[00:05:41] [bz] (NEW - created by: Peter Bena, priority: Unprioritized - normal) [Bug 48058] /etc/fstab contain non existent partition - https://bugzilla.wikimedia.org/show_bug.cgi?id=48058 [00:06:06] MaxSem: filesystem size fixed in new precise image [00:06:12] it's not in the lucid image yet [00:06:22] Ryan_Lane, whee, thanks! [00:06:25] yw [00:06:31] let me know if you run into any other image issues [00:42:10] * Hazard-SJ says hello and asks if bots-bnr4 is down [00:44:03] legoktm? [00:44:11] Hi [00:44:14] I wouldn't know [00:44:17] Ask petan or addshore [00:45:23] I don't even see it on [[Nova Resource:Bots#Instances for this project]] :S [00:45:54] o.O [00:46:19] legoktm: Try to log in to it :/ [03:47:08] [bz] (NEW - created by: Riley Huntley, priority: Unprioritized - normal) [Bug 47808] internal_api_error when running bot on en.wikipedia.beta.wmflabs.org - https://bugzilla.wikimedia.org/show_bug.cgi?id=47808 [07:21:00] @seen Hazardsj [07:21:00] petan: I have never seen Hazardsj [07:21:05] @seen HazardSJ [07:21:05] petan: I have never seen HazardSJ [07:21:13] @seenrx Hazard [07:21:14] petan: Last time I saw Hazard-SJ they were quitting the network with reason: Quit: Ping timeout: 1 second at 5/7/2013 5:31:02 AM (01:50:12.0900310 ago) (multiple results were found: Hazard-SJ_, Hazard-Away, Hazardd-SJ) [07:21:26] LOL [07:21:44] freenode has hardcore ping timeouts [07:30:30] [12:31:02 AM] Hazard-SJ (~Hazard-SJ@wikimedia/Hazard-SJ) left IRC. (Quit: Ping timeout: 1 second) [07:43:50] legoktm and yes, bnr4 was removed as we didn't need so many of them, considering to remove bnr3 as well [07:44:08] people shouldn't care if we remove them or not, if they run stuff properly on grid [07:44:08] tell hazard that! [07:44:14] how, he's not in here [08:07:14] are there ongoing, known issues regarding the access to the instances ? [08:10:48] no idea but it seems that there are ongoing issues with users staying on channel until they get an answer [08:11:11] LOL [08:11:19] petan: nice one [08:38:01] !log deployment-prep Created deployment-nginx-test to try out the nginx manifests for SSL. [08:38:03] Logged the message, Master [09:31:03] Coren|Sleep how would you list all jobs and the memory the requested? [09:45:27] [bz] (NEW - created by: Niklas Laxström, priority: Unprioritized - major) [Bug 48203] Purging does not work on deployment-prep / beta labs - https://bugzilla.wikimedia.org/show_bug.cgi?id=48203 [09:51:41] !ping [09:51:41] pong [09:52:18] !ping [09:52:18] pong [09:53:25] jak malo staci k lidskemu stesti... ;-) [09:55:51] testuju :P [10:57:35] Warning: There is 1 user waiting for shell: Fischx (waiting 0 minutes) [11:02:20] petan: .. [11:02:27] bash: php: command not found ;p [11:11:04] Warning: There is 1 user waiting for shell: Fischx (waiting 13 minutes) [11:24:29] Warning: There is 1 user waiting for shell: Fischx (waiting 27 minutes) [11:38:03] Warning: There is 1 user waiting for shell: Fischx (waiting 40 minutes) [11:51:29] Warning: There is 1 user waiting for shell: Fischx (waiting 54 minutes) [12:04:54] Warning: There is 1 user waiting for shell: Fischx (waiting 67 minutes) [12:18:25] Warning: There is 1 user waiting for shell: Fischx (waiting 80 minutes) [12:20:45] @notify Coren [12:20:45] I will notify you, when I see Coren around here [12:31:52] Warning: There is 1 user waiting for shell: Fischx (waiting 94 minutes) [12:45:13] Warning: There is 1 user waiting for shell: Fischx (waiting 107 minutes) [12:45:18] o/~ Here I come to save the daaaay! o/~ [12:55:01] Coren: install php on tools-dev? ;p [12:55:19] o_O It's not there? How silly. :-) [12:55:23] indeed ;p [12:55:30] Gimme a min. [12:55:42] cheers :) [12:58:44] Warning: There is 1 user waiting for shell: Fischx (waiting 121 minutes) [13:01:53] addshore: {{done}} [13:01:59] cherrs :) [13:12:16] Warning: There is 1 user waiting for shell: Fischx (waiting 134 minutes) [13:23:15] hello [13:24:09] if I connect to bots-gs theire is a message, that this server was replaced [13:24:19] what does that mean? [13:24:26] [bz] (NEW - created by: Antoine "hashar" Musso, priority: Unprioritized - normal) [Bug 48210] nginx fails: could not build the types_hash - https://bugzilla.wikimedia.org/show_bug.cgi?id=48210 [13:25:42] [bz] (ASSIGNED - created by: Antoine "hashar" Musso, priority: High - enhancement) [Bug 36648] replicate HTTPS architecture - https://bugzilla.wikimedia.org/show_bug.cgi?id=36648 [13:26:56] does that mean that I should move my bot to bots-login? if that, are theire differences? [13:27:27] Pyfisch: you shouldnt be running your bot on bots-gs at all.. [13:27:39] heh [13:27:39] but yes bots-gs is being replaced by bots-login [13:27:54] addshore: yes yes but start it with qsub [13:27:56] which should be used to submit your jobs to the grid servers [13:28:01] [= [13:28:09] Pyfisch that makes it run on nodes not -gs at all [13:28:23] Pyfisch basically, you should move your cron from -gs to -login [13:28:36] but no worries, -gs will be there for some time anyway [13:28:38] petan: you should add that the the motd ;p [13:28:43] I did [13:28:56] addshore add what? [13:29:04] addshore you know you are root on bots? you can do that as well [13:29:09] just say move your cron ;p [13:29:51] i know :P currently trying to wrap my head around the wikibase repo ;p [13:45:30] [bz] (ASSIGNED - created by: Antoine "hashar" Musso, priority: High - normal) [Bug 48210] [upstream] nginx fails: could not build the types_hash - https://bugzilla.wikimedia.org/show_bug.cgi?id=48210 [14:13:25] [bz] (NEW - created by: Antoine "hashar" Musso, priority: Normal - minor) [Bug 48210] [upstream] [worked around] nginx fails: could not build the types_hash - https://bugzilla.wikimedia.org/show_bug.cgi?id=48210 [14:23:27] hi; anyone around who has a clue why ee-prototype (labs instance) tells me it "Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock'"? [14:27:53] [bz] (NEW - created by: Chris McMahon, priority: Unprioritized - major) [Bug 47893] fully support Wikilove on beta (2 issues) - https://bugzilla.wikimedia.org/show_bug.cgi?id=47893 [14:34:37] Getting closer to Amsterdamn.... When is replication supposed to go live? [14:35:01] greenrosetta still few weeks to amsterdam [14:35:03] ;) [14:35:58] Is there a DB to test yet? [14:36:18] test what [14:37:22] test a copy of the en-wiki DB [14:39:23] not that I know of [14:39:30] hi there greenrosetta [14:39:39] hello [14:39:53] greenrosetta: I know Coren and https://wikitech.wikimedia.org/wiki/ToolLabsDatabasePlan are the person and doc to consult on this [14:40:42] greenrosetta: The objective is by Amsterdam. We'd /like/ to have it sooner, but there's no promise. [14:40:54] NOW NOW NOW [14:40:55] :D [14:41:08] Coren any progress on dedicated mysql? I tried to catch binasher but I failed [14:41:12] he doesn't respond to my messages [14:41:58] TBH it should be a matter of 1 day or 2 to configure a mysql server for user db's only problem is to get a hardware running... [14:42:00] petan: I guess you could cc him on email to labs-l? :/ [14:42:12] I don't even know his e-mail [14:48:39] petan: you can usually do firstletter of firstname + lastname @ wikimedia [14:48:59] petan: Asher Feldman [14:49:01] example: sharihareswara@wikimedia, afeldman, etc [14:50:50] petan: What does dedicated MySQL mean? A dedicated machine instead of a virtual one? [14:55:52] yes [14:57:07] petan: Yes, that would be very useful. [16:05:52] beta labs just went down? [16:06:10] petan: Could you install the package lintian on tools-dev, please? [16:13:11] scfc_de: {{done}} [16:14:00] Coren: Thanks. [16:58:06] Coren or Ryan_Lane either of you have contact with Sumana? [16:58:26] probably all the same ways you do :) [16:58:27] why's that? [16:58:58] I haven't been able to get a hold of her, and I emailed her sometime last week. [16:59:42] i've been checking IRC too, haven't seen her [17:10:35] Izaroundz: She was here at 14:49Z. [17:11:06] Izaroundz: Type in "@notify sumanah". [17:11:21] ok [17:11:27] @notify sumanah [17:11:27] This user is now online in #mediawiki so I will let you know when they show some activity (talk etc) [17:19:59] Izaroundz: She was