[00:11:38] 05/04/2012 - 00:11:38 - Creating a project directory for preflights [00:11:39] 05/04/2012 - 00:11:38 - Creating a home directory for lcarr at /export/home/preflights/lcarr [00:11:39] 05/04/2012 - 00:11:38 - Creating a home directory for mark at /export/home/preflights/mark [00:12:41] 05/04/2012 - 00:12:40 - Updating keys for lcarr [00:12:41] 05/04/2012 - 00:12:40 - Updating keys for mark [00:15:38] Damianz: any idea why the mailman webui is not working? I think it might be to do with that lighttpd config but not sure [00:16:32] Yes [00:16:37] For the 237985723987234987236 time [00:16:49] lighttpd won't get configured until the htdigest file exists in the private repo [00:16:55] Which I can't create because it's broken [00:17:00] until then puppet is broken on that box [00:17:04] So it won't work ui wise. [00:17:09] * Damianz rangs randomly [00:17:18] the Damianz has been created [00:17:30] not the Damianz file, the htdigest file* :P [00:17:48] that's what maplebed did [00:04:09] New patchset: Bhartshorne; "adding an empty placeholder file for lighttpd/htdigest so that mailman will install in labs" [labs/private] (master) - https://gerrit.wikimedia.org/r/6566 [00:18:19] the puppetd -tv worked without any errors showing up either [00:18:30] * maplebed prefers "the Daimianz has been created." [00:19:28] * Damianz eats maplebed [00:21:12] I've heard that I don't taste very good. [00:21:24] Thehelpfulone: Need to edit the config [00:21:51] yeah that's what I thought, should I try copying the one from puppet with the tweaks to the hard coded URLs? [00:22:00] or I can let you do it ;) [00:22:38] eh [00:22:56] 1min [00:23:27] sure [00:24:29] Damianz: that seriously was some serious jacuzzing... [00:35:44] Totally missing a config but meh still fixing config stuff [00:39:23] Damianz: https://gerrit.wikimedia.org/r/gitweb?p=operations/puppet.git;a=blob_plain;f=files/lighttpd/50-mailman.conf;hb=HEAD could help [00:41:07] 1hy [00:41:11] just go away for a in [00:41:13] :P [00:44:07] hey, I'm getting failures to create some new instances ? [00:45:57] Ryan_Lane: is something up with creation ? i am making a moderate amount of instances … is there some speed limit or did i bork the virts ? [00:48:50] LeslieCarr: what kind of failures? [00:49:19] just "Failed to create instance." [00:49:33] that's abnormal [00:49:45] yay, i kill everything ! [00:50:15] Wait till you break puppet [00:50:18] that was a fun day [00:50:35] Ryan_Lane: Does the precise image in labs work at this time? [00:52:14] New patchset: DamianZaremba; "Disable https redirection for labs" [operations/puppet] (test) - https://gerrit.wikimedia.org/r/6584 [00:52:28] New review: gerrit2; "Lint check passed." [operations/puppet] (test); V: 1 - https://gerrit.wikimedia.org/r/6584 [00:54:46] Thehelpfulone: That should be working now atm me thinks but it iwll borke when puppet runs until that changes passes [00:55:51] ok [00:57:23] err: Could not request certificate: getaddrinfo: Name or service not known [00:57:28] LeslieCarr: it's possible you hit your quota [00:57:32] LeslieCarr: can you try again? [00:57:35] Thehelpfulone: that's normal [00:57:39] wait [00:57:40] oh I see [00:57:40] no that's not [00:57:45] o.0 [00:57:46] Thehelpfulone: where are you seeing that? [00:57:55] trying to run puppetd -tv [00:57:56] on mailman-01 [00:58:06] ok [00:58:11] LeslieCarr has awesome hair :o Really need to fix mine [00:58:22] LeslieCarr: quota [00:58:27] damn [00:58:33] can you up mine plz ? [00:58:44] yeah [00:58:46] Damianz: I'm totally willing to help play hairstylist ! [00:58:47] which? [00:59:03] LeslieCarr: btw, you're spiking the load on the systems [00:59:09] can you space