Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2013-10-28

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

All times shown according to UTC.

Time Nick Message
00:00 packeteer ^^
00:08 VSpike joined #salt
00:19 __number5__ Is there an easy way to update pillar within module on minion?
00:24 hazzadous joined #salt
00:24 UtahDave __number5__: not really. Pillar data is controlled pretty tightly.  You might be able to send an event with    event.fire_master   and then have the reactor add some data to the pillar.
00:25 __number5__ hmm, what I'm looking for is actually to make a dynamic variable available to all states/templates on minion
00:26 __number5__ maybe custom grains is better way to go?
00:27 noob2 joined #salt
00:27 UtahDave where does the dynamic variable come from?
00:28 __number5__ via a salt-call in cloud-init script, the minions is running masterless mode
00:28 noob2 can i use flat minion id's with the salt LocalClient?  like client.cmd('host01,host02,host03', 'cmd.run', ['uptime'])
00:32 mgw joined #salt
00:34 UtahDave __number5__: well, pillar data comes from yaml files in /srv/pillar/*  so I guess you could manually update those files, since in masterless mode they are on the same machine.
00:34 UtahDave noob2: yeah, but I think have you to add add   , expr_form=list   or something like that
00:34 noob2 ok i'll give that a shot :)
00:36 __number5__ UtahDave: cool, thanks
00:37 noob2 UtahDave: yeah i see it now, it should be 'list,of,stuff', expr_form=list i think
00:38 UtahDave noob2: Yeah, I always have to go track down that syntax,
00:50 noob2 UtahDave: that worked.  This is awesome :D
00:50 noob2 i'm feeling the power
00:50 UtahDave doing some cool stuff, huh?
00:57 foxx joined #salt
00:59 jcockhren noob2: aren't you the one working on deis?
01:01 noob2 jcockhren: yup i'm working on that also
01:02 noob2 UtahDave: running some distributed commands with the salt python client.  It's so damn easy
01:03 noob2 jcockhren: i'm trying to figure out how i'm going to make the data bags work still
01:03 ak__ joined #salt
01:03 UtahDave nice
01:03 noob2 pillar is the right thing to use.  I just don't know how i'm going to do the yaml parsing code yet
01:04 BrendanGilmore joined #salt
01:04 noob2 UtahDave: have other people told you about cool things they're doing with the salt client?  It's powerful
01:05 UtahDave noob2: A few.  People use it for integrating other apps, among other things
01:06 noob2 nice
01:06 UtahDave noob2: what are you doing with it?
01:07 bemehow joined #salt
01:07 honestly this is what I use salt for: https://github.com/duk3luk3/dirty-user-sync-formula/
01:07 honestly lol
01:08 jcockhren noob2: can there be a data bag to yaml translator
01:08 jcockhren ?
01:08 jcockhren honestly++ lol
01:11 noob2 jcockhren: probably haha.  it's just json <-> yaml
01:11 jcockhren noob2: I'm sure that's been solved to some degree
01:11 nocturn joined #salt
01:12 noob2 google knows, lets see
01:12 jcockhren yeah. but that's heavy nodejs
01:13 jcockhren I guess because json is used heavily in JS
01:13 m_george|away joined #salt
01:13 jcockhren *shurgs*
01:13 jcockhren shrugs*
01:14 noob2 i'm just not sure a lot of the deis stuff is needed with salt's powerful command execution.  chef is lacking that and therefore they had to build an entire set of stuff around that
01:14 UtahDave json is valid yaml, btw
01:14 ak__ Is it possible to have a minion manage 2 instances of MYSQL server (say MySQL1 and MYSQL2). If so how is the db credentials (user/password) kept secured such that MYSQL1::DBA can't peek into MYSQL2 and vice-versa
01:14 noob2 UtahDave: cool, i didn't know that
01:14 jcockhren UtahDave: quit playing
01:14 taylorgumgum joined #salt
01:14 jcockhren you serious?
01:14 UtahDave yep.  Valid json can be loaded just fine by pyyaml
01:15 noob2 heh, that's neat
01:15 jcockhren "this changes everything"
01:15 noob2 brb gotta chop and onion and garlic up
01:15 jcockhren noob2: need salt?
01:16 jcockhren lol
01:16 UtahDave lol
01:18 mafrosis joined #salt
01:18 mafrosis lo
01:21 mwillhite joined #salt
01:26 Nexpro joined #salt
01:26 taylorgumgum joined #salt
01:28 pdayton joined #salt
01:35 malinoff joined #salt
01:35 ak__ joined #salt
01:36 noob2 lol
01:37 noob2 actually i did :D
01:38 bemehow joined #salt
01:41 malinoff noob2, nice nickname
01:41 noob2 malinoff: thanks haha.  It's the first lame nick that popped into my head
01:42 noob2 i'm not sure what happened to noob1, he registered the nick and i couldn't use it anymore
01:42 malinoff noob2, what about just noob? :)
01:42 noob2 yeah someone stole that also.  i was using that for about a month and then someone registered it
01:43 noob2 i should really get around to registering something
01:44 deepakmd_oc joined #salt
01:46 HeadAIX joined #salt
01:46 __number5__ guess what happen to number5?
01:52 malinoff __number5__, i thought, you simply like python's magic :(
01:54 ajw0100 joined #salt
01:54 __number5__ malinoff: that's also part of the reason :)
02:06 oz_akan_ joined #salt
02:12 dcrouch joined #salt
02:16 xl1 joined #salt
02:18 tulu joined #salt
02:25 luminous does file.directory really fail if you put a trailing slash??
02:26 mgw joined #salt
02:28 bemehow joined #salt
02:39 anuvrat joined #salt
02:42 vipul joined #salt
02:45 Linz joined #salt
03:04 berto- joined #salt
03:04 MTecknology holy crap... the file module is huge
03:04 luminous hah! so /foo/bar/ fails with file.directory unless you have makedirs: True
03:05 luminous where as /foo/bar is ok
03:05 luminous MTecknology: yea, esp with the updates for 0.17.x
03:07 elfixit joined #salt
03:11 malinoff http://pastebin.com/kHLMaXGY
03:12 malinoff http://pastebin.com/BhJ2kScp - with makedirs: True
03:13 malinoff It works as expected
03:13 cachedout joined #salt
03:13 forrest get outta here cachedout, it's not monday
03:14 cachedout True, but hacking on Salt is fun even when it isn't Monday. :]
03:14 forrest very true!
03:17 luminous malinoff: did /tmp/test exist before?
03:17 luminous in my case it does. /tmp/test/1 would succeed, /tmp/test/2/ would fail, unless makedirs: True
03:19 malinoff luminous, No, it did not
03:19 malinoff luminous, what version of salt do you use?
03:20 luminous malinoff: this test was on 0.16.x
03:20 luminous 0.17.x still breaks for me
03:20 luminous LocalClient specifically
03:20 malinoff luminous, do you use custom python script to run this state?