around in and out, but she never stays long in any one place. :-) [17:20:20] Coren: thanks got her :) [17:24:02] in and out [17:24:08] A Clockwork Orange [17:46:00] addshore ping [17:46:29] yup [17:46:47] addshore I was thinking, that bouncer I am working on, we could eventually install it somewhere on labs, so that any wikimedia user could use it... that would be like quassel for anyone :D [17:46:52] it's open source in the end [17:47:20] you know how it works? [17:47:21] would be epic [17:47:27] do u know quassel a bit? [17:47:38] not quassel but i have used other bouncers [17:48:02] I don't want to compare it with irc cloud because it's not browser based (though web browser interface is possible as well if someone make it) [17:48:12] but it's somewhat like ircloud just in application [17:48:34] and it's all in c# XD [17:48:56] btw did you see a source code? you could learn something and use it in huggle ^^ [17:49:14] most hardcore part is that scrolback control [17:49:45] xD [17:50:04] * addshore is currently trying to work out how to use i18n within mediawiki. [17:50:27] * addshore has never fiddled with mediawiki itself before [17:58:01] addshore or we could create alternative of ircloud on labs XD [18:09:48] lol Coren why so many ip [18:09:50] on tools [18:10:30] petan: There's a good reason for it. Ask Leslie eventually. :-) [18:10:45] Leslie isn't here what is that good reason [18:11:00] @seen LeslieCarr [18:11:00] petan: Last time I saw LeslieCarr they were talking in the channel, they are still in the channel #wikimedia-operations at 5/7/2013 4:20:31 PM (01:50:28.7227040 ago) [18:35:42] Coren there is something wrong in mono we have in labs, all programs I run on labs take 10 times more virtual memory than on my own machines [18:36:02] I guess that the version we have packged in repository is very old or broken [18:36:58] hm --version produces the same [18:37:36] * Coren knows ~0 about .net or mono. [18:38:02] problem is that it eats only more virtual memory, so it's something inside of mono [18:38:30] my bot on my laptop eats some 50mb of ram, on grid it eats 800mb [18:38:40] resident memory is same [18:38:45] just virtual is huge [18:39:19] mhhhm [18:39:22] only difference I see is that I am running kernel 3.8 and on labs we have 3.2 but I don't think it changes the way how memory is handled [18:39:33] how to submit to gerrit for code review from windows when you cant get git review? :P [18:39:47] addshore you need to get a better OS [18:44:08] gragh [18:44:20] there must be somethign similar for doze.. [18:44:41] * addshore doesnt want to have to somehow use cygwin for it :P [18:45:13] !!!!! http://www.mediawiki.org/wiki/Thread:User_talk:Sumanah/Installing_Git_review_on_windows !!!! [18:45:14] ;p [18:51:07] addshore: there's a way to push without git review [18:51:11] ^demon: ^^ [18:51:22] ^demon doesn't use git-review [18:51:49] Isn't it just pushing to master or something? [18:52:01] not really [18:52:08] push refs/for/master [18:52:09] ^^ [18:52:10] * addshore has no clue [18:52:10] I think [18:52:40] we disallow direct pushes [18:52:55] because that would be too easy :P [18:53:06] well, because it would bypass review [18:53:19] we can allow direct pushes for specific repos [18:53:26] that don't do code review [18:53:46] Yes, so says http://www.mediawiki.org/wiki/Git/Workflow#git_review_doesn.27t_like_merge_commits. But I thought that Gerrit doesn't use a "regular" Git server, but does its own thing anyhow. [18:53:51] we can also allow direct pushes to branches other than master [18:54:08] scfc_de: it's a git implementation [18:54:37] <^demon> `git push origin HEAD:refs/for/master` [18:54:45] it supports most normal git server things, but it also has "magic" refs that push into changesets [18:54:54] which get reviewed, then merged into the branch [18:55:21] [bz] (NEW - created by: Antoine "hashar" Musso, priority: Lowest - enhancement) [Bug 41530] setup redis on beta - https://bugzilla.wikimedia.org/show_bug.cgi?id=41530 [18:57:08] ^demon, just that? nothing else? :D [18:57:22] no silly pip, python