, private and special to 1.22wmf2 [20:14:17] Logged the message, Master [20:16:12] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: Revert that due to large number of warnings... [20:16:19] Logged the message, Master [20:17:06] New patchset: Hashar; "zuul: migrate git dir in production to the ssd" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/58899 [20:17:06] New patchset: Hashar; "zuul: support specifying the git directory" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/58898 [20:18:29] !log stopping Zuul gracefully [20:18:36] Logged the message, Master [20:19:13] PROBLEM - MySQL Slave Delay on db1025 is CRITICAL: CRIT replication delay 201 seconds [20:20:14] RECOVERY - MySQL Slave Delay on db1025 is OK: OK replication delay 10 seconds [20:21:33] PROBLEM - zuul_service_running on gallium is CRITICAL: PROCS CRITICAL: 0 processes with regex args ^/usr/bin/python /usr/local/bin/zuul-server [20:24:45] * hashar jenkins jobs are on hold for an emergency fix sorry. [20:24:50] Reedy: meow? [20:25:04] Have you turned into a cat? [20:25:19] what warnings? [20:26:40] ottomata: could you merge in some Zuul related changes for me please? https://gerrit.wikimedia.org/r/58898 https://gerrit.wikimedia.org/r/58899 :D [20:26:46] ottomata: need to switch some directory [20:26:53] the changes are straight forward [20:28:07] Change merged: Ottomata; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/58898 [20:28:31] Change merged: Ottomata; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/58899 [20:28:35] thhhx :-} [20:28:48] done! [20:28:54] merged on sockpuppet ? [20:29:47] !log Switching zuul dir to /srv/ssd/zuul ( {{gerrit|58898}} and {{gerrit|58899}} ) [20:29:54] Logged the message, Master [20:30:03] PROBLEM - Puppet freshness on cp3003 is CRITICAL: No successful Puppet run in the last 10 hours [20:31:32] New patchset: Lwelling; "Set up email addresses for sending notifications from Echo for enwiki and mediawiki.org" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59717 [20:33:37] New patchset: Hashar; "zuul: fix missing git_dir class parameter" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59718 [20:33:55] ottomata: and the lame follow up :(( https://gerrit.wikimedia.org/r/59718 [20:34:05] ottomata: sorry I missed a line when doing my quick rebase :/ [20:34:25] Change merged: Ottomata; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59718 [20:34:42] done [20:34:45] \O/ [20:35:10] the changed dependended on some other dependencies, so I rebased them to make a clean and fast to merge patch [20:35:14] missed something :( [20:35:41] https://bugzilla.wikimedia.org/show_bug.cgi?id=47332 [20:35:51] would there by any significant reason not to do that ^^ ? [20:41:44] !log reedy synchronized php-1.22wmf2/includes/db/DatabaseMysql.php 'debugging' [20:41:50] New patchset: Dereckson; "(bug 47325) Rights configuration on es.wikivoyage" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59712 [20:41:51] Logged the message, Master [20:42:35] New review: Dereckson; "Ok." [operations/mediawiki-config] (master) C: 1; - https://gerrit.wikimedia.org/r/59712 [20:42:57] New review: Dereckson; "In the previous review comment, by "commit message" I meant "the first line of the commit message"." [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59712 [20:44:29] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: Some wikis to create some noise [20:44:38] Logged the message, Master [20:46:15] Reedy: I think someone removed the group :/ [20:46:24] I'm not seeing it in conf anymore [20:48:57] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wikivoyage, wikimania and wikimedia to 1.22wmf2 [20:49:04] Logged the message, Master [20:49:57] New patchset: Hashar; "zuul: drop dupe file reference in production" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59721 [20:50:14] ottomata: and got that one https://gerrit.wikimedia.org/r/59721 (duplicate file{} sorry) [20:50:30] Reedy: I'll add it back [20:50:54] apergos: I read https://bugzilla.wikimedia.org/show_bug.cgi?id=21117 but I guess enwikivoyage changing its default thumbnail size wouldn't have too much