Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-05-30

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

All times shown according to UTC.

Time Nick Message
00:17 onlyanegg joined #salt
00:44 johnkeates joined #salt
00:47 tracphil joined #salt
00:52 DEger joined #salt
00:53 dyasny joined #salt
00:54 DEger joined #salt
00:57 euidzero joined #salt
01:00 euidzero joined #salt
01:02 euidzero joined #salt
01:04 euidzero joined #salt
01:08 fwoosh joined #salt
01:08 fwoosh anyone know how I can do the equivalent of a "chmod go=u-w" kind of thing?
01:08 euidzero joined #salt
01:09 fwoosh I have a tarball that I need to extract, the source tarball is out of my control, and I need to set the permissions on a fair number of files within
01:09 catpiggest joined #salt
01:11 amcorreia joined #salt
01:25 euidzero joined #salt
01:27 euidzero joined #salt
01:28 euidzero joined #salt
01:47 ilbot3 joined #salt
01:47 Topic for #salt is now Welcome to #salt! | Latest Versions: 2015.5.10, 2015.8.10, 2016.3.0 | Support: https://www.saltstack.com/support/ | Logs: http://irclog.perlgeek.de/salt/ | Paste: https://gist.github.com/ (please don't multiline paste into channel) | See also: #salt-devel, #salt-offtopic | Ask with patience as we are volunteers and may not have immediate answers
01:54 favadi joined #salt
01:56 k_sze[work] joined #salt
01:57 beardedeagle joined #salt
01:57 catpig joined #salt
02:02 catpig joined #salt
02:13 tracphil joined #salt
02:14 hasues joined #salt
02:14 hasues left #salt
02:14 johnkeates joined #salt
02:15 favadi joined #salt
02:26 danslimmon joined #salt
02:28 danslimmon Sup? I'm trying to run unit tests on Ubuntu 16.04, Python 2.7.11, Salt v2015.8.10, and almost all my tests are failing with MagicMock "calls not found" AssertionErrors
02:28 danslimmon Here's an example (of a new test I just wrote): https://gist.github.com/danslimmon/f3002f225c7766e93e8d5138315e3021
02:29 danslimmon It almost seems like MagicMock behavior changed since tests.unit.modules.mysql_test was written? I dunno what I'm missing
02:33 rgrinberg joined #salt
02:33 danslimmon Oh, I guess this should be in #salt-devel. My bad.
02:35 cyborg-one joined #salt
02:35 danslimmon_ joined #salt
02:41 k_sze[work] joined #salt
02:48 tuxx joined #salt
02:51 favadi joined #salt
03:09 sjorge joined #salt
03:10 ramteid joined #salt
04:02 Sammichmaker joined #salt
04:08 lorengordon joined #salt
04:27 keimlink joined #salt
04:27 mannynotfound joined #salt
04:31 Sammichmaker joined #salt
04:36 rdas joined #salt
04:44 chrismoos left #salt
04:51 racooper joined #salt
04:54 armyriad joined #salt
04:57 noraatepernos joined #salt
05:12 netcho joined #salt
05:16 mowntan joined #salt
05:21 evle joined #salt
06:05 felskrone joined #salt
06:06 obimod joined #salt
06:09 DEger joined #salt
06:15 ravenx joined #salt
06:18 colttt joined #salt
06:33 DEger joined #salt
06:33 duncanmv joined #salt
06:38 DEger joined #salt
06:42 dmaiocchi joined #salt
06:50 dmaiocchi joined #salt
06:54 fracklen joined #salt
06:58 fracklen joined #salt
06:59 martoss joined #salt
07:00 deniszh joined #salt
07:02 upb any way to save output from salt command to a file per minion?
