Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2014-02-28

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

All times shown according to UTC.

Time Nick Message
00:02 manfred does anyone remember/know where __salt__ is injected into the environment?
00:02 manfred i can't find it
00:05 kintel joined #salt
00:06 snuffeluffegus joined #salt
00:10 clearcla1 joined #salt
00:14 budrose joined #salt
00:16 markm joined #salt
00:16 ninkotech__ joined #salt
00:17 jacksontj_ joined #salt
00:17 dave_den manfred:  loader.py
00:17 manfred thanks
00:17 justBob joined #salt
00:19 hunter joined #salt
00:21 ipalreadytaken joined #salt
00:21 pydanny joined #salt
00:21 themadcanudist hrm… dave_den… can you mix jinja with py and yaml? if so, how do you "brace" the py code?
00:23 pdayton joined #salt
00:23 ndrei joined #salt
00:26 schristensen joined #salt
00:28 fauxtoniche joined #salt
00:29 taion809 joined #salt
00:31 Zuru joined #salt
00:32 srage joined #salt
00:32 zzzirk joined #salt
00:39 ninkotech__ joined #salt
00:39 pcarrier_ joined #salt
00:39 kainswor any changes to multimaster for v2014? I'm getting an exception in tune_in that prevents the minion daemon from running
00:39 scalability-junk joined #salt
00:39 jinnko_ joined #salt
00:39 melinath_ joined #salt
00:39 ajw0100_ joined #salt
00:39 wendall911 joined #salt
00:39 goki_ joined #salt
00:39 justBob joined #salt
00:40 jeblair joined #salt
00:40 twinshadow joined #salt
00:42 abele joined #salt
00:42 aurigus_ joined #salt
00:42 marcinkuzminski joined #salt
00:42 jeff joined #salt
00:42 mortis_ joined #salt
00:42 TamCore joined #salt
00:42 mgw joined #salt
00:51 ilbot3 joined #salt
00:51 Topic for #salt is now Welcome to #salt | Thanks to everyone who made SaltConf 2014 amazing! | 2014.1.0 is the latest | Please be patient when asking questions as we are volunteers and may not have immediate answers | Channel logs are available at http://irclog.perlgeek.de/salt/
00:51 hunter_ joined #salt
00:51 Zuru_ joined #salt
00:51 rgbkrk_ joined #salt
00:52 Sypher joined #salt
00:52 darrend_ joined #salt
00:52 njpatel joined #salt
00:53 Valda joined #salt
00:53 chrisjones joined #salt
00:54 schristensen joined #salt
00:56 ninkotech joined #salt
00:56 beardo_ joined #salt
00:56 eculver joined #salt
00:56 eculver joined #salt
00:57 baoboa joined #salt
00:57 kballou joined #salt
00:58 ahammond joined #salt
00:59 vipul joined #salt
00:59 Valdo joined #salt
00:59 mikkn joined #salt
01:00 ajw0100 joined #salt
01:00 svx joined #salt
01:01 markm joined #salt
01:01 bashcoder joined #salt
01:02 d10n joined #salt
01:02 flebel joined #salt
01:03 gnugnu6 joined #salt
01:04 jpcw joined #salt
01:04 nn0101 joined #salt
01:05 think-free joined #salt
01:06 kermit joined #salt
01:07 Nazca__ joined #salt
01:07 JordanRinke joined #salt
01:07 ajw0100 joined #salt
01:08 sulky joined #salt
01:08 munhitsu_ joined #salt
01:08 erjohnso joined #salt
01:08 neilf_ joined #salt
01:08 abele joined #salt
01:08 rgbkrk_ joined #salt
01:09 ninkotech__ joined #salt
01:10 jforest joined #salt
01:10 MTecknology joined #salt
01:12 hardwire joined #salt
01:12 timoguin_ joined #salt
01:12 kaictl joined #salt
01:13 kaictl manfred made me do it
01:14 themadcanudist1 joined #salt
01:16 Katafalkas joined #salt
01:17 timoguin joined #salt
01:20 liwen joined #salt
01:22 xzarth_ joined #salt
01:24 kintel_ joined #salt
01:24 ninkotech joined #salt
01:24 rgbkrk joined #salt
01:25 saurabhs1 joined #salt
01:25 n8n joined #salt
01:25 vbabiy_ joined #salt
01:26 SEJeff_work joined #salt
01:26 Ryan_Lane1 joined #salt
01:27 zain joined #salt
01:28 dwfreed_ joined #salt
01:28 think-free_ joined #salt
01:28 schristensen_ joined #salt
01:28 rreboto joined #salt
01:29 kedo40 joined #salt
01:30 jnials joined #salt
01:30 kaictl joined #salt
01:30 melinath joined #salt
01:30 troyready joined #salt
01:30 scarcry_ joined #salt
01:31 abele_ joined #salt
01:31 andersb joined #salt
01:31 vipul joined #salt
01:32 patrek_ joined #salt
01:33 skynet joined #salt
01:33 NotreDev joined #salt
01:33 utahcon joined #salt
01:33 jeff joined #salt
01:33 jeff joined #salt
01:33 kossy joined #salt
01:34 hardwire joined #salt
01:34 rmnuvg joined #salt
01:35 SmileyChris joined #salt
01:35 jmccree joined #salt
01:37 SmileyChris joined #salt
01:37 SmileyChris Hi, having a strange failure with the file.directory state failing with "Failed to change mode to 750;"
01:38 SmileyChris https://dpaste.de/buoa
01:38 pydanny joined #salt
01:39 kballou joined #salt
01:39 gasbakid joined #salt
01:39 SmileyChris is there an easy way to try and get a bit more info about the error?
01:39 jinnko_ joined #salt
01:40 mgw joined #salt
01:41 toothrot joined #salt
01:41 schimmy1 joined #salt
01:42 toothrot left #salt
01:43 kedo39 joined #salt
01:44 schimmy joined #salt
01:45 gasbakid joined #salt
01:49 gasbakid joined #salt
01:51 synfin In the top.sls file, is 'base' a keyword?
01:51 fridiculous joined #salt
01:54 liwen joined #salt
01:54 manfred synfin: i believe dev and qa are keywords for environments?
01:54 manfred err and base
01:55 themadcanudist1 left #salt
01:55 gasbakid joined #salt
01:58 vejdmn joined #salt
01:58 synfin ty
02:00 gasbakid joined #salt
02:01 alexj joined #salt
02:04 kermit1 joined #salt
02:04 gnugnu6 joined #salt
02:04 Networkn3rd joined #salt
02:08 budrose joined #salt
02:09 jacksontj_ joined #salt
02:10 timoguin_ joined #salt
02:11 snuffeluffegus joined #salt
02:14 timoguin joined #salt
02:22 njpatel joined #salt
02:23 ravibhure joined #salt
02:28 ravibhure left #salt
02:28 l0x3py joined #salt
02:32 Ryan_Lane joined #salt
02:32 kolaman joined #salt
02:33 kolaman hi all, what is the latest version (stable) of salt stack ?
02:35 mgw joined #salt
02:36 joehh 2014.1 has just been released
02:36 joehh packages are not yet available for all distros/oses
02:36 joehh debian ubuntu coming shortly
02:37 ravibhure1 joined #salt
02:37 patrek_ joined #salt
02:37 kolaman joehh: actually when I install from epel repo (as described in doc), that is showing up 0.17.5-1.el6 as version release . .so is that the latest ?
02:37 ravibhure1 left #salt
02:37 kaictl joined #salt
02:40 rgbkrk joined #salt
02:40 jinnko_ joined #salt
02:41 Ahlee latest packaged, probably. It takes a relatively long time for releases to trickle up
02:41 Ahlee Released latest is Hydrogen, which is 2014.1.0
02:44 dimeshake joined #salt
02:44 zaz\ joined #salt
02:44 jaimed joined #salt
02:46 ipmb joined #salt
02:47 faldridge joined #salt
02:57 HeadAIX joined #salt
02:58 Ryan_Lane joined #salt
03:04 pdayton joined #salt
03:13 anuvrat joined #salt
03:18 fllr joined #salt
03:23 forresta Ahlee, the stable should be there tomorrow or Saturday depending on how long the mirrors take to update. We made an effort to vote it out of the test repo
03:23 rostam joined #salt
03:24 abele joined #salt
03:25 srage joined #salt
03:25 Ahlee forresta: woot
03:26 forresta yup
03:28 fllr joined #salt
03:29 yomilk joined #salt
03:31 njpatel joined #salt
03:31 NotreDev i'm getting a false error with Salt: http://pastebin.com/YriKPV8k
03:37 n8n joined #salt
03:41 jinnko_ joined #salt
03:42 pdayton joined #salt
03:48 schimmy joined #salt
03:51 dccc joined #salt
03:52 mattmtl joined #salt
03:55 malinoff joined #salt
04:03 rojem joined #salt
04:08 snuffeluffegus joined #salt
04:12 thayne joined #salt
04:19 frasergraham joined #salt
04:23 sgviking joined #salt
04:34 _fllr_ joined #salt
04:34 yomilk_ joined #salt
04:34 ravibhure joined #salt
04:34 Katafalk_ joined #salt
04:34 ravibhure left #salt
04:35 jacksontj__ joined #salt
04:35 pdayton joined #salt
04:36 dmalinovsky joined #salt
04:36 abele_ joined #salt
04:37 patrek__ joined #salt
04:37 kedo39 joined #salt
04:37 rgbkrk joined #salt
04:38 rostam joined #salt
04:38 kaictl joined #salt
04:39 kossy joined #salt
04:40 pdayton joined #salt
04:40 davet joined #salt
04:42 dmalinovsky_ joined #salt
04:44 pdayton joined #salt
04:44 think-free joined #salt
04:45 jpcw joined #salt
04:45 nn0101 joined #salt
04:45 kermit joined #salt
04:46 rgbkrk_ joined #salt
04:47 kintel joined #salt
04:48 cofeineSunshine joined #salt
04:48 abele joined #salt
04:50 pdayton joined #salt
04:50 malinoff joined #salt
04:50 kedo39 joined #salt
04:50 rostam joined #salt
04:50 kossy joined #salt
04:51 kaictl joined #salt
04:54 malinoff_ joined #salt
04:55 abele_ joined #salt
04:56 svx joined #salt
04:56 eculver joined #salt
04:56 eculver joined #salt
04:59 cofeineSunshine joined #salt
04:59 rgbkrk joined #salt
04:59 pdayton1 joined #salt
05:00 jacksontj_ joined #salt
05:00 NotreDev_ joined #salt
05:00 patrek_ joined #salt
05:01 malinoff joined #salt
05:01 patrek_ joined #salt
05:01 malinoff_ joined #salt
05:01 malinoff joined #salt
05:02 scarcry joined #salt
05:02 zzzirk_ joined #salt
05:02 xzarth joined #salt
05:02 ravibhure1 joined #salt
05:02 abele_ joined #salt
05:03 ravibhure1 left #salt
05:03 kaictl joined #salt
05:04 yomilk joined #salt
05:06 kossy joined #salt
05:06 frasergraham joined #salt
05:08 andersb joined #salt
05:08 frasergraham joined #salt
05:10 foresj joined #salt
05:10 dN0t joined #salt
05:12 pdayton joined #salt
05:12 eculver_ joined #salt
05:13 MTecknology joined #salt
05:13 scarcry_ joined #salt
05:14 abele__ joined #salt
05:15 kaictl joined #salt
05:15 danielbachhuber joined #salt
05:19 cofeineSunshine joined #salt
05:21 Katafalkas joined #salt
05:22 eculver joined #salt
05:28 ewong_ joined #salt
05:28 ckao joined #salt
05:28 jcockhren joined #salt
05:28 dN0t joined #salt
05:28 renoirb joined #salt
05:28 rhand joined #salt
05:28 jforest joined #salt
05:29 monokrome joined #salt
05:29 jmccree joined #salt
05:30 kaictl joined #salt
05:30 rojem joined #salt
05:31 anuvrat joined #salt
05:32 MTecknology joined #salt
05:42 Gordonz joined #salt
05:43 jinnko_ joined #salt
05:44 jacksontj_ joined #salt
05:51 jmccree_ joined #salt
05:51 kaictl joined #salt
05:51 forresta Had to tweet back at saltstack about the lack of the toasters in this newest salt air, no toaster, 0/10.
