[00:07:53] How can I join to nginx project on lab? [00:08:08] magic [00:09:04] ask Ryan_Lane paravoid or ottto or someone else I don't know of in that project to add you [00:09:23] devunt: what would you like to do in it? [00:11:59] Ryan_Lane, make and build some nginx module, integration with mediawiki. [00:13:36] hmm do I fancy a trip to china this year... might just do ams [00:14:00] devunt: done [00:14:13] thanks toy [00:14:16] you [00:14:36] csteipp: it's not being sent data properly [00:14:44] csteipp: that's been broken for a bit now [00:14:44] yw [00:15:11] maybe we should just replace it with salt runners on a schedule and graphite ;) [00:15:21] down again [00:15:21] heh [00:16:39] it pretty much breaks on every puppet run [00:17:31] yep [00:17:50] I'm working on other things right now, or I'd push fixes in [00:18:16] and I'm going to bed so I don't care to fix it atm :P [00:18:34] heh [00:18:36] Damianz: night [00:18:40] * Damianz waves [00:20:10] One of these day's I'll figure puppet out. Restarted it again (for now) [00:22:51] csteipp: restarting it probably won't help [00:36:43] Damianz, still awake perchance? [00:37:00] I'm reading salt code again, could use a guide [00:37:09] * Damianz licks andrewbogott behind the ear [00:37:34] So… there's a python client. It talks to salt via REST? [00:37:49] And is the stuff in the auth/ dir used for that, or for something else? [00:38:36] no python client I know of -that would be cool [00:38:57] There's `salt` which talks to the master locally via functions that does zeromq magic to talk to the minions [00:39:06] http://docs.saltstack.org/en/latest/ref/python-api.html <- this is something else? [00:39:11] then there's the rest api code, which uses the same functions to provide a rest (http api) [00:39:26] that is the 'internal' api [00:39:40] Ah, ok. So there's nothing that wraps the rest api atm? [00:39:55] not that I know of [00:40:06] basically that api there is what the salt-api code uses to provide a http interface [00:40:12] the idea being rest is easy to talk to from anything [00:40:15] !logs [00:40:15] logs http://bots.wmflabs.org/~wm-bot/logs/%23wikimedia-labs [00:40:53] RECOVERY Free ram is now: OK on sube.pmtpa.wmflabs 10.4.0.245 output: OK: 22% free memory [00:41:01] The stuff in the auth directory is confusing - it's mostly used from the salt-api code to do authentication, but shares configuration settings and maybe (eventually) auth calls from salt its self [00:41:20] Since the authorization is within the salt-master and authentication in the salt-api code (confusingly) [00:41:53] I think I understand the difference... [00:42:22] So, one dumb one: Where is the rest implementation? I don't see an obvious directory name [00:42:43] is 'wheel' the same thing as the rest api? [00:42:46] Or somehow related? [00:43:12] seperate repo - https://github.com/saltstack/salt-api/ [00:43:28] oooh, no wonder I can't find it! [00:43:33] https://github.com/saltstack/salt-api/blob/develop/saltapi/__init__.py is the salt chunky code, https://github.com/saltstack/salt-api/blob/develop/saltapi/netapi/rest_cherrypy/__init__.py the restish code [00:43:40] it's really, really, really simple code [00:43:53] so what the heck is wheel then? [00:43:59] wheel? [00:44:14] oh [00:44:33] I /think/ it's like ENC in puppet [00:45:08] Under 'wheel' the docs say "A new API was added to the Salt Master which allows the master to be managed via an external API." [00:45:11] see there's also 'external auth' which ties into the master to support acl stuff [00:45:15] Which sounds like... [00:46:33] At a guess I think wheel is for managing the config and the rest api for interacting with minions.... like