07:02 upb in case multiple minions are targeted
07:05 saltuser joined #salt
07:06 KermitTheFragger joined #salt
07:09 linjan_ joined #salt
07:10 DEger joined #salt
07:10 dynamicudpate joined #salt
07:12 oida joined #salt
07:15 ivanjaros joined #salt
07:17 josuebrunel joined #salt
07:19 DEger joined #salt
07:21 kbaikov joined #salt
07:22 Sylvain31 joined #salt
07:22 kawa2014 joined #salt
07:23 slav0nic joined #salt
07:29 DEger joined #salt
07:29 kbaikov joined #salt
07:35 joakkinen joined #salt
07:41 DEger joined #salt
07:44 toanju joined #salt
07:44 lero joined #salt
07:47 majikman joined #salt
07:47 dgutu joined #salt
07:48 manji joined #salt
07:52 watersoul joined #salt
07:56 krymzon joined #salt
07:57 saltsa joined #salt
07:58 cwright joined #salt
07:58 XenophonF joined #salt
08:01 mackripeum joined #salt
08:10 kawa2014 joined #salt
08:12 favadi joined #salt
08:19 bdrung_work joined #salt
08:19 GreatSnoopy joined #salt
08:21 CeBe1 joined #salt
08:21 s_kunk joined #salt
08:25 punkoivan joined #salt
08:29 N-Mi joined #salt
08:42 Sylvain31 hi
08:42 colinstgeorge joined #salt
08:42 Sylvain31 I would like to manage my customer with a higher level pillar, like this one: http://hastebin.com/aduyuceqap
08:42 oida joined #salt
08:43 baffle joined #salt
08:43 babilen pillarstack
08:43 babilen https://github.com/saltstack/salt/issues/23910#issuecomment-112355187
08:43 saltstackbot [#23910]title: Please implement static pillars | Hi,...
08:44 Sylvain31 https://github.com/bbinet/pillarstack
08:46 Sylvain31 babilen: thanks, but I can't see the "why" of static pillars, yet. It could be acheived with any kind of pillar, no?
08:48 jhauser joined #salt
08:49 N-Mi joined #salt
08:49 N-Mi joined #salt
08:50 kawa2014 joined #salt
08:51 babilen Sylvain31: Pillars can't be self-referential ... And the question is *what* do you want to achieve? I simply used my crystal ball to guess your question, but what is your actual question?
08:54 Sylvain31 babilen: I do an exstensive use of my brain, and I sometime forgot some already exposed question/affirmation. I guess we do already spoke about that… I already saw that link. I remember now, than my brain is warmer… ;) your crystal ball is a souvenir. :))
08:55 babilen Hehe
08:55 babilen You can't express "Use foo:bar as mysql:bar pillar value" in normal pillars
08:56 babilen This would be a very nice feature to have as it would allow you to design specific pillars for a project, while still using "standard" formulas that live in a more global namespace
09:05 pseudonymous joined #salt
09:05 yuhlw joined #salt
09:14 fracklen joined #salt
09:14 fracklen joined #salt
09:18 permalac joined #salt
09:18 pseudonymous joined #salt
09:20 felskrone can i check for the existence of a custom (which might have not been synced yet) in a state?+
09:21 jamesog A custom what?
09:23 felskrone err, good question, module…
09:23 yuhlw Hi, Is it possible to have a file managed which contains a section that can be modified locally. This needs come frome the /etc/ssh/sshd_config which does not support ".d" config files and where we would like to let the sysadmins to add some content.
09:24 jamesog You can try: if 'module' in __salt__
09:25 felskrone ill try that, sec
09:25 jamesog It'll probably have to be: if 'module.function' in __salt__ actually
09:26 jamesog yuhlw: Uhh, I don't think that's possible. Can't the sysadmins just make a commit to the template?
09:27 yuhlw jamesog: not really.
09:27 jamesog As a sysadmin, that's how I do it *shrug*
09:29 felskrone jamesog: you mean like this, right? https://gist.github.com/felskrone/3c14e9952ded7883e8ae7511ebb8e849
09:29 ravenx left #salt
09:30 yuhlw jamesog: can you think of any workaround to do such job ?
09:30 yuhlw such task !
09:30 yuhlw left #salt
09:30 yuhlw joined #salt
09:31 netcho how can i setup salt-cloud to use custom minion bootstrap script from master?
09:32 jamesog felskrone: Yeah I guess
09:33 jamesog yuhlw: Not really. The point of config management is to keep the system in a consistent state. Local modifications aren't "consistent".