and git reivew? :O [18:57:32] <^demon> That's what I use every day. [18:57:39] <^demon> Plus a ton of aliases because typing is hard :) [18:57:51] * addshore starts to wonder why this is not written anywhere :P [18:57:53] I used to use aliases. git review is easier :) [18:58:16] <^demon> addshore: Because people say things like ^ ;-) [18:59:20] hehe, auth failed [19:00:53] now that is interesting as its definatly the right password! [19:01:52] <^demon> Wait, username and password? [19:01:56] <^demon> Are you trying to push over https? [19:05:09] * addshore would like to point out this is the first time he has tried to push to gerrit or infact ever use gerrit! I am just doing what I am told :O [19:11:24] so although you clone from https i need to set my ssh url somewhere ^demon (and key)? :) [19:17:24] gragh, i might consider rewriting all of these pages into a simple windows one after all of this [19:36:09] addshore, you there? [19:36:41] Ryan_Lane: any idea if you can remove your first identity from gerrit? i.e. I have two email addresses when I should only have one, but the one I wish the remove doesnt have a tickbox next to it to remove it >.< (looking at https://gerrit.wikimedia.org/r/#/settings/web-identities) [19:36:44] Cyberpower678: yes [19:37:12] addshore, https://github.com/cyberpower678/Peachy will host Peachy. I intend to commit on the third alpha release. [19:37:27] :> [19:39:10] addshore: you need to change your email address via wikitech [19:39:21] that email address comes from ldap [19:39:22] and wikitech can modify it [19:39:28] hmm, it just let me change it through gerrit [19:39:48] andrewbogott: we're seeing some weird behavior on openid-wiki and openid-wiki2 [19:39:55] may be related to your puppet changes [19:40:15] Which ones? [19:40:21] not sure [19:40:37] did you move orig to _orig? [19:41:06] not on purpose... [19:41:06] $wgScriptPath = "/srv/mediawiki"; [19:41:10] that was also set [19:41:13] which is… odd [19:42:03] The scriptpath thing might be me… previously I was setting it explicitly in the pupetized settings file… but then I noticed that it gets set during wiki setup anyway so I removed the explicit setting. [19:42:33] /srv/mediawiki seems right, though? [19:42:41] it's definitely not :) [19:42:49] it should be /w [19:43:09] ok, hang on... [19:44:10] andrewbogott: mind taking a look at openid-wiki2? it's all screwed up [19:44:11] yeah, it's getting set to /w in my test cases. [19:44:28] php is complaining about being out of memory as well [19:44:56] /usr/bin/php: error while loading shared libraries: libasn1.so.8: failed to map segment from shared object: Cannot allocate memory [19:45:47] I have to add myself to the project, one minute... [19:46:01] If 'orig' is moved to '_orig' that suggests someone was monkeying with it and made a backup first [19:46:02] maybe [19:46:28] * Ryan_Lane nods [19:46:29] hm [19:46:35] there's something funky going on [19:47:24] so, yeah, looks like _orig is a copy and not being loaded. [19:47:28] So, can ignore that for now [19:47:36] * Ryan_Lane nods [19:48:27] mhhhm I get error: failed to push some refs, any ideas? [19:49:16] Ryan_Lane, what kind of bad behavior are you seeing? First glance it looks ok. [19:53:01] hm… oom even for a failed login [19:53:07] So it's the identity checker that's ooming [19:53:30] right [19:53:32] not sure why [19:53:44] it wasn't fairly recently [19:54:00] this seemed to have changed after an upgrade and a reboot [20:01:48] googling suggests it might be a wonky ulimit [20:04:11] …but they look fine [20:06:20] I'm also seeing gluster weirdness [20:06:32] oh? [20:06:35] but it's weird [20:06:41] as my user I get gluster errors [20:06:50] as root or any other user I don't [20:06:58] I hate this damn filesystem [20:10:00] Ryan_Lane are you sure it is gluster which suck or we at configuring it properly? [20:10:09] there's not much to configure [20:10:19] mhm... I should try gluster myself [20:10:20] somewhere [20:10:37] it is a big project I can't believe it's so broken [20:10:38] it's generally stable if you only