Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2013-07-26

| Channels | #salt index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:03 fivethreeo joined #salt
00:05 ostercloud joined #salt
00:08 aat joined #salt
00:16 bhosmer joined #salt
00:20 nmistry joined #salt
00:22 efixit joined #salt
00:23 jslatts joined #salt
00:36 _jps joined #salt
00:37 _jps joined #salt
00:41 helderco joined #salt
00:46 UtahDave joined #salt
00:53 DanGarthwaite joined #salt
00:54 DanGarthwaite left #salt
00:54 DanGarthwaite joined #salt
00:56 jeddi joined #salt
00:57 jheise joined #salt
00:57 sturdy joined #salt
00:58 jheise is there any way to use pillars or grains to get dynamic data out of a minion
01:02 iquaba joined #salt
01:03 UtahDave jheise: that's what modules are for.
01:03 jheise is there a way to use that data to select a group of minions ?
01:03 UtahDave Ah, I see.
01:04 UtahDave You can use the Salt Mine for that.
01:04 UtahDave the mine runs any module you want on a regular basis, and you can target on that.
01:04 jessep joined #salt
01:06 UtahDave actually, maybe I'm wrong on that.  I thought you could target on mine data, but I can't find any confirmation on that.
01:06 UtahDave Salt Mine is very new.
01:08 goozbach UtahDave: can you give a quick explaination of the differences between mine, pillar, and grain?
01:08 jheise yeah, i want to be able to execute a module on a group of minions but their state could change without interaction
01:09 UtahDave pillar is data that you keep centrally on your master. Each minion is only given data that you specifically assign to it. It's encrypted so that only the assigned minion can see it.
01:10 UtahDave grains are bits of information about each minion. Each minion figures out its grains on startup. Grains are generally pretty static things.
01:10 UtahDave The mine is a system where you specify functions that then get run at a specified interval.  The Master then keeps the very last execution of the mine that you can then query whenever you want.
01:11 UtahDave So the mine provides a really fast way to query dynamic data about a minion that will only be slightly stale
01:12 jesusaurus oh, so the mine is similar to grains? but a bit more dynamic?
01:12 goozbach that's what I thought, thanks for the confirmation
01:13 UtahDave jesusaurus: yep!  Except that I thought you could match on the mine, but I appear to be incorrect on that.
01:13 UtahDave you're welcome, goozbach
01:16 jheise UtahDave: thanks for the info
01:17 eskp happy sysadmin day ya'll
01:18 UtahDave yep1
01:24 Nexpro1 joined #salt
01:29 cocoy joined #salt
01:32 jpeach joined #salt
01:32 prologic joined #salt
01:33 prologic Does anyone have a demo of the salt-ii running?
01:34 UtahDave whiteinge: do you?
01:34 UtahDave prologic: I think they're going to release a demo in the next couple of days
01:35 prologic sweet that would be good
01:35 prologic Also I get the distinct impress (from warnings and the defaults) that you really ought to run up a salt-master on a real server dedicated to being the "salt master" and run the salt-master daemon as root?
01:36 prologic I've been playing around with running up a salt-master on my Mac as a regular user using -c option (installed salt with pip install salt)
01:38 mackstic1 joined #salt
01:38 aat joined #salt
01:38 jpeach is there any way to debug the gifts fileserver?
01:38 jpeach I think it is seeing a stale checkout of the repo for some reason, but not sure why that would be
01:38 jpeach I cranked up the log level in /etc/salt/master, but there's no output from gitfs :(
01:38 z0rkito joined #salt
01:39 mgw joined #salt
01:41 UtahDave prologic: There shouldn't be any problem running the salt-master as a non-root user on your mac
01:41 cxz fuck yea sysadmin day today
01:41 cxz i got a slushie from my boss
01:42 UtahDave SLUSHIES FOR EVERYONE!!!    :)
01:42 UtahDave jpeach: have you tried deleting the cache?
01:44 goozbach left #salt
01:45 dthom91 joined #salt
01:45 cxz UtahDave: it's cold where i am this morning
01:45 cxz :P
01:45 jessep joined #salt
01:45 cxz so i think it was more of a mock than an actual gift
01:45 mmilano joined #salt
01:45 cxz lol
01:45 UtahDave lol
01:45 UtahDave nice
01:49 jpeach UtahDave: yep, I deleted the cache and then did it again when I saw your suggestion
01:49 jpeach the second time got it working, so I'm not sure what went wrong
01:50 UtahDave do new changes to the repo appear now?
01:50 jpeach possibly some combination of flaky gitlab instance and ssh debug logging?
01:50 jpeach yep, I'm seeing the expected states now
01:50 jeddi joined #salt
01:51 UtahDave ok, good
01:51 UtahDave gitfs is still newish. We definitely need better logging/visibility in there.
01:53 napperjabber joined #salt
01:53 andrew joined #salt
01:54 jpeach gifts is great though; it makes for a really nice workflow :)
01:54 jheise is there anyway to target minions based a value from a module
01:55 _jps joined #salt
01:56 UtahDave jpeach: yep!
01:57 UtahDave jheise: I don't think so:  http://docs.saltstack.com/topics/targeting/index.html
01:57 UtahDave jheise: Can you explain in more detail what you're trying to do?  maybe I can help you find a way around that
01:59 jheise so i have a cluster of databases that can change mastership without human interaction, i want to be able to execute a module across all of the current database slaves in the cluster
02:06 nmistry joined #salt
02:06 liuyq joined #salt
02:08 xl1 joined #salt
02:09 jdenning joined #salt
02:09 Gifflen joined #salt
02:10 prologic Ok so I have the salt mater running as root now on my Mac. Can't run salt commands as a non-root users though (tried setting client_acl to no avail) -- sudo salt … works though
02:21 raydeo joined #salt
02:24 UtahDave jheise: do you have any control over how the mastership changes?
02:25 UtahDave like, can you make a cli call to change grains?
02:28 zirpu left #salt
02:29 joeTSUNAMI joined #salt
02:51 Joe630 joined #salt
02:53 Furao joined #salt
02:53 avienu joined #salt
02:53 Joe630 hello all
02:56 UtahDave hey, Joe630
02:56 Joe630 Just got back from oscon
02:56 Joe630 had good talks with Mr. Hatch
02:59 UtahDave nice!
03:00 UtahDave He loves getting out and talking to people
03:01 Joe630 Can't wait to fire off some pull requests
03:02 rberger joined #salt
03:06 HaxCore joined #salt
03:07 UtahDave nice!
03:08 UtahDave What do you plan on working on, Joe630?
03:08 andyshinn` joined #salt
03:09 jshare joined #salt
03:11 Joel_re joined #salt
03:12 Joe630 UtahDave: not sure yet, looking tru to see whats needed
03:16 jaequery joined #salt
03:17 UtahDave cool. I'm heading home now, but let me know if you have any questions
03:18 jaequery joined #salt
03:19 jshare joined #salt
03:22 mel joined #salt
03:23 aat joined #salt
03:25 jaequery joined #salt
03:29 napperjabber joined #salt
03:36 mel does anyone know if someone's working on a systemd state module?
03:36 syngin erf.
03:38 syngin hmm
03:39 syngin i might be able to work around this...
03:43 auser joined #salt
03:52 dthom91 joined #salt
03:54 auser joined #salt
03:57 prologic left #salt
04:08 Lue_4911 joined #salt
04:11 nmistry joined #salt
04:38 Joel_re joined #salt
04:51 cxz hey guys i need to do a dns lookup for a jinja template in a state
04:51 cxz what's the best way to go about this
05:03 Shell joined #salt
05:08 dthom91 joined #salt
05:22 jalbretsen joined #salt
05:24 cxz can we get some better documentation around the salt mine
05:24 cxz it's really not that clear at all
05:43 koolhead17 joined #salt
05:45 Shell left #salt
05:46 Ryan_Lane joined #salt
05:48 scooby2 joined #salt
05:56 Joel_re joined #salt
05:56 az87c joined #salt
05:57 az87c joined #salt
06:08 Joel_re joined #salt
06:09 dthom91 joined #salt
06:11 Newt[cz] joined #salt
06:25 woyerbse joined #salt
06:33 davromaniak joined #salt
06:39 dthom91 joined #salt
06:58 carlos_ joined #salt
06:59 woyerbse left #salt
06:59 woyerbse joined #salt
07:02 Joel_re joined #salt
07:02 carmony joined #salt
07:08 kstaken joined #salt
07:12 matanya joined #salt
07:20 bud joined #salt
07:20 bud Is anybody of you people running Salt with GitFS backend on Debian Squeeze?
07:34 arthurlutz joined #salt
07:34 arthurlutz hi, saltsprint about to begin in Paris, France
07:35 Ryan_Lane joined #salt
07:35 arthurlutz1 joined #salt
07:36 ml_ joined #salt
07:36 arthurlutz2 joined #salt
07:39 dthom91 joined #salt
07:41 logilab joined #salt
07:42 logilab hi from paris salt sprint
07:46 scott_w joined #salt
07:53 jcristau joined #salt
08:00 sturdy joined #salt
08:02 ptonelli joined #salt
08:05 LucasCozy joined #salt
08:05 LucasCozy joined #salt
08:19 krak3n` joined #salt
08:28 zooz joined #salt
08:36 Xeago joined #salt
08:37 felixhummel joined #salt
08:44 Anb joined #salt
08:52 woyerbse joined #salt
09:00 alain joined #salt
09:01 backjlack joined #salt
09:11 techdragon joined #salt
09:13 krak3n` joined #salt
09:17 krak3n` joined #salt
09:21 whiskybar joined #salt
09:22 arthurlutz2 follow the salt stack sprint in Paris : http://framapad.org/saltsprintparis
09:39 tomeff joined #salt
09:46 bejer Hi, how would I use the data in grains as the content of a jinja template variable in an sls file? (I would like to pull the fqdn grain from the minion and have it placed in a configuration file - preferbly by using jinja templating) - I know from the examples that I can do flow control on the grains data, but how do I "output" the data?