a http interface to the `salt` command line [00:46:58] ok. I will do my best to ignore wheel then. [00:47:13] looking at the functions it's stuff like accepting/deleting keys etc, which is more management [00:47:30] So, just to make sure I'm clear… code in salt/auth is used to authenticate users of salt-api? Despite being in a different repo? [00:47:46] yes [00:47:50] no idea why [00:48:24] ok. [00:48:28] So, I will… read. [00:48:42] I think auth is in salt because it's also used by 'external auth' [00:48:45] Is it semi-obvious how to get the rest api up and running? [00:49:05] pretty much - download it to a master, run the bin file call salt-api or something [00:49:39] http://salt-api.readthedocs.org/en/latest/#getting-started ;) [00:49:47] the docs are actually a bit crap tbf [00:49:57] probably better than nothing. [00:50:29] Thanks for the primer! No doubt I will have more questions tomorrow if I am not distracted by other things like every other time I've started to read this code... [00:50:35] So really we'd probably use salt-api for the interface, salt/auth for keystone token based authentication and external auth for acls (needs group support adding) to allow users (authenticated via auth from the api) to run 'stuff' [00:50:45] yay to confusion [00:51:20] I think I follow all that. I mean, salt-api will manage the authentication and use salt/auth on the backend, right? [00:51:40] And then external auth controls what is allowed to, once has been identified. [00:51:47] …? [00:52:13] yeah so salt-api is literally just an interface and the meat is done within the main salt code [00:52:17] and yep [00:52:31] cool [00:52:39] ideally (maybe in .13 release) eauth will support groups so we can make auth return the users project membership and control acls on that (hopefully) [00:55:36] Hmm I might throw a pr to salt tomorrow, want sub-module support in modules so I can do cool monitoring stuff [01:03:52] PROBLEM Free ram is now: WARNING on sube.pmtpa.wmflabs 10.4.0.245 output: Warning: 13% free memory [04:38:52] RECOVERY Free ram is now: OK on sube.pmtpa.wmflabs 10.4.0.245 output: OK: 21% free memory [04:51:52] PROBLEM Free ram is now: WARNING on sube.pmtpa.wmflabs 10.4.0.245 output: Warning: 13% free memory [06:29:24] PROBLEM Total processes is now: WARNING on parsoid-roundtrip4-8core.pmtpa.wmflabs 10.4.0.39 output: PROCS WARNING: 152 processes [06:34:23] RECOVERY Total processes is now: OK on parsoid-roundtrip4-8core.pmtpa.wmflabs 10.4.0.39 output: PROCS OK: 149 processes [07:09:23] PROBLEM Current Load is now: WARNING on parsoid-roundtrip7-8core.pmtpa.wmflabs 10.4.1.26 output: WARNING - load average: 5.06, 5.16, 5.07 [12:41:52] RECOVERY Free ram is now: OK on sube.pmtpa.wmflabs 10.4.0.245 output: OK: 21% free memory [12:59:52] PROBLEM Free ram is now: WARNING on sube.pmtpa.wmflabs 10.4.0.245 output: Warning: 13% free memory [13:22:12] petan: I usually like to avoid subpages because it makes it difficult to move things around [13:22:41] but it likely makes sense for a lot of the docs [13:22:46] like bots and tools [13:23:07] it's better than namespaces, for sure [14:25:17] Ryan_Lane: I would like to add some documentation to the Help:Self-hosted puppetmaster page. (A bit more for newbies, hehe) [14:25:27] Silke_WMDE: have at it [14:26:04] Am I correct that right now the way to submit one's own stuff to gerrit is via one's local computer? [14:26:17] (That's what I did.) [14:26:36] yea, you basically just edit stuff in /etc/puppet/ [14:27:09] the drawback is there is no easy way to switch back from self-hosted to the "real" puppetmaster, besides using a fresh instance [14:27:17] yes [14:27:26] We should really have a way of