side-effects, would it? since it's a relatively small wiki... [20:51:11] !log reedy synchronized php-1.22wmf2/includes/db/DatabaseMysql.php 'debugging' [20:51:17] Logged the message, Master [20:52:06] Reedy: hrm, wfErrorLog( 'Some error', 'udp://10.0.5.8:8420/filename' ); might be easier ;) [20:52:35] New patchset: Ottomata; "Setting setgid bit on $fundraising_log_directory/logs so that file_mover process can do its thang." [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59724 [20:53:10] Change merged: Ottomata; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59724 [20:54:33] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wikinews.dblist and special wikis to 1.22wmf2 [20:54:41] Logged the message, Master [20:56:05] New patchset: Aaron Schulz; "Added a temp debug log back." [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59725 [20:56:13] Quick! [20:56:19] Seems it's something in special.dblist [20:56:41] Change merged: Reedy; [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59725 [20:56:56] Reedy: it's temp-debug now, not tempDebug :) [20:57:10] mmm, will update [20:57:17] !log reedy synchronized wmf-config/InitialiseSettings.php [20:57:24] Logged the message, Master [20:57:27] * AaronSchulz likes to throw people off [20:58:59] !log reedy synchronized php-1.22wmf2/includes/db/DatabaseMysql.php 'moar' [20:59:06] Logged the message, Master [20:59:19] aude: It's Wikidata [20:59:28] !log stopping Jenkins that somehow restarted [20:59:34] Logged the message, Master [20:59:41] Reedy: ORM? ;) [21:00:04] AaronSchulz: How did you guess!? [21:00:10] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wikidatawiki back to 1.22wmf1 [21:00:17] Logged the message, Master [21:00:24] http://p.defau.lt/?jUnh3ng0cDb_p3APQPEmAg [21:00:24] * AaronSchulz actually looks [21:00:38] heh [21:01:23] PROBLEM - jenkins_service_running on gallium is CRITICAL: PROCS CRITICAL: 0 processes with regex args ^/usr/bin/java .*-jar /usr/share/jenkins/jenkins.war [21:01:41] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wikibooks and wikiversity to 1.22wmf2 [21:01:47] New patchset: Ori.livneh; "Add supervisord configuration for IPython Notebook" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59727 [21:01:48] Logged the message, Master [21:03:53] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wikisource and wikiquote to 1.22wmf2 [21:03:55] New patchset: Lwelling; "Set up email addresses for sending notifications from Echo for enwiki and mediawiki.org" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59717 [21:04:00] Logged the message, Master [21:05:42] New patchset: MaxSem; "WIP: OSM module" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/36222 [21:06:41] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wiktionaries to 1.22wmf2 [21:06:48] Logged the message, Master [21:07:29] New patchset: Reedy; "Everything non wikipedia (bar wikidatawiki) to 1.22wmf2" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59728 [21:07:30] New review: Dzahn; "to fix a duplicate definition and unbreak zuul" [operations/puppet] (production) C: 2; - https://gerrit.wikimedia.org/r/59721 [21:07:54] Change merged: Reedy; [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59728 [21:08:31] New review: Dzahn; "to fix a duplicate definition and unbreak zuul, manual verify (jenkins/zuul is being fixed)" [operations/puppet] (production); V: 2 - https://gerrit.wikimedia.org/r/59721 [21:08:32] Change merged: Dzahn; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59721 [21:08:38] !log Zuul ETA: it is copying MobileFrontend extension right now so half way done [21:08:45] Logged the message, Master [21:08:46] New patchset: Ori.livneh; "Add supervisord configuration for IPython Notebook" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59727 [21:14:08] Ryan_Lane: KEYSTACK [21:14:20] hahaha [21:14:21] yep [21:14:24] it's not a joke [21:16:06] !log finished migrating zuul git repositories ( rsync -av /var/lib/zuul/git /srv/ssd/zuul ) [21:16:13] Logged the message, Master [21:19:16] !log restarted Zuul [21:19:23] Logged the message, Master [21:19:27] !log restarted Jenkins [21:19:34] RECOVERY - zuul_service_running on gallium is OK: PROCS OK: 1 process with regex args ^/usr/bin/python /usr/local/bin/zuul-server [21:19:34] Logged the message, Master [21:20:03] !log stopped Zuul. Will wait a bit till jenkins start up [21:20:09] Logged the message, Master [21:20:24] RECOVERY - jenkins_service_running on gallium is OK: PROCS OK: 1 process with regex args ^/usr/bin/java .