09:47 LucasCoz joined #salt
09:48 Newt[cz] joined #salt
09:50 ggoZ joined #salt
09:51 sw_ bejer: {{ grains['fqdn'] }} in your jinja template
09:54 bejer oh okay, thanks
09:55 sw_ don't forget "template: jinja" in your file.managed statement
09:55 bejer yeah
10:13 cocoy left #salt
10:17 scalability-junk sw_: but isn't jinja default?
10:19 xl2 joined #salt
10:40 krissaxton joined #salt
10:47 karlp scalability-junk: you'd think so, but .... I needed to add template:jinja to make it actually work...
10:47 scalability-junk mhh only in config files or in .sls files too?
10:48 krak3n` joined #salt
10:50 karlp I had to be explicit when I was passing values into a generated template _from_ the sls,
10:51 mikedawson joined #salt
10:52 scalability-junk yeah but you can also use jinja within sls files and I was wondering of config files are interpreted per default differently than sls ones
10:53 david_a joined #salt
10:56 giantlock joined #salt
10:56 krak3n` joined #salt
11:05 younqcass joined #salt
11:31 lemmings scalability-junk: AFAIK, sls are jinja yaml by default
11:32 lemmings configs, however, are NOT templates by default, they're plain text
11:32 lemmings if you want to use variables in it, you have to change the template directive, like he said
11:33 lemmings also, protip, consider using stateconf for sls... you get sls_dir for free and some other goodies. Portability :D
11:33 lemmings wow... I feel like I know stuff now. This means I'm probably wrong on something :D
11:37 lemmings btw, anyone knows how absent.sls works?
11:38 lemmings is it something like "if absent is present and server doesn't match, remove package?" and is is a mirror of init?
11:48 LucasCoz joined #salt
11:52 SpX joined #salt
11:52 blee joined #salt
11:54 DredTiger joined #salt
11:55 [diecast] joined #salt
12:07 zooz joined #salt
12:09 lemao joined #salt
12:09 krak3n` joined #salt
12:24 JasonSwindle joined #salt
12:26 jslatts joined #salt
12:38 kenbolton joined #salt
12:41 jeffrubic joined #salt
12:44 Joel_re joined #salt
13:08 anteaya joined #salt
13:08 JasonSwindle left #salt
13:11 Gifflen joined #salt
13:17 jeddi for a require: apparently you can use the id of a stanza, but I'm trying with either just the name (not a list error) or "- id: make_moindir" ... but then says it can't find that id.     any suggestions?
13:23 xl2 Hi everyone! Is it possible to use the "py" renderer for states.file's template? jinja works fine (and gets the default template variables), but I need some advanced logic.
13:23 xl2 The error is "...File "/usr/lib/pymodules/python2.7/salt/utils/templates.py", line 183, in py     data = mod.run()     TypeError: run() takes exactly 1 argument (0 given)"
13:23 Kholloway joined #salt
13:23 xl2 I have a run(some_variable) in the template file
13:24 mmilano joined #salt
13:26 dthom91 joined #salt
13:26 mgw joined #salt
13:26 toastedpenguin joined #salt
13:27 brianhicks joined #salt
13:30 p3rror joined #salt
13:32 mikedawson joined #salt
13:34 ipmb joined #salt
13:34 racooper joined #salt
13:36 juicer2 joined #salt
13:39 isomorphic joined #salt
13:40 aat joined #salt
13:40 chubrub joined #salt
13:41 chubrub hello everyone! Is there an option to run states in ordered way, not using overstate or require, but ordering states in some file? It's easier to me to order sls files, then building complicated dependency tree.
13:43 Gifflen @chubrub At the talk I saw the salt guys give they mentioned this was possible but I'm not sure of the way to activate it off the top of my head
13:45 Gifflen I am seeing this http://docs.saltstack.com/ref/states/ordering.html#the-order-option
13:47 chrisgilmerproj joined #salt
13:48 chubrub #Gifflen: thx! Saw this option before but it's not perfect. anyway and I don't know how could i use it in top.sls file
13:50 Jose__ joined #salt
13:52 mikedawson joined #salt
13:52 _jps joined #salt
13:54 DanGarthwaite joined #salt
13:57 robertkeizer joined #salt
13:59 m_george left #salt
14:00 mannyt joined #salt
14:01 guest32768 joined #salt
14:03 mgw joined #salt
14:03 krak3n` joined #salt
14:04 mikedawson joined #salt
14:07 jeddi joined #salt
14:08 SJR40 joined #salt
14:09 napperjabber joined #salt
14:10 jaequery joined #salt
14:19 dthom91 joined #salt
14:20 jeddi seriously frustrating .. there's a *hint* in the documentation that you can - require: / - id:  stanza_name .... but it says it can't find the id.
14:21 mikedawson joined #salt
14:21 xl2 left #salt
14:22 [diecast] joined #salt
14:22 [diecast] joined #salt
14:23 pentabular joined #salt
14:23 SEJeff_work jeddi, Can you put a minimal example on gist.github.com please?
14:26 jeddi SEJeff_work:  here's what i've been playing with.  http://rn0.ru/show/JpYOtYjRwySUbV13GtcS/
14:27 jeddi SEJeff_work: i'm probably better off pushing out a small bash script and then running it rather than these linked dependencies, i know.
14:28 jeddi being told by salt that neither name: or id: items are recognised in the require: sub-para.
14:28 techdragon joined #salt
14:32 jeddi just noted my dir.present structure was broken .. i've fixed that .. same error(s).
14:33 jeddi revised http://rn0.ru/show/IKdpjnWGDRXvNJYI8lWb/
14:33 mr_chris left #salt
14:34 _jps_ joined #salt
14:35 cnelsonsic joined #salt
14:35 jalbretsen joined #salt
14:38 hypyrdon joined #salt
14:42 Ahlee whiteinge: oi, do you document out anywhere why you changed the /minions return in 0.8.2?
14:43 pentabular1 joined #salt
14:45 avienu joined #salt
14:46 jslatts joined #salt
14:47 sturdy joined #salt
14:49 dthom91 joined #salt
14:53 waverider joined #salt
14:54 ostercloud joined #salt
14:54 teskew joined #salt
14:55 ostercloud joined #salt
14:55 jslatts joined #salt
14:57 hazzadous joined #salt
14:58 logix812 joined #salt
15:00 bastion2202 joined #salt
15:03 StDiluted joined #salt
15:03 hypyrdon joined #salt
15:04 bastion2202 Hey Guys, what is the RAM requirement for salt-master ? around 70 minions
15:04 hypyrdon joined #salt
15:05 sturdy joined #salt
15:05 SEJeff_work very minimal
15:07 denstark joined #salt
15:09 krissaxton joined #salt
15:10 conan_the_destro joined #salt
15:11 Ahlee after upgrading to salt 0.16.0, salt-master is no longer present?
15:12 Ahlee times like this I wish my systems had a conection to the internet.
15:13 felixhummel is there some way like watch files on some minions like salt 'foo*' tail.files /some/file ?
15:17 dthom91 joined #salt
15:19 arthurlutz joined #salt
15:21 lemmings quick question... let's say I want to install cacti, and this depends on nginx/apache... how can I say that?
15:22 mikedawson joined #salt
15:22 StDiluted you can use a require or a require_in
15:23 StDiluted joined #salt
15:23 lemmings isn't that just for packages in the same sls?
15:24 lemmings if I have an sls that just installs apache
15:24 StDiluted use an include statement
15:24 lemmings and now add one to install cacti...
15:24 StDiluted in your cacti state
15:24 lemmings ah
15:24 lemmings good point
15:24 lemmings I can do that :D
15:24 lemmings and with the include, I can then use require, right?
15:24 lemmings you're awesome
15:24 StDiluted yes, I believe so
15:27 lemmings StDiluted: crap... include needs a relative path. I was using stateconf to get sls_dir and make the sls files more location independent....
15:28 StDiluted hrm
15:28 lemmings StDiluted: I assume there is nothing akin to "search for the package declaration called X somewhere....
15:29 jschadlick joined #salt
15:29 lemmings well... I wanted to publish some of these later and wanted to make this location independent (more modular, if you will)
15:29 StDiluted maybe extend?
15:29 Ahlee well, rm -rf venv, recreating created salt-master that time. NOt sure why pip install localmirror/salt-0.16.0.tar.gz didn't create
15:29 lazyguru joined #salt
15:32 lemmings StDiluted: humm... include and extend are top level declarations... methinks the look at the state tree and use the name from there
15:32 lemmings will check
15:32 lemmings thanks
15:34 chrisgilmerproj who's the expert on user permissions here? client_acl?
15:43 herlo joined #salt
15:46 lemmings StDiluted: apparently, Unknown include: Specified SLS base: nginx is not available on the salt master in environment(s): base
15:46 lemmings damn
15:46 StDiluted boo
15:47 lemmings StDiluted: in stateconf, using .nginx instead of nginx works, however
15:48 lemmings dunno about normal jinga sls thingy... still, works. Will test portability later
15:48 redbeard2 joined #salt
15:50 brianhicks joined #salt
15:58 lemmings StDiluted: since you're an expert... I have this on my top file: "base:   'srvtstminion':      - test"
15:58 StDiluted <- not an expert! ;)
15:58 StDiluted but I will try to help
15:58 lemmings srvtstminion does not match. the other one (I'm testing with 2) does!