03:21 mwillhite joined #salt
03:21 malinoff luminous, I'm using 0.17.1 and it works as expected, as you see
03:22 luminous no custom python script, no
03:22 luminous but I think you're missing one detail
03:23 luminous in your test /tmp/test did not exist, correct?
03:24 luminous malinoff: ^^
03:25 malinoff luminous, yep
03:25 luminous so that's not the same as what I'm saying
03:26 luminous malinoff: create /tmp/test, then run a state with /tmp/test/1 and /tmp/test/2/ - ignore makedirs for now
03:26 bemehow joined #salt
03:26 luminous with 0.16.x, /tmp/test/1 would succeed, /tmp/test/2/ would fail,
03:27 luminous if you add makedirs: True to /tmp/test/2/ it'll work
03:30 Furao joined #salt
03:32 malinoff luminous, interesting. I don't know what's wrong but right now i have [WARNING ] TypeError encountered executing state.sls: 'NoneType' object is not iterable. See debug log for more info.  Possibly a missing arguments issue:  ArgSpec(args=['mods', 'env', 'test', 'exclude', 'queue'], varargs=None, keywords='kwargs', defaults=('base', None, None, False))
03:33 malinoff ah
03:34 malinoff luminous, http://pastebin.com/wWEZwbVq
03:34 malinoff It works :)
03:35 luminous malinoff: awesome, so this is a bug in 0.16.x and fixed in 0.17.x - thanks for confirming
03:36 malinoff luminous, no problem :)
03:49 woebtz joined #salt
04:04 jacksontj joined #salt
04:06 jacksontj_ joined #salt
04:25 Viraj joined #salt
04:27 jkleckner joined #salt
04:29 jkleckner joined #salt
04:30 taylorgumgum joined #salt
04:31 jkleckner joined #salt
04:35 xl1 The bug https://github.com/saltstack/salt/issues/7933 breaks a couple of things, what is the best way to work around it? Salt master and minions are installed from Ubuntu PPA (I'm considering manually patching pkg.py on minions, but not sure if this is a safe thing to do)
04:40 taylorgumgum joined #salt
04:40 bhosmer joined #salt
04:42 xl1 I mean apt.py
04:46 joehh xl1: I think you can put it into _modules and sync it out to the minions
04:48 xl1 joehh: Thanks!
04:49 jalbretsen joined #salt
04:53 joehh xl1: though be carefull there is states/apt.py and modules/apt.py so make sure you put the right one in
04:53 joehh ie if it is states/apt.py, put it in _states/
05:03 xl1 joehh: I checked it is _modules for this patch. Thanks for your help :)
05:13 joehh no worries
05:29 summernguyen joined #salt
05:37 cachedout joined #salt
05:41 jkleckner joined #salt
05:44 Khazix joined #salt
05:44 malinoff Guys, can somebody help? What's the best way to get file.managed with a git repository as source?
06:14 jjasinski_ joined #salt
06:16 jjasinski_ HI all, I'm new to Salt and am looking for a DRY way to set a default directory location that I can use all throughout my salt states and optionally override in a pillar
06:16 jjasinski_ any thoughts as to where I might start looking?
06:17 jjasinski_ for example, I'd like to set a project directory to /www/sites/ by default, but could set to something else via a pillar
06:23 woebtz joined #salt
06:26 jkleckner joined #salt
06:33 anuvrat joined #salt
06:36 jjasinski_ Should I try setting these paths in the top file ?
06:47 pears jjasinski_: in your position I would set a default location in pillar and then a second pillar variable that overrides the default if present
06:47 sebgoa joined #salt
06:49 jjasinski_ Thanks pears, good to know I should try setting the default value in the pillars
06:51 pears or you could just have the default in your states and do a jinja template thing that checks if there's an override available in pillar
06:53 pears like uh {% set project_dir = pillar.get('project_dir', '/www/sites' %}
06:53 pears and then you can use {{ project_dir }} in whatever state
06:53 pears if that makes sense
06:55 jjasinski_ ya, i think so.  If I perform that set high up in the tree, do the included states get that variable? Or is there a way to explicitly pass the variable to included states ?
06:56 jjasinski_ or is the idea to use that set statement in every sls file that I need to use it in?
06:57 pears I had in mind the second way
06:58 pears not sure if you can pass data from the top file down
06:59 jjasinski_ Ok - good to know.
07:00 jjasinski_ Thanks for the recommendations.
07:03 linjan_ joined #salt
07:09 futrboy joined #salt
07:11 futrboy hello! can i create grains on an individual minion by not having to append to the grains: section within /etc/salt/minion?
07:11 futrboy im looking for something like this in /etc/salt/minion, grains_root: /etc/salt/grains.d
07:12 futrboy and /etc/salt/grains.d/foo would contain bar: baz
07:14 pears futrboy: http://docs.saltstack.com/topics/targeting/grains.html#grains-in-etc-salt-grains ?
07:14 pears or do you specifically need multiple files
07:15 futrboy multiple files would be nice to have pears.
07:16 * futrboy has found grains.setval
07:17 pears yeah I use grains.setval myself
07:17 pears the idea of manually setting grains on all the hosts is unappealing
07:17 * futrboy ponders
07:17 pears so I have a script that pushes out all the grains to the relevant hosts
07:17 * futrboy nods
07:17 futrboy problem im trying to solve is, im doing a yum update --assumeyes
07:17 futrboy i'd like to have grain tell me which host has this completed on.
07:18 futrboy sorry, yum update is done via a state.
07:18 futrboy i only need to do it once, when a host is provisioned
07:21 futrboy i'll just append to /etc/salt/grains for now. thanks pear! :-)
07:28 darrend joined #salt
07:35 gildegoma joined #salt
07:42 balboah joined #salt
07:50 mikko__ joined #salt
07:56 ollins joined #salt
08:12 matanya joined #salt
08:17 bhosmer joined #salt
08:17 permalac joined #salt
08:27 Destro joined #salt
08:27 linjan joined #salt
08:32 sebgoa joined #salt
08:33 carlos joined #salt
08:43 derelm joined #salt
08:47 faldridge joined #salt
09:12 scott_w joined #salt
09:17 networkpadawan joined #salt
09:17 krissaxton joined #salt
09:29 krissaxton joined #salt
09:35 lemao joined #salt
09:39 bhosmer joined #salt
09:44 kopernikus joined #salt
09:45 zooz joined #salt
09:46 kopernikus Hi all, how can I debug the "state xyz found in sls foo is unavailable" message?
09:47 kopernikus getting this for ssh_auth.present and mysql_grant.present
09:47 kopernikus mysql_database.present works however....