have a very small number of volumes [20:10:45] did you report all these issues to their bugzilla? [20:10:47] I'm using it elsewhere [20:10:49] yes [20:10:51] they get ignored [20:10:55] o.O [20:11:13] their devs don't really hang out in their support channel, either [20:11:25] they have a developer channel and you aren't supposed to discuss support issues there [20:11:26] is there anything better than gluster? [20:11:33] even if they are related to bugs [20:11:40] we're switching to NFS [20:11:42] where is that channel here on freenode? [20:11:48] #gluster [20:11:51] and #gluster-dev [20:12:08] I don't see the point of bothering with them anymore [20:12:12] hm, and nfs is better? :P [20:12:19] we're switching away and have already switched a few projects [20:12:32] at our current usage, yes, nfs is better [20:12:37] yes I know, just wondering if nfs is going to be permanent [20:12:42] you said it's not well scalable [20:12:46] it isn't [20:13:06] so we are waiting for them to fix gluster or try something else? [20:13:12] but we can scale it for probably a year or two [20:13:24] there's a number of things they are fixing in gluster [20:13:29] almost 200 people in their channel... [20:13:33] we're also discussing things with the cephfs folks [20:13:49] cephfs is what? [20:14:06] a shared posix filesystem on top of ceph [20:14:19] gerrit is now just pissing me off >.< [20:14:23] with a kernel client [20:14:28] rather than a fuse one [20:16:09] Ryan_Lane: any idea what I would be getting [remote rejected] HEAD -> refs/for/master (change 61379 closed) ? [20:16:27] as far as I can tell thats talking to me about https://gerrit.wikimedia.org/r/#/c/61379 which is not anything to do with my commit [20:16:45] ^demon: ^^ [20:16:57] :P [20:17:02] does your commit have a changeid in it? [20:17:25] no, its a new change [20:17:42] there's a pre-commit hook you need to add [20:17:54] <^demon> That usually happens when trying to submit the same change-id to a different branch where it's already been merged. [20:17:56] that'll add a changeid to your changes [20:38:09] New patchset: Andrew Bogott; "Added fake mongodb::eventlogging passwords." [labs/private] (master) - https://gerrit.wikimedia.org/r/62722 [20:38:27] ori-l: For your reference, in case it's useful: ^ [20:45:42] Change merged: Andrew Bogott; [labs/private] (master) - https://gerrit.wikimedia.org/r/62722 [21:00:33] any idea where the pre commit hook is? [21:02:41] add-shore install git-review, and then run git-review -s in your repo [21:02:56] um… addshore: ^ [21:04:52] got it from scp ;p and i dont want gitreview ;p (+ im using windows) [21:29:55] [bz] (NEW - created by: Niklas Laxström, priority: High - major) [Bug 48203] Purging does not work on deployment-prep / beta labs - https://bugzilla.wikimedia.org/show_bug.cgi?id=48203 [21:34:18] Now just to fidn out how to push a second patch to the same gerrit change [21:36:13] addshore: You mean a new patch set? [21:36:20] yes :P [21:36:26] without using git review [21:38:18] addshore: amend your commit [21:38:38] since it's using the same change-id, it'll see that it needs to go into the same change as a new patchset [21:41:11] lovaly :D [21:50:39] Warning: There is 1 user waiting for shell: Indranisen (waiting 0 minutes) [22:04:09] Warning: There is 1 user waiting for shell: Indranisen (waiting 13 minutes) [22:17:39] Warning: There are 2 users waiting for shell, displaying last 2: Indranisen (waiting 27 minutes) Ji-eun.lee (waiting 11 minutes) [22:31:13] Warning: There are 2 users waiting for shell, displaying last 2: Indranisen (waiting 40 minutes) Ji-eun.lee (waiting 25 minutes) [22:44:43] Warning: There are 2 users waiting for shell, displaying last 2: Indranisen (waiting 54 minutes) Ji-eun.lee (waiting 38 minutes) [22:58:13] Warning: There are 2 users waiting for shell, displaying last 2: Indranisen (waiting 67 minutes) Ji-eun.lee (waiting 52 minutes) [23:24:50] Warning: There is 1 user waiting for shell: Anboia (waiting 0 minutes) [23:38:12] Warning: There is 1 user waiting for shell: Anboia (waiting 13 minutes)