the creations out a little? :) [00:59:17] until we replace shitty, shitty glusterfs [00:59:30] LeslieCarr: Urgh you totally don't want to see my current hair then :P [00:59:59] LeslieCarr: which project? [01:00:10] Ryan_Lane: preflights [01:00:14] oh sorry [01:00:15] hehehe [01:00:21] preflights? [01:00:24] yeah [01:00:53] * Damianz stops trolling wikipedia for vandals and random labs people and goes back to work [01:02:11] Damianz: http://mailman.wmflabs.org/ is a 403 forbidden instead of the it works page, to do with this certificate error? [01:03:03] ooh, http://mailman.wmflabs.org/mailman/listinfo works :D [01:03:16] I know [01:03:18] no mailing lists there yet, but it's a start! [01:03:24] cause I'm just lame like that [01:10:34] Ryan_Lane: any idea how to fix that error? we need a master puppeteer ;) [01:11:51] Ryan_Lane: am I killing labs again with my deletions ? :) [01:12:12] Thehelpfulone: can you paste into pastebin the full output when you run puppet ? I am leaving in 5 minutes but can try to help :) [01:12:23] LeslieCarr: it's spiking again, yeah [01:12:31] ok, I stop the deleting [01:12:31] sue [01:12:32] Thehelpfulone: What error? ligttpd fails to restart? That's a ssl issue [01:12:32] maybe wait 10-15 mins or so [01:12:56] http://pastebin.com/cy3eFbeW [01:12:57] * Damianz gives LeslieCarr 512kb's of bw rather than a gig :D [01:13:23] You're not root [01:13:28] oh duh [01:13:44] * Thehelpfulone *facepalm* [01:14:14] wtf [01:14:25] Do people no longer have sudo on instances they creatE? [01:14:33] yeah it's a bug [01:14:45] you need to give yourself sudo through the group policy RoanKattouw [01:14:57] I don't think the default one is working - Ryan_Lane ^ bug report #2 [01:15:08] How do I do that exactly? [01:15:19] The project list doesn't let me do it [01:15:25] Oh "sudo policies" [01:15:25] https://labsconsole.wikimedia.org/wiki/Special:NovaSudoer [01:15:26] Sudo policy [01:16:26] sorry if this is a stupid question Damianz, but I thought that the whole point of puppet was that it would get all the existing lists and thus http://mailman.wmflabs.org/mailman/listinfo would be populated? [01:16:43] no? [01:19:37] hmm.. so can you get those lists into labs? [01:27:11] RoanKattouw: ah. crap. right. I need to fix that bug [01:27:40] RoanKattouw: go to "Manage sudo policies" [01:27:50] RoanKattouw: create a policy called "default" [01:27:56] with ALL for hosts, users, and commands [01:31:23] Oooh commands==ALL too [01:35:43] !log visualeditor Installing nodejs on ve-nodejs [01:37:26] nodejs? Arrrrrrgggggggghhhhhhhhhhh [02:13:30] Ryan_Lane: Special:NovaProjects doesn't list all the products and who's got access where anymore, I remember you said that you'd played around with that - what's the new page to see all of that information? [02:14:28] well, you can see who's members of projects, but not sysadmins or netadmins [02:14:33] I need to fix that [02:15:01] * Damianz yawns [02:16:25] I can't even see who members of projects are if I'm not a part of that project [02:16:32] where should I be looking? [02:16:52] !projects [02:16:52] https://labsconsole.wikimedia.org/wiki/Special:Ask/-5B-5BResource-20Type::project-5D-5D/-3F/-3FMember/-3FDescription/mainlabel%3D-2D [02:17:07] it's also on the front page. It's linked from "Number of projects" [02:17:09] ah that's easy to remember.. :P [02:17:31] can you add a link to the sidebar? [02:20:45] Ryan_Lane: are there any requirements to get a git/labs account? [02:21:12] ask for one and have a reason to have one [02:21:18] agree to the terms of use [02:21:39] https://www.mediawiki.org/wiki/Developer_access - no reason required there? [02:21:58] nope. just ask [02:22:32] so it's when