09:50 carlos joined #salt
09:55 malinoff kopernikus, -l debug
09:57 ronc joined #salt
09:57 Savagedlight joined #salt
09:58 baffle Hi, I have a jinja template that is managed by "base.sls". "base.sls" is included by various other state-files, like "api.sls", "scheduler.sls" and "compute.sls". But the content of the file on disk needs to be different based on wich state-files that includes it; I.e. on one minion I might use "api.sls"and "scheduler.sls", but on another I just use "compute.sls".
09:58 baffle What is the best way to define variables I can use in the template to selectively include portions of the template? For now I have made a pillar-file pr. configuration, and define these for each host... Is there a way to see wich SLS files are in use somehow when generating a template?
09:59 * NV throws a fish at Savagedlight
09:59 malinoff baffle, Such question should be asked on Salt-users mailing list
09:59 baffle malinoff: Ok.
10:00 * Savagedlight throws a salt shaker at NV
10:01 NV :D
10:12 joehh joined #salt
10:16 swa joined #salt
10:21 kopernikus @malinoff: thanks, -l revealed several PEBKAC issues
10:23 woebtz joined #salt
10:23 baffle Is it "okay" to use/parse the output of salt.modules.state.show_top on a minion, to discover wich states it has, and use that to genereate a template? :-)
10:38 giantlock joined #salt
10:45 ricicle joined #salt
10:45 jcsp joined #salt
10:47 ricicle Hi all. I'm trying to build an rpm for centos 6.4 with Python 2.7 from the salt git source.
10:49 ricicle I'm getting RPM build errors: File not found: /srv/foo/incoming/git/saltstack/salt/build/bdist.linux-x86_64/rpm/BUILDROOT/salt-0.17.0_1599_g3d29e9b-1.x86_64/usr/share/man/man1/salt-master.1
10:49 ricicle The actual files end .gz
10:50 ricicle That's when building with /usr/local/bin/python27 setup.py bdist_rpm
10:50 pkimber joined #salt
10:50 ricicle when building with fpm it builds but then I hit https://github.com/saltstack/salt/issues/4454
10:51 ricicle I tried applying the patches for rpms but they would not apply
10:51 ricicle Is there a how to I'm missing please?
10:51 ricicle I'm no packaging expert so feel free to ask me dumb questions :)
10:52 ricicle Or if there's a 0.17-dev rpm I could pinch that would work too
11:00 Ixan there should be a srpm up for salt 0.17 on fedora epel 6
11:00 Ixan you can alter the spec yourself and remove the .gz and it should build fine
11:00 ricicle So https://github.com/saltstack/salt/issues/2869 says to build a sdist, this seems to work. Now to re-package that as n rpm somehow. Anyone find it ironic that fpm doesn't build an rpm of itself?
11:02 ricicle Ixan: Thanks, I'll try and find that one. Our (re-)packaging system is trying to be fairly generic so I was avoiding overlays for now. There's a method for adding rpms to a custom repo so your suggestion is probably the easiest
11:02 Ixan if you're making a sdist from git, make sure you checkout the 0.17 tag first
11:03 ricicle Ixan: I would prefer to push the spec changes upstream if I went that way.
11:04 ricicle Ixan: I was looking for the very latest (given that 0.17.1 should be released soon). Are you saying there be dragons in revisions after the 0.17 tag?
11:04 Ixan no
11:05 Ixan just making sure you make the sdist from the tag and not from head
11:09 elfixit joined #salt
11:10 ricicle I'm checking out the head of the 'develop' branch
11:12 Furao joined #salt
11:15 blee joined #salt
11:27 blee joined #salt
11:51 swa baffle: I've asked the same question on the mailing list.. my idea was to generate a MOTD showing which configs were managed by Salt
11:53 jasiek_ joined #salt
11:53 mackstic1 joined #salt
11:56 simonmcc joined #salt
12:03 baffle swa: I'm getting closer to a solutions, using salt['state.show_top']..
12:03 baffle -s
12:04 BogdanR Hi
12:05 BogdanR I am trying to use the winrepo from git but I have a problem which I don't understand
12:05 BogdanR http://pastie.org/pastes/8437055/text
12:06 BogdanR As you can see, genrepo didn't work
12:06 BogdanR What is the issue here?
12:09 swa baffle: what do you have so far ? I can try something on my side
12:15 hazzadous joined #salt
12:17 LLckfan2 joined #salt
12:17 LLckfan2 I have an android tablet that I have put youtube app on. When I try to sign into youtubr it says an error occurred. Is there anything I can do to fix this? I did not get the app from Google Play as Google Play does not support my brand of tablet
12:18 swa LLckfan2: wrong channel :)
12:19 LLckfan2 The right channel does not help
12:19 LLckfan2 But excuse me for being here
12:19 LLckfan2 left #salt
12:23 oz_akan_ joined #salt
12:25 oz_akan_ joined #salt
12:28 carlos joined #salt
12:29 baffle swa: I have a file I include, wich iterates over state.show_top on the minion, and uses that information in ugly for-loops. It's not very purdy (or working) right now, I'll get back to you... :)
12:31 dpoirier left #salt
12:35 diegows joined #salt
12:36 nkuttler joined #salt
12:44 ricicle Ixan: I've tried the http://arm.koji.fedoraproject.org/packages/salt/0.17.0/1.fc19/noarch/salt-0.17.0-1.fc19.noarch.rpm and its master and minion rpms. Looks like it might be crufty: Requires: systemd-units (which has been replaced by systems for over a year) and also Requires: python(abi) = 2.7\n Installed: python-2.6.6-37.el6_4.x86_64 (installed)\n python(abi) = 2.6 so I'm not sure how to tell it to use python27 which is a
12:44 ricicle spec file and giving you a PR is the way to go.
12:47 Ixan i suggest you use the epel package for el6 instead of fedora
12:47 Ixan 0.17.1 is in epel-testing now
12:48 Ixan if you want to build the packages from scratch, download the srpms for that
12:48 Ixan building fedora 19 packages on centos 6.3 will end in tears
12:49 brimpa joined #salt
12:49 arapaho joined #salt
12:51 brianhicks joined #salt
12:51 kopernikus left #salt
12:52 amahon joined #salt
12:56 oz_akan__ joined #salt
12:57 anteaya joined #salt
12:59 cron0 joined #salt
13:01 jslatts joined #salt
13:02 mwillhite joined #salt
13:04 mgw joined #salt
13:05 oz_akan_ joined #salt
13:06 mpanetta joined #salt
13:07 Gifflen joined #salt
13:08 racooper joined #salt
13:08 xl1 left #salt
13:09 juicer2 joined #salt
13:12 cachedout joined #salt
13:12 cowboy joined #salt
13:16 timoguin joined #salt
13:17 snikkers joined #salt
13:19 _ilbot joined #salt
13:19 Topic for #salt is now Welcome to #salt - SaltConf in January!! http://saltconf.com | 0.17.1 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/
13:20 carlos joined #salt
13:23 colinbits joined #salt
13:24 mgw joined #salt
13:27 terminalmage joined #salt
13:31 kaptk2 joined #salt
13:32 krissaxton joined #salt
13:35 toastedpenguin joined #salt
13:36 toastedpenguin left #salt
13:39 isomorphic joined #salt
13:44 balboah joined #salt
13:45 racooper Is there a salt module to manage scheduled tasks on Windows Server 2008?