disabling it [14:28:55] And I was also wondering about the variables thing: I called my variables $wikidata_something. Why did I used "wikidata"? My project is called "wikidata-dev", my main puppet file is called wikidata.pp. Is the latter where I derive the prefix for my variables from? [14:29:24] Or could I use anyting? [14:29:44] you could just have 2 instances, one self-hosted and one with ext. master, initial tests on the self-hosted one, and after things work you submit them to gerrit [14:30:38] i don't think the name of the .pp file matters at all to variable names [14:30:55] I haven't really understood what it needs to make labsconsole talk to the variables in my puppet files. They are working but I just followed your working example. [14:30:55] besides it is maybe a nice covention for humans [14:32:01] ok [14:32:54] the variables are added to labsconsole via webui, the same way puppet classes are added to appear in "puppet groups" [14:33:19] eh..right Ryan.. did things change there [14:34:37] I wouldn't mind for someone to make that not suck :) [14:34:43] ok, so my "wikidata" prefix for variables is not a technical necessity to make them be recognized but it's to make them more readable. [14:34:49] that's my least favorite interface in labsconsole [15:31:11] Change on 12mediawiki a page Wikimedia Labs/Toolserver features needed in Tool Labs was modified, changed by Johannes Kroll (WMDE) link https://www.mediawiki.org/w/index.php?diff=642504 edit summary: [+984] /* Toolserver features needed in Tool Labs */ [15:31:50] Change on 12mediawiki a page Wikimedia Labs/Toolserver features needed in Tool Labs was modified, changed by Johannes Kroll (WMDE) link https://www.mediawiki.org/w/index.php?diff=642505 edit summary: [+2] /* Render */ [15:32:35] Change on 12mediawiki a page Wikimedia Labs/Toolserver features needed in Tool Labs was modified, changed by Johannes Kroll (WMDE) link https://www.mediawiki.org/w/index.php?diff=642507 edit summary: [-115] /* Render */ [15:35:34] @rss-on [15:35:34] Rss feed has been enabled on channel [15:58:03] ping jcmish [16:08:39] [bz] (8NEW - created by: 2Nemo, priority: 4Unprioritized - 6normal) [Bug 44756] [Errors 301] Many entry URLs to be updated - https://bugzilla.wikimedia.org/show_bug.cgi?id=44756 [16:30:47] Hi! I'm trying to use /data/project/pywikipedia, but I can't seem to make it use my own user-config.py file. Does anyone knows how to do this properly? [16:34:06] Dakdada which project [16:34:23] Bots [16:34:37] tbh I am no expert on python, maybe andrewbogott_afk knows [16:34:52] if you type @notify andrewbogott_afk bot will ping you when he's around [16:35:34] Wait, I'll ask the #wikipediabot first [16:39:52] RECOVERY Free ram is now: OK on sube.pmtpa.wmflabs 10.4.0.245 output: OK: 21% free memory [16:55:03] <^demon> I just noticed wm-bot does BZ changes too. Maybe we should kill wikibugs. [17:02:52] PROBLEM Free ram is now: WARNING on sube.pmtpa.wmflabs 10.4.0.245 output: Warning: 13% free memory [17:07:37] Alright, it looks like it is not possible to do what I want. For now it seems that all pywikipedia users will have to have their own copy. [17:07:54] Change on 12mediawiki a page Wikimedia Labs/Toolserver features needed in Tool Labs was modified, changed by Johannes Kroll (WMDE) link https://www.mediawiki.org/w/index.php?diff=642567 edit summary: [+1] s/is/are [17:09:23] RECOVERY Current Load is now: OK on parsoid-roundtrip7-8core.pmtpa.wmflabs 10.4.1.26 output: OK - load average: 4.48, 4.85, 4.98 [17:17:23] PROBLEM Current Load is now: WARNING on parsoid-roundtrip7-8core.pmtpa.wmflabs 10.4.1.26 output: WARNING - load average: 6.99, 5.94, 5.39 [17:42:52] PROBLEM Current Load