it comes to assigning projects to your labs account that someone is actually checked for trustworthiness, or is that also a "just ask"? [02:23:31] * Damianz finds his gun and checks if Thehelpfulone is a un-helpful troll [02:23:57] Damianz: I was trying to figure out how you got access :P [02:24:37] I had sex with a computer and it resulted in bots to look after [02:48:58] THO|Cloud: almost back [02:49:12] err, you're gone [05:36:00] New review: Hashar; "Indeed mark does not want that utility in production which I can understand. But I have had it insta..." [operations/puppet] (test); V: 0 C: 0; - https://gerrit.wikimedia.org/r/6468 [09:41:48] mutante: I am trying to mkdir -p /home/wikipedia/syslog on an instance [09:41:54] and get a file or directory not found [09:41:55] :-( [09:42:12] could it be caused by the virtualization? [09:42:55] root@i-000000d2:~# mkdir -p /home/wikipedia [09:42:55] mkdir: cannot create directory `/home/wikipedia': No such file or directory [09:42:56] hashar: yea, i guess because /home dirs are shared [09:43:22] how is that shared ? According to `mount` that is on /dev/vda [09:43:27] on the glusterfs [09:43:31] oh my god [09:43:40] I am not going to touch that [09:43:57] anyway I end up being in need of a shared /home/wikipedia directory [09:44:22] or any over shared directory, then I will just update any reference to /home/wikipedia in puppet :-] [09:44:23] create a wikipedia user via puppet? [09:44:46] i did that on an instance, made a system user via puppet, gave it a home [09:44:52] (which wasnt in /home though) [09:45:34] hashar: gotta lookup the automount stuff Ryan implemented.. [09:45:58] you can mount more space on central storage [09:47:13] ideally I would like to have /home/wikipedia to avoid rewriting classes in puppet [09:47:24] hashar: https://gerrit.wikimedia.org/r/#change,533 [09:47:58] automount based on projects.. i didnt look at the details, but .. hmm [09:48:24] and things are changing [09:49:04] the more I look at the beta project the more stuff I have to figure out :-( [09:50:11] hashar: just add a systemuser via puppet, with /home/wikipedia.. does it work? [09:50:22] or does it fail inside /home [09:50:28] I have no idea :D [09:50:47] would have to edit puppet, submit change, ask you to merge it, wait for cronjob to kick in, run puppetd -tv [09:50:50] ;-] [09:51:21] i'll merge:) [09:51:31] looking for a wikipedia user [09:52:32] systemuser { wikistats: name => 'wikistats', home => '/var/lib/wikistats', groups => [ 'project-wikistats' ] } [09:52:48] <-- i don't use /home , but i do try to put it into the project- labs group [09:53:07] but that won't work. i opened a bug about it [09:53:43] because if not in that group, currently not allowed to run cron [09:57:18] figuring out where to put that now :-D [10:00:48] which classes are you using in this case? [10:01:26] mediawiki.pp ? [10:04:03] going to be nfs::home::wikipedia [10:04:52] New patchset: Hashar; "nfs::home::wikipedia describe /home/wikipedia" [operations/puppet] (test) - https://gerrit.wikimedia.org/r/6545 [10:05:02] mutante: ^^^ [10:05:07] New review: gerrit2; "Lint check passed." [operations/puppet] (test); V: 1 - https://gerrit.wikimedia.org/r/6545 [10:13:53] New review: Dzahn; "inline comments" [operations/puppet] (test); V: 0 C: 0; - https://gerrit.wikimedia.org/r/6545 [10:16:29] mutante: I have duplicated fenari configuration [10:16:55] [10:16][hashar@fenari(mw-inst):~]$ ls -ld /home/wikipedia [10:16:56] drwxr-xr-x 29 root root 4096 2012-04-13 13:04 /home/wikipedia/ [10:17:44] New review: Hashar; "(no comment)" [operations/puppet] (test); V: 0 C: 0; - https://gerrit.wikimedia.org/r/6545 [10:18:34] New review: Dzahn; "ok, the case for labs and prod looks good, given that the directory with root owner matches fenari s..." [operations/puppet] (test); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6545 [10:18:38] Change merged: Dzahn; [operations/puppet] (test) - https://gerrit.wikimedia.org/r/6545 [10:18:55] mutante: while you are at it, look at https://gerrit.wikimedia.org/r/6546 [10:19:00] that creates /home/wikipedia/syslog :-D [10:19:03] New review: Dzahn; "we also want to see if we can create the home in /home with puppet" [operations/puppet] (test) - https://gerrit.wikimedia.org/r/6545 [10:19:04] might want to merge both [10:19:15] or first check if puppet actually create the homedir [10:19:30] I am pretty sure it is not going to work though ;-( [10:21:10] hashar: project "beta" right [10:21:17] or has it been renamed already [10:21:48] should be betalabs [10:21:57] though in Nova it is known as deployment-prep [10:23:35] ah, thats what i meant [10:23:44] has it been renamed from deployment-prep to something shorter [10:23:47] ok [10:25:37] !log deployment-prep Creating 5 new m1.large instances hosting apaches and named deployment-apacheXXX [10:27:01] of course it does not log anything [10:27:02] !log [10:27:07] hashar: issue with project filter i think :p [10:27:25] hashar: i can't see deployment-prep in the puppet group list , but it is enabled in project filter [10:27:44] arg, and log bot down?:) [10:27:57] ok..one by one.. [10:27:58] !log deployment-prep foo [10:28:24] * mutante gets the bot docs from last time [10:29:01] forget about the bots :-D [10:30:11] https://labsconsole.wikimedia.org/wiki/Nova_Resource:Bots/Documentation#labs-morebots [10:30:24] fixing [10:32:43] hashar: sigh.. more breakage detected:) [10:33:01] bots-2 instance... and User: talk pages on labsconsole wiki empty [10:33:09] \O [10:34:57] !log bots restarted labs-morebots on bots-2 [10:34:58] Logged the message, Master [10:35:50] hashar: feel free to repeat logging depl-prep stuff [10:36:12] next on the issue why i cant see the project to add the puppet class [10:36:21] !log deployment-prep Creating 5 new m1.large instances hosting apaches and named deployment-apacheXXX [10:36:22] Logged the message, Master [10:36:24] danke [10:36:36] can we try the /home/wikipedia hack ? [10:36:53] yea, thats what i want to do, but you need the class to select in labsconsole [10:37:10] and when i want to add it for you.. i see like all projects BUT the one i need [10:37:19] it is already there IIRC [10:37:27] it is? but you just created it [10:37:31] nfs::home::wikipedia being a dependency of misc::syslog-server [10:37:48] checking [10:38:11] ok, so you did not expect to select it as a group in labsconsole [10:38:12] misc::syslog-server checked [10:38:20] but it to be pulled in automatically as dependency [10:38:31] uh, why another bot [10:40:13] so misc::syslog-server does create a /home/wikipedia/syslog directory which in turn requires nfs::home::wikipedia [10:40:30] !log bots killed duplicate adminbot procs, restarted one [10:40:31] Logged the message, Master [10:43:03] start-bot conflict zzz [10:46:47] !log deployment-prep added myself to admin groups to add/change puppet groups [10:46:48] Logged the message, Master [10:48:12] !log deployment-prep added class nfs::home::wikipedia to puppet group list in "beta-labs" [10:48:14] Logged the message, Master [10:48:36] hashar: you could now try it with, or without dependency, by selecting/deselecting it for an instance directly [10:48:51] hashar: wanna test that? i have other stuff waiting..kind of... [10:49:23] I got an instance with misc::syslog-server already [10:49:29] will run puppet on it to see [10:49:36] cool [10:51:05] New patchset: Hashar; "class to install the ack-grep utility" [operations/puppet] (test) - https://gerrit.wikimedia.org/r/6467 [10:51:19] New review: gerrit2; "Lint check passed." [operations/puppet] (test); V: 1 - https://gerrit.wikimedia.org/r/6467 [10:52:05] New review: Hashar; "patchset 3 is just a rebase" [operations/puppet] (test); V: 0 C: 0; - https://gerrit.wikimedia.org/r/6467 [10:56:49] mutante: I needed https://gerrit.wikimedia.org/r/#change,6546 which implements the dependency [10:56:56] will be for next week :-D [10:57:14] need to get out for some appointments then prepares my SF trip [11:01:34] hashar: ok, fine, ttyl [11:01:54] still on server upgrades and they also wait for new OS installs [11:01:57] bbl [11:02:06] thanks mutante! [11:02:08] see you next week [11:02:13] yw.cya [12:12:30] petan|wk: seen your bug about deployment.wmflabs.org not redirecting [12:12:46] petan|wk: I am in a hurry this afternoon, have several appointments and need to prepare a trip to SF [12:19:57] commented on it [12:20:10] deployment.wmflabs.org virtual host does not exist anymore [12:20:14] and need to be recreated [12:20:22] I am off! [13:13:49] mutante: these 5 new isntances is there any plan with them [13:20:22] petan|wk: i don't know. hashar made them. (yeah, i saw it says my name next to one of them in RC, but i don't know why, because all i did was add my user to roles and a puppet class, but not instances [13:20:41] ok [13:20:50] since he is going to be travelling I will prepare them... [13:20:59] cool [13:21:10] I am wondering why he made 5 of them [13:21:18] Ryan was like: be carefull with resources [13:21:27] i dont know why 5 [13:21:33] and site isn't having troubles with performance [13:58:28] New review: Demon; "(no comment)" [operations/puppet] (test); V: 0 C: 2; - https://gerrit.wikimedia.org/r/6541 [14:02:11] Change abandoned: Demon; "No longer needed, was submitted as an amend to I4199c995ac37b87a81dc03a0800f718d67622220" [operations/puppet] (test) - https://gerrit.wikimedia.org/r/6543 [14:02:38] New review: Demon; "I meant I66ac39a6f19e043c218d4ba480bd9731f81365dd." [operations/puppet] (test) - https://gerrit.wikimedia.org/r/6543 [14:13:15] maxsem@bastion1:~$ ssh mobile-enwp [14:13:16] ssh: connect to host mobile-enwp port 22: No route to host [14:13:22] what's going on? [14:14:22] no idea [14:14:28] but I have same problem [14:14:34] connectivity is down [17:02:31] MaxSem: check its console log [17:02:34] it OOM'd [17:02:41] MaxSem: you need to reboot it [17:02:53] I'll do that for you [17:15:50] <^demon> Ryan_Lane: So robla pencilled us un for like...right now. But we've got the whole day--when works best for you to knock this out? [17:18:06] I think now is fine [17:18:11] have you backed up the database? [17:18:20] how long will the outage be? [17:18:37] no clue [17:18:41] ^demon: short, right? [17:18:45] a minute or two? [17:19:01] <^demon> Um, like 4-5 minutes tops. [17:19:05] ^demon: are the repos backed up too? [17:19:13] <^demon> I'll go ahead and back everything up [17:19:19] ok [17:19:58] y'all should probably at least give a heads up on #wikimedia-dev and #mediawiki [17:20:27] <^demon> nobody likes gerrit anyway, they'll cheer ;-) [17:23:26] <^demon> Probably good to go ahead and shut down gerrit before I back stuff up, so nobody does anything after the backup. [17:23:37] want me to do so? [17:23:52] <^demon> Yes, please. [17:26:20] <^demon> Database backed up. [17:27:46] <^demon> Git repos backed up too. [17:31:36] ok [17:31:40] force running puppet [17:32:04] ^demon: how do I need to run gerrit the first time? [17:32:10] also, do I need to run it as the gerrit2 user? [17:32:23] <^demon> Second