*-jar /usr/share/jenkins/jenkins.war [21:22:34] PROBLEM - zuul_service_running on gallium is CRITICAL: PROCS CRITICAL: 0 processes with regex args ^/usr/bin/python /usr/local/bin/zuul-server [21:24:29] hashar: you're debugging that, right [21:24:45] yeah [21:24:49] and fyi..the monitoring fix has been merged the other day as you see:) [21:24:55] emergency maintenance :/ [21:25:16] tell me if you need to merge a hotfix or something [21:25:43] luckily I had all the patches prepared :-] [21:25:49] :) [21:25:55] ottomatta kindly merged them on request :-] [21:26:04] but if there is any follow up I will make sure to ping you! [21:28:57] ooo [21:29:06] https://bugzilla.wikimedia.org/show_bug.cgi?id=27839 [21:29:19] so https://bugzilla.wikimedia.org/show_bug.cgi?id=47332 is a WONTFIX? [21:29:35] hashar: ^^ :) [21:30:14] odder: can't look tonight sorry [21:30:51] Dereckson: thanks for posting the see also [21:35:01] New patchset: Dereckson; "(bug 47337) Flagged Revisions configuration for ru.wikipedia" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59732 [21:43:53] PROBLEM - Puppet freshness on virt3 is CRITICAL: No successful Puppet run in the last 10 hours [21:44:08] New patchset: RobH; "RT 4965 adding antimony to netboot.cfg" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59734 [21:45:26] New patchset: Demon; "Base puppet setup for antimony" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59735 [21:45:35] New review: RobH; "wait for jenkins to work, nahhhh" [operations/puppet] (production); V: 2 C: 2; - https://gerrit.wikimedia.org/r/59734 [21:45:35] Change merged: RobH; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59734 [21:45:42] <^demon|busy> RobH: Went ahead and did that much for now ^ [21:45:58] cool, i wasnt touchin site.pp [21:46:05] if it had nothing it would just get default [21:46:09] standard i mean. [21:46:41] <^demon|busy> I gave it enough to get working. The rest will come as soon as I'm done testing/puppetizing my work. [21:47:08] what server does this now, the same gerrit host? [21:47:12] (gallium?) [21:47:15] just cuirous [21:47:18] curious evne [21:47:19] <^demon|busy> manganese now. [21:47:24] bleh, fuckin new keyboard [21:47:24] <^demon|busy> gallium is jenkins + zuul. [21:48:02] so [21:48:16] turns out a ssd is nice [21:48:33] cloning TBytes of data from a disk to another is not :D [21:51:59] New patchset: MaxSem; "WIP: OSM module" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/36222 [21:53:22] odder: sorry [21:53:31] odder: so yeah we do not want to vary the thumbnails sizes :/ [21:53:47] odder: that does not scale very well right now. [21:59:35] New patchset: Dzahn; "change user prefs thumbnail sizes for en.wikivoyage (bug 47332)" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59739 [21:59:50] hashar: thanks, I asked above before even taking this bug, but since no-one replied, I went ahead... and then Dereckson added the see also :) [21:59:57] closed the bug as a WONTFIX already [22:00:07] but mutante is very helpful I see :) [22:01:03] New review: Odder; "This doesn't scale, at the bug was closed as a WONTFIX anyway." [operations/mediawiki-config] (master) C: -1; - https://gerrit.wikimedia.org/r/59739 [22:01:57] arr, hashar, you won't fixed it:) (thumbnail sizes:) [22:02:39] hmm @ CPU cost [22:02:41] odder: yeah that was the right decision. Thank you for baby siting our bugs :-] [22:03:01] https://meta.wikimedia.org/w/index.php?title=Limits_to_configuration_changes&diff=5401760&oldid=5394408 hashar [22:03:46] ohh [22:03:51] that is a useful page [22:04:00] it is awesome how much can be done in our community [22:04:04] communities [22:04:59] ^demon|busy: I think I will get Jenkins upgraded next week. [22:05:15] ^demon|busy: that slow start up is ruining the 'fun' [22:05:56] !log jenkins restarted. Restarting Zuul. [22:06:03] Logged the message, Master [22:06:30] RECOVERY - zuul_service_running on gallium is OK: PROCS OK: 1 process with regex args ^/usr/bin/python /usr/local/bin/zuul-server [22:07:09] \o/ [22:07:20] nice [22:07:42] New review: MaxSem; "Per the outcome of bug 41712 and bug 47332, we shouldn't do it." [operations/mediawiki-config] (master) C: -2; - https://gerrit.wikimedia.org/r/59739 [22:07:45] New review: Hashar; "recheck" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59656 [22:11:30] PROBLEM - zuul_service_running on gallium is CRITICAL: PROCS CRITICAL: 0 processes with regex args ^/usr/bin/python /usr/local/bin/zuul-server [22:13:54] New review: Lcarr; "if you fix the one problem, this is good to submit...." [operations/puppet] (production) - https://gerrit.wikimedia.org/r/56107 [22:14:11] New patchset: Lcarr; "erb expander for testing purposes" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/55304 [22:14:21] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/55304 [22:15:26] New patchset: Lcarr; "tell users which username to use for LDAP auth" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/57754 [22:16:43] New review: Ryan Lane; "It's just a dialog. See: http://httpd.apache.org/docs/2.2/mod/core.html#authname" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/57754 [22:17:21] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59735 [22:18:40] New patchset: RobH; "RT 4965 antimony addition" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59745 [22:19:20] PROBLEM - MySQL Slave Delay on db1025 is CRITICAL: CRIT replication delay 235 seconds [22:19:47] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/57754 [22:19:48] Change merged: RobH; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59745 [22:20:20] RECOVERY - MySQL Slave Delay on db1025 is OK: OK replication delay 10 seconds [22:20:28] !log Jenkins repointing all jobs to use the new Zuul repo ( /srv/ssd/zuul/git ) {{gerrit|59744}} [22:20:35] Logged the message, Master [22:20:59] Shouldn't the icinga alerts for db stuff be change to "MariaDB Slave Delay"...? ;) [22:21:10] s/change/changed/ [22:21:58] wow, how do I get CR+2 from *2* people??! :) [22:22:57] you are very special [22:23:33] Change abandoned: Dzahn; "won't fixed" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59739 [22:25:42] :) [22:26:43] New patchset: Lcarr; "fixing missing comma" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59747 [22:28:14] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59747 [22:30:10] New review: Lcarr; "about to check this out but the fact that someone made this monitoring setup without checking it in ..." [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59059 [22:50:33] mutante: planet isn't still running on singer is it? [22:50:54] New review: Lcarr; "If these aren't going to be used in the near future, I would like to delete instead of just comment ..." [operations/puppet] (production) - https://gerrit.wikimedia.org/r/58922 [22:51:12] nope it's on zirconium [22:51:34] [23:46:42] singer is a Wikimedia Planet weblog aggregator (misc::planet). [22:51:34] [23:46:42] singer is a Wikimedia secure.wikimedia.org (misc::secure). [22:52:18] New patchset: Lcarr; "systemuser learned 'managehome' (default true)" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/53879 [22:52:33] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/53879 [22:52:54] well the motd is never updated [22:53:05] and it could still have those packages on it [22:53:28] secure.wm is still on singer [22:54:28] I thought it was moved [22:55:30] New patchset: Lcarr; "create jenkins user with systemuser" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/53880 [22:56:40] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/53880 [22:59:28] !log restarting Zuul [22:59:30] RECOVERY - zuul_service_running on gallium is OK: PROCS OK: 1 process with regex args ^/usr/bin/python /usr/local/bin/zuul-server [22:59:34] Logged the message, Master [23:00:22] New review: Hashar; "recheck" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59656 [23:00:34] Krinkle: zuul restarted [23:01:13] New patchset: Lcarr; "sql script no more need /etc/cluster" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/55877 [23:01:14] New review: Faidon; "(3 comments)" [operations/debs/kafka] (master) C: -1; - https://gerrit.wikimedia.org/r/53170 [23:01:20] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/55877 [23:01:52] New patchset: Lcarr; "Remove unused junk from gerrit config" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59403 [23:02:07] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59403 [23:02:59] New patchset: Lcarr; "Gerrit, now with 50% more memory" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59411 [23:03:42] New patchset: Asher; "db1028 to mariadb" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59753 [23:03:47] New review: Lcarr; "confirmed that we have 1gig left of free memory for at least the past month without going into swap." [operations/puppet] (production); V: 2 C: 2; - https://gerrit.wikimedia.org/r/59411 [23:03:48] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59411 [23:04:04] Reedy: no, it's not, it's on zirconium in eqiad [23:04:26] Reedy: what did you just say about contacts.wm? [23:04:31] mutante: https://wikitech.wikimedia.org/w/index.php?title=Singer&diff=67033&oldid=47857 [23:04:34] You're welcome [23:04:41] ;) [23:04:55] thanks:) [23:05:35] New patchset: Odder; "(bug 46944) Allow users to save books to userspace on enwikipedia" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59756 [23:05:48] greg-g: I'm ready to push my changes out whenever [23:05:49] New patchset: Lcarr; "Fixing comment in puppet-merge" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59519 [23:05:49] * hashar jenkins is back up [23:05:57] hashar: *high five* [23:06:05] huzzah [23:06:20] I am not sure for how long though [23:06:26] :p [23:06:33] cause Gerrit gives me error: [Errno 111] Connection refused [23:06:39] from time to time [23:06:53] mwalker: go for it [23:07:38] BTW, that we have '*' => array( 'createpage' => false ), for enwikipedia is a fscking disgrace. [23:08:29] Change merged: Asher; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59753 [23:10:38] <^demon|busy> hashar: We passed off connection throttling from jetty to apache this morning. [23:10:51] \O/ [23:10:56] <^demon|busy> So was that apache or gerrit giving you the errno 111? [23:10:57] just add a couple connection refused [23:11:00] New patchset: Lcarr; "Don't link to draft patches" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59628 [23:11:20] ^demon|busy - i'm doing a lot of reviews on your small patches today :) [23:11:23] ^demon|busy: the ssh stream I think [23:11:32] <^demon|busy> Oh, that's not apache, nvm. [23:11:32] !log mwalker synchronized php-1.22wmf1/extensions/CentralNotice 'Beginning update of CentralNotice and ContributionTracking to latest' [23:11:35] <^demon|busy> LeslieCarr: <3 [23:11:37] New patchset: Asher; "db1004 to mdb, not 1028" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59759 [23:11:40] Logged the message, Master [23:12:08] !log mwalker synchronized php-1.22wmf1/extensions/ContributionReporting/ [23:12:15] Logged the message, Master [23:12:33] Change merged: Asher; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59759 [23:13:51] !log If jenkins/zuul goes wild overnight: connect to gallium, killoff puppet in crontab then /etc/init.d/zuul stop and drop me an email. Will take of it whenever I wake up. [23:13:58] Logged the message, Master [23:15:05] binasher: you know anything about an uncommitted change to DatabaseMysql.php in php-1.22wmf2 on fenari? [23:15:16] nope [23:15:22] greg-g: ^ you got anything? [23:16:17] <^demon|busy> Reedy: That yours ^ from the escaping error you pasted earlier? [23:16:33] Yeah [23:16:42] Feel freeeee to revert it. Or I can if you want [23:16:46] heh [23:17:09] Reedy: I can do it [23:19:35] !log mwalker synchronized php-1.22wmf2/extensions/CentralNotice/ [23:19:43] Logged the message, Master [23:20:52] !log mwalker synchronized php-1.22wmf2/extensions/ContributionReporting 'Finished update of CN and CR to