05:52 jforest joined #salt
05:52 malinoff +
05:52 malinoff Ups
05:52 malinoff Missclick
05:53 MTecknology joined #salt
05:55 kossy joined #salt
05:58 Gifflen joined #salt
06:07 monokrome joined #salt
06:08 jforest joined #salt
06:08 jeffro joined #salt
06:09 MTecknology joined #salt
06:10 frasergr_ joined #salt
06:11 Gordonz joined #salt
06:14 jmccree joined #salt
06:15 kossy joined #salt
06:26 linjan_ joined #salt
06:28 kermit joined #salt
06:39 ndrei joined #salt
06:45 jinnko_ joined #salt
06:46 leonardinius joined #salt
06:50 jacksontj_ joined #salt
06:52 rojem joined #salt
06:57 bhosmer joined #salt
07:01 jacksontj_ joined #salt
07:03 saurabhs1 left #salt
07:05 ipalreadytaken joined #salt
07:08 linjan joined #salt
07:14 malinoff joined #salt
07:19 Shenril joined #salt
07:21 hardwire joehh: google fails me (irc logs) when I try to figure out how debian packaging with 2014.1.0 will be approached due to the version schema change.
07:22 hardwire whats the haps?  any way I can assist?
07:22 hardwire also.. bed time.
07:23 che-arne joined #salt
07:25 leonardinius joined #salt
07:31 jacksontj_ joined #salt
07:32 NotreDev joined #salt
07:33 Iwirada joined #salt
07:38 tr_h joined #salt
07:44 fllr joined #salt
07:44 astol joined #salt
07:45 Ryan_Lane joined #salt
07:46 _fllr_ joined #salt
07:46 jinnko_ joined #salt
07:47 ml_1 joined #salt
07:49 jinnko_ joined #salt
07:56 abele joined #salt
08:02 lessismore joined #salt
08:04 slav0nic joined #salt
08:06 alexj joined #salt
08:08 harobed_ joined #salt
08:09 fllr joined #salt
08:09 balboah joined #salt
08:10 fllr joined #salt
08:13 Kenzor joined #salt
08:19 miguitas joined #salt
08:22 chuffpdx_ joined #salt
08:26 zooz joined #salt
08:31 giantlock_ joined #salt
08:35 yomilk joined #salt
08:38 kadel joined #salt
08:42 aleszoul3k joined #salt
08:44 fllr joined #salt
08:46 scott_w joined #salt
08:46 _fllr_ joined #salt
08:56 johtso joined #salt
09:06 ndrei joined #salt
09:07 alexj joined #salt
09:08 MrTango joined #salt
09:20 ggoZ joined #salt
09:25 jinnko_ joined #salt
09:34 xet7 joined #salt
09:35 favadi joined #salt
09:40 bhosmer joined #salt
09:41 astol joined #salt
09:44 fllr joined #salt
09:48 symroe joined #salt
09:50 martoss joined #salt
09:50 symroe I'm running "salt-call pillar.items --local" on my laptop (brand new MacBook Pro ssd/16gb ram) and finding it *very* slow (taking about 5 seconds).  Is there something I can do to speed this up?
09:50 symroe I've tried disabling all the modules I don't need, but that doesn't seem to help at all
09:51 symroe (On 0.17.5)
09:54 nn0101 joined #salt
10:00 symroe joined #salt
10:08 babilen I am using a, very simple, recipe http://paste.debian.net/84375/ and started playing with different environments. Unfortunately I get a "Detected conflicting IDs, SLS IDs need to be globally unique" error. How would I deal with this most idiomatically?
10:10 symroe I don't think you need to include global.packages twice
10:13 babilen symroe: Yes, I do not, but I am including other (identical) states in both environments, so I will probably run into this sooner or later.
10:13 astol joined #salt
10:14 babilen I just don't quite understand what exactly I'm doing wrong. It should be possible to include two similar states in two different environments or do I have this all wrong?
10:15 n8n joined #salt
10:15 symroe Well, the wildcard there will import in every env, so you'll get a name overlap.  Change the wildcard, or namespace the states?
10:18 dave4925 joined #salt
10:19 babilen symroe: Okay, the general plan is to have two git branches (master (aka base) and dev) and to merge tested new configurationx from dev into base. I'm just not sure how I can test changes to states that I already use in base if I can't have "duplicates" in dev.
10:19 babilen symroe: Could you elaborate on the "namespace the states" part? Does that mean that I would call all "dev" states something else? Doing that would make it impossible to simply merge commits from the dev branch to the master branch though.
10:20 symroe Yeah, I guess it would :/
10:20 babilen Was that the idea?
10:20 symroe Erm, Maybe wait for someone else to reply, I'm not sure I know how to deal with the git branches stuff
10:20 symroe Are you using gitfs?
10:21 babilen So, the main problem is that I cannot target the same minion with the same state (by state ID not content) in more than one environment?
10:21 babilen Yes, I am indeed using GitFS
10:21 babilen Just trying to combine a sensible git workflow with salt :)
10:23 babilen The setup uses vagrant locally on our developer boxes for the "dev" branch and corresponding minions and "base"/"master" is essentially the production system. Might introduce additional layers, but that's it for now.
10:25 joehh hardwire: uploaded to unstable (NEW queue) now
10:25 joehh backporting and uploading to debian.saltstack.com now
10:25 babilen joehh: Thank you!
10:25 babilen Much appreciated and a nice gift before the weekend
10:26 joehh no worries
10:27 xl1 joined #salt
10:27 xl1 left #salt
10:27 babilen symroe: The problem seems to be that the ID I generate is not unique, would it make sense to simply include the environment name in the ID and slightly rewrite that recipe so that it uses enumerates the packages to install explicitly?
10:28 babilen Background that this question is that the SLS seems to be generated regardless of me targetting a minion and I therefore will always run into this problem with generated state-ids
10:35 nn0101 joined #salt
10:36 yomilk joined #salt
10:44 fllr joined #salt
10:46 babilen How can I get the name of the current environment in jinja? (or where can I find documentation to answer questions like this myself?)
10:47 symroe joined #salt
10:49 diegows joined #salt
10:54 jeffro joined #salt
10:55 faldridge joined #salt
10:56 faldridg_ joined #salt
11:06 nn0101 joined #salt
11:07 jrdx joined #salt
11:07 babilen Okay, I ended up using: http://paste.debian.net/84547/ -- Why is pciutils installed twice?
11:11 yomilk joined #salt
11:16 ggoZ joined #salt
11:18 alexj joined #salt
11:19 valgrind joined #salt
11:23 NotreDev joined #salt
11:24 ndrei joined #salt
11:28 carlos_ joined #salt
11:43 elfixit joined #salt
11:44 fllr joined #salt
11:45 bhosmer joined #salt
11:51 Ztyx1 joined #salt
11:52 Ztyx1 Godday, I've implemented an external pillar module. How can I make the master pick it up? I put directly under a folder that I defined as `extension_modules` in the master config?
11:55 middleman_ joined #salt
12:07 psyl0n joined #salt
12:09 whyzgeek joined #salt
12:14 Ztyx joined #salt
12:17 unicoletti joined #salt
12:18 ndrei joined #salt
12:25 srage joined #salt
12:27 cnf left #salt
12:30 zzzirk joined #salt
12:34 glen_ joined #salt
12:43 srage joined #salt
12:44 jeddi I thought I'd found one ages ago, but no luck re-finding it --- I'm looking for a good pillar / networking (state) combination for letting me have n numbers of interfaces configured per pillar/hostname.sls ... so i can dynamically have eth0, eth1, eth2, etc -- and I can generate a (debian-friendlly) /etc/network/interfaces from same, regardless of which parameters I can provide (ie. *normally* I don't have gateway on anything other
12:44 jeddi than eth0)
12:44 fllr joined #salt
12:49 n8n joined #salt
12:59 valgrind joined #salt
13:00 unicoletti joined #salt
13:00 MrTango joined #salt
13:01 valgrind joined #salt
13:05 Ztyx left #salt
13:07 bhosmer joined #salt
13:15 jcsp joined #salt
13:19 babilen Argh! It seems to be impossible to combine a sane git workflow in which one can merge between branches with salt and gitfs. How do you deal with this? How do you stage changes to production (base) and test them?
13:23 alexj joined #salt
13:33 timoguin joined #salt
13:35 unicoletti left #salt
13:36 Katafalkas joined #salt
13:41 faldridge joined #salt
13:44 fllr joined #salt
13:45 jslatts joined #salt
13:46 leonardinius joined #salt
13:49 ndrei joined #salt
13:51 Sypher__ joined #salt
13:51 martoss joined #salt
13:55 symroe joined #salt
14:00 R_Shackleford joined #salt
14:01 martoss left #salt
14:02 ipmb joined #salt
14:02 terminalmage joined #salt
14:03 symroe joined #salt
14:03 rojem joined #salt
14:05 toastedpenguin joined #salt
14:07 vejdmn joined #salt
14:08 Ahlee babilen: salt subsetofminions state.sls my.state env=<branch> test=True
14:13 rgbkrk joined #salt
14:13 bhosmer_ joined #salt
14:14 racooper joined #salt
14:15 fragamus joined #salt
14:17 ninkotech joined #salt
14:21 quickdry21 joined #salt
14:21 middleman_ joined #salt
14:22 sroegner joined #salt
14:23 R_Shackleford joined #salt
14:25 faldridge joined #salt
14:27 Networkn3rd joined #salt
14:27 pidah joined #salt
14:28 dyim joined #salt
14:28 pass_by_value joined #salt
14:29 pidah Hi,  just a quick question, does gitfs work with a minion in masterless mode ?
14:31 babilen Ahlee: I don't quite see how that helps me with merging in git branches (as I would still have to explicitly define environments in top.sls that I cannot merge between branches for example)
14:32 Ahlee Every branch is an environment, you don't need to specify them in top.sls
14:32 Linz joined #salt
14:32 Ahlee when I want to do something new, I branch, target the branch, when i'm happy i merge that branch in
14:33 babilen Would you mind elaborating on that? I don't quite get it yet :)
14:34 babilen What I should maybe add is that I use a "playground" in vagrant in which I have a salt master + the minions that I want to test.
14:34 Ahlee babilen: so this minion here, it happens to be in environment=Staging, but say I want to change the state packages to include a new package.  I create a branch feature_newpackage, modify the state, push it up to the git server
14:35 Ahlee i then salt mytestminions packages env=feature_newpackage
14:35 mpanetta joined #salt
14:35 babilen Ahlee: How does it happen to "be" in environment=Staging? Do you set that in the minion config?