15:59 StDiluted what does it say in salt-key -L
15:59 lemmings StDiluted: well, more expert than me, that's for sure. And thanks
15:59 lemmings Accepted Keys: srvtstminion.gca.ad.root srvwiki.gca.ad.root
16:00 lemmings wait... I think _I know the issue
16:00 lemmings i ran a sls.state
16:00 bud joined #salt
16:00 lemmings and that bypasses the top file, apparently
16:00 lemmings salt '*' state.sls test
16:00 lemmings still, should match both...
16:00 lemmings weird
16:00 StDiluted what you have in the '' is a regular expression for matching
16:01 StDiluted so 'srvstminion' wouldn't match anything
16:01 StDiluted because you have the fqdn
16:01 Rog joined #salt
16:01 lemmings yeah, notices, need an * at the end
16:01 StDiluted 'srvstminion*' would match
16:01 lemmings matches but doesn't run
16:01 lemmings salt 'srvtstminion*' state.sls test
16:01 lemmings will try highstate
16:02 lemmings nop
16:02 krissaxton joined #salt
16:02 bud Hello all. I have a problem with a custom module I have created. I want to pass it two parameters gathered via salt['network.interfaces']()['eth0'][inet'][0]['(address|netmask'].
16:02 bud In a file templated using jinja.
16:02 lemmings fqdn... nope
16:02 StDiluted what do you get back, just nothing?
16:03 bud *(address|netmask) above :-).
16:03 lemmings StDiluted: takes time, then nothing. Rebooted the salt-minion on the target machine
16:03 lemmings and still nothing
16:04 StDiluted hm, try a salt-call with -l debug on the minion
16:04 StDiluted see what you get there
16:04 lemmings humm
16:04 lemmings salt-call on it's own queries the master?
16:04 StDiluted yes
16:05 Rog Anybody know of anyone working on a salt module that uses git-tf or some other method in order to use MS TFS (as opposed to git) as a file repository?  Right now i have to mirror TFS to a git repo and then pull from there, I'd like to use TFS directly.
16:06 arthurlutz salt sprint ending in Paris, we had a major electricity cut off all afternoon so some things didn't go as planned
16:07 arthurlutz we'll report back on monday on the things that were done
16:07 StDiluted cool, we are looking forward to ours tomorrow
16:07 lemmings StDiluted: salt-call returned the help, salt-minion reports the master key has changed... can't figure out how!
16:08 StDiluted lemmings: that explains why though
16:08 lemmings ah... wait... yes I can
16:08 arthurlutz joined #salt
16:08 lemmings I retargeted this one to a new master
16:08 StDiluted ah there you go
16:08 lemmings it registered the key without error though and the fqdn is different
16:08 lemmings should at lreat give a warning
16:08 lemmings and rstarting the service works ok
16:09 lemmings this kind of error should be an error signal throwing one... without mutual auth, it doesn't work
16:09 StDiluted agreed
16:10 shane joined #salt
16:10 lemmings any idea how I can file a bug report?
16:11 StDiluted you can do it in github issues
16:11 StDiluted github.com/saltstack
16:12 lemmings can't remember my github auth... will see if I do it later
16:12 lemmings utahdave could be online... :D
16:12 StDiluted he's not online at the moment
16:13 bud I have pasted the templated file, well just a part of it, on pastebin http://pastebin.com/JhX2agas. If somebody could take a look I would be grateful.
16:14 UtahDave joined #salt
16:14 bud The thing I want to achieve is to create the network interfaces file on Debian Squeeze with data gathered from the running system via salt modules.
16:15 bud I have found how to template the eth0 Internet address here http://cnygaard.blogspot.ro/2012/11/how-to-template-eth0-address-with.html.
16:16 nmistry joined #salt
16:16 lemmings UtahDave: quick question. When you retarget the minion to another master, if you run salt-minion, it says there's a key mismatch and doesn't work. However, you can register the key on the master and /etc/init-d/salt-minion restart has no errors. Is this by design?
16:16 StDiluted hey UtahDave
16:16 jdenning joined #salt
16:17 UtahDave hey, StDiluted and lemmings
16:17 Lue_4911 joined #salt
16:17 bud Hey UtahDave
16:18 UtahDave lemmings: I believe that /etc/init.d/salt-minion restart does have errors, they're just in the background.  if you run   salt-minion -l debug, you'll get the mismatch error
16:18 UtahDave hey, bud!
16:18 bud If you have a couple of minutes to look over something I would be grateful.
16:18 UtahDave lemmings: and the minion service will immediately shut down
16:18 UtahDave bud: sure
16:18 bud http://pastebin.com/JhX2agas
16:19 bud I am trying to build the network interfaces file of a Debian Squeeze host using a template file.
16:19 bud Using modules I would like to gather information from the running system and stuff it in the file.
16:20 bud Here is the whole file http://pastebin.com/gkyZApKe.
16:21 bud The subnet module is built by hand and distributed to minions
16:22 bud I apologize, I am new to pastebin, the whole file is here http://pastebin.com/eNEb3pRY.
16:22 UtahDave ok, and so what error are you running into?
16:23 bud http://pastebin.com/Gbjz0egm this one.
16:24 bud Using salt-call with debug log level on the minion.
16:25 UtahDave Oh, ok. I think what's happening is that you've got several "salt" calls inside one {{ var }}  jinja statement
16:27 bud Can I fix that somehow. I honestly tell I am out of ideas.
16:27 krak3n` joined #salt
16:28 dthom91 joined #salt
16:28 UtahDave yeah, let me pastebin it.
16:30 UtahDave bud: is line 23 supposed  to be 3 items? or is the second item supposed to be an argument to the subnet.find?
16:31 bud subnet.find takes to arguments the ip address and the subnet mask.
16:31 lemmings UtahDave: I believe so, but the script exits with ok.. kinda confusing. Still, makes sense
16:32 lemmings thanks for the explanation
16:32 bud *two
16:32 UtahDave Hm. lemmings, that might be worth opening an issue over.  what's the output of   echo $?  after it fails to connect to the new master?
16:33 lemmings well... kinda... still prefered if it didn't start at all, I guess... restart fails, you check the logs. Restart oks, you assume minion is running...
16:33 lemmings wait, I fixed the issue... let me break it again
16:33 bud UtahDave: I have pastebined the _modules/subnet.py file it case it helps - http://pastebin.com/sWN4LAkW.
16:34 UtahDave bud: try this: http://pastebin.com/HcURBBtb   see how I made each var into it's own {{ }}   ?
16:34 JasonSwindle joined #salt
16:34 UtahDave bud: ah, wait.  I totally misunderstood what you were doing
16:34 nmistry joined #salt
16:34 UtahDave your last pastebin helped.  let me redo that.
16:34 JasonSwindle is there a way with grains to do grain:grain_self?
16:34 JasonSwindle Like for example; I have node_type:web
16:35 JasonSwindle I want to have node_type:myself
16:35 lemmings UtahDave: here http://pad.yohdah.com/155/32d89153-155f-48fb-92ea-99d0040cd3a8
16:35 lemmings just changed a char on the master key, tried it again
16:35 UtahDave bud, I think you just need to put those two arguments inside the the ()
16:36 lemmings UtahDave: and now I have to go... still, you can easily reproduce the issue, I think
16:36 lemmings hope it helps
16:36 UtahDave lemmings: Yeah, it makes sense to me that 1 should be the result in that instance.
16:37 lemmings http://pad.yohdah.com/156/0deece0e-7b80-4283-8c5d-8c0f0f8c1061
16:38 lemmings just to see that it's not the restart (it starts, creates a pid file, hence the cannot kill process X)
16:38 lemmings I assume it's be more clear, yes...
16:38 lemmings anyway, gotta go
16:38 lemmings hope I helped.
16:38 lemmings regards
16:38 bud UtahDave: your pastebin gives the same error.
16:39 bud I will try what you suggested now, to put in ().
16:39 conan_the_destro hello any FreeBSD person having this probelm when trying to create a user: http://pastebin.com/U6ezXGnZ
16:39 UtahDave cedwards: have you seen conan_the_destro's error?
16:39 UtahDave thanks, lemmings
16:40 conan_the_destro I saw that Solaris has this issue and there was a PR for it but didn't see a reported issue for FreeBSD
16:40 UtahDave if cedwards isn't around to reply soon, would you mind opening an issue for this?
16:41 mgw joined #salt
16:41 JasonSwindle akoumjian: BTW, that worked great!
16:42 conan_the_destro will do UtahDave
16:42 JasonSwindle akoumjian: https://gist.github.com/JasonSwindle/4552855c7e7c35069a5e <- Works flawlessly
16:42 UtahDave thanks, conan_the_destro!
16:43 akoumjian JasonSwindle: \0/
16:43 UtahDave JasonSwindle: is that in response to your grains question?
16:43 JasonSwindle yes
16:44 JasonSwindle I can now setup per node minion files
16:44 JasonSwindle My vagrantfile is getting ever more complex
16:44 UtahDave nice
16:45 JasonSwindle akoumjian: I cannot wait for Salt to be added into Vagrant in the 1.3.x line.