question: yes [17:32:46] <^demon> First question: `java -jar new-war-file.war init -d /var/lib/gerrit2/review_site` [17:34:51] all defaults? [17:34:58] to the questions? [17:35:20] ^demon: ? [17:35:35] <^demon> It'll pull your current config, so defaults should all be fine. [17:35:40] ok [17:35:47] <^demon> But it's always good to look before blindly pressing enter :) [17:36:11] hm. it didn't put a value for the user [17:36:14] ldap user [17:36:23] <^demon> Grrr. [17:36:35] oh [17:36:45] <^demon> Config gets overwritten by puppet anyway, so we could just re-force a puppet run after we're done with init. [17:36:51] <^demon> The main thing we want init for is the schema changes. [17:36:53] yeah [17:37:51] ok [17:37:54] it's upgrading the schema [17:39:02] <^demon> We could theoretically make the debian package run init for us, right? [17:39:21] no. I have puppet doing that [17:39:28] I hate when packages do that [17:39:30] <^demon> Ah ok. [17:39:33] it installs it, though [17:39:43] <^demon> "Something automated" was what I was going for :) [17:39:49] yeah [17:39:51] it's automated [17:39:54] oh [17:39:58] you mean the upgrade? [17:39:59] fuck that [17:40:07] I don't trust packages or puppet to do that [17:40:14] I thought you meant the init script [17:40:14] heh [17:40:29] <^demon> Noo, the war's init thingie :) [17:40:34] definitely not [17:40:43] <^demon> Fair enough. [17:41:07] <^demon> Still schema updating? [17:41:26] mother fucker [17:41:39] we didn't shut down formey [17:41:50] <^demon> Oh, whoops. forgot all about the mirror. [17:42:11] the schema update failed due to a concurrency exception [17:42:13] lemme try that again [17:42:27] <^demon> That would explain why gerrit is still pretending to work? [17:42:50] maybe [17:42:59] I shut down formey, though [17:43:16] wait [17:43:21] is it actually running on manganaese too? [17:43:29] ugh [17:43:30] puppet [17:43:35] it probably started it [17:43:47] <^demon> Heh, puppet didn't want you to shut it off :p [17:44:05] well. I'm going to wait for this init to stop [17:44:12] if it fails I'll stop the server [17:44:39] since I'm not totally sure it isn't just the init process running [17:46:20] meh [17:46:23] I just stopped it [17:46:28] schema update is going on now [17:46:53] <^demon> Ok, I only see the one process belonging to gerrit2 now. [17:46:55] <^demon> Instead of 2 [17:46:59] yep [17:47:00] me too [17:47:05] I stopped puppet too [17:47:25] I need to remember to install gerrit 2.3 on formey too :D [17:47:43] <^demon> Yeah. Won't have to do init though :) [17:47:46] yeah [17:48:40] well, this is taking a while [17:50:13] I wish this was more verbose [17:50:36] ^demon: 4-5 minutes tops :-P [17:50:47] <^demon> Pfft. Puppet interrupted us :p [17:51:00] the schema upgrade is taking ages [17:51:09] <^demon> What number is it on? [17:51:20] 59->60 [17:51:36] <^demon> I think there's only 61 on 2.3 [17:52:37] <^demon> Their schema stuff is totally ORMd...it's really hard to figure out what does what since it's so freakin' abstract. [17:54:13] <^demon> Oh damn...schema change 60 sucks. It iterates all changes and makes sure they have proper Change-Id's. [17:54:17] ok, going to 60 -> 61 [17:56:30] seems to be working [17:56:49] <^demon> I'm hanging on "Loading Gerrit Code Review..." [17:57:02] it made me log back in, but it's back up for me [17:59:21] <^demon> Hmm, still hanging for me :\ [18:01:16] hm for me too [18:01:21] wait [18:01:21] no [18:01:24] it's working for me [18:01:25] <^demon> Works in Firefox. [18:01:32] <^demon> Must've just been chrome stuck caching something. [18:01:36] working for me in Chrome [18:01:44] same [18:02:46] easy enough [18:03:42] <^demon> We can