14:35 Ahlee babilen: Yes, but that's mostly irrelevant since I'm telling it with the target what environment to use
14:36 babilen Okay, but that doesn't work with highstate
14:36 Ahlee the same is true for me in production wich resolves salt's base to master branch
14:36 babilen Ahlee: The best I could find for that is to set gitfs_base to the branch, but that will become tedious for short-lived feature branches
14:36 Ahlee Is highstate still lacking env tracking?
14:37 Ahlee I don't use highstate myself, sorry
14:37 babilen Okay - Well, I am pretty new and just want to figure out a reasonable workflow :)
14:39 babilen Ahlee: With highstate it seems to be impossible to use the same top.sls file for different branches as I can't seem to be able to specify the branch explicitly (unless I set gitfs_branch), but would have to do that via targetting in top.sls
14:39 babilen That, however, makes it impossible to merge changes on the feature branch back into master as the top.sls files are imcompatible.
14:40 babilen And if that weren't enough already it seems as if top.sls files are simply merged between branches with a "base/master wins in case of conflicts" strategy
14:41 kintel joined #salt
14:42 Ahlee That is true, all top.sls files are merged into a single dictionary
14:42 Ahlee top.sls is awkward with gitfs. it's recommended to only have one
14:42 babilen How can I only have one? I mean I branch in git and therefore start with a copy of the original branch :)
14:42 Ahlee which makes any type of branching strategy wierd
14:42 Ahlee right.
14:42 babilen exactly
14:43 babilen This just doesn't seem to fit together and my wonderful plan of "push to vagrant git to test" and "push to production repo to get it live" is somehow impossible :)
14:44 fllr joined #salt
14:45 babilen I don't really care *how* I configure it, but it would just be very nice to be able to replicate the production environment with a bunch of local VMs with which I can play without the fear to break anything
14:45 HeadAIX joined #salt
14:45 alexj joined #salt
14:47 ninkotech joined #salt
14:51 timoguin babilen, I put an environment-specific top.sls in each branch and merged all the way from dev to master, with each branch keeping its own top.sls
14:51 timoguin of course this makes merging the top.sls difficult
14:52 timoguin but once it's set, merging other changes works as expected
14:52 babilen As soon as I make changes in one branch and commit them I will merge those changes. That includes changes to top.sls ... I therefore don't quite follow
14:52 timoguin I'm still trying to figure out a good workflow for developing locally with a Vagrant setup and pushing it out to a cloud-based setup
14:53 babilen I am now trying to play with "gitfs_base: dev", but that doesn't seem to play nice with an external git pillar.
14:53 babilen (as I don't seem to be able to set "base == dev" for that pillar too)
14:53 zzzirk joined #salt
14:54 babilen So all I am seeing for hours now is "in environment 'base' is not available on the salt master" with s/base/dev and various states :-/
14:54 gasbakid joined #salt
14:54 n8n joined #salt
14:58 jaimed joined #salt
14:59 Mr_N morning
14:59 babilen heh :)
15:01 kedo39 joined #salt
15:01 juicer2 joined #salt
15:02 danielbachhuber joined #salt
15:02 GradysGhost joined #salt
15:03 jeremyBass joined #salt
15:05 dyim hi
15:05 dyim am i doing something wrong using a map.jinja file and in the pkg list, i'm doing {{ map.pkgs|json }}
15:06 dyim I"m getting a TypeError Undefined is not JSON serializable
15:06 Gifflen joined #salt
15:07 fishpen0 joined #salt
15:07 NotreDev joined #salt
15:07 fishpen0 left #salt
15:07 vejdmn joined #salt
15:09 Gifflen joined #salt
15:11 CeBe joined #salt
15:11 gnugnu6 joined #salt
15:16 gnugnu8 joined #salt
15:17 blee joined #salt
15:19 cron0 joined #salt
15:20 gnugnu8 joined #salt
15:20 HeadAIX joined #salt
15:21 babilen Is there something comparable to gitfs_base for salt.pillar.git_pillar ?
15:21 babilen It's the one thing that I am missing now ...
15:22 johtso joined #salt
15:23 rostam joined #salt
15:23 dyim crap... that is so BS... I changed the quotes from single quote to double quote, per the JSON standard and it is working now
15:24 dyim lesson learned i guess
15:24 repl1cant joined #salt
15:25 naiyte joined #salt
15:28 alexj joined #salt
15:30 babilen dyim: Exactly that mistake just came up on the mailing list too. Don't feel too bad, it seems to be a common one :)
15:31 toastedpenguin joined #salt
15:31 babilen dyim: But I didn't see a single ' in your "i'm doing {{ map.pkgs|json }}" -- Would have helped to see more data :)
15:39 Katafalkas joined #salt
15:40 vlcn morning guys :)
15:40 dyim it was in the map.jinja, i didn't realize that file was actually taking in real JSON rather than python quotes
15:40 vlcn say I have a state that needs to make use of the contents of a file on the minion as a variable -- whats the best way I could go about making that work?
15:41 kaptk2 joined #salt
15:42 dyim is the map.jinja file actually a JSON file? should i be formatting this file to be compliant with JSON?
15:42 Mr_N vlcn: Use the pure python renderer, IMHO.
15:42 Ztyx joined #salt
15:42 Mr_N and essentially write a python script ><
15:42 vlcn ick
15:42 opapo joined #salt
15:43 Ztyx joined #salt
15:45 forrest joined #salt
15:45 blee joined #salt
15:45 penguin_dan joined #salt
15:46 Ztyx I'm trying to get gitfs to work together with a gitfs_root. Seems like only some files have been checked out (like top.sls). Running 17.5. Anyone with some input?
15:47 abe_music joined #salt
15:47 madduck so with the new versioning scheme, does this mean saltstack guarantees a new release every year from now on? Or will we find ourselves in a situation of using Salt 2020 in 2028, when Salt XP comes out, just before it's renamed Salt 7 and Salt 8?
15:47 madduck . o O (http://semver.org)
15:50 Nazca joined #salt
15:50 Nazca joined #salt
15:51 johtso What's the best way to clean up and detect unresponsive minions?
15:51 johtso I'm sure I saw something in the docs at somepoint
15:51 ndrei joined #salt
15:51 johtso I thought there was even salt-key command maybe..
15:53 johtso Ah, the FAQ is what I was looking for! google letting me down
15:53 Rasathus joined #salt
15:55 sfello joined #salt
15:56 Rasathus Hello, Im interested to understand how people are currently managing ntp on their linux machines via salt.  Are you simply managing the ntp.conf file with preset values, or have you written your own module to keep things in step.
15:56 madduck Rasathus: I always suggest to manage the whole file for maximum control
15:56 madduck for pretty much anything.
15:57 Mr_N madduck: I've recently become pretty enamored with regard to the block replace functionality.
15:57 Mr_N It makes configuring the DNS/DHCP records in the hilarious way they're set up at this job not an impossible task.
15:58 Rasathus madduck: it does make sense when you start to think how you would go about modifying an existing one with new server information.  For example the default ubuntu one has a number of comments that are associated with the server entries, so would look very untidy if just the server lines were removed and replaced.
15:58 Rasathus Mr_N: sorry, whats the block replace functionality you're reffering to ?
15:59 pass_by_value johtso: https://salt.readthedocs.org/en/v0.17.0/ref/wheel/all/salt.wheel.key.html#salt.wheel.key.list_all should show you all minions whose keys have been accepted. salt-run manage.present will show you the ones that are currently connected. You can salt-key -d to "cleanup" the difference of those two commands.
15:59 madduck Mr_N: sure, until you find that it doesn't work anymore and that your configs have diverged across the infrastructure
15:59 Mr_N madduck: Wait; is there something I should be worried about?  (I've only recently started playing with it)
15:59 johtso pass_by_value, cheers
15:59 Mr_N Rasathus: It's one of multiple mechanisms for modifying files, such as managed, append, etc.
16:00 madduck well, I can tell you that I've been playing with system automation for about 15 years now and for many years I tried to maintain my systems automatically in the same way a Debian sysadmin would do it manually
16:00 madduck and that ended up being *way* more work than time saved
16:00 madduck Now I just say: thanks Debian, but I'll manage those files by brute-force and then I know what I can expect
16:00 Mr_N My big goal at this point is to unify "deploy new machine into networking" a one step instead of multi multi step.
16:00 Gifflen_ joined #salt
16:00 Rasathus Mr_N: thanks, I think I've found it "salt.states.file.blockreplace"
16:00 madduck it's not as elegant, but then again, onceyou have automation working, you won't do things manually anymore
16:00 Mr_N Since we have a lot of really programmatically replacable steps, mostly along the lines of "ensure there exists this line in this file"
16:01 madduck if that is the sort of configuration you are looking for, then fine
16:01 madduck for me, I generally want to ensure a line exists, and no other line I don't want, and this, and that
16:01 madduck so by the time I told augeas or AppendUnlessPresent or whatever they are all called all the things I want
16:02 rgbkrk joined #salt
16:02 madduck I will have (a) overlooked some aspects that will come to haunt me later and (b) will have spent 20 times as much time on it
16:02 Mr_N So what; you'd just like, scrape ALL of your data together and serialize it to a managed file every time you made a modification?
16:02 madduck so instead, I just paste over the files and make sure that the files are generated from smart templates and the data comes from pillar which is smartly assembled (i.e. reclass)
16:02 Mr_N I can see it being quite sane in terms of not getting into broken states
16:02 Mr_N But it seems like a lot of overhead.
16:03 jalbretsen joined #salt
16:03 madduck forget the overhead. you are making a computer do it.
16:03 Iwirada left #salt
16:03 madduck it'll still be bored most of the time ;
16:03 madduck )
16:03 Mr_N I know I'm making a computer do it, but if we're managing a list of ~3000 hosts in DNS, and I'm building that from an ext_pillar, I want it to maintain some level of responsiveness.
16:04 madduck i don't think pillar assembly will be your problem and you need that anyway
16:04 Mr_N It'll certainly "get done" but in the current workflow, you modify the conf files, do a ./remake in NIS and restart dns/dhcp, and "everything is there."
16:04 madduck and templating should happen on the client, no?
16:04 thayne joined #salt
16:04 madduck Mr_N: parts of me say you should reboot after every change too ;)
16:05 Mr_N Sure.  But that's the thing; if I'm treating my DNS server as a salt minion, for which I generate the proper managed DNS files.
16:05 madduck else you might find it no longer working after the next power failure.
16:05 Mr_N so the work will be done there; it might just take a bit.
16:05 Mr_N And unfortunately, I can't reboot after changes, we run _way_ too much production shit.
16:05 madduck come on. 3000 entries? it won't take a second or two…
16:05 Mr_N madduck: Ok; maybe most of the salt latency I've seen thus far is just constant time overhead then
16:05 madduck i would reboot while you're there and memory of what you are doing is still fresh especially because of "production shit" ;)
16:06 johtso Do you have to restart anything after accepting minion keys?
16:06 madduck and if that means lost revenue, you need a better HA solution
16:06 madduck johtso: no
16:06 ahctpac joined #salt
16:06 Mr_N madduck: Revenue?  Hah.  No, this is academia.  Meaning we don't have the money for a better HA solution, but I still have researchers bitching to me if their DB machines go down.