13:46 Ahlee anybody got the magic snippet to get custom _modules to log to salt minion's log?
13:48 Ahlee I thought I could just log = logging.getLogger(__name__)
13:49 Ahlee but, I don't see items tossed at log showing up
13:49 timoguin_ joined #salt
13:50 dpoirier joined #salt
13:50 vkurup joined #salt
13:51 sgviking joined #salt
13:52 ipmb joined #salt
13:52 jumperswitch joined #salt
13:55 micah_chatt joined #salt
13:57 abe_music joined #salt
13:59 gildegoma joined #salt
13:59 ashtonian joined #salt
14:01 gmcwhistler joined #salt
14:08 ricicle joined #salt
14:18 c0bra joined #salt
14:22 az87c_ joined #salt
14:22 az87c joined #salt
14:26 alunduil joined #salt
14:30 pmrowla joined #salt
14:31 smccarthy joined #salt
14:35 vagrant_ joined #salt
14:39 juicer2 joined #salt
14:39 bin_1 hi folks
14:40 jimallman joined #salt
14:40 bin_1 is there an option in remote command execution module to print only if a certain criteria matches?
14:41 bin_1 say print the node name which has ulimit core size 0
14:45 jankowiak joined #salt
14:50 mgw joined #salt
14:53 BogdanR So, anyone can help me with this: http://pastie.org/pastes/8437055/text ?
14:54 HeadAIX joined #salt
14:57 jalbretsen joined #salt
14:58 pdayton joined #salt
15:00 mwillhite joined #salt
15:00 samsalt joined #salt
15:01 cachedout joined #salt
15:02 noob2 joined #salt
15:02 amahon joined #salt
15:02 matanya joined #salt
15:03 mua joined #salt
15:04 amahon joined #salt
15:04 noob21 joined #salt
15:07 matanya joined #salt
15:07 pentabular joined #salt
15:09 UtahDave joined #salt
15:09 honestly o/ UtahDave
15:10 mapu joined #salt
15:11 bemehow joined #salt
15:12 UtahDave hola, honestly!
15:12 gmcwhistler left #salt
15:14 matanya joined #salt
15:15 lineman60 joined #salt
15:16 teskew joined #salt
15:17 linuxnewbie joined #salt
15:19 BogdanR UtahDave: Can you please give me a hint on how to fix this? http://pastie.org/pastes/8437055/text
15:20 baffle swa: Here is what I'm trying to do: http://paste2.org/K4pY5XNG
15:21 linuxnewbie left #salt
15:21 danielbachhuber joined #salt
15:21 baffle swa: It seems it doesn't really iterate over the state.. Very confused..
15:24 UtahDave baffle: you probably need to have state.show_top return json or something
15:25 UtahDave BogdanR: Yeah, I ran into that not too long ago
15:25 UtahDave BogdanR: I think it's because one of the packages in winrepo is malformed.  Would you mind opening an issue on that?
15:26 UtahDave BogdanR: I want to change the behavior to skip malformed packages and log them and then continue on
15:26 baffle UtahDave: Oh? Any idea why this would work for swa then? https://groups.google.com/forum/#!topicsearchin/salt-users/author$3Ame%7Csort:date/salt-users/mFP8RzQ65tk
15:26 jdenning joined #salt
15:27 baffle UtahDave: And I tought it would want an interable python list, not JSON? I'm so confused. :)
15:27 baffle iterable
15:29 UtahDave baffle: well, have you looked at the text it is outputting?  It might iterate over each line of text
15:32 srage joined #salt
15:32 baffle UtahDave: I can't see any errors. And it seems to kindof iterate as well. Updated example: http://paste2.org/LhzAscJg
15:34 holmb mikko__: I sent PR #8136 which gives you a log of which minions are not upgraded to 0.17.1.
15:35 baffle Is there any easy way to do "printf-debugging" in a template/sls? I.e. to output the content of a variable at any time to debug-output.. :)
15:35 mikko__ holmb: thnx
15:35 ajw0100 joined #salt
15:36 JesseC joined #salt
15:37 BogdanR UtahDave: Sure, I will open an issue
15:39 jkleckner joined #salt
15:39 xmltok joined #salt
15:41 mwillhite joined #salt
15:43 mwmnj joined #salt
15:43 mwmnj I am trying to go through the salt walk through and it seems I must have flubbed a step somewhere
15:43 mwmnj http://docs.saltstack.com/topics/tutorials/walkthrough.html#minion-id-generation
15:44 mwmnj I am trying to connect a minion to a master
15:46 UtahDave mwmnj: did you set    master: <ip or hostname>  in the minion's config?
15:48 druonysus joined #salt
15:48 druonysus joined #salt
15:48 xet7 joined #salt
15:49 UtahDave Hey, everyone, today is the last day to propose a talk for SaltConf!   saltconf.com
15:49 holmb UtahDave: Uhm, I have been thinking about it but don't think I will have time. :(
15:50 holmb UtahDave: Have you received many talk proposals?
15:50 mwmnj UtahDave: yes, i set it to the amazon ec2 public ip
15:50 mwmnj UtahDave: but when I run $salt-key -L on the master
15:50 mwmnj There are no unaccepted keys listed
15:51 UtahDave mwmnj: have you made sure ports 4505 and 4506 are open on the master?
15:51 mwmnj I am running salt on the minion with $ salt-minion also
15:52 UtahDave holmb: Are you going to be able to attend?
15:52 mgw joined #salt
15:52 abe_music joined #salt
15:54 Jak joined #salt
15:54 bemehow joined #salt
15:55 bemehow joined #salt
15:56 Guest98862 Hi guys, can anyone answer a question about file declarations in sls files please?
15:57 m_george left #salt
15:58 UtahDave Guest98862: ask away
15:59 _ilbot joined #salt
15:59 Topic for #salt is now Welcome to #salt - SaltConf in January!! http://saltconf.com | 0.17.1 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/
16:00 Guest98862 Thanks. I've noticed that in the documentation files are always delcared with their target location as their name, but can I simply declare "file:" then specify "- name:" in its scope?
16:00 UtahDave use    - name: /tmp/file/location
16:00 mwmnj UtahDave: UtahSave: that was it, thanks!