mark operations/private as hidden now :) [18:03:50] \o/ [18:04:44] <^demon> And my typos :p [18:05:48] robla: we still made it below the end of the window :D [18:06:09] <^demon> Ugh, we need to shut off those `repo` instructions on changes. [18:06:12] <^demon> None of us use repo. [18:06:56] heh [18:07:04] puppet changes just restarted gerrit [18:07:12] <^demon> repo is a huge PITA. [18:07:16] <^demon> I tried playing with it. [18:07:23] gerrit review is a lot better, eh? [18:12:34] <^demon> Hrm? [18:12:49] than repo [18:13:36] <^demon> Eh, git-review's ok. I think it could be better. [18:13:50] <^demon> There is a git-gerrit I saw the other day that might be worth trying. [18:14:30] <^demon> I *still* can't get chrome to work with gerrit. [18:14:54] close and reopen? [18:14:56] it works for me [18:15:21] <^demon> I closed the browser, cleared the cache. Nothing. [18:15:27] heh. I just told you to try turning it off and on again [18:15:44] clear your cookies? [18:16:06] <^demon> Yup [18:16:10] <^demon> Some stuck cookie [18:16:32] * Ryan_Lane nods [18:25:44] Thehelpfulone: what's the story? [18:25:58] so http://mailman.wmflabs.org/mailman/listinfo is up thanks to Damianz [18:26:08] but I'm not sure how to get all the lists into that [18:26:38] uhh, what about the whitespace? [18:28:18] that's what I meant, I'm trying to figure out how to populate it with all the lists from https://lists.wikimedia.org/mailman/listinfo [18:28:40] uhhh, you could scrape? [18:28:50] idk what you're trying to do [18:29:19] oh, it looks like ^demon just fixed it for you [18:29:23] i was thinking about that ;P [18:31:13] <^demon> Ryan_Lane: https://gerrit.wikimedia.org/r/#/c/6620/ fixes the download options [18:31:18] oh, i can tell this must be a new gerrit. it's behaving differently [18:31:33] download options? [18:32:03] no. [18:32:06] no http [18:32:16] no anon http either [18:32:19] it's insecure [18:32:36] also, how does this fix it? [18:32:40] gerrit is being proxied to [18:32:45] unless http is code for https? [18:32:54] you'd need to fix it at apache and gerrit [18:33:03] gerrit is running on 127.0.0.1 [18:33:10] ^demon: I'm rejecting this [18:33:37] <^demon> It should be based on the canonicalUrl, I thought. [18:33:52] apache is force redirecting http to https [18:35:50] <^demon> I can resubmit with https instead of http. But nothing should need to be done at the apache level. [18:36:03] what's broken right now? [18:36:46] <^demon> Pfft. Nothing. [18:36:55] so what's the purpose for the change? [18:37:05] <^demon> This is why we need to reskin gerrit so I can remember when I'm looking at the real gerrit vs. my local copy vs. gerrit's gerrit [18:37:29] heh [18:40:19] Thehelpfulone: so... what are you tyring to do? why copy those in? [18:56:08] RoanKattouw: how do you use that variable in the prompt? [19:09:18] jeremyb: sorry, was eating - so basically the plan is to make the mailman list info pages look nicer - something more like https://lists.wikimedia.org/mailman/listinfo/commons-poty [19:09:43] but do it globally across all of them, this is to test that doing that won't break anything for future upgrades [19:14:35] Oh god that's ugly [19:14:57] Ryan_Lane: function git_dirty as in your code [19:15:04] Then: [19:15:06] function dirty_git_prompt { [19:15:07] git_dirty && __git_ps1 [19:15:09] } [19:15:10] function clean_git_prompt { [19:15:12] git_dirty || __git_ps1 [19:15:14] } [19:15:35] If you want it without dirty/clean checking, you can just use $(__git_ps1) [19:15:42] (it's just a command) [19:16:55] Damianz: of course that's will be on git for people to adjust/tweak at will [19:17:05] or completely redesign if you want to! [19:19:45]