16:06 Mr_N I pushed for 2 years JUST to get our core infrastructure on a single replicated master setup.
16:07 johtso madduck, hmm, having trouble talking to my minion after its IP address changed, so I accepted the new IP address, but it's still complaining that its key hasn't been accepted
16:07 dangra joined #salt
16:08 madduck Mr_N: academia is where BOFHs live. You gotta learn that! ;)
16:08 Mr_N It's funny that you mention doing a restart though, since restarting salt-minion is the only way I found to successfully refresh pillars...
16:08 Mr_N BOFH?
16:08 Mr_N Not an acronym I'm familiar with :x
16:08 johtso madduck, hah, just stumbled across this issue of yours https://github.com/saltstack/salt/issues/2358
16:08 madduck johtso: IP address changes are a nightmare with Salt. I think you need to dig deep and write a mail to the list with all details.
16:08 madduck Mr_N: seriously?
16:08 mpanetta Bas*ard Operator From Hell
16:08 Mr_N :x
16:08 Mr_N Ah.
16:09 madduck that's a legendary concept
16:09 johtso madduck, argh, do you know of any quick "get it working again" solutions?
16:09 madduck johtso: not really, and I am about to run out the door, sorry. :/
16:09 johtso np
16:09 Mr_N I stay in my little sysadmin hole and scream and shout about things like running 5 year old systems, only coming out when the power substation next door catches fire.
16:09 Mr_N keeping up with acronyms is too much to ask ><
16:09 Mr_N :P
16:10 mpanetta It is an old one.
16:10 madduck 5 year old systems. Someone should put you into an AIX-run university for a refresher of your standards.
16:10 mpanetta You should read the stories :P
16:10 forrest BOFH is a terrible concept
16:10 forrest reading those stories makes me cringe
16:10 mpanetta haha
16:10 Mr_N madduck: My problem isn't so much with the systems being old, but with the expectation that they should still act and be robust like new ones.
16:10 alunduil joined #salt
16:11 Mr_N Especially when my team is spending a good few man hours a week just swapping out the perpetually failing ancient drives and controllers.
16:12 Mr_N Luckily, we _FINALLY_ got a budget allocation to replace all the drives at once, so I'm counting my blessings.  (although they still think throwing new drives into a decaying cluster is a "good idea...")  /rant
16:12 linjan_ joined #salt
16:12 Mr_N I'm a FOFP.  a frustrated op from philly. :(
16:13 seapasulli_ joined #salt
16:13 funzo joined #salt
16:13 Ashu joined #salt
16:14 Gifflen joined #salt
16:14 frasergraham joined #salt
16:17 pdayton joined #salt
16:17 babilen Hooray! The combination of gitfs_base and a local git repository that I manually handle (rather than using a git ext_pillar) did the trick :)
16:18 babilen If only there were an equivalent of gitfs_base for git ext_pillars (will file a bug later)
16:18 johtso madduck: pebcak.. my minion ended up on a different subnet :(
16:19 frasergr_ joined #salt
16:20 lessismore left #salt
16:20 Ztyx1 joined #salt
16:22 dyim i'm working on salt-ssh, the highstate is returning { "local": {} }
16:22 dyim empty dictionary... do you guys have anything I can try?
16:25 fllr joined #salt
16:25 kaptk2 joined #salt
16:26 ipmb I'm struggling with the chicken/egg problem where I have a pillar that depends on stuff defined in another pillar. How have you been working around this?
16:26 ipmb Specifically, the "schedule" pillar needs to know what services are running on a given minion.
16:27 forrest basepi, you guys should start putting the toaster into the salt air videos like an easter egg, similar to the snail in adventure time.
16:35 baniir joined #salt
16:39 gnugnu8_ joined #salt
16:39 yomilk joined #salt
16:39 Ztyx1 left #salt
16:39 Gareth forrest: the toaster?
16:41 forrest Gareth, yea back during the salt-ssh video that was made you could see the toaster and toaster oven near Dave's desk in the background. Mark didn't really like that it was in the vid since it was kind of unprofessional, but I keep telling him it's funny, just makes it more start up esque.
16:41 forrest sad to say the latest vid did not contain my favorite actors
16:41 Gareth forrest: ahh.
16:42 Gareth didn't notice the toaster :) Will have to go back and look.
16:42 forrest yea the toaster and toaster oven are on the left right next to Dave's head in the video, you can't miss them. They're the stars of the show!
16:43 vegardx_ Don@t underestimate the toaster. I've made proper dinners in them.
16:43 forrest vegardx_, I'm not, that's why I had to tweet at them saying I was sad they were missing from the latest vid!
16:44 vegardx_ I'll support that! Fuck looking professional ;-)
16:45 njpatel joined #salt
16:47 xmj so many mentions of toaster and NetBSD nowhere in sight..
16:47 xmj what is this? some windows shop?
16:48 gnugnu8_ joined #salt
16:48 jdenning joined #salt
16:48 forrest xmj, I think we get fewer BSD related questions than windows
16:48 xmj forrest: cedwards and terminalmage do an incredible job documenting things.
16:48 jacksontj_ joined #salt
16:51 forrest xmj, yea I know, I'm just joking
16:51 terminalmage xmj: as always, patches are welcome :)
16:51 xmj forrest: the interesting thing is, you're right
16:52 xmj forrest: truth is most often the better joke
16:52 nes13 joined #salt
16:52 jeremyfelt joined #salt
16:52 xmj terminalmage: ill gladly help with the port, but i only ever get to salt things up on RHEL boxen
16:54 terminalmage cedwards manages the port. Has it been updated yet? He submitted the update last Friday or Thursday.
16:54 ml_1 joined #salt
16:56 xmj yes, no
16:56 xmj http://www.freebsd.org/cgi/query-pr.cgi?pr=ports/186950
16:58 blee joined #salt
17:02 Mr_N assuming ZOL doesn't make me hate my life, I hopefully won't ever have to worry about salt on BSD, or else I'd be more interested in this port ><
17:02 srage joined #salt
17:02 tomas joined #salt
17:03 fxhp 2014.1.0 is the latest
17:03 fxhp w00t
17:03 cedwards I submitted the FreeBSD patch for 2014.1.0 last weekend. It shouldn't be much longer.
17:03 cedwards I'll follow up on it now
17:04 Guest82171 Hi everyone! :) Question: Is there a way how to have a state that is not run in the highstate? I.e. I want to have a state that ensures that the maintenance mode is enabled, but I obviously do not want it to be a part of highstate
17:05 forrest Guest82171, just don't include it in your top.sls
17:05 Mr_N As long as it's reachable in some way from file_roots, you're fine.
17:05 R_Shackleford joined #salt
17:07 tomas1 joined #salt
17:07 xmj cedwards: depending on how lazy you are when it comes to submitting new patches, look into ports-mgmt/porttools
17:07 xmj cedwards: makes life easier :)
17:08 cedwards xmj: thanks
17:08 xmj yw
17:08 ipalreadytaken joined #salt
17:08 vejdmn Hey folks .. I see 2014.1.0 is not in http://ppa.launchpad.net.  Is there a timeframe following a release when it can be expected to make it to the ppa?
17:09 xmj joehh: hah, do yourself a favor and put the estimated dates into the Topic
17:10 meteorfox joined #salt
17:10 vejdmn ppa
17:12 gildegoma joined #salt
17:13 budrose joined #salt
17:17 tomas1 joined #salt
17:17 joehillen joined #salt
17:17 blee joined #salt
17:19 saurabhs joined #salt
17:19 nkuttler joined #salt
17:20 Linz joined #salt
17:22 KyleG joined #salt
17:22 KyleG joined #salt
17:22 Linz joined #salt
17:24 smcquay joined #salt
17:27 sfmoe joined #salt
17:33 frasergraham joined #salt
17:34 kedo40 joined #salt
17:36 danielbachhuber- joined #salt
17:40 abele_ joined #salt
17:41 lahwran_ joined #salt
17:48 srage joined #salt
17:48 harobed joined #salt
17:48 sroegner joined #salt
17:48 MindDrive joined #salt
17:49 ckao joined #salt
17:50 codysoyland joined #salt
17:50 jeffro joined #salt
17:50 racooper joined #salt
17:50 jcsp joined #salt
17:50 meteorfox joined #salt
17:50 etw joined #salt
17:50 GradysGhost joined #salt
17:50 jmccree joined #salt
17:50 vejdmn joined #salt
17:50 Katafalk_ joined #salt
17:50 jaimed joined #salt
17:50 KyleG joined #salt
17:50 abe_music joined #salt
17:50 R_Shackleford joined #salt
17:50 ipalreadytaken joined #salt
17:50 bhosmer joined #salt
17:50 saurabhs joined #salt
17:50 Gifflen joined #salt
17:50 funzo joined #salt
17:50 pass_by_value joined #salt
17:50 Gifflen joined #salt
17:50 R_Shackleford joined #salt
17:50 meteorfox joined #salt
17:50 faldridge joined #salt
17:50 gasbakid joined #salt
17:51 dN0t joined #salt
17:51 njpatel joined #salt
17:51 KyleG joined #salt
17:51 jeffro joined #salt
17:51 Nazca joined #salt
17:51 Nazca joined #salt
17:52 kaictl joined #salt
17:52 kossy joined #salt
17:54 kballou joined #salt
17:55 scalability-junk joined #salt
17:55 meteorfo_ joined #salt
17:59 djaime joined #salt
17:59 schimmy joined #salt
17:59 Doormatt joined #salt
18:01 Doormatt I just upgraded my minions and master to 2014.1.0, and now every command I send to the minions results in a zombie minion process.
18:01 Doormatt Reverting to 0.17.5 solves it.
18:01 Doormatt I see many issues open (and closed) around this zombie issue - is it worthwhile opening up a new one?
18:02 slav0nic joined #salt
18:04 kaictl joined #salt
18:04 xet7 joined #salt
18:04 Katafalkas joined #salt
18:05 forrest Doormatt, depends, if it is the same issue as one of the existing ones, re-open that
18:05 forrest if it's new, open a new one
18:05 racooper joined #salt
18:05 Doormatt Seems to be new - it's hard to say.
18:05 johtso joined #salt
18:05 forrest then collect as much information as you can and create a new one
18:05 jeffro joined #salt
18:05 Doormatt Will do!  Just was unsure of proper protocol at this point.
18:06 bhosmer joined #salt
18:06 miguitas left #salt
18:06 srw_ joined #salt
18:06 forrest Doormatt, for sure, getting the issue out there with details is more important than not creating it because of uncertainty! :D
18:08 zzzirk joined #salt
18:08 srw_ is there a simple way to pass vars between states?  I have one state that creates a VM with an autogenerated macaddress, in the next state I want to add the VM and mac address into inventory
18:08 Doormatt srw_: have you tried pillars?
18:09 Doormatt or grains?
18:09 pass_by_value joined #salt
18:09 srw_ i dont know what the mac address is at run time
18:09 srw_ so i cant define a static value
18:09 Doormatt grains will pull it dynamically from the running machine.
18:10 srw_ I thought that the state.sls are rendered templates?
18:10 Mr_N They are.
18:10 Doormatt You lost me there.
18:10 srw_ im new sorry, ...
18:10 Mr_N Pretty much everything in salt can be treated as a rendered template
18:10 Mr_N including source files
18:10 Doormatt S'all good!  We've all been there at one point or another :)
18:10 Mr_N this does not impact which can be used for what, really.