16:45 akoumjian JasonSwindle: You and me both. Been a long time waiting
16:46 koolhead17 joined #salt
16:46 koolhead17 joined #salt
16:46 JasonSwindle Yeah.  Feel free to use my Vagrantfile as an example if you wish
16:46 JasonSwindle It is pretty decked out
16:48 KyleG joined #salt
16:50 bud UtahDave: I have tried a couple of variations among the following: {{ subnet.find() {{ salt call }} {{ salt call }} }},  {{ subnet.find() {{ salt call }}, {{ salt call }} }},  {{ subnet.find( {{ salt call }} {{ salt call }} ) }}, {{ subnet.find( {{ salt call }}, {{ salt call }} ) }}.
16:50 bud None of them seem to work :|.
16:51 UtahDave bud, ok, let me try something
16:51 bud Thank you :-).
16:52 UtahDave try this line, bud: http://pastebin.com/V5RqFnQS
16:56 andrew joined #salt
16:56 devinus joined #salt
16:56 JasonSwindle Is it possible to do grain:grain_self?
16:57 JasonSwindle like for example, I have node_type:web
16:57 JasonSwindle but I want to be able to do node_type:What_I_AM
16:58 KyleG Happy Sys Admin Day #salt!
16:59 whiteinge Ahlee: the returned data structure was inconsistent with the return for all the other URLs (a list instead of a dict)
16:59 whiteinge i could definitely have gone into more detail on that in the release notes...
17:00 JasonSwindle I ask because it would take my top.sls from / to https://gist.github.com/JasonSwindle/9b3b5cd4b7c488d97877
17:02 tomeff_ joined #salt
17:02 UtahDave Hm. There are some yaml constructs that *might* do what you're trying to do
17:02 JasonSwindle UtahDave: Oh?  And, can that be used in top.sls?
17:03 UtahDave yeah, if you can get the yaml to work correctly, then Salt just takes it in as a datastructure
17:03 JasonSwindle Hmm, I would not even know what to google for or etc.
17:05 conan_the_destro UtahDave cedwards: https://github.com/saltstack/salt/issues/6330
17:05 Ahlee whiteinge: ok, so it was just a consistency thing, that's what we were wondering
17:05 bud UtahDave: Thank you! It works like a charm. Now I can go back to my DNS supermaster :).
17:06 Ahlee fully agree with the change, just the wrong person saw it so they wanted more details than i had
17:06 Ahlee whiteinge: thanks
17:06 UtahDave thanks, conan_the_destro!
17:06 UtahDave you're welcome, bud!  glad it worked
17:08 JasonSwindle conan_the_destro: BTW, this will make the issue easier to read
17:08 JasonSwindle http://pastebin.com/raw.php?i=3jUM4pEp
17:08 timl0101 joined #salt
17:09 conan_the_destro oh, ok.
17:09 JasonSwindle ``` CODE HERE ``` is a code block on github :)
17:09 bud I have to tell guys I find SaltStack really awesome and uber powerful :-). I enjoy coding infrastructure and I find the learning curve really easy.
17:09 UtahDave :)  thanks, bud!
17:09 UtahDave A lot of people have worked hard to make it awesome!
17:10 JasonSwindle conan_the_destro:  Also, can you supply a salt-call --versions-report in the ``` code block? :)
17:10 JasonSwindle I know this information helps them a lot :)
17:10 whit joined #salt
17:10 bud I am sure a lot of work has been put in by a lot of awesome people with a lot of experience :-).
17:11 bud I hope tomorrow I can bring a small contribution.
17:12 JasonSwindle conan_the_destro:  I hope that helps. :)
17:12 UtahDave that would be great, bud!  Which sprint location are you attending?
17:12 bud I want to configure a tinc VPN state so machines can connect automatically to the VPN. I plan to use the mine to gather the data from the minions and use it when a new minion comes online.
17:12 conan_the_destro JasonSwindle: issue updated. thank you for the tips
17:12 bud VP. I am from Romania and I have managed to gather 3 people only.
17:13 UtahDave thanks, conan_the_destro! JasonSwindle's tips do help.
17:13 bud We have a location and is pretty real though, just that we are not official sort of speak :).
17:13 JasonSwindle conan_the_destro:  Any time.  :)  I remember my first bug report was so horrible and I got flamed so badly for it.   But #salt is a friendly place.
17:13 UtahDave bud: Ah, cool!  3 people is great!
17:14 JasonSwindle Well, my first bug report was not here….. it was many moons ago.
17:15 JasonSwindle UtahDave:  Do you know what the YAML option you are talking about is called?  Or how to do it?
17:15 JasonSwindle :)
17:16 UtahDave JasonSwindle: just a sec
17:16 JasonSwindle Sure
17:16 jpeach joined #salt
17:17 JasonSwindle UtahDave:  BTW, I may pop-in to the Hackathon this weekend.  Joys of working at the same place.  I maybe late because I will be camping in Austin tonight.
17:17 jeddi anyone have any clue how you can do a require: to an 'id' of a stanza?   i find the idea of having sensible stanza names but then doing a require to the *content* of that stanza a bit messy, especially with cmd.run's.
17:18 jeddi i have a fragment here of my sls - http://rn0.ru/show/IKdpjnWGDRXvNJYI8lWb/  it's very procedural in a way, but i want to require that a previous paragraph has run, and identify it via the id .. but id: and name: don't work.  any clues?
17:22 UtahDave jeddi: just use the same state dec.
17:22 UtahDave jeddi: - name: make_moindir    should be   - cmd: make_moindir
17:22 UtahDave jeddi: you can refer to either the ID declaration or the name
17:23 UtahDave JasonSwindle: awesome!  It would be great to see you there.
17:23 UtahDave JasonSwindle: try this example: https://gist.github.com/ddlsmurf/1590434#file-output-txt-L63
17:23 hypyrdon i am going thru the salt state tut.  why is installing a pkg on a minion where it is already installed an error? (centos6,salt-0.15.3)
17:25 JasonSwindle Hmmm, I still have to set it.  I am trying to have salt ask the minion what is thyselves.
17:27 _jps joined #salt
17:27 DanGarthwaite hypyrdon: It should not be an error.
17:27 JasonSwindle So on the web node, node_type:@SELF becomes web; and node_type:@SELF on database is database.  Or whatever the grain node_type has
17:28 hypyrdon i didn't think so.
17:28 UtahDave JasonSwindle: I guess you could drop into jinja and access the grains variable to place it there.
17:28 dthom91 joined #salt
17:29 JasonSwindle UtahDave:  Something like; {% set self_env = grains.env %} ?
17:29 JasonSwindle I was hoping Salt had Automagic for it. :)
17:29 DanGarthwaite hypyrdon: What is the error?  Could you post the state file and error on pastie?
17:30 hypyrdon the first time i ran "salt '*' state.sls vim", the minon installed vim-enhanced, as expected.  i did note that there was no green return text on the master.
17:30 hypyrdon pastie.. yes. jas.
17:30 StDiluted I've had no return from high states now and then
17:30 UtahDave JasonSwindle: sure'
17:31 JasonSwindle I was trying to remove down the number of lines an let Salt handle it
17:33 Ahlee any plans to allow you to block/deny/auto deny based on wildcard?  somehow I'm getting prod hosts talking to my staging salt master, and I want those keys to be automatically dropped if hostname matches a known regex for prod hosts
17:34 Ahlee right now i'm just running a cron job to loop over them and delete them, but i'd rather not risk them being presented in case of an overzealous salt-key -A gets in there
17:37 jdenning UtahDave:  When using the external pillar module cmd_json, what is the order of arguments passed to the external cmd?  E.g. say my external command (which returns json) is 'pillar $nodeID', (where $nodeID would be the name of the node for which I am returning data) - is the configuration statement simply            "cmd_json: foo" ?
17:37 jdenning oops, make that "cmd_json: pillar"
17:38 UtahDave Ahlee:  you can reject them
17:39 UtahDave jdenning: can you pastebin what you've got so far?
17:40 Ahlee UtahDave: right, but that includes having previous knowledge once the key is presented
17:41 Ahlee host unintendedprod01.domain.tld shows up in list, right now i'm looping over the keys, greping for prod and then removing it
17:41 jdenning UtahDave: http://pastebin.com/LPTJJs3Z
17:41 Ahlee i'd rather just only allow intendedenvX.domain.tld be allowed, or the negative on that
17:41 jdenning (Just the relevant fragment of the master config)
17:41 UtahDave Ahlee: good point. There's an auto_accept file, it makes sense that an auto_reject might be nice.  Would you mind opening a feature request on the issue tracker?
17:41 Ahlee sure.
17:42 Ryan_Lane joined #salt
17:42 Ryan_Lane joined #salt
17:42 UtahDave thanks, Ahlee
17:42 UtahDave jdenning: have you read this document?  It will probably help: http://docs.saltstack.com/topics/development/external_pillars.html
17:42 _jps left #salt
17:43 Newt[cz] joined #salt
17:43 jeddi thanks UtahDave .. I thought I'd tried that approach , but note that i need to do file.(dependency name) ... in that instance.  working now, in any case, so all good.  :)
17:43 UtahDave ok, cool
17:44 jdenning UtahDave: I did read that, but I'm not clear on how salt actually calls the external command.  In the example cmd in that doc (`"echo {'arg':'value'}"`)  where is the actual nodeID passed?
17:45 jdenning i.e. If I configure cmd_json: get_pillar   , will salt call the command like `get_pillar nodeID` ?
17:46 UtahDave jdenning: have you read the cmd_json pillar code?  It's really short: https://github.com/UtahDave/salt/blob/develop/salt/pillar/cmd_json.py
17:47 bastion2202 joined #salt
17:47 jdenning UtahDave: Just checked it out - so the minion_id is *not* passed to the external command at all?
17:48 jdenning (maybe I also need to look at cmd.run code?)