latest' [23:20:59] Logged the message, Master [23:21:17] greg-g: ok; I'm out and testing [23:22:02] PROBLEM - Puppet freshness on ms-fe3001 is CRITICAL: No successful Puppet run in the last 10 hours [23:22:51] New patchset: Lcarr; "Don't link to draft patches" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59628 [23:22:54] mwalker: awesome, thanks [23:22:59] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59628 [23:24:50] New patchset: Asher; "pulling db1004" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59761 [23:25:17] Change merged: Asher; [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59761 [23:25:40] New patchset: Lcarr; "Swap string "true" for boolean true in $ganglia_aggregator" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59613 [23:26:29] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59613 [23:27:27] greg-g: it looks stable [23:27:29] so yay! [23:28:43] New patchset: Lcarr; "fixing last non-boolean ganglia_aggregator" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59763 [23:29:01] mwalker: better be! ;) [23:29:18] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59763 [23:29:23] yay jenkins is so muhc smoother right now [23:29:45] <^demon|busy> Whoops, missed that one. [23:31:17] all good [23:31:30] New review: Lcarr; "rebasing fail can you fix?" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/58737 [23:31:50] New patchset: Lcarr; "correct jenkins master system role" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59124 [23:32:04] Change merged: Lcarr; [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59124 [23:33:12] New patchset: Lcarr; "Fixing comment in puppet-merge" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/59519 [23:35:43] New review: Lcarr; "+2 however path conflict during rebase. please rebase." [operations/puppet] (production) C: 2; - https://gerrit.wikimedia.org/r/57302 [23:38:48] ^demon|busy - can you chime in on https://gerrit.wikimedia.org/r/#/c/58082/ if you are unbusy and lying in your nick? [23:39:25] !log asher synchronized wmf-config/db-eqiad.php 'pulling db1004' [23:39:33] Logged the message, Master [23:40:31] <^demon|busy> I really don't know. Krinkle might be able to advise :\ [23:40:44] * Krinkle peeks [23:41:10] New patchset: Krinkle; "Prevent gerrit logo from pushing the search bar outside the screen" [operations/puppet] (production) - https://gerrit.wikimedia.org/r/58082 [23:41:25] ^demon|busy: LeslieCarr: What is the question? [23:41:41] <^demon|busy> https://gerrit.wikimedia.org/r/#/c/58082/ [23:41:51] about the css on that patch [23:41:56] my css-foo is weak [23:42:04] <^demon|busy> As is mine [23:43:49] PROBLEM - mysqld processes on db1004 is CRITICAL: PROCS CRITICAL: 0 processes with command name mysqld [23:47:43] New review: MZMcBride; "From my very quick look at InitialiseSettings.php, it seems no other public Wikimedia wiki uses this..." [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59756 [23:48:29] PROBLEM - DPKG on db1004 is CRITICAL: DPKG CRITICAL dpkg reports broken packages [23:48:42] css-fu * [23:48:51] https://en.wiktionary.org/wiki/-fu :-) [23:50:38] Susan: true, but this is hardly controversial. [23:51:06] I don't think it's controversial. [23:51:16] I was just worried it used some crazy code path. [23:51:27] I haven't looked at the extension very closely. [23:52:26] We could even change it for all wikis, but since nobody requested that... [23:52:29] RECOVERY - DPKG on db1004 is OK: All packages OK [23:53:01] New patchset: MZMcBride; "Always redirect wikimediafoundation.org to https (RT-4830)" [operations/apache-config] (master) - https://gerrit.wikimedia.org/r/56062 [23:53:14] New patchset: MaxSem; "Tweak wgLoadScript in startup module" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59767 [23:53:22] odder: I wonder why it's disabled by default. [23:53:35] I have no idea, to be frank. [23:53:50] not like it would cause massive horrors on any wikis, is it? [23:54:41] Dunno! [23:56:58] New patchset: Catrope; "Set $wgVisualEditorParsoidProblemReportURL explicitly" [operations/mediawiki-config] (master) - https://gerrit.wikimedia.org/r/59768