18:10 saurabhs joined #salt
18:11 Mr_N Doormatt: Some of us are still there :P
18:11 Mr_N (despite our best efforts ><)
18:11 Doormatt Pfft - I consider myself to know next to nothing about Salt.  The fact I have it working I attribute mostly to luck.
18:11 srw_ ill post an esample
18:11 jmccree joined #salt
18:11 Mr_N Yeah; I'm with you on that...
18:11 KyleG joined #salt
18:11 KyleG joined #salt
18:12 penguin_dan joined #salt
18:13 srw_ esample of what im trying to do
18:13 srw_ http://pastebin.com/qRP5zp48
18:14 srw_ state: esx.present: creates a VM and autogenerates a MAC
18:14 srw_ state: infoblox.present im trying to get the MAC address from the new VM
18:15 kossy joined #salt
18:15 srw_ I could hack around this issue creating a new state file that does both esx and infoblox... but dont like that idea
18:16 srw_ im sure this is a common issue that has a nice solution
18:16 __alex joined #salt
18:16 rojem joined #salt
18:16 srw_ im just not seeing it in the docs
18:16 sinh joined #salt
18:17 hardwire so can reactors be handled by minions as well as masters?
18:19 fllr joined #salt
18:19 hardwire I'd like to get to a point where I'm mounting an encrypted volume before spinning up some services..  so I'd like the minion to attempt to mount (i.e. wait) until the encrypted volume is mounted before doing a highstate.
18:19 manfred i believe right now the reactor is only on the master
18:19 hardwire I understand this can all be done during boot as well.. it's just more difficult to handle that way.
18:19 hardwire ok
18:20 manfred i believe the events were in there for a while before the reactor was actually written
18:20 manfred but you could write anything you wanted to key off of the reactor
18:20 hardwire and if I had a bunch of services in the disabled state that are marked as to be running.. only a highstate should bring them up at that point.. it's not automatic for a salt minion to see if things are running without a highstate command.
18:20 manfred or that event feed
18:20 bemehow joined #salt
18:21 manfred you wouldn't have to use a highstate, you could just use a regular state ?
18:21 manfred service.running:
18:21 manfred names:
18:21 hardwire see.. now I need to learn the difference :)
18:21 manfred - httpd
18:21 manfred so a highstate keys off of the top.sls
18:21 rgbkrk joined #salt
18:21 srw_ does any salt wizards have any thoughts on my issue?  or am i trying to do something outside the scope of salt support?
18:21 manfred to figure out what states to apply to the minions
18:21 hardwire ah ok
18:21 manfred but you could just salt '*' state.sls services
18:22 hardwire thanks
18:22 manfred and have a services.sls that just checks that all of your services are running
18:22 Doormatt srw_: So, you generate the mac address in one state, and need to use it in another, correct?
18:22 manfred hardwire: you coudl even do that from the minion with salt-call
18:22 hardwire when the minion runs does it attempt to achieve state all on its own?
18:22 srw_ dormatt yes
18:22 manfred hardwire: if you tell it to do state.sls services, then it will make sure it is in that state
18:23 babilen Is there an equivalent to gitfs_base for a git ext_pillar ?
18:23 R_Shackleford joined #salt
18:23 hardwire ah.  in my case I'm trying to keep redis from running via init and have the minion attempt to deal with that.  I'm just not sure if when the salt minion daemon is started if it will attempt to run it without explicitely being told to try to resolve the state.
18:24 hardwire I'll test it out.
18:24 hardwire I have a looong list of tests I need to do.
18:24 manfred hardwire: you could have this
18:24 srw_ doormatt: yes
18:24 manfred startup_states: sls.services
18:24 srw_ Doormatt: yes
18:24 Gareth morning
18:24 Doormatt srw_: I think you're going to find that you'll have to use pillars to accomplish this.
18:24 manfred hardwire: in your /etc/salt/minion
18:24 manfred hardwire: and then it would run that state to make sure that redis is running
18:24 hardwire manfred: ah ok.  That's one thing the docs are sparse on.. where things go (into what files)
18:24 Doormatt srw_: Then use a rendered template on the second state which references the pillar data that your previous state created.
18:25 manfred hardwire: http://docs.saltstack.com/ref/states/startup.html
18:25 hardwire thanks manfred
18:25 manfred np
18:25 Gareth manfred: Did you see my latest comments re: the help function in loader.py?
18:25 ipmb joined #salt
18:25 srw_ Doormatt: but would that not render a blank macaddress on the first run
18:26 srw_ and require a second run
18:26 hardwire are you mostly arch or are you familiar with debian.  I've had a fun time setting up policies using policy-rc.d to deal with not automatically starting installed packages that have init scripts.
18:26 manfred Gareth: nah, i have been working today, but as i mentioned in that, i am off now :)
18:26 manfred hardwire: /whois manfred
18:26 hardwire yeh I saw that.
18:26 manfred i mostly have been using ubuntu and fedora
18:26 mgw joined #salt
18:26 hardwire [root@archlinux/trusteduser/gtmanfred]
18:26 abe_music joined #salt
18:26 manfred si
18:27 Doormatt srw_: Well, your second state would depend on the first state, so it wouldn't run until there was a mac address ready, and then you could use saltutil.refresh_pillar to make sure the new pillar data exists on the minion.
18:27 manfred and no, i am not actually running irc as root
18:27 hardwire yes you are
18:27 hardwire nobody can verify you are fibbing about it
18:27 manfred hardwire: /ctcp manfred os
18:27 manfred or version
18:27 manfred i also think i still have finger set
18:27 hardwire heh'
18:28 manfred yup that one is too
18:28 hardwire ctcp version me
18:28 hardwire with znc you'll get back all of your connected clients info
18:28 manfred non interesting
18:28 manfred hardwire: i use znc, and then just made them weird to make people angry
18:29 srw_ Doormatt: sure but I would have to call highstate twice...
18:29 Gareth manfred: okay cool.  I managed to get it to be available but it doesn't like the call to sys.doc.  And fails when called.
18:29 Doormatt srw_: Why?  First state runs, and creates mac address - stores in pillar.  Second state runs, updates pillar data to get the MAC address, then runs.
18:29 manfred Gareth: kk, i will look
18:30 manfred Gareth: can you push it to github on a seperate branch?
18:30 srw_ Doormatt: Ok, so the pillar data is no rendered like I have in my example?
18:30 ashb wow salt.loader is *complex*
18:31 Doormatt srw_: Your example confuses me a little - it doesn't look like jinga templating to me - are you using another templating engine?
18:31 srw_ Doormatt: im refruing to {{ salt['esx.get_macaddress'](test2) }}
18:31 jforest joined #salt
18:31 17SAAMJQX joined #salt
18:31 racooper joined #salt
18:31 smcquay joined #salt
18:31 MTecknology joined #salt
18:31 alunduil joined #salt
18:31 repl1cant joined #salt
18:31 aleszoul3k joined #salt
18:31 rhand joined #salt
18:31 dren_ joined #salt
18:31 twiedenbein joined #salt
18:31 zloidemon joined #salt
18:32 hardwire how ephemeral is pillar data?
18:32 Doormatt srw_: Ahh - I think pastebin mangled some of the formatting.
18:32 hardwire like all memory?
18:32 srw_ Doormatt: na i think it was my issue
18:32 Doormatt hardwire: Could you ask that a different way?
18:32 srw_ Doormatt: was missing the {{  }}
18:32 Doormatt srw_: That'll do it!
18:32 srw_ Doormatt: but you see my issue?
18:32 racooper joined #salt
18:33 diegows joined #salt
18:33 Doormatt I still think you just need to split this into two states, and have them use pillar to pass the information back and forth.
18:33 Doormatt The first state would be "static", and the second state would be rendered to include the new MAC.
18:33 Gareth manfred: yeah.  give me a few minutes.
18:33 manfred kk
18:33 srw_ if i have them in two states the data dont get rendered all in one go?
18:34 gasbakid joined #salt
18:34 Doormatt srw_: AFAIK, each state gets rendered just before it's applied.
18:34 srw_ SWEET thanks
18:35 hardwire Doormatt: is pillar data by default stored as a pickle or something on disk or is it stored in memory in the salt-minion process
18:36 manfred hardwire: pretty sure it is stored in memory, but it should be requesting the information from pillar.get, or you can refresh with saltutils.refresh_pillar
18:36 manfred i don't see pillar data in /var/cache/salt/minion
18:37 druonysus joined #salt
18:37 Doormatt Looking at the code, it depends on which pillar file driver you're using.
18:37 hardwire ah good call
18:38 Doormatt Looks like it's stored as an encrypted dictionary.
18:38 hardwire to which the decryption key is on disk .. i'm not sure if the master has to help unlock it.
18:38 elfixit joined #salt
18:39 hardwire it's a safe.. but with the key taped to the side.
18:39 Doormatt It's easier that that - the key is "pillar".
18:39 Doormatt ret_pillar = self.sreq.crypted_transfer_decode_dictentry(load, dictkey='pillar', tries=3, timeout=7200)
18:39 * hardwire giggles.
18:39 manfred lol
18:41 vlcn can anyone think of a way that cmd.run would incorrectly interpret a return code?
18:41 hardwire I'm really curious if it goes further than that as far as ephemeral unlock keys via negotiated pairs or something.
18:41 vlcn when I run a command locally via the cli the return code is 0, but salt reports a return code of 1
18:41 Doormatt vlcn: It is possible it's due to an environment variable or something?
18:41 Doormatt vlcn: As it's running in a different shell?
18:42 vlcn hmm, possibly.
18:42 ipalreadytaken joined #salt
18:43 Corey I need to figure out what the saltstack/pagerduty integration looks like.
18:43 Corey "Hey I just woke you, and this is crazy. But your infrastructure ate it. So log in maybe?"
18:44 hardwire vlcn: I'm not entirely sure what happens if the process died an unnatural death
18:44 Mr_N Hm.  pkg says that it's not finding salt-minion in the repository, so clearly I'm doing some win_repo stuff wrong
18:44 Mr_N Anyone use a win_repo whose brain I could pick?
18:45 hardwire Corey: ooh neat
18:45 vlcn Doormatt, that was exactly it.  thanks.
18:45 hardwire I've been using twilio and some custom apps for that
18:45 anuvrat joined #salt
18:45 hardwire and.. email (to sms addresses)
18:45 Doormatt vlcn: No worries - that one's bitten me a lot lately.
18:48 FarrisG Is there a decent method for managing standalone daemons/services that don't have system packages? I can manually create an init/upstart script, but how would I manage that via state?
18:49 manfred FarrisG: file.managed ?
18:50 vlcn okay, heres another... Is there any way I can use cmd.wait and watch without specifying a specific state?
18:50 manfred Gareth: any news?
18:52 krow joined #salt
18:52 Mr_N ls
18:52 Mr_N Asdfwrongterminal.
18:52 FarrisG manfred: Just use file.managed to make sure all the executables (including init) are in place? After that, can service.running be used?
18:53 Gareth manfred: sorry.  $dayjob stuff.  pushing now.
18:53 manfred FarrisG: that i do not know.
18:53 manfred Gareth: cool
18:54 Mr_N Anyone here familiar with the windows package repository before I run off to lunch?
18:55 Gareth manfred: https://github.com/garethgreenaway/salt/commit/43f0e0b103282f0c1799a8538390e8eb1c333865
18:55 blee joined #salt
18:56 Doormatt vlcn: I don't think so...what are you trying to accomplish?