16:01 baffle UtahDave: Any idea why my "huh" variable is not set here? http://paste2.org/LhzAscJg
16:01 pears Guest98862: you can do that but you'll still need to give the declaration some sort of name, I think
16:02 troyready joined #salt
16:04 Guest98862 So I could do something like this, if I were using pillars? http://pastebin.com/v2aY6WXf
16:05 pears looks reasonable to me
16:06 UtahDave yep, Guest98862 that looks great
16:06 KyleG joined #salt
16:06 KyleG joined #salt
16:06 Guest98862 Great, thanks very much guys :-)
16:07 UtahDave baffle: I'm not entirely sure.
16:07 pipps joined #salt
16:08 Guest98862 left #salt
16:08 baffle UtahDave: Good to see that you're unsure; That means I have a reason to be confused :-)
16:09 UtahDave baffle: are you sure "huh" is getting set in line 6?
16:09 swa baffle, i've tried on my end and still no go in init.sls, but works in the template
16:09 baffle UtahDave: It doesn't seem to be set, no.
16:11 baffle UtahDave: I see that it matches the "if" once, so that works. It is just the "set" that doesn't work.
16:12 baffle UtahDave: I.e. if I just write some text instead of the {% set %} statement, I see the text in the output from the template. Once. So the if works.
16:13 honestly UtahDave: how's my bug :3
16:13 baffle UtahDave: I assume the variable isn't local for the loop? Right?
16:14 UtahDave baffle: I'm not sure, actually. I'm not a jinja expert.  What happens if you set your variable first above the loop?
16:15 pipps joined #salt
16:16 baffle UtahDave: It value becomes what I set it to before the loop. So it just doesn't get set inside the loop.
16:17 baffle I am suspecting some kind of scope issues now.
16:18 honestly jinja is not a programming language
16:18 honestly try mako instead
16:18 honestly or py or pydsl
16:19 jdenning joined #salt
16:19 carmony jinja can be odd to use at times
16:19 honestly but sometimes it's worth investing some extra thinking effort to make it work in jinja
16:19 honestly because it will be cleaner
16:19 baffle Yeah, if I use the "huh" variable inside the {% if %} block it contains the correct value.
16:19 carmony UtahDave: btw, I fixed that issue with digital-ocean
16:22 sgviking joined #salt
16:23 UtahDave Yeah, I saw that, carmony!  Thanks!
16:23 baffle So, anyone has a good way to define what states a minion has, expose that as variables or something, so that I can make decisions in a jinja template on what parts of it to include? Right now I set some nasty global pillar variables pr. host, and use those inside the templates.. But it is static and not very purdy.
16:24 Guest49917 joined #salt
16:25 carmony UtahDave: I'm sure python has this, but is there a way to step through a cli command so I don't have to just pprint all over the place to debug? :P
16:25 N-Mi joined #salt
16:25 N-Mi joined #salt
16:26 baffle My usecase is this: I want to generate OpenStacks "nova.conf". This is used by a multitude of deamons, but let's just say it is used by "api", "compute" and "scheduler". Each of these deamons is configured by a separate sls file. Each of these includes a "base.sls" that "owns" the nova.conf. I need to know at generation time wich sections to include in nova.conf, based on what services that minion runs...
16:26 UtahDave carmony: there's pdb, but I haven't used it much.  I tend to pprint all over the place. :)
16:27 baffle Maybe better suited for the ml. :)
16:27 UtahDave baffle: have you looked at file.accumulated?  http://docs.saltstack.com/ref/states/all/salt.states.file.html#salt.states.file.accumulated
16:27 carmony lol, it feels "wrong" from a elegancy stand point, but it is just so effective
16:29 baffle pprinting random debug information is how it has always been done. Everywhere. :)
16:30 Trebortech left #salt
16:32 baffle UtahDave: I don't really understand what it does/how to use it based on the documentation.
16:39 xmltok for people out there using salty vagrant in their workflow: do you have a single vagrant configuration with a /srv/salt containing all of your formulas (pulled in from other git repos) or do you have a vagrant configuration in each of your formula repos?
16:39 xmltok the first method is how i am using to using vagrant with berkshelf from my chef experiences, the second is how it seems to work with salt
16:40 akoumjian xmltok: We clone our formulas to a separate folder on the host machine and share that in the Vagrantfile config
16:41 redondos joined #salt
16:41 xmltok ok, so sounds like #2. do you guys do automated testing of the formulas? I am thinking I could keep a jenkins.sls in each formula repo that could replace the top.sls in the vagrant config repo for jenkins builds
16:41 xmltok so the jenkins job is run on a commit, fetches the vagrant repo, runs with the new formula repo commit, and copies the jenkins.sls out of there into the top.sls so that you can customize what jenkins should test
16:42 ctdawe joined #salt
16:42 xmltok it also solves my formula dependency problem, since more formulas could be cloned into that directory if dependencies exist
16:43 toastedpenguin joined #salt
16:44 xmltok sounds reasonable?
16:44 StDiluted joined #salt
16:46 jkleckner joined #salt
16:54 vipul joined #salt
16:56 sebgoa joined #salt
16:57 taylorgumgum joined #salt
16:58 pentabular joined #salt
16:58 berto- joined #salt
16:59 lineman60__ joined #salt
16:59 jcsp joined #salt
17:00 dpoirier left #salt
17:01 dvogt joined #salt
17:01 dvogt left #salt
17:03 dvogt joined #salt
17:03 cowmix3 joined #salt
17:04 ninkotech joined #salt
17:07 Ryan_Lane joined #salt
17:07 kyusan joined #salt
17:07 srage joined #salt
17:12 ipmb joined #salt
17:13 wibberwock joined #salt
17:13 carmony xmltok: that sounds reasonable to me
17:14 brimpa joined #salt
17:15 Ryan_Lane joined #salt
17:15 jkleckner joined #salt
17:17 alunduil joined #salt
17:17 srage joined #salt
17:19 pkimber joined #salt
17:21 pdayton joined #salt
17:21 ninkotech joined #salt
17:27 Linz joined #salt
17:28 wibberwock is there a definitive list of master-side modules?  things that can be called from a pillar template?
17:29 Corey UtahDave: Bah.
17:29 lineman61 joined #salt
17:29 ninkotech joined #salt
17:31 UtahDave wibberwock: You can pretty much call any module that appears in __salt__  but just remember that they work from the Salt Master's perspective.  So salt['network.ip_addrs'] will give you the Salt Master's ip addresses
17:31 * UtahDave hugs Corey
17:32 wibberwock UtahDave: for example, the publish module doesn't seem to work
17:33 UtahDave wibberwock: are you running a minion on the master?  and have you given that minion rights to do what you're trying to do with the publish module?
17:34 Corey UtahDave: Work is thumbtwisting me into getting the SSCE. I blame you.
17:34 UtahDave :)
17:35 Corey UtahDave: And uh... you aren't giving me a whole lot of time. Looks like the conference ends Thursday. Friday is a travel day, then FOSDEM kicks off in Brussels.