17:48 brianhicks joined #salt
17:48 UtahDave not in cmd_json and cmd_yaml.   Those are really mostly for examples.
17:48 UtahDave I think you ought to create your own custom pillar.  It's really easy and you have an awful lot of power
17:49 bubba_ joined #salt
17:49 berto- joined #salt
17:49 jdenning UtahDave - I actually am trying to use reclass (by madduck), but while I can get it to export either yaml or json, I can't seem to get it working with salt, so I was trying to use the cmd_json as a workaround..but if you can't pass the minion_id, that obviously won't work..
17:52 jdenning UtahDave: This (http://pastebin.com/Fux0Y35Q) is what I was trying to do originally, but again, salt does not seem to pick up the pillar data
17:52 hypyrdon @dangarthwaite pastie.org/8178541  this is the sls, the command output on the master and minion log
17:53 Thiggy joined #salt
17:54 DanGarthwaite hypyrdon: Thanks, reviewing
17:54 jdenning madduck: Any chance I could pick your brain about reclass / external pillar stuff?
17:55 DanGarthwaite strange yaml bug.   Change   "pkg.installed" in vim.sls to    "pkg.installed: []"
17:55 jschadlick How would I call the master's file roots in a runner? would it be __salt__['config.option']('file_roots.base')  ?
17:55 jdenning UtahDave: Any tips for troubleshooting this stuff?  The reclass external pillar *does* seem to be working as advertised, but salt does not seem to be grabbing the pillar data.
17:55 Thiggy I'm having some trouble getting gem.installed to respect requirements. I have this statefile (https://gist.github.com/anonymous/6090839) and when it runs, sass and compass try to install before the gem command is installed/available.
17:55 hypyrdon ok
17:57 UtahDave jdenning: are you sure you're set up correctly with reclass and that your pillar/top.sls is matching correctly?
17:57 hypyrdon @dangarthwaite: no change. same error
17:57 timidshark joined #salt
17:59 DanGarthwaite hypyrdon: is therea package called just 'vim'  Can you  "sudo yum install vim"  ?
17:59 DanGarthwaite hypyrdon: Actually, you need more debug info.  Do this:   sudo salt-call state.sls vim -l debug
17:59 jdenning UtahDave: I'm pretty sure that reclass is set up correctly - it seems to be returning top and pillar data as expected when I use the reclass-salt command.  I am trying to use just the ext_pillar aspect of reclass - do I need to use the external top system as well?  (I am sure that there are the exact same nodes in my top.sls as in my reclass inventory - I set up a completely fresh saltmaster and a single test node just for testing this)
18:01 Newt[cz] joined #salt
18:01 hypyrdon on redhat-based machines "vim" is a yum alias for "vim-enhanced". now that i think that thru that could be the reason the first run installs the package, but sucessive runs fail to detect it.
18:01 dave_den thiggy: i have had some problems with the rvm module in the past. can you post the output from your state run with '-l debug'?
18:01 Thiggy moment plz.
18:01 DanGarthwaite Try the salt-call state.sls vim -l debug  <-- much better run through of what is happening
18:02 hypyrdon debug, right. jas ...
18:02 Thiggy crap thing is that subsequent runs succeed because by that time gem is installed, I'll provision a new machine to get the logs
18:02 UtahDave hypyrdon: yeah, you must use vim-enhanced
18:03 UtahDave jdenning: You're going to have to contact the developer of reclass, which is madduck
18:03 UtahDave I haven't had the chance to use reclass yet
18:03 jdenning UtahDave: Yeah, that's why I was trying to use cmd_json or cmd_yaml :)  I didn't realize they were unable to target a specific node..
18:04 hypyrdon that is a lot of output. anything i should look for?
18:04 jdenning UtahDave: IMO, that makes them kind of worse-than-useless as examples (since I don't see how pillar data could be usefull without the minion_id of interest)
18:05 UtahDave jdenning: just create your own.  copy cmd_yaml and then load a yaml file an use the minion_id as the key in the loaded yaml.
18:07 cnelsonsic joined #salt
18:08 fxhp Hey UtahDave
18:08 hypyrdon_ joined #salt
18:08 UtahDave hey, fxhp.  you coming tomorrow?
18:08 jdenning UtahDave:  Yeah looks like that's my next course..Where should I put the custom_ext_pillar.py file?
18:08 fxhp UtahDave:  coming where, I'm in CT
18:08 jdenning (UtahDave: I know it's documented somehwere, just trying to save some time :) )
18:08 hypyrdon_ i see where yumpkg is loaded as a virtual service
18:09 UtahDave fxhp: there's a Salt Sprint location in New york!
18:09 fxhp hmmm
18:09 Xeago joined #salt
18:09 jMyles joined #salt
18:09 UtahDave At Columbia University Library
18:09 DanGarthwaite hypyrdon_: UtahDave confirmed you have to use vim-enhanced, but it is good to know _how_ to get the further debug info in case you need it
18:10 DanGarthwaite hypyrdon_:  Try   salt-call state.sls vim  without the  -l debug for stil-better-but-not-as-much info.  :)
18:11 hypyrdon_ thx. i looked ahead in the tut. later there is an example for aliasing based  on os family.  that is good because we have deb boxes too.  thanks for your help.
18:11 fxhp UtahDave: I wasn't planning, didn't know about it.  Wonder if my wife will let me go
18:11 fxhp UtahDave: lol
18:11 dthom91 joined #salt
18:11 hypyrdon_ my batt is down to %7 so, i have to go. thanks again.
18:11 DanGarthwaite hypyrdon_:  Good luck
18:11 dave_den is there going to be a google hangout for virtual participants, or should we just be here in #salt?
18:12 fxhp UtahDave: 2 hour drive
18:12 UtahDave fxhp: lol.  Well it would be awesome if you could make it.
18:12 UtahDave Yeah, we'll hve a google hangout for everyone.
18:12 * fxhp nods
18:13 fxhp UtahDave: do you know if a particular file in a file.recurse can be watched by a cmd.wait?
18:13 fxhp AKA I want to use file.recurse to place a drop directory down, but watch different files to trigger different commands
18:14 jdenning UtahDave: Sorry to keep bugging you on this - but where am I supposed to put code for a custom external pillar?  The "Extending Salt" doc references "Minion Execution Modules", "Grains", "States", "Renderers", and "Returners", but not Pillars?
18:15 UtahDave jdenning: I think that right now you can just drop it in the pillars directory in your python install.
18:15 UtahDave I don't think we have a generic spot for them yet
18:15 UtahDave fxhp: all files will be watched in a file.recurse
18:15 fxhp bummer
18:16 jdenning So it should be like /usr/lib/python2.7/pillars ??
18:16 fxhp Thats what I though, currently watching the recursed directory and running all commands
18:16 UtahDave fxhp: you can try specifying a specific file, but i'm not sure it would work.
18:16 oz_akan_ joined #salt
18:16 fxhp UtahDave: doesn't seem to work
18:17 UtahDave jdenning: no, where ever your distro installs salt in site-packages
18:22 jdenning UtahDave: So I put it in the installed salt-common directory (On ubuntu 12.04 - /usr/share/pyshared/salt/pillar/) ?  Wow, not a lot of people are doing this, eh? :)
18:22 UtahDave Just a few!  I think we have an open issue to create a place to drop custom pillars.
18:24 dave_den fxhp: what if you did a file.managed on the particular file, then have your cmd.wait watch that file.managed and require: the directory.recurse?
18:25 dave_den you'd have to make sure your file.managed ran before the recurse, though
18:26 fxhp dave_den: that could work, I'll give it a shot.  It feels like duplicate work
18:26 jdenning UtahDave: Is this stuff only working on the develop branch?  I am running 0.16.0 and looking at the code for cmd_json.ext_pillar and the function signature is different!  - def ext_pillar(pillar, command)  as opposed to def ext_pillar(minion_id, pillar, command) ???
18:27 fxhp dave_den: the drop dir only have 4 files, maybe I put it back to call out file.managed explicitly
18:27 dave_den fxhp: indeed.
18:27 UtahDave jdenning: What stuff exactly?
18:28 dave_den fxhp: this might be even more messy, but you could create meta functions that get the contents of the directory on the master and then write the file.managed functions dynamically… :D
18:28 Gifflen_ joined #salt
18:28 jdenning UtahDave: cmd_json.external_pillar()
18:28 madduck jdenning: hit me! ;)
18:30 UtahDave cmd_json has been there for quite a while, I believe
18:30 oz_akan_ joined #salt
18:30 jdenning madduck: Cool maybe you can help me with this - I have reclass set up, and it seems to be working, but can't seem to get salt to use the pillar data ($ salt 'test1' pillar.items doesn't return any of the data I set in reclass)
18:30 jdenning UtahDave: Yeah, but in 0.16.0 it does not take the minion_id as an input..that is what is confusing the heck out of me..
18:30 LyndsySimon joined #salt
18:31 jdenning madduck: I will get a pastebin up for you in just a sec..
18:31 madduck jdenning: cool. It's probably all my fault. ;)
18:32 madduck jdenning: show me the master config, give me some information about paths i.e. where you installed salt and reclass, and run the master with -l DEBUG and show me some output…
18:32 jdenning madduck: So here's my master config: http://pastebin.com/DyuWT78h
18:33 jdenning reclass node & class definitions under /srv/reclass/nodes & /srv/reclass/classes
18:33 madduck jdenning: "output: json" does not work, and that should get you an error in the master log
18:33 madduck the output must be yaml for Salt, and hence this is not configurable.
18:33 madduck does the master log show this as an error?
18:33 jdenning Hmm, let me check..