18:56 manfred ahh cool
18:56 manfred gimme a bit
18:57 vlcn Doormatt, I have about 500 states that are generated within a jinja loop -- only want to restart a service if any of them change/are new
18:57 vlcn as an alternative to restarting the service every time
18:57 manfred i wish github wasn't show dumb and had an easy way for me to copy a raw diff of commits
18:57 vlcn manfred, you can
18:57 manfred vlcn: how
18:57 Doormatt vlcn: Forgive me for being dense, but why can't you just watch the states?
18:57 manfred vlcn: how do i find it from this https://github.com/garethgreenaway/salt/commit/43f0e0b103282f0c1799a8538390e8eb1c333865
18:58 vlcn https://github.com/garethgreenaway/salt/commit/43f0e0b103282f0c1799a8538390e8eb1c333865.patch
18:58 manfred nice
18:58 manfred is there a button for that or do you just have to know to add patch
18:58 manfred ?
18:58 vlcn Doormatt, because they are generated within the loop.  There certainly are a few ways I can work around it
18:59 vlcn https://github.com/kelchm/salt-examples/blob/master/centreon-clapi/add-nodegroup.sls
18:59 Doormatt vlcn: Couldn't you just use the same loop to generate the entries for the watch statement?
18:59 vlcn this is a simplified version of what I am working with
18:59 vlcn Doormatt, I suppose I can do that but things are starting to get a bit messy.
19:00 Doormatt vlcn: What's the larger goal here?
19:00 Doormatt vlcn: I guess I'm not seeing why every state has to be unique.
19:01 Doormatt vlcn: For example, why use "{{ hostname }}-nagios-add" instead of just "nagios-add"
19:01 vlcn Doormatt, don't they have to be unique?
19:02 Doormatt Yes - but the contents will change as they're rendered.
19:02 vlcn hmm, as I recall that wasn't my experience
19:02 Doormatt You can use the same state ID, and each minion will render the SLS file for themselves.
19:02 Ryan_Lane joined #salt
19:03 vlcn Doormatt, ah -- thats not how this is used
19:03 vlcn this is ran on a single minion
19:03 Doormatt Ah.  Boourns.
19:03 Mr_N No one?  win_repo?
19:03 vlcn notice I'm using publish.publish
19:04 Doormatt vlcn: My apologies - I've never used salt that way.
19:04 xerxas joined #salt
19:04 vlcn Doormatt, no worries at all.  It's a more unusual way to use it but it works nicely
19:04 Doormatt vlcn: I think I'm starting to see your dilemma.
19:05 Doormatt vlcn: So, ideally - what would the watch trigger on?
19:05 Ryan_Lane1 joined #salt
19:07 n8n joined #salt
19:08 vlcn Doormatt, basically any time salt detects that any one of those generated states has changed
19:08 vlcn so yeah, I could do another loop
19:08 vlcn its just messy
19:08 vlcn if you want to get an idea of what my production version looks like, here is an updated copy: https://github.com/kelchm/salt-examples/blob/master/centreon-clapi/add-nodegroup.sls
19:09 Doormatt vlcn: Off the top of my head, could you use a text file with file.managed, and simply change its contents, and then have the other state watch the file?
19:09 Doormatt Not pretty, I agree.
19:10 Doormatt vlcn: Or, could you store all the information in a pillar, and use that to query...
19:10 MTecknology joined #salt
19:10 Mr_N Man.  I need to file a whole fucking PILE of issues on documentation being quite lacking...
19:11 Doormatt Mr_N: Frustrating isn't it?
19:11 * vlcn nods
19:11 Mr_N I've been stuck on this win_repo issue for a day.  Compared to how long I sunk into getting multimastering stuff to work, and realizing that pillars were actually LESS dynamic than salt...
19:11 Mr_N grains*
19:11 Mr_N The frustration is only just beginning.
19:12 Mr_N (the strange part is, it's still FAR more intuitive than most of the other CMS systems out there...)
19:12 vlcn thats the truth
19:12 Mr_N I
19:13 Mr_N I've foud that if I get really really stuck, the best thing to do is idle here, occasionally poke people, and try to come up with a workaround.
19:13 Mr_N Salt is usually robust enough to have a million ways of doing any one thing (which is something I'm not sure is good or bad...)
19:13 vlcn this IRC channel is exceptionally good
19:13 Doormatt Mr_N: Which I find funny, since that "million ways of doing something" is completely opposite to how Python programming works ;)
19:13 Mr_N Certainly much more tolerant of my constant stream of "x isn't working as I would expect"
19:14 Mr_N ><
19:14 vlcn and Salt gets enough things right that I can tolerate lacking documentation
19:14 Mr_N Doormatt: As a primarily python guy, yeah, it was quite the 180 :P
19:14 Mr_N "do one thing well" HAH
19:14 Doormatt Mr_N: Seems like we're in the same boat then.
19:14 vlcn I'm really, truly awful at Python still.
19:14 Mr_N vlcn: Documentation will come, salt is "relatively young."  just means I have to put up with more of the headache.
19:14 Mr_N And to be fair, despite how old it is, puppet docs are often quite... not fun.
19:14 Doormatt vlcn: I don't think I'll ever consider myself "proficient" at anything.
19:14 Mr_N old/mature*
19:14 vlcn I just realized that indentation matters in python about 15 minutes ago ;)
19:15 Mr_N vlcn: And then you started writing jinja, which throws all that out the window.
19:15 Mr_N ENJOY.
19:15 Doormatt vlcn: Ha!  That'll make life easier.
19:15 Mr_N Doormatt: Something like that ><  The more I learn about something the more I know I don't know.
19:15 vlcn Mr_N, hey I feel like I have an 'okay' handle on jinja at this point
19:15 Mr_N vlcn: I only like a week ago fully wrapped my head around needing to do "set" and "do" in certain situations
19:16 vlcn yeah, I hear you.  It took me a lot of banging my head on the keyboard to get a basic functional understanding
19:17 Mr_N But that's the thing.  I see multimastering as windows repos as "basic functionality", so the frustration hits full force when even with the documentation I'm hitting a wall
19:17 Mr_N at least with Jinja there's more mature documentation
19:17 Mr_N and more people are able to help you
19:17 Mr_N and windows repos*
19:17 Doormatt Mr_N: While I'm not using win_repo at all, what's the problem you're running into?
19:18 Mr_N Doormatt: Very simple; pkg claims it can't find any of the packages I tell it to instal in the repository.
19:18 Mr_N I configure the repository EXACTLY as in documentation, utilize repo sls file from the official saltstack git
19:18 Doormatt Mr_N: Can you share your config?
19:18 Mr_N and despite trying every permutation on the package name I could think of, I get package "x" not found.
19:18 Mr_N Sure, I can throw the relevent parts up in a pastebin.  One sec.
19:19 Doormatt Mr_N: And I'm assuming you've ruled out permissions?
19:19 Mr_N Also; for the record; running pkg.install raw from the command prompt generates NO errors whatsoever, but calling a state that calls the pkg.install does
19:19 Mr_N and yes, everything is root.
19:19 JordanRinke Morning
19:20 vlcn anyone know where I can find more examples of using the python renderer?
19:20 Doormatt vlcn: I'm assuming you found this: http://docs.saltstack.com/ref/renderers/all/salt.renderers.py.html
19:21 vlcn indeed
19:21 Doormatt vlcn: Then I'm useless :(
19:21 manfred Gareth: i got it to kind of work, but you have to loop through all modules instead of just doing one overall alias.
19:21 manfred Gareth: cause it loads everything up at the start of the command, instead of when that module is started running
19:21 Doormatt vlcn: It looks like it's just using dictionaries - anything specific you need it to do?
19:22 Gareth manfred: hmm.
19:23 manfred i do have another idea
19:23 Gareth manfred: got a code example?
19:23 manfred but i don't know how this is going to work
19:23 manfred Gareth: i do not right this second, i am just editing it on the fly on a salt minion
19:23 hunter joined #salt
19:24 Mr_N Doormatt: This should be enough info http://pastebin.com/3XrCEvk9
19:24 Mr_N to see where I'm at.
19:24 Mr_N Let me know if I could be more clear, and thanks for taking a peek :)
19:24 vlcn Doormatt, just trying to get a better feel for it without spending all day on it ;)
19:24 ipalreadytaken joined #salt
19:25 Doormatt Mr_N: What have you got for your minion config?
19:25 Gareth manfred: got it.
19:25 manfred you did?
19:25 Mr_N It's an exceedingly vanilla windows config.  I _only_ set the master and minion names in the installer.
19:25 Mr_N completely untouched other than that.
19:25 krow joined #salt
19:26 Mr_N (by the way, running all of the above calls in -ldebug gives no more useful information)
19:26 Doormatt Mr_N: That may be the issue - from (https://github.com/saltstack/salt/blob/b08d5916e3283fab99d89cb19a3c4434ac6f8b50/salt/modules/win_repo.py)
19:26 Doormatt The following options must be set in the Minion config:
19:26 Doormatt file_client: local
19:26 Doormatt win_repo_cachefile: c:\salt\file_roots\winrepo\winrepo.p
19:26 Doormatt win_repo: c:\salt\file_roots\winrepo
19:27 manfred Gareth: it is also adding the module and exectuing the whole _help_fun from the environment inside of loader.py, so certain things don't exist
19:27 Gareth manfred: nod. thats what I was seeing too.
19:28 Mr_N Doormatt: Interesting.  That's literally nowhere in the official windows-package-manager documentation, AFAIK
19:28 manfred i wonder,
19:28 manfred Gareth: one second
19:28 Mr_N Thanks for the pointer, I'll try tweaking that.
19:30 Mr_N Where typically is the minion file on win... >_>
19:31 Mr_N Ohey, right under C
19:32 Mr_N and here I was digging through program files
19:32 hardwire oh oh.. I got to use salt (masterliss) on the trueability linuxshowdown 6
19:32 hardwire and I need to submit a bug report on salt-bootstrap now :)
19:33 Mr_N Doormatt: AH
19:33 Mr_N That's only for a standalone minion
19:33 Mr_N it wouldn't make any sense on a mastered minion
19:33 Mr_N I was like "I don't have that filepath..."
19:34 Mr_N Ok; I'm going to a very very late lunch.  I've already put off way too much to keep banging on tihs...
19:34 Mr_N hopefully if I keep an eye on the room someone who has dug through this already will appear ><
19:34 Mr_N Thanks again for your help Doormatt
19:35 zain joined #salt
19:36 lkong joined #salt
19:36 dyim does anyone know how the grains work with salt-ssh? It seems to be missing the target OS.
19:36 lkong morning guys
19:38 bhosmer joined #salt
19:39 ml_1 joined #salt
19:39 db joined #salt
19:40 manfred Gareth: i don't see a good way to make this work, sorry
19:41 timoguin joined #salt
19:41 vlcn would anyone be interested in helping me with making an equivalent of this for the python renderer? https://gist.github.com/kelchm/9229997
19:42 Doormatt vlcn: Let me take a shot at it :)
19:44 Gareth manfred: its okay :)  thatch wasn't totally for it anyway.  The sys.doc method does work, I just thought it would be nice to have .help available.
19:45 ajw0100 joined #salt
19:47 Doormatt vlcn: How's this:  http://pastebin.com/L7sPp9tV
19:48 vlcn ah, forgot you could chain the renderers
19:48 vlcn nice!