17:36 backjlack joined #salt
17:36 wibberwock UtahDave: no, no minion running on the master.  Let me backtrack and explain what I'm trying to accomplish.  I have several clusters, each with 3-4 roles and 2-3 nodes per role.  I'd like to have per-cluster information available to the sls, like how many nodes in the cluster, how many nodes per role, ips per role. i would also like to share the information over sls files (an iptables sls file needs the ips, an app role sls needs the number of db roles,
17:36 wibberwock etc.).
17:36 UtahDave that soon?  I wonder if we're going to fosdem this year.  We were planning on it
17:36 Linz joined #salt
17:36 wibberwock UtahDave: It seems like pillars are the right fit, but I'm having trouble collecting per cluster information on the master.
17:37 druonysus joined #salt
17:37 amckinley joined #salt
17:38 swa UtahDave and Corey, I'm belgian, and I owe you some beers :)
17:38 UtahDave wibberwock: well, the master can just ask directly.
17:38 UtahDave swa: :)  We'll have to get together out there!
17:39 swa sure thing
17:39 wibberwock UtahDave: I've had to resort to using {{ salt['salt.cmd']('salt 'G@cluster:my cluster' network.ipaddrs) }}, which is a bit hacky, no?
17:39 Corey swa: Ugh! I wound up having uh... a bit too much beer last year.
17:39 Corey It tastes like normal beer except for the part where it's over 12% alcohol.
17:39 Corey It's the stuff poor decisions are made of!
17:39 swa haha
17:40 swa we have light 9% beers as well
17:41 jkleckner joined #salt
17:42 swa I'll be the guy wearing a t-shirt saying : salt '*' free.beer
17:42 Gifflen joined #salt
17:42 bhosmer joined #salt
17:42 UtahDave nice, swa!
17:42 swa maybe not *, gotta do some targeting
17:44 carmony LOL
17:45 carmony carmony.sls - extend: free.beer: -name: root-beer    :P
17:46 totte joined #salt
17:46 krissaxton joined #salt
17:47 druonysus joined #salt
17:47 druonysus joined #salt
17:49 networkpadawan joined #salt
17:50 ssbarnea joined #salt
17:51 ssbarnea hi! I'm quite new to salt and have a question: how do make the deployment fail if a command fails (return code != 0) ?
17:53 bhosmer_ joined #salt
17:55 faust joined #salt
17:56 bhosmer joined #salt
18:01 amahon joined #salt
18:01 pipps1 joined #salt
18:02 bhosmer joined #salt
18:02 indymike joined #salt
18:03 wibberwock How would I convert the command salt '*' network.ipaddrs to a pillar jinja template? {{ salt['cmd.run']('salt * network.ipaddrs') }} times out
18:07 pdayton joined #salt
18:07 sifusam joined #salt
18:08 micah_chatt Does anyone have a good solution to removing all unmanaged files in a directory with some managed files in it? For example like in an NGINX sites-enabled directory
18:09 micah_chatt besides maybe running "rm -rf"  before setting up my managed files on every highstate
18:10 swa that's a bad idea, this would trigger at each highstate
18:10 swa if you know the name of file you want to get rid of, just put that in your state
18:10 mmarch joined #salt
18:11 sifusam or just pull all your files into a directory at the source and use file.directory with recurse and clean.
18:11 bwq joined #salt
18:12 micah_chatt sifusam: not really possible (that I'm aware of), since each file in the directory is configured for that host specifically
18:12 micah_chatt using it's unique grains and pillars
18:13 sifusam ok I see.
18:13 ajw0100 joined #salt
18:13 ckao joined #salt
18:13 micah_chatt swa: I'm getting list of sites dynamically
18:13 micah_chatt so salt is dynamically adding sites based on external pillar data
18:13 micah_chatt read from a database
18:13 redondos joined #salt
18:13 redondos joined #salt
18:14 micah_chatt ahhh I may have just found it
18:15 micah_chatt see the "clean" param under http://docs.saltstack.com/ref/states/all/salt.states.file.html#salt.states.file.directory
18:15 lineman61 joined #salt
18:16 sifusam you're still using file.directory
18:16 * Gareth waves
18:16 lineman60__ joined #salt
18:17 micah_chatt sifusam: you're right, I just didn't get that when you said 'clean' it was an actual param :)
18:17 sixninetynine left #salt
18:18 bretep[hirevue] joined #salt
18:20 bretep[hirevue] So I've run into something that I haven't encountered before. I need to pass in user specific information into a high state to be used in templates. The data changes per user. Specifically the state is being used in a Vagrant setup. So ideally user would create a file called, user_data.yml  and have user specific data in that file. The highstate would load that in as grains or global variables. Thoughts?
18:23 wibberwock if anyone was having similar problems using salt cmd.run in a pillar template, the correct module to use is salt saltutil.cmd
18:23 pipps joined #salt
18:25 bretep[hirevue] Found what I needed. Thanks!
18:25 bretep[hirevue] https://github.com/saltstack/salty-vagrant#miscellaneous Pillar Data
18:27 snikkers joined #salt
18:33 tempspace Has anybody gotten this working? http://salt.readthedocs.org/en/v0.17.1/topics/pillar/index.html#including-other-pillars
18:39 pipps joined #salt
18:39 cnelsonsic joined #salt
18:40 m_george|away joined #salt
18:40 jkleckner joined #salt
18:40 tempspace UtahDave: Is this including and overriding pillar value supposed to be in 0.17? http://salt.readthedocs.org/en/v0.17.1/topics/pillar/index.html#including-other-pillars
18:42 m_george left #salt
18:43 bemehow joined #salt
18:48 craig joined #salt
18:48 mannyt joined #salt
18:51 forrest joined #salt
18:54 ggoZ joined #salt
18:54 ninkotech joined #salt
18:56 jergerber joined #salt
18:56 giantlock joined #salt
18:58 bemehow joined #salt
19:05 noob2 joined #salt
19:06 micah_chatt actually sifusam: clean isn't working quite like I'd liek
19:06 micah_chatt since the files in the directory are dynamic, I don't list them under 'require' in the directory
19:07 noob2 left #salt
19:08 jkleckner1 joined #salt
19:08 berto- joined #salt
19:09 amartin joined #salt
19:10 redondos_ joined #salt
19:10 forrest Does anyone know what Warby Parker's Salt implementation looks like?
19:15 alunduil joined #salt
19:16 forrest Dan Garthwaite are you in here? If so, high five for the comment on 6655
19:21 ashtonian joined #salt
19:27 atx_sono joined #salt
19:27 jslatts joined #salt
19:29 m_george joined #salt
19:29 sifusam @micah_chatt I didn't think it would.  a patch to file.directory to allow for templating would be needed
19:30 __number5__ joined #salt
19:30 _FL1SK joined #salt
19:30 UtahDave tempspace: it should be. The doc says it's new in 0.16.0, so it should be in 0.17.1
19:30 tempspace UtahDave: the docs said it was in 0.16.0 and it wasn't
19:31 tempspace someone merged in the docs before the code last time
19:31 redondos joined #salt
19:31 UtahDave tempspace: Hm. I'm not sure on that.