18:34 Gifflen joined #salt
18:34 madduck btw, you don't need nodes_uri and classes_uri if they are just nodes/classes under base_inventory_uri
18:34 jdenning madduck: Yeah, I've been trying a bunch of different stuff over the past couple of days :)
18:35 EntropyWorks for the git state module is there a way to set GIT_SSL_NO_VERIFY=1
18:35 madduck jdenning: you are the first user who seeks my help, so i am all motivated. ;)
18:35 Gifflen_ joined #salt
18:35 jdenning madduck: Ah, maybe this is the issue (from master.log): Specified ext_pillar interface reclass is unavailable - maybe it is not finding the package?
18:35 jdenning I installed via "sudo python setup.py install" (Ubuntu 12.04)
18:36 madduck jdenning: or it fails to load the module
18:36 madduck brb
18:37 JasonSwindle UtahDave:  I was trying JINJA in top.sls; and it bombed out.  Am I doing it wrong?  https://gist.github.com/JasonSwindle/f3e4bb064b24c5f628b6
18:38 Xeago joined #salt
18:38 Thiggy @dave_den: I got the logs. Interesting it looks like it is respecting the requiremnets, but gem still isn't available. It looks a lot like when you first load rvm and you have to logout and log back in for it to be available to you. https://gist.github.com/anonymous/79f95fd6fcacbaeeef22#file-minion-L16011
18:38 UtahDave JasonSwindle: line 1 should be: {% set self= salt['node_type']  %}
18:38 UtahDave no need for those braces
18:38 JasonSwindle Let me try
18:40 madduck jdenning: what version of Salt are you running?
18:40 jdenning 0.16.0
18:41 jdenning revised master config: http://pastebin.com/1xndeH6s
18:41 JasonSwindle UtahDave:  Different error, closer :)
18:41 JasonSwindle https://gist.github.com/JasonSwindle/f3e4bb064b24c5f628b6
18:41 madduck jdenning: does /path/to/salt/source/salt/pillar/reclass_adapter.py contain reclass_source_path ?
18:41 jdenning madduck: I'm not sure what you mean?
18:42 madduck jdenning: are you running salt from source or from a package?
18:42 jdenning madduck: Salt is from the package
18:42 madduck jdenning: does /srv/reclass-src/reclass contain a directory reclass? if not, then change the config to just /srv/reclass-src
18:42 madduck jdenning: which package? which distro?
18:42 oz_akan__ joined #salt
18:42 jdenning madduck: /srv/reclass-src/ is the path to my git clone from your repo
18:43 jdenning madduck: Ubuntu 12.04, standard Salt ppa
18:43 madduck jdenning: then you should use that as reclass_source_path
18:43 dave_den thiggy: yes indeed. you might try setting the rvm loading function in something like /etc/bashrc.d/90-rvm before installing rvm, or doing a cmd.wait that watches the ruby-1.9.3 install function that then runs the rvm load/reload
18:43 madduck the reclass source path is the root of the clone
18:43 dave_den i can't remember exactly how i fixed my issue, but it was something similar
18:43 madduck i should probably make this clearer
18:43 jdenning madduck: Gotcha, I'll try that..
18:44 cnelsonsic joined #salt
18:45 Thiggy @dave_den whacky. I'll try the cmd.wait and see what happens.
18:46 Thiggy unrelated: but does anyone know why my ubuntu boxen don't have locale's set for the salt-minion processes? when I do a cmd.run on a minion on an ubuntu box, there's no locale set, and django has an epic hissy fit because #reasons
18:47 JasonSwindle UtahDave:  any idea?  Or am I just doing it wrong? lol :)
18:47 jdenning madduck: Still not working - command output: http://pastebin.com/F0YKyzfQ
18:47 EugeneKay Probably because Ubuntu has funky ideas about setting env variables
18:47 oz_akan_ joined #salt
18:47 chrisgilmerproj hey, how do i modify the PATH environment variable on my salt minions?
18:48 madduck jdenning: what does -l DEBUG output on the master say now?
18:48 EugeneKay cmd.run doesn't spawn a shell, it just executes the given command
18:48 EugeneKay (AFAIK)
18:48 madduck jdenning: do you feel empowered to do some printf-debugging?
18:48 jdenning madduck: Got this error: [salt.pillar      ][CRITICAL] Specified ext_pillar interface reclass is unavailable
18:48 UtahDave JasonSwindle: Assuming that you've set a grain named 'node_type' then use this: {% set self = grains['node_type'] %}
18:48 madduck jdenning: yeah, that's not ideal. It means that Salt failed to load reclass. I wonder why.
18:49 jdenning madduck: Sure, I'm up for some debugging, at least until my morning meeting :)
18:49 Thiggy @EugeneKay yeah that's what I kind of gathered, though I didn't dig into source to confirm. I had to do some weird bash-fu to get some stuff to work because of that.
18:49 JasonSwindle I think that did it!  salt / grains / pillar
18:49 madduck jdenning: can you give me the output of dpkg -S reclass_adapter.py?
18:49 JasonSwindle too many things to remember
18:49 JasonSwindle lol
18:50 Thiggy @EugeneKay Do you have any suggested reading for Ubuntu's env variable weirdness?
18:50 jdenning dpkg-query: no path found matching pattern *reclass_adapter.py*.
18:50 EugeneKay Thiggy - the back of a rum bottle
18:50 Thiggy @eugenekay I prefer to read the bottom.
18:51 dave_den thiggy: also, the default shell for ubuntu is 'dash', so if salt is running /bin/sh (symlink to dash), it may not be reading your rvm load functions (rvm install puts it in your ~/.bashrc)
18:51 jdenning madduck: Is that supposed to install w/ salt?
18:51 EugeneKay The point stands
18:51 madduck jdenning: yes.
18:51 Thiggy dash? how have I never known this?
18:52 EugeneKay Because nobody in their right mind uses it
18:52 EugeneKay Again, Ubuntu has some funny ideas.
18:53 madduck jdenning: step one, patch the file salt/pillar/reclass_adapter.py with http://slexy.org/raw/s219PggsAo, start the salt master from the command line with -l DEBUG, reproduce the problem, and paste the entire output.
18:57 jdenning madduck: I don't think I have the reclass_adapter.py file to patch - http://pastebin.com/9As6GPzp  ..maybe this is just an issue with the ubuntu PPA?
18:57 kermit joined #salt
18:58 madduck jdenning: uh, yeah… you're right. 0.16.0 does not include reclass. :(
18:58 madduck jdenning: you now have two options: install it by hand, or run salt from source.
18:59 madduck i'd do the first.
19:00 jdenning madduck: Do I just need to drop the file into the right spot in the salt directory structure?
19:00 madduck jdenning: i.e. clone the salt repo and copy the file from the pillar and tops directories to the appropriate directories in /usr/lib (see dpkg -L output)
19:00 madduck jdenning: yes, it should just work
19:00 jdenning madduck: Cool, I'll try that..
19:00 madduck and when salt PPA includes it, then it'll just overwrite.
19:01 jdenning ls
19:01 jdenning oops :)
19:01 Thiggy @dave_den I think I'm just gonna use a cmd.wait with the rvm gem path hardcoded. This seems troublesome.
19:02 Thiggy thanks for your help
19:02 dave_den thiggy: just do a test with /bin/sh linked to /bin/bash in a test minion - it might be as simple as that
19:03 opapo joined #salt
19:03 dave_den i'm not sure there's an option for the default shell in the salt minion config - this might be a good thing to have
19:04 Gifflen joined #salt
19:06 juicer2 cp.push is available in 0.15.?  or only in 0.16?
19:10 jdenning madduck: copied the files over to pillar and tops respectively, applied the patch to salt/pillar/reclass_adapter.py , but I'm still getting the error "[CRITICAL] Specified ext_pillar interface reclass is unavailable" when I try to load the pillar
19:11 berto- joined #salt
19:12 madduck jdenning: just to be sure: where are the files you just patches, full path please?
19:12 jeffasinger joined #salt
19:12 jdenning madduck: I patched /usr/share/pyshared/salt/pillar/reclass_adapter.py
19:12 madduck are they readable by the user that runs the salt master?
19:13 jdenning madduck: yes (running as root)
19:13 madduck ew!
19:13 madduck ;)
19:13 madduck and you are running from the command-line?
19:13 jdenning madduck: (For testing! :) )
19:13 Thiggy @dave_den I can't get it to fail except on the first highstate. Even if I do a rvm implode and then do the symlink you suggested it succeeds
19:13 madduck i.e. output to stdout/stderr?
19:13 jdenning madduck: yes, from cli
19:13 Thiggy @dave_den (with or without symlink)
19:14 madduck jdenning: and there is no output resembling e.g. sys.path output?
19:14 * madduck fails to imagine that
19:15 dave_den thiggy: well, you can be sure it's the rvm loading issue. sorry i can't help much more than that
19:15 jdenning madduck: Checking.. you mean output from salt-master, correct?
19:15 madduck yes
19:17 Thiggy @dave_den no worries. knowing that is a huge help.
19:20 Joe630 is utahdave aboot
19:20 Joe630 and is mr. hatch an IRCer?
19:20 StDiluted joined #salt
19:20 jdenning madduck: Ok, this looks like a more sensible path: /usr/lib/pymodules/python2.7/salt/ ..I'll try to copy & patch again in that path.. dpkg -S does give the wierd /usr/share/pyshared/ stuff though..
19:20 StDiluted Hello all...