19:48 Doormatt vlcn: Yeah, that was a nice suprise that you could do that.
19:51 dave_den Doormatt: i gave you the wrong info about mod_watch the other day. hope you didn't spend tons of time on it
19:51 dave_den sorry bout tht
19:52 Doormatt dave_den: No worries!  How did you mislead me?
19:53 dave_den i believe you asked if a module must implement a mod_wtch() in order for other states to be able to watch it. the answer is no
19:53 dave_den you were trying to watch a pip state
19:54 rlarkin joined #salt
19:54 Doormatt dave_den: Ah - well that DOES change things then!  How does watch work if a module hasn't implemented mod_watch()?
19:54 hardwire 01:25 <joehh> hardwire: uploaded to unstable (NEW queue) now
19:54 hardwire 01:25 <joehh> backporting and uploading to debian.saltstack.com now
19:54 bdf do 0.17.5 minions work with 2014.1.0 masters?
19:54 hardwire yay
19:54 Doormatt bdf: Yup!
19:55 dyim hi, is it me or does grains not work with salt-ssh?
19:55 rlarkin Is 2014.1.0 actually released now?
19:55 dyim salt-ssh "*" grains.items returns values
19:55 Doormatt rlarkin: Yup!
19:56 dave_den Doormatt: the official docs are here http://docs.saltstack.com/ref/states/ordering.html#watch-and-the-mod-watch-function
19:56 rlarkin Nice.  I've been just checking the /debian folder once in a while and so didn't notice.
19:56 Doormatt dave_den: doesn't this line: The watch requisite only works if the state that is watching has a mod_watch function written. If watch is set on a state that does not have a mod_watch function (like pkg), then the listed states will behave only as if they were under a require statement." disagree with you though?
19:57 bemehow_ joined #salt
19:57 aleszoul3k joined #salt
19:57 rlarkin Now I have to balance my desire to update *right now* with my desire to only use .deb's.  Life is so hard.
19:58 rlarkin virtualization to the rescue
20:00 frasergr_ joined #salt
20:00 dave_den Doormatt: the mod_watch function just allows the state to implement different functionality when a watched state has changes. for example, if the service.running state is watching a configuration file, then if the configuration file changes then mod_watch is called within the service state. this allows the service state to reload the service whereas if no mod_watch was implemented in service, then it would only make sure the service was r
20:01 dave_den i think you were thinking mod_watch was supposed to be implemented in the state being watch (in this example, the file module)
20:01 linjan_ joined #salt
20:01 ajw0100 joined #salt
20:01 Doormatt dave_den: Not trying to argue with you, but IMHO, I think you've got the mod_watch backwards - if file is watching pip, then pip needs to implement mod_watch, not file, yes/no?
20:01 dave_den right.
20:01 frasergr_ joined #salt
20:01 dave_den err, no
20:02 Doormatt dave_den: You're saying the watching state needs mod_watch(), not the watched state?
20:02 dave_den right. it doesn't make sense for file to have a mod_watch, so in the absence of a mod_watch in file, it just acts like a regular include
20:02 Doormatt dave_den: Gah - you're completely right.  How did I repeatedly miss that.
20:03 Doormatt I figured the watched state needed it, so it could determine if anything had changed, and relay that info back to the watching state.
20:03 Doormatt instead, it's more so that the watching state knows what to do when/if the watched state changes.
20:03 dave_den mod_watch is to make the *watching* state do something different than it usually would if it were executed without watching anything
20:04 naiyte joined #salt
20:05 Doormatt dave_den: Thanks again for being so patient!
20:05 dave_den yeah no problem. helps me understand salt better, too
20:06 toastedpenguin joined #salt
20:14 farra joined #salt
20:14 saint-n does salt take exclusions on the cli?
20:14 Doormatt saint-n: Not 100% sure what you're asking.
20:15 saint-n like salt '*dev*' AND NOT 'test.dev' cmd.run
20:15 saint-n host exclusions sorry
20:15 saint-n sorry for the splunk syntax ;)
20:15 Doormatt saint-n: Yup - http://docs.saltstack.com/topics/targeting/compound.html
20:15 saint-n excellent thanks
20:16 Doormatt saint-n: Anytime!
20:16 schimmy joined #salt
20:17 meteorfox joined #salt
20:18 meteorf__ joined #salt
20:20 dyim i think i have a weird problem. my sls file is not being rendered by jinja but my map.jinja works. as a workaround, i'm passing the grain information from the map.jinja to the sls file
20:21 Doormatt dyim: Firstly, why isn't your file being rendered by jinja?
20:22 dyim that's what i want to know. :) i have been bashing my head into the wall and finally was like "let me see if i can pass it through the *.jinja file" and that worked
20:23 dyim Doormatt: i wasted so much time on this thinking it was my syntax. I'm new to salt and would love to implement it in my enviornment here
20:23 Doormatt dyim: Have you tried putting #!jinja at the top of the sls file?
20:23 dyim Doormatt: Let me try that now
20:24 dyim Doormat: why would the "from map...." line work? maybe it is just grains are not being translated in my sls files
20:26 dyim Doormatt: i that that back about jinja not working. It is only accessing grains that seems to be the problem.
20:27 dyim Doormatt: is there a reason why this would be? I'm trying to get stuff like grains['os'] and it isn't doing anything or returning the wrong value and therefore entering the wrong if clause
20:27 Doormatt dyim: Can you post your config to pastebin?
20:32 dyim Doormatt: Sure, here you go http://pastebin.com/1YkJ5KSe
20:34 racooper howdy. Is there a way, in a state file, to conditionally include another state file based on grains?
20:35 saurabhs joined #salt
20:37 johtso joined #salt
20:38 ipmb any tips on how to dynamically build the schedule pillar? Basically, I need pillar data to determine the schedule in the pillar
20:39 srw_ inside a state.sls is there a template variable for 'name'
20:39 ipmb right now I'm setting grains, but it takes 2 highstates to get in the right state (first to set the grain, second to set the schedule)
20:42 TomasNunez joined #salt
20:42 TomasNunez Hi! I'm having a weird error...
20:43 TomasNunez I just rebuilt a server, and when trying to apply state.highstate, I get "       Comment:   No Top file or external nodes data matches found
20:43 TomasNunez "
20:43 TomasNunez but it's working with the rest of servers...
20:43 Doormatt dyim: Can't see a problem - sorry :(
20:44 TomasNunez and there is nothing either in minion log or master log
20:44 TomasNunez I'm using 0.17.5 from epel-testing
20:44 Doormatt racooper: Yup - you'd use the template engine to process the state file before it's sent to the client.
20:44 Doormatt http://docs.saltstack.com/ref/renderers/index.html
20:45 n8n joined #salt
20:46 gasbakid joined #salt
20:47 mattmtl is it possible to import a custom module in a custom grain?
20:47 frasergraham joined #salt
20:48 racooper Doormatt,  so something like this won't work? (well, obviously, it's not , but I can't find any examples like what I want to do yet...) https://gist.github.com/racooper/9279589
20:50 Doormatt racooper: You want this - http://docs.saltstack.com/topics/tutorials/states_pt3.html (3.3.4.2 - Using Grains in SLS modules)
20:50 QuantumRiff joined #salt
20:50 QuantumRiff i'm looking at an example of a salt file online: http://serverfault.com/questions/524201/how-do-you-execute-a-cmd-script-once-the-pre-requisite-files-have-been-downloade  and I'm curious, will this script execute every time state.highstate is run?
20:51 oraqol joined #salt
20:51 Doormatt QuantumRiff: Yes.
20:51 fllr joined #salt
20:51 Doormatt As long as it's assigned to a minion via top.sls or some other method.
20:51 gnugnu8 joined #salt
20:51 oraqol Hi guys, this may be a stupid question, but here goes: is there anyway to tell whether a state has been successfully applied to minions?
20:52 dave_den mattmtl: no, i don't think so
20:52 oraqol without having to actually reach out and check the minion manually?
20:52 Doormatt oraqol: Check the response of the job?
20:52 gasbakid joined #salt
20:53 oraqol Doormatt: responses come back when I run highstate in batch mode, but not when I run it with '*'
20:53 mattmtl dave_den: is there a reason?
20:53 oraqol just returns to newline, but some of the minions have the state applied so I know the jid hit a number of them, however some of them dont get it applied
20:53 Doormatt oraqol: You can handle the failure at the minion level - have a state that runs last and checks the other states?
20:55 dave_den mattmtl: IIRC, it's a chicken/egg thing. a lot of modules use grains info, so if salt tried to load modules before grains things would blow up.
20:55 dave_den mattmtl: see here: https://github.com/saltstack/salt/blob/develop/salt/grains/core.py#L33
20:55 mgw joined #salt
20:56 oraqol Doormatt: I just touch a file in tmp to verify that jid ran on minion, but I have to verify that file exists manually, which is a pain.  Are the returns from the minions saved somewhere on the master, even if the highstate command doesnt return anything to CLI?
20:56 Ryan_Lane joined #salt
20:56 dave_den mattmtl: and if you look a fiew lines down, you will see how that is worked around in the core grains
20:56 Ryan_Lane joined #salt
20:56 synfin Guys, noob question here.  I have several environments, defined by file_roots.  How do I tell salt to use an environment other than 'base'?
20:56 oraqol Also, why is that happening?  Why is running global highstate not returning anything to screen even if the jids were applied?
20:56 Doormatt synfin: http://docs.saltstack.com/ref/states/top.html
20:57 Doormatt oraqol: Good questions - you might try messing around with this: http://docs.saltstack.com/ref/output/all/salt.output.highstate.html
20:57 dave_den oraqol: because by default the salt master cli only waits a few seconds for return data before exiting. increase your timeout to tell it to wait longer.
20:58 dave_den -t 300 or so
20:58 mattmtl dave_den: I get the workaround... problem is how you get the import path to the custom modules?
20:59 oraqol ok, thanks
20:59 dave_den mattmtl: i believe __opts__ may have that info
21:01 quickdry21 joined #salt
21:01 jeremyfe_ joined #salt
21:03 racooper Doormatt,  thanks for the help. I was able to get it working once I had a correct example to see.
21:06 smcquay joined #salt
21:06 oraqol One more question: is there anyway to schedule highstate in the master config file to run in batched mode?
21:08 zain joined #salt
21:09 schimmy joined #salt
21:09 zooz joined #salt
21:09 srw_ inside a state.sls is there a template variable for the current state 'name'?
21:09 TomasNunez I can "cmd.run" commands to a server (mysql-slave), and "state.highstate" to the neighbour server (mysql-master), that should have the same salt config... but I can't "state.highstate" mysql-slave... What can be wrong? Where can I look?
21:13 Vye joined #salt
21:15 srw_ Doormatt: it seems that the state.sls is completely rendered, instead of rendering each state one at a time on each run
21:15 dyim hi, what is the difference between salt['grains.get']('os', '') and grains['os']? It doesn't seem to be returning the same result.
21:16 manfred dyim: you can do salt['grains.get']('os:something:else', '')
21:16 manfred dyim: grains[] actually references the dictionary, grains.get() calls the built in get function, and salt['grains.get'] calls the grains.get module that does magic so i can easily have 11 layers to a dictionary
21:17 gasbakid joined #salt
21:17 baniir joined #salt
21:18 chrisjones joined #salt
21:19 dyim the salt['grains.get'] is returning the correct value that I'm after. Not sure why grains['os'] isn't working.