19:31 tempspace and it's still not working for me in 0.17.1, but the example it gives isn't the greatest, so I don't know if a) it's there or b) if I'm doing it right
19:31 mgw joined #salt
19:32 indymike joined #salt
19:33 jeblair joined #salt
19:33 alunduil joined #salt
19:34 oberstille joined #salt
19:34 micah_chatt_ joined #salt
19:35 scalability-junk joined #salt
19:36 ajw0100 joined #salt
19:37 amartin left #salt
19:38 adamisos joined #salt
19:40 bwq so uhm
19:40 bwq gitfs, is that only for a master-minion config?
19:40 forrest it works on a masterless minion bwq
19:40 forrest I'm using it for my blog
19:40 bwq cool, thanks
19:40 forrest yep, let me know if you need an example
19:41 bwq there's so much stuff in Salt that i find it pretty overwhelming
19:41 forrest Yea it takes a bit to get the hang of
19:41 forrest but once you get that you're up to speed pretty quickly
19:41 godber joined #salt
19:41 bwq yeah, just a lot of new concepts
19:42 forrest Have you used config management before?
19:42 bwq states, pillars, formulas, etc.
19:42 bwq nope
19:42 bwq but Salt looked really nice
19:42 forrest ahh yea, that makes it a bit tough then
19:42 forrest it is
19:42 bwq jinja2 in the templates is very nice and yaml makes the states very easy to look at and understand what they're doing
19:42 carmony formulas are just sets of states, correct?
19:42 monokrome joined #salt
19:43 Nazca joined #salt
19:43 bwq yeah
19:43 forrest no, formulas are pre-written salt states.
19:43 bwq i'm sure there's a small difference.. somewhere
19:43 bwq what does pre-written mean?
19:43 forrest they don't have to include multiple states if that isn't required
19:43 forrest plug and play
19:44 bwq but aren't lots of salt states plug 'n play?
19:44 forrest yes, but formulas aim to be OS agnostic most of the time
19:44 carmony oh, so they added a formal way of doing it
19:44 carmony in .17
19:44 forrest carmony, ?
19:44 forrest is there actually documentation on it?
19:45 forrest I just started copying what whiteing did because it's awesome, and that's what the existing formulas were getting changed to
19:45 forrest and I apologize bwq, I thought carmony said just 'solo' states
19:45 bwq yeah, there's docs on it :p
19:45 bwq a single page iirc
19:45 bwq np
19:45 forrest yea there is http://docs.saltstack.com/topics/conventions/formulas.html
19:45 forrest but that doesn't talk about writing formulas
19:45 forrest oh it does now
19:45 forrest sweet
19:45 forrest high five to whoever made that commit
19:47 carmony I should spruce up your guy's php formula :)
19:47 bwq so hmm, with a masterless minion, does the local /etc/salt/master act as the "master" config?
19:47 forrest carmony, yea I'm sure the salt guys would love that. It doesn't have multi-OS support right now it looks like
19:47 bwq as in, can i add my file_roots to that?
19:48 forrest bwq, it would be in /etc/salt/minion
19:49 forrest there is already a pre-built commented out section for the file_roots in there as well
19:49 bwq ah cool, indeed
19:49 bwq figured it'd act as the local master
19:49 forrest yea, I haven't tested how that works, so I don't know if it just operates just like on the master, but I'd imagine it does
19:53 aserdp joined #salt
19:55 pipps joined #salt
19:55 dave_den joined #salt
19:55 ctdawe joined #salt
19:55 pkimber joined #salt
19:57 gmoro joined #salt
20:01 mapu joined #salt
20:02 toofer joined #salt
20:06 aserdp_ joined #salt
20:09 hjubal joined #salt
20:09 hjubal joined #salt
20:16 diegows joined #salt
20:17 mr_chris So now that I
20:18 mr_chris Err...pressed enter too soon.
20:19 mr_chris So now that I've got a rather large-ish salt deployment I'm finding that "salt '*' state.highstate" from the master no longer returns what happening on the minions. It just times out with no data. I can do it on smaller groups of servers or on specific sls files.
20:19 pdayton joined #salt
20:19 redondos joined #salt
20:22 supplicant mr_chris: http://docs.saltstack.com/ref/cli/salt.html
20:22 supplicant you probably want the --timeout option or maybe --async
20:23 mr_chris supplicant, I did increase the timeout in the master settings conf but I haven't tried it as a flag yet. It may need to be increased some more. I haven't tried --async yet.
20:23 mr_chris Thanks.
20:24 KyleG joined #salt
20:24 KyleG joined #salt
20:24 redondos joined #salt
20:27 vipul joined #salt
20:27 mhayden joined #salt
20:29 mhayden just started using salt today and i'm having a bunch of trouble getting augeas.setvalue working: http://www.fpaste.org/50036/82992145/
20:30 mhayden but augeas is installed and available
20:30 mhayden having a very similar issue to https://github.com/saltstack/salt/issues/6082 but i'm on CentOS 6
20:31 amckinley joined #salt
20:32 bitz joined #salt
20:37 Corey UtahDave: OOH you're near Park City!
20:37 Corey I know what I'll be doing out of hours at SaltConf!
20:39 mmarch joined #salt
20:39 cowmix3 joined #salt
20:45 Ahlee Have fun, I'll be belly'd up at the bar, as per usual.
20:45 cewood joined #salt
20:46 Ahlee suppose i should finish registering
20:47 lyddonb joined #salt
20:48 ricicle left #salt
20:48 bemehow joined #salt
20:48 timoguin joined #salt
20:50 supervacuo joined #salt
20:50 KyleG joined #salt
20:50 KyleG joined #salt
20:50 supervacuo left #salt
20:50 supervacuo joined #salt
20:51 supervacuo Hi all... been running salt for a while, and just installed minion on my Raspberry Pi (ARM SOC), running Fedora 18. I get "Function pkg is not available" trying to run pkg.update, and no other log output. Any hints?
20:52 bemehow joined #salt
20:52 supervacuo (test.ping works fine...)
20:57 Gareth Corey: You're not going to be running around Park City scaring little old ladies again, are you?
20:57 pears supervacuo: does "salt $minion sys.doc pkg" work?
20:57 druonysuse joined #salt
20:57 druonysuse joined #salt
20:57 mackstick joined #salt
20:57 alekibango joined #salt
20:58 pipps joined #salt
20:58 ashtonian joined #salt
21:02 bemehow joined #salt
21:02 rlarkin joined #salt
21:03 Fandekasp joined #salt
21:05 Corey Gareth: Are you kidding me? I'll be at High West!