is now: WARNING on ve-roundtrip2.pmtpa.wmflabs 10.4.0.162 output: WARNING - load average: 7.55, 7.00, 5.54 [17:46:52] PROBLEM Current Load is now: WARNING on parsoid-roundtrip3.pmtpa.wmflabs 10.4.0.62 output: WARNING - load average: 8.09, 7.16, 5.71 [17:56:53] RECOVERY Current Load is now: OK on parsoid-roundtrip3.pmtpa.wmflabs 10.4.0.62 output: OK - load average: 2.91, 3.86, 4.78 [18:02:52] RECOVERY Current Load is now: OK on ve-roundtrip2.pmtpa.wmflabs 10.4.0.162 output: OK - load average: 3.61, 3.94, 4.82 [18:04:22] PROBLEM Current Load is now: WARNING on parsoid-roundtrip6-8core.pmtpa.wmflabs 10.4.0.222 output: WARNING - load average: 5.34, 5.54, 5.11 [18:12:43] PROBLEM Free ram is now: WARNING on bots-4.pmtpa.wmflabs 10.4.0.64 output: Warning: 12% free memory [18:17:42] PROBLEM Free ram is now: CRITICAL on bots-4.pmtpa.wmflabs 10.4.0.64 output: Critical: 4% free memory [18:32:42] RECOVERY Free ram is now: OK on bots-4.pmtpa.wmflabs 10.4.0.64 output: OK: 83% free memory [18:39:23] RECOVERY Current Load is now: OK on parsoid-roundtrip6-8core.pmtpa.wmflabs 10.4.0.222 output: OK - load average: 2.57, 3.30, 4.63 [19:09:52] PROBLEM Current Load is now: WARNING on parsoid-roundtrip3.pmtpa.wmflabs 10.4.0.62 output: WARNING - load average: 6.07, 5.84, 5.25 [19:25:53] PROBLEM Current Load is now: WARNING on ve-roundtrip2.pmtpa.wmflabs 10.4.0.162 output: WARNING - load average: 6.47, 5.86, 5.25 [19:45:43] PROBLEM Free ram is now: WARNING on bots-4.pmtpa.wmflabs 10.4.0.64 output: Warning: 17% free memory [20:10:53] RECOVERY Current Load is now: OK on ve-roundtrip2.pmtpa.wmflabs 10.4.0.162 output: OK - load average: 3.83, 4.13, 4.76 [20:32:22] PROBLEM Current Load is now: WARNING on parsoid-roundtrip6-8core.pmtpa.wmflabs 10.4.0.222 output: WARNING - load average: 5.59, 5.84, 5.41 [20:37:23] RECOVERY Current Load is now: OK on parsoid-roundtrip6-8core.pmtpa.wmflabs 10.4.0.222 output: OK - load average: 1.21, 3.19, 4.43 [20:37:52] RECOVERY Free ram is now: OK on sube.pmtpa.wmflabs 10.4.0.245 output: OK: 21% free memory [20:45:42] RECOVERY Free ram is now: OK on bots-4.pmtpa.wmflabs 10.4.0.64 output: OK: 86% free memory [21:10:53] PROBLEM Free ram is now: WARNING on sube.pmtpa.wmflabs 10.4.0.245 output: Warning: 13% free memory [21:30:40] [bz] (8NEW - created by: 2Matthias Mullie, priority: 4High - 6major) [Bug 42188] Allow for something other than master to be deployed - https://bugzilla.wikimedia.org/show_bug.cgi?id=42188 [21:33:24] <^demon> marktraceur: etherpad on labs is down :\ [21:33:40] Augh. [21:34:02] ....wfm [21:34:09] <^demon> And seems to be back now. [21:34:15] That happens sometimes [21:34:16] <^demon> Was definitely giving 503s a minute ago during a meeting. [21:34:42] Sorry about that [21:34:55] <^demon> np [21:39:32] PROBLEM Disk Space is now: WARNING on wikidata-dev-9.pmtpa.wmflabs 10.4.1.41 output: DISK WARNING - free space: / 573 MB (5% inode=71%): [22:35:32] PROBLEM Free ram is now: WARNING on nova-precise2.pmtpa.wmflabs 10.4.1.57 output: Warning: 16% free memory [22:38:22] PROBLEM Total processes is now: WARNING on nova-precise2.pmtpa.wmflabs 10.4.1.57 output: PROCS WARNING: 159 processes [22:53:24] RECOVERY Total processes is now: OK on nova-precise2.pmtpa.wmflabs 10.4.1.57 output: PROCS OK: 150 processes [23:16:23] PROBLEM Total processes is now: WARNING on nova-precise2.pmtpa.wmflabs 10.4.1.57 output: PROCS WARNING: 153 processes [23:51:13] PROBLEM Disk Space is now: CRITICAL on scribunto.pmtpa.wmflabs 10.4.0.144 output: DISK CRITICAL - free space: / 282 MB (2% inode=74%): [23:56:13] PROBLEM Disk Space is now: WARNING on scribunto.pmtpa.wmflabs 10.4.0.144 output: DISK WARNING - free space: / 298 MB (3% inode=74%):