21:20 forrest Gareth, I added you to an issue on the bind formula
21:21 gasbakid joined #salt
21:22 oraqol left #salt
21:23 NotreDev joined #salt
21:23 gildegoma joined #salt
21:23 tr_h joined #salt
21:23 Gareth forrest: Just saw it.
21:24 codysoyland joined #salt
21:24 manfred dyim: you might need to use saltutil.refresh_grains to get it to refresh the dictionary, i think it is loaded when the minion starts, whereas salt['grains.get'] might actually check what is set in the files?
21:24 manfred uhh not refresh_grains
21:24 manfred sync_grains
21:24 Shish_ joined #salt
21:25 forrest Gareth, ok cool, how does that work?
21:25 krow joined #salt
21:25 forrest Gareth, oh I see your update already, thanks
21:25 sashka_ua joined #salt
21:25 andyshinn` joined #salt
21:26 manfred hrm, maybe I am just wrong
21:26 nhubbard_ joined #salt
21:26 gasbakid joined #salt
21:27 d10n joined #salt
21:28 ampex joined #salt
21:28 forrest Gareth, that makes me feel less stupid
21:28 forrest so thanks
21:28 rofl____ joined #salt
21:29 seapasulli joined #salt
21:29 Valdo joined #salt
21:29 Gareth forrest: :)
21:30 egads joined #salt
21:30 Gareth listening to Corey's salt talk from FOSDEM...was the talk taking place in a bar?
21:30 tobiassjosten joined #salt
21:31 forrest Gareth, I was wondering the same thing, seems like maybe people were just rude? It is the same way during the puppet talk right after.
21:31 forrest or maybe the camera was near a door? I don't know. I don't get why there is so much background audio.
21:32 bejer joined #salt
21:32 emostar joined #salt
21:33 Gareth forrest: seems like it was near the door.  but wow.  a lot of background noise.
21:34 forrest Gareth, maybe Corey just talks quietly, we know how softly spoken he is.
21:34 Gareth haha
21:34 hunter The videos made it online?
21:34 djinni` joined #salt
21:35 forrest hunter, this is the fosdem one
21:35 forrest I spoke with Rhett last week about the saltconf videos and he says they're working on them
21:35 forrest I believe he is doing all the editing in addition to his normal job currently
21:35 hunter o
21:35 hunter oh
21:36 hunter forrest: I don't know that I've seen the fosdem one - url?
21:36 tomas1 joined #salt
21:41 NotreDev joined #salt
21:42 Gareth hunter: https://www.youtube.com/watch?v=Le07vJqjkYU&amp;list=UUpveTIucFx9ljGelW63-BWg
21:42 hunter Gareth: thank you
21:45 zirpu left #salt
21:47 baniir joined #salt
21:49 Ryan_Lane joined #salt
21:51 blee_ joined #salt
21:53 mgw1 joined #salt
21:53 gildegoma_ joined #salt
21:53 frasergr_ joined #salt
21:53 djinni` joined #salt
21:53 timoguin joined #salt
21:53 Shish joined #salt
21:54 schimmy joined #salt
21:54 jcsp joined #salt
21:55 abe_music joined #salt
21:55 Vye joined #salt
21:55 johtso joined #salt
21:56 anuvrat joined #salt
21:56 ahctpac_ joined #salt
21:57 Shish_ joined #salt
21:57 sashka_u1 joined #salt
21:59 Rusty_Shacklefor joined #salt
22:00 djinni`_ joined #salt
22:02 jesusaurus did the salt documentation adopt the dewey decimal system?
22:02 jesusaurus http://docs.saltstack.com/ref/states/all/salt.states.file.html
22:02 jesusaurus whats the 19.25.16. ?
22:02 meteorfox joined #salt
22:03 Gifflen_ joined #salt
22:04 gnugnu8_ joined #salt
22:04 tomas2 joined #salt
22:04 n8n joined #salt
22:04 jmccree_ joined #salt
22:06 superdug joined #salt
22:06 NotreDev joined #salt
22:08 d10n_ joined #salt
22:09 tobiassjosten joined #salt
22:11 bhosmer joined #salt
22:11 Ryan_Lane1 joined #salt
22:11 funzo joined #salt
22:11 saurabhs joined #salt
22:11 jaimed joined #salt
22:11 jeffro joined #salt
22:11 meteorfox joined #salt
22:11 zain joined #salt
22:12 kossy joined #salt
22:13 QuantumRiff joined #salt
22:13 penguin_dan joined #salt
22:13 jacksontj_ joined #salt
22:13 alekibango joined #salt
22:13 fllr joined #salt
22:13 linjan joined #salt
22:14 littleidea joined #salt
22:14 sfello joined #salt
22:15 res0nat0r joined #salt
22:16 res0nat0r left #salt
22:16 res0nat0r joined #salt
22:16 smcquay joined #salt
22:16 chrisjones joined #salt
22:17 dN0t left #salt
22:18 QuantumRiff So i'm looking at the salt reactor.. am I right in assuming that runner.* runs on the salt master server itself, and cmd.* runs on the minion I specify in tgt: * ?
22:18 bretep Does anyone have suggestions on howto include 50+ sls files in a pillar. /pillar_root/users/user{n}.sls
22:18 bretep Each user{n}.sls has the users configuration
22:19 kossy joined #salt
22:20 bretep Better example: /pillar_root/users/${username}.sls
22:20 penguin_dan joined #salt
22:22 bejer joined #salt
22:24 kermit joined #salt
22:24 basepi forrest: only just saw your mention (been moving today) -- did the toaster make it into a salt-air previously or something?  =P
22:25 sfello joined #salt
22:27 Gareth Next salt-air.  Just the toaster.
22:28 dyim hi guys, do you use the salt-formulas on github? If you do, how do you get the file templates to work? I'm trying to get. I checked the debug output and it did read in my pillar data but the formula i'm using (openssh) doesn't seem be be making any changes to the sshd_config
22:29 naiyte joined #salt
22:29 pydanny joined #salt
22:30 valgrind joined #salt
22:31 schimmy1 joined #salt
22:32 zzzirk joined #salt
22:32 basepi Gareth: we can make this happen.  20 minutes of pure toaster.
22:32 ggoZ joined #salt
22:33 forrest basepi, it was in the salt-air that was for salt-ssh
22:33 forrest basepi, I saw your tweet, are you guys almost moved?
22:34 funzo_ joined #salt
22:34 saurabhs1 joined #salt
22:35 Ryan_Lane joined #salt
22:35 quantumriff joined #salt
22:36 hunter joined #salt
22:36 bhosmer_ joined #salt
22:37 Gareth basepi: The toaster should talk about Salt.
22:38 basepi Gareth: Finally, a chance to practice my ventriloquy!
22:38 smcquay_ joined #salt
22:38 basepi forrest: yes, the movers just arrived at the old office.  It has begun!
22:38 sfello joined #salt
22:38 schimmy joined #salt
22:40 chrisjones joined #salt
22:40 yomilk joined #salt
22:40 hunter forrest: ugh - that audio is terrible.
22:43 meteorfox joined #salt
22:44 NotreDev joined #salt
22:45 diegows joined #salt
22:45 forrest basepi, nice!
22:45 forrest hunter, I think it was just a more open space for that fosdem one
22:45 forrest It happens, there could be no video at all!
22:46 forrest at least it isn't like the puppetconf videos
22:46 forrest filmed in 240p potato vision
22:48 hunter potato vision?
22:48 hunter I shudder to ask
22:49 zzzirk joined #salt
22:50 forrest puppetconf has been streamed every year by some random third party streaming service which is terrible. The video only goes up to 240p
22:50 forrest so you can't read the slides.
22:51 linjan joined #salt
22:51 hunter oh - god lord
22:52 kermit joined #salt
22:52 kossy joined #salt
22:56 elfixit joined #salt
22:57 VertigoRay joined #salt
22:57 VertigoRay Hey guys ... any word on debian repo getting update with 2014.1?
22:57 alunduil joined #salt
22:58 vejdmn joined #salt
22:59 AdamSewell joined #salt
22:59 AdamSewell joined #salt
23:00 pdayton1 joined #salt
23:02 frasergraham joined #salt
23:04 baniir joined #salt
23:04 bretep VertigoRay: I asked joehh a few days a go, still no answer.
23:05 VertigoRay @bretep Thx!
23:05 bretep He is the one that maintains it
23:06 srage joined #salt
23:06 sroegner joined #salt
23:08 Gareth http://paradoxxxzero.github.io/2014/02/28/butterfly.html?utm_content=buffer3dff5&amp;utm_medium=social&amp;utm_source=twitter.com&amp;utm_campaign=buffer
23:09 lester_ joined #salt
23:16 frasergraham joined #salt
23:17 kballou joined #salt
23:18 baniir joined #salt
23:18 rowleyaj joined #salt
23:23 ipalreadytaken joined #salt
23:27 monokrome joined #salt
23:29 gariful joined #salt
23:29 martoss joined #salt
23:29 Ryan_Lane1 joined #salt
23:33 jeffro heyoo, i have a newb question. im trying to set up an additional service that runs in my dev environment on a webapp im working on. i've tried include: - base: browserify and extend: but it complains about conflicting ids
23:33 jeffro what horrible newb mistake am i making :)
23:35 pdayton joined #salt
23:35 forrest jeffro, can you do a pastebin/gist of your state?
23:36 jeffro yes, one sec
23:37 alunduil_ joined #salt
23:37 srage joined #salt
23:37 linjan_ joined #salt
23:37 AdamSewell_ joined #salt
23:38 jeffro http://pastebin.com/2jUWJNv5
23:38 sroegner_ joined #salt
23:38 martoss1 joined #salt
23:39 forrest jeffro, so which state is giving an error?
23:39 forrest usually that error would literally mean you have an ID naming conflict somewhere.
23:39 jeffro right
23:40 jeffro the duplicates are browserify and nodejs
23:40 elfixit1 joined #salt
23:40 jeffro do i need to make dev versions of both of those? i was hoping to keep it like prod is the base env, and dev extends it
23:41 forrest jeffro, and you have these configured as actual environments with file_roots and everything?
23:41 MTecknology all states need to have unique identifiers (this only applies to what is sent to the box)
23:42 jeffro right
23:42 jeffro yes the file roots are like so:
23:42 kermit1 joined #salt
23:42 jeffro prod: - /srv/salt/base
23:42 jeffro dev: - /srv/salt/dev
23:42 jeffro - /srv/salt/base
23:43 forrest gotcha
23:43 Ryan_Lane2 joined #salt
23:43 forrest jeffro, I'm gonna head home real quick, I'll be back in 20, try to see if you can pastebin the error, and if it turns up in a quick google search.
23:43 forrest otherwise I'll try to help when I am back online
23:43 jeffro forrest: will do, thanks
23:44 kossy joined #salt
23:46 alexthegraham joined #salt
23:48 krow joined #salt
23:49 jeffro MTecknology: do you have an example somewhere of extending the base environment inside another, or is this a bad pattern for salt?
23:51 sroegner_ joined #salt
23:51 cjbarnes18 joined #salt
23:55 yomilk joined #salt
23:55 littleidea joined #salt
23:56 MTecknology jeffro: I've not seen it done; usually just kept completely separate
23:57 MTecknology might not be bad at all
23:57 sinh joined #salt
23:58 jeffro MTecknology: i see

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