21:05 Corey Ahlee: ^
21:05 supervacuo pears: ty for reply. not really. it gives output for "pkg_resource", e.g. add_pkg, check_targets, compare, find_changes etc.
21:06 supervacuo (but crucially no "install", "upgrade")
21:06 UtahDave Corey: Yep!  Park City isn't very far
21:07 supervacuo I'm on salt-minion-0.12.0-1 -- is this too old? I notice I'm on at least 0.16 elsewhere
21:07 UtahDave supervacuo: yeah, 0.12.0 is VERY old
21:07 supervacuo UtahDave: OK, perhaps that's it, then.
21:08 * supervacuo sets about recompiling 0.16 for ARM
21:08 UtahDave supervacuo: follow these instructions for Debian Wheezy on your pi
21:08 UtahDave http://docs.saltstack.com/topics/installation/debian.html#wheezy-stable
21:08 UtahDave I did this last week and it worked perfectly
21:08 amckinley joined #salt
21:08 supervacuo UtahDave: Thanks! Sadly, I'm using Fedora (Pidora). Surplus of masochism, perhaps...
21:09 UtahDave supervacuo: OK, well you could clone the saltstack git repo and install over the top.
21:09 Ahlee I don't know why, but I thought you were referring to a ski resort Corey
21:09 Ahlee no idea where that came from.
21:09 supervacuo UtahDave: Seems the work packaging for Fedora has been done already; F19 repos have 0.17. I just need to rebuild on the Pi...
21:10 supervacuo UtahDave: many thanks for the suggestions tho!
21:12 Gareth supervacuo: pip install salt works well on the Pi too, so that might be an option :)
21:13 apergos joined #salt
21:13 supervacuo Gareth: If only I weren't such a distribution package zealot :P
21:14 supervacuo Gareth: also, in before pyp2rpm ;)
21:14 supervacuo (which I somehow doubt would work very well for salt, anyway)
21:14 supervacuo Cheers for help, all!
21:16 Gareth supervacuo: use what works :)
21:19 aserdp joined #salt
21:19 ajw0100 joined #salt
21:20 auser joined #salt
21:21 ajw0100 joined #salt
21:22 aserdp_ joined #salt
21:23 bretep[hirevue] Anyone know how to set a variable in a jinja2 template that can be accessed in a for loop?
21:25 Gareth {% set iirc
21:26 Gareth bretep[hirevue]: {%- set file = salt['pillar.get']("available_zones:" + key + ":file") %}
21:26 mgw joined #salt
21:26 pdayton joined #salt
21:27 enertek joined #salt
21:27 Ahlee joined #salt
21:28 KyleG joined #salt
21:28 KyleG joined #salt
21:28 mwillhite joined #salt
21:29 jcsp when I listen for events using MasterEvent.iterevents, and specify a tag, is the tag only an exact match or is any globbing possible?  Tried something with an appending asterisk and it didn't seem to match anything, but optimistically wondering anyway?
21:29 pdayton joined #salt
21:29 bretep[hirevue] Gareth: here is an example http://pastebin.com/PAfe01Gi
21:30 bretep[hirevue] I get the "Undefined jinja variable" exception
21:30 rgarcia_ joined #salt
21:30 foxx joined #salt
21:30 krissaxton joined #salt
21:31 rgarcia_ joined #salt
21:35 jpcw joined #salt
21:36 Gareth bretep[hirevue]: not sure if it matters...but your set might need the {%-
21:39 pears bretep[hirevue]: what line does line 50 translate to in this pastebin?
21:39 bretep[hirevue] Line 7:     - name: echo {{ my_needed_var }}
21:40 pears forgive the obvious question but where is my_needed_var supposed to come from?
21:41 Gareth pears: he has a set line.
21:41 pears ohh I see it
21:41 bretep[hirevue] Line 2
21:41 bretep[hirevue] http://pastebin.com/AUG6mQxS
21:41 pears sorry, missed that
21:41 bretep[hirevue] No problem
21:42 newbie-lurker joined #salt
21:42 Gareth thats weird.
21:42 Gareth bretep[hirevue]: you're code ran fine here.  what version of salt?
21:43 CerealBox joined #salt
21:43 enertek hi all, is there any mechanism for inheritance or overriding in pillars? even on a file level rather than an id level (i.e., if multiple pillar_roots are defined, should the 'first found' be used?)
21:43 bretep[hirevue] Gareth salt 0.17.0-1610-g614a009
21:44 oz_akan_ joined #salt
21:44 Gareth bretep[hirevue]: 0.17.1 here.  is that the whole file?
21:45 bretep[hirevue] That's not the whole file... Yeah, when I isolate it to that it works
21:45 danielbachhuber joined #salt
21:47 sifusam joined #salt
21:51 Gareth bretep[hirevue]: can you pastebin the whole file?
21:52 krissaxton joined #salt
21:55 pipps joined #salt
21:58 kermit joined #salt
21:59 rgbkrk joined #salt
22:02 pentabular left #salt
22:02 dvogt joined #salt
22:03 adamisos joined #salt
22:06 xmltok_ joined #salt
22:07 jslatts joined #salt
22:08 pipps joined #salt
22:10 bhosmer joined #salt
22:10 adamisos joined #salt
22:19 goki joined #salt
22:19 scalability-junk joined #salt
22:20 octarine joined #salt
22:22 cbloss joined #salt
22:24 aserdp joined #salt
22:31 aserdp_ joined #salt
22:31 Gifflen joined #salt
22:32 alunduil joined #salt
22:33 auser joined #salt
22:33 alunduil joined #salt
22:35 ctdawe joined #salt
22:54 aserdp joined #salt
22:59 dvogt joined #salt
23:00 c0bra joined #salt
23:04 c0bra joined #salt
23:09 bwq so uh
23:09 bwq GitPython==0.3.2.RC1
23:09 bwq Git fileserver backend is enabled in configuration but GitPython version is not greater than 0.3.0, version 0.1.6 detected
23:09 bwq loading returner in ['/var/cache/salt/master/extmods/returners', '/usr/lib/python2.6/dist-packages/salt/returners'
23:09 bwq why is my salt installed in python2.6?
23:12 MTecknology joined #salt
23:13 pmrowla joined #salt
23:19 c0bra joined #salt
23:21 ajw0100 joined #salt
23:24 mwmnj joined #salt
23:25 krissaxton joined #salt
23:30 jdenning joined #salt
23:31 ajw0100 joined #salt
23:32 cachedout joined #salt
23:33 fragamus joined #salt
23:40 lesnail joined #salt
23:41 rgarcia_ joined #salt
23:42 CheKoLyN joined #salt
23:42 bemehow joined #salt
23:51 xmltok joined #salt

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