19:20 StDiluted I have an ordering problem…
19:20 madduck jdenning: the /usr/lib/pymodules/python2.7/salt/ path should just be symlinks
19:20 StDiluted https://gist.github.com/dginther/6091522
19:21 StDiluted Trying to get the repo to get added before the package gets installed
19:21 jdenning madduck - yep, you are correct..
19:21 madduck OF COURSE
19:21 madduck ALWAYS
19:21 jdenning madduck: Ok, I'll try to make the symlinks in that dir and see if that works..
19:21 StDiluted even though i am requiring the repo before the package, the package is getting installed first, resulting in the official ubuntu redis 2.2.12 getting installed instead of 2.6.x like I want from the repo
19:22 madduck jdenning: good point, yeah, that's required.
19:23 berto- StDiluted: line 5 looks wrong … i think it should be - require_in:\n  - pkg: redis
19:23 StDiluted ah, yeah just saw that myself, I'll fix that and give it another go
19:23 berto- StDiluted: but, i don't think you need the require_in on the pkgrepo _and_ require in pkg
19:23 StDiluted yeah, seems redundant but I was desperate ;)
19:24 pjs joined #salt
19:24 juicer2 cp.push is available in 0.15.?  or only in 0.16? I have file_recv: true set on my master
19:24 berto- in the salt docs that talk about the python client api, it mentions the need to run the program as root.  is it possible for the api to use the client_acl setting in /etc/salt/master ?
19:25 jdenning madduck: Ok, made the symlinks, here is the import error: http://pastebin.com/hEq0QdBX
19:26 madduck jdenning: you also need to install and symlink salt/utils/reclass.py
19:26 madduck sorry…
19:27 jdenning madduck: No worries :) (Thanks for releasing the code *and* help me troubleshoot! )
19:27 madduck wait until it works… ;)
19:28 madduck then you can send me beer.
19:29 dthom91 joined #salt
19:30 jdenning madduck: Ok, it loaded reclass that time! Now the error is that 1 argument is given but it is expecting 2.. http://pastebin.com/xLeEvusY
19:35 Thiggy If there's a choice between "- user: foo" and "- runas: foo", I will choose the wrong one every time.
19:38 [diecast] is it recommended to use a dedicated account for running salt-minion
19:39 StDiluted berto: thank you, that fixed it.
19:40 SEJeff_work [diecast], Well depends on what you want to use it for. If you want to run remote executions as that user only and not use the state stuff much at all, sure
19:40 [diecast] that's the issue im having
19:40 SEJeff_work if you want to use salt to configure your minions using states, run it as root (which is the default as most people use it for this as well as remote execution)
19:40 [diecast] i need both
19:40 SEJeff_work so run it as root
19:40 SEJeff_work [diecast], And if you want to get fancy, you could create a very small module which would emulate cmd.run, but always pass the runas argument
19:41 SEJeff_work Then set the ACL to allow the users who want to use it to only run the custom module that enforces they run as a certain user
19:41 [diecast] what if i want to get data from pillar.data as non-root while root is running salt-minion
19:41 madduck jdenning: i am confused by that. but I think this is my fault. github is not loading right now, but this sounds like a pull request i made since then against salt. Would you consider running Salt from source instead?
19:41 SEJeff_work pillar.data is stored in memory. I think you're confused a bit
19:41 [diecast] salt-call seems to fail as it wants to write to /etc/salt/pki/minion
19:42 SEJeff_work [diecast], Thats just a permissioning thing. You can run salt-call as non-root depending on what you want to do
19:42 jdenning madduck:  I've got to run to a meeting in about 5 min, but I'll set up a source install and get back to you after it's done..
19:42 [diecast] ok, so i could chown someone /etc/salt/pki/minion and salt-call should work
19:42 madduck jdenning: i'll be asleep then, sorry.
19:42 SEJeff_work [diecast], I'd likely use extended posix acls, but yes
19:43 [diecast] i would like to keep it as secure as possible, which is why chowning sounded iffy
19:43 SEJeff_work [diecast], Hence the acls
19:43 [diecast] SEJeff_work thanks for the info
19:43 SEJeff_work setfacl -m u:someuser:rx /etc/salt/pki/minion/foo.pem or whatever
19:43 madduck jdenning: I think this is commit 7dc623861469ec1a5677f313978c4abb30f30a0e and if so, then I screwed it up, which is a bit weird, but I'd appreciate if we debugged it.
19:44 madduck jdenning: just send me mail or message here or privmsg with anything you have.
19:44 jdenning madduck: No prob, you got me much further along.  I'll poke at it a bit more this afternoon..if you want to PM with your email add, I'll send you an email with what I find out..
19:44 SEJeff_work that allows user someuser read and execute (not really necessary) on /etc/salt/pki/minion/foo.pem (just an example)
19:44 madduck i will get back to you, but not this weekend… :/
19:44 [diecast] SEJeff_work excellent.. it probably just wants read
19:44 SEJeff_work [diecast], exactly
19:44 madduck jdenning: echo mailto: !#^."<*>"|tr "<*> mailto:" net@madduck
19:44 jdenning madduck: Sounds good - let's touch base on Monday :)  Thanks again for your help!
19:45 jdenning madduck: got it, I'll be in touch.
19:45 SEJeff_work [diecast], You can also create a group "localsalt" and do something like: setfacl -m g:localsalt:rx /etc/salt/pki/minion*
19:45 SEJeff_work I believe that will work
19:51 jslatts joined #salt
19:53 Fin1te joined #salt
19:53 teskew i've got a password stored as pillar.  it has two dollar signs back to back, like  pass$$word ...it's being used on the rabbitmq_user state module.  salt doesn't escape the dollar signs, but when we try to manually escape them as such  pass\$\$word , salt then escapes the escapes! :)  as such  pass\\$\\$word
19:54 teskew so, salt should either take our dollar signs as it from the pillar, or it should escape them correctly.  it does neither
19:55 teskew going to open an issue if nobody has a solution
19:57 SEJeff_work teskew, Have you tried quoting the string?
19:57 SEJeff_work Salt should never mess with pillar values that are in strings
19:57 SEJeff_work ie:
19:57 Xeago joined #salt
19:57 SEJeff_work password: "P@$$word"
19:58 teskew we have, indeed
19:59 SEJeff_work teskew, Odd I'd file an issue and then mail the salt-users mailinglist
19:59 SEJeff_work that sounds wrong
20:00 SEJeff_work From a python interpreter: >>> yaml.load('pass: foo$$bar')
20:00 SEJeff_work {'pass': 'foo$$bar'}
20:00 SEJeff_work >>> yaml.load("pass: foo$$bar")
20:00 SEJeff_work {'pass': 'foo$$bar'}
20:00 SEJeff_work >>> yaml.load('pass: "foo$$bar"')
20:00 SEJeff_work {'pass': 'foo$$bar'}
20:00 SEJeff_work def a bug in salt
20:00 zooz joined #salt
20:01 madduck why would salt worry about $?
20:02 [diecast] (╯°□°)╯︵ ┻━┻
20:02 SEJeff_work let me test
20:02 [diecast] acls not working out
20:02 jschadlick joined #salt
20:03 jschadlick1 joined #salt
20:04 chrisgilmerproj joined #salt
20:04 [diecast] interesting…. the plot thickens...
20:04 [diecast] salt python module owned by root
20:06 teskew so, the rabbitmq salt state module is trying to run something like 'rabbitmqctl user_add username password'..so we need the dollar signs escaped, because back to back dollar signs are interpreted as a set of numbers, obviously. when we try to escape the dollar signs (since salt won't leave our string alone) it then escapes our escapes :)
20:07 SEJeff_work [diecast], I just tested by creating a pillar with:
20:07 SEJeff_work test: P@$$word_encoded!
20:07 SEJeff_work worked perfect
20:07 SEJeff_work A set of numbers == the current pid of the shell
20:07 SEJeff_work in bash do: echo $$
20:07 [diecast] SEJeff_work https://gist.github.com/garlandkr/6091849
20:07 SEJeff_work [diecast], You need to fix the rabbitmq state to wrap the password in single quotes
20:08 SEJeff_work Has absolutely nothing to do with pillar
20:08 oz_akan_ joined #salt
20:08 [diecast] teskew
20:08 [diecast] ^^
20:08 SEJeff_work [diecast], Ok so give rx perms on the directory
20:08 SEJeff_work thanks
20:09 [diecast] i gave rwx to /etc/salt/pki* recursively
20:09 SEJeff_work [diecast], Oh but you didn't :)
20:09 oz_akan_ joined #salt
20:09 [diecast] ok, ill check and try again
20:09 SEJeff_work setfacl has to be explicitly set on the directory
20:09 [diecast] im willing to accept fault =)
20:09 cnelsonsic joined #salt
20:09 SEJeff_work run getfacl /etc/salt/pki
20:09 SEJeff_work Well setfacl -R doesn't work like chown/chmod -R
20:09 SEJeff_work because it is stupid
20:09 [diecast] user:jenkins:rwx
20:09 SEJeff_work you have to explicitly set the facls on the parent directory
20:10 jaequery joined #salt
20:10 cnelsonsic left #salt
20:10 SEJeff_work getfacl /etc/salt/pki/minion
20:10 [diecast] same
20:11 SEJeff_work Actually the code is explicitly running os.chmod
20:11 SEJeff_work But it doesn't have access
20:11 SEJeff_work os.chmod(dir_, 448)
20:11 SEJeff_work OSError: [Errno 1] Operation not permitted: '/etc/salt/pki/minion'
20:11 [diecast] ya
20:11 SEJeff_work I think there was something in the minion config you had to do which would chmod/chown it over. I can't remember exactly
20:12 SEJeff_work Actually in the minion config, set verify_env = False
20:12 SEJeff_work and re-run
20:12 SEJeff_work silly I know, but it will likely work then
20:12 [diecast] for a,b,c in os.walk('/etc/salt/pki/minion'): print a;
20:12 [diecast] i tried that in python and it works
20:12 [diecast] ok ill try
20:13 [diecast] negative
20:13 [diecast] back to original error now
20:14 teskew SEJeff, like  password: '{{ pillar['rabbitmq_password'] }}'   within the state?
20:14 [diecast] correct teskew
20:14 teskew doesn't work.  it removes the quotes when it tries to run the rabbitmqctl command via the state
20:14 SEJeff_work teskew, That might work. I actually meant the rabbitmq state python module should be fixed to always quote the pass. That might allow it
20:15 [diecast] you dont need to quote the {{}}
20:15 [diecast] just the python dict key
20:15 teskew i understand that [diecast]
20:15 [diecast] ok
20:16 teskew the rabbitmq state module needs to be fixed
20:16 Nexpro1 joined #salt
20:16 SEJeff_work teskew, What version of salt?
20:16 [diecast] SEJeff_work i restarted minion afterwards incase you wondered
20:17 teskew 15.3
20:17 SEJeff_work teskew, https://github.com/saltstack/salt/blob/develop/salt/states/rabbitmq_user.py#L68 so here you see it is calling the rabbitmq.change_password module
20:18 SEJeff_work teskew, Do you know any python and git?
20:18 teskew i do
20:18 SEJeff_work teskew, It would be trivial to fix this: https://github.com/saltstack/salt/blob/develop/salt/modules/rabbitmq.py#L128
20:18 teskew the problem is not just the quotes, though
20:18 SEJeff_work I think it is because you said $$ is being turned into a number, which (in bash) expands to the current shell's pid
20:18 SEJeff_work teskew, This line is blatantly wrong by not quoting the password: https://github.com/saltstack/salt/blob/develop/salt/modules/rabbitmq.py#L137
20:19 SEJeff_work Change {1} to \'{1}\'
20:19 teskew because if you run rabbitmqctl user_add username 'pass$$word'  from the command line, it still doesn't work. it needs not to escape the escapes when you escape a password in pillar
20:19 SEJeff_work how can it not work then?
20:20 SEJeff_work if you manually run that command, it should work
20:20 teskew nope.
20:20 SEJeff_work That will pass 'pa$$word' to rabbitmq
20:20 baniir joined #salt
20:20 teskew yeah, and that doesn't work from the command line
20:20 teskew only pass\$\$word  works from the command line
20:21 jeffmendoza joined #salt
20:21 teskew but if you put pass\$\$word in the pillar, salt escapes the escapes as such  pass\\$\\$word
20:27 [diecast] SEJeff_work so tedious… getting closer
20:27 [diecast] needed write to /var/log/salt/minion and /var/cache/salt/minion
20:29 [diecast] that was it.. working now
20:29 whiskybar joined #salt
20:30 dthom91 joined #salt
20:30 jslatts joined #salt
20:40 hazzadous joined #salt
20:40 felskrone joined #salt
20:48 robertkeizer left #salt
20:49 MZAWeb joined #salt
20:50 MZAWeb Hi! Just starting with Salt and I'm wondering if there's a way to create a State for an old PHP version (ie: 5.2, 5.3) to spawn QA sites. Can't find anything? Even compiling from source would be fine, I guess.
20:52 napperjabber joined #salt
20:57 Ahlee restarting salt-minions from the salt-master - do people just background the service.restart?
21:00 dthom91 joined #salt
21:03 jdenning joined #salt
21:03 Xeago joined #salt
21:06 jslatts joined #salt
21:08 brutasse joined #salt
21:11 auser joined #salt
21:11 auser hey all
21:11 auser anyone here in San Diego, by chance?
21:13 nmistry joined #salt
21:19 qba73 joined #salt
21:32 pentabular joined #salt
21:37 pentabular joined #salt
21:39 chrisrebert joined #salt
21:39 pentabular1 joined #salt
21:40 chrisgilmerproj any idea how to change the PATH environment variable for cmd.run?
21:40 chrisgilmerproj on the command line?
21:40 chrisgilmerproj or better yet, permanently for salt
21:44 DanGarthwaite left #salt
21:48 zooz joined #salt
21:49 racooper chrisgilmerproj,  it's probably better to use full paths in scripts and remote commands
21:50 chrisgilmerproj sure, but it would be nice to include /usr/local/bin as a standard path
21:50 chrisgilmerproj so many things come out of that directory that its a convenience to have it available
21:51 chrisgilmerproj and it certainly is using some kind of environment variable for the path
21:52 chrisgilmerproj aha, finally
21:52 chrisgilmerproj its in the salt.utils default_path
21:53 [diecast] anyone else find that using .sls file extensions under pillar is confusing sometimes
21:55 chrisgilmerproj and there's an article in the man page on this
21:55 Ryan_Lane joined #salt
21:56 olive_ joined #salt
21:57 olive_ Hello, new to salt, I have a basic question to control dpkg/apt on Debian
21:57 olive_ Is it possible to specify to refresh the dpkg/apt database in a SLS file ?
21:58 chrisgilmerproj what's the debian equivalent of the freebsd rc.conf file?
21:59 stephen__ joined #salt
21:59 Ryan_Lane joined #salt
21:59 UtahDave joined #salt
22:00 dthom91 joined #salt
22:01 Ryan_Lane joined #salt
22:02 auser hey all
22:02 auser anyone here in San Diego, by chance?
22:03 Joe630 i'v been TOO san diego!
22:03 Joe630 to.
22:03 Joe630 to san diego.
22:03 carlos joined #salt
22:03 Joe630 good tacos
22:03 Ryan_Lane joined #salt
22:03 auser yeah?
22:04 auser we are vacationing here for a few days
22:04 auser unfortunately that means we can't attend the salt hackathon tomorrow
22:10 dave_den olive_: yes, you can set refresh = True - see http://docs.saltstack.com/ref/states/all/salt.states.pkg.html
22:11 olive_ @dave_den: Thanks !
22:11 dave_den refresh is listed in the pkg.installed args, but it looks like it's missing from the arg descriptions
22:17 jacksontj joined #salt
22:18 jacksontj UtahDave: hi :)
22:20 kermit joined #salt
22:20 stevetodd joined #salt
22:24 mikedawson joined #salt
22:24 oz_akan_ joined #salt
22:26 MZAWeb Hey, any way to select a specific version for a pkg in a Salt?
22:26 MZAWeb Trying to create QA environments for PHP and I need to install 5.2, 5.3 and 5.4 in different minions
22:30 baniir joined #salt
22:33 go8ose joined #salt
22:34 go8ose Can someone clear up something for me on using github. If I think of something to fix, and I write a pull request, is there any point in also lodging an issue about it?
22:35 go8ose Or are issues ways to submit ideas/problems that you don't have a patch for?
22:35 kenbolton joined #salt
22:39 iquaba joined #salt
22:40 p3rror joined #salt
22:41 sciyoshi2 joined #salt
22:41 UtahDave MZAWeb: yeah, you can select a specific version
22:42 UtahDave go8ose: If it's a quick fix, go ahead and just send in a pull request
22:42 UtahDave go8ose: It's often good to open up an issue in case someone else has noticed the bug and has started working on it.  The issue tracker can help a little bit to avoid duplicating effort.
22:45 kenbolton joined #salt
22:45 KyleG joined #salt
22:47 Newt[cz] joined #salt
22:51 chrisrebert Is there a way, when installing a daemon package, to prevent the daemon from being started by the postinst hook?
22:53 olive_ Trying to install uwsgi in a virtualenv and start the service. I got "uwsgi: Package already installed" but pip freeze doesn't show uwsgi. Is there some subtetly to install in a virtualenv ?
22:53 pentabular1 left #salt
22:53 chrisrebert olive_: perhaps it's installed as a system site-package?
22:54 olive_ I used. uwsgi: pip.installed: pkgs: uwsgi
22:54 olive_ apt-cache policy uwsgi gives: Installed: (none)
22:56 olive_ Nothing also in the log with salt-call -l debug state.highstate > /tmp/log.txt 2>&1
22:56 olive_ Nothin interesting I mean
22:56 chrisrebert olive_: If it's in a virtualenv, looks like you gotta specify `bin_env` in the pip.installed
23:01 whit joined #salt
23:04 olive_ chrisrebert: that was it ! Thanks ! But why didn't I get an error ? And why was uwsgi reported as Installed ? Was it installed in a hidden virtualenv ??
23:06 chrisrebert olive_: I think perhaps it was installed in the global site-packages by *pip*, not by apt. check /usr/lib/pythonN.N/{dist,site}-packages/ for uwsgi entries
23:09 olive_ find . -iname "*wsgi*" in /usr/lib doesn't give anything
23:11 olive_ Found it: usr/local/lib/python2.7/dist-packages !
23:11 StDiluted joined #salt
23:13 andrew joined #salt
23:17 sciyoshi2 left #salt
23:27 krissaxton joined #salt
23:31 jacksontj joined #salt
23:31 chrisrebert Is there a way, when installing a daemon package, to prevent the daemon from being started by the postinst hook?
23:32 chrisrebert left #salt
23:39 ipmb joined #salt
23:42 StDiluted joined #salt
23:43 jMyles joined #salt
23:52 jacksontj joined #salt
23:55 jacksontj_ joined #salt
23:59 jschadlick1 left #salt

| Channels | #salt index | Today | | Search | Google Search | Plain-Text | summary