09:33 jamesog Potentially you could store the local mods in another file and use file.append
09:34 jamesog that's a hack though
09:35 felskrone jamesog: got it, {% if not salt.get('proftpd.regen_all', False) %}, that works, but the module is not available in the same run, too bad
09:36 Antonio_ joined #salt
09:37 jamesog Yeah because jinja already rendered the state file
09:38 Antonio_ HI
09:38 Antonio_ i need a small help
09:38 Antonio_ how i can run a program like "C:\Program Files (x86)\JetBrains\PhpStorm 8.0.3\bin\PhpStorm.exe" with GUI form cmd.run
09:39 Antonio_ i see program has process on Taskmanager (windows) but no GUI is displayed for current logged in user
09:41 tracphil joined #salt
09:44 GothAck joined #salt
09:49 toanju joined #salt
09:52 Amnesia question, is there anything that has been changed since 2016.3.0 in terms of delivering grains to pillar files?
09:59 Amnesia in my current pillar, __grains__ is empty
09:59 Amnesia using 2016.3.0 *
10:00 MadHatter42 joined #salt
10:04 mage_ joined #salt
10:12 Amnesia https://docs.saltstack.com/en/latest/topics/development/dunder_dictionaries.html#grains states that __grains__ should be available in external pillars
10:14 patarr joined #salt
10:22 fredvd joined #salt
10:55 permalac Hello guys, I would like to install networker with salt, but there are many interactions to take in account.
10:56 permalac can I use expect with salt ?   Or something alike. ?
10:56 punkoivan joined #salt
10:57 SWA permalac: yeah, just use expect with a response file
10:58 punkoivan joined #salt
10:58 permalac I think I have jumped over some part of the docs, I'll check .
10:58 SWA permalac: i don't think there is a expect module, but you can manage the response file and use cmd.run
10:59 permalac https://docs.saltstack.com/en/latest/ref/states/all/salt.states.cmd.html   OK!
10:59 permalac SWA, thanks.
11:04 mage_ what's the best/most secure option to auto-accept minion keys for instances (VM, ...) created "on-the-fly" ?
11:04 mage_ I used the reactor with a custom event tag but I wonder if it's the best
11:10 SWA mage_: considered using salt-cloud? It takes care of accepting a new minion automatically
11:11 mage_ I never used salt-cloud, does it works for FreeBSD jails ?
11:11 SWA dunno, I used it on RHEL only
11:14 fracklen joined #salt
11:14 DEger joined #salt
11:14 fracklen joined #salt
11:17 amcorreia joined #salt
11:23 toanju joined #salt
11:25 bmcorser hey guys
11:25 bmcorser i'm getting an error message when i try to use a runner i wrote
11:25 bmcorser 'myrunner.function' is not available
11:26 bmcorser i have runner_dirs configured in /etc/salt/master
11:26 bmcorser and i can see the myrunner.py in the directory that i have configured
11:26 bmcorser do i need to refresh something?
11:26 bmcorser i've tried service salt-master restart a few times
11:27 bmcorser (i'm running on ubuntu, salt version 2015.8.10)
11:29 bmcorser ahfuk i was missing a dep, running with -l debug showed me
11:32 lilvim joined #salt
11:34 bernieke joined #salt
11:37 bernieke joined #salt
11:37 bernieke left #salt
11:38 jhauser joined #salt
11:52 av_ joined #salt
12:02 evle1 joined #salt
12:04 west575 joined #salt
12:09 josuebrunel joined #salt
12:11 JohnnyRun joined #salt
12:12 kbaikov hello
12:13 kbaikov should all states in salt that accept a boolean "True" also accept "yes", "on", "1" ?
12:13 Amnesia babilen: https://github.com/saltstack/salt/issues/33614 that's similar to the thing I experienced last friday
12:13 saltstackbot [#33614]title: grains.has_key()  always returns false in 2016.3.0 | ### Description of Issue/Question...
12:13 Amnesia grains == {} -> empty
12:13 Amnesia in the context of the pillar
12:14 Amnesia so the pillar doesn't have access to the grains data of the minion
12:16 babilen Amnesia: Well, not quite .. Knuta's problem is that specifically the .has_key() dictionary method isn't working
12:17 Amnesia are you sure?
12:17 babilen No
12:17 babilen I'm sure of very few things
12:18 babilen But given that ".. is defined" works, it seems to be a different issue
12:28 Amnesia hm teh wonky
12:28 Amnesia I'll ask one of the devs later on
12:38 cyborg-one joined #salt
12:43 abednarik joined #salt
12:55 evle joined #salt
13:02 Sylvain31 I'm trying to configure pillarstack. I produced http://hastebin.com/jezuvotaqe How do I get at line 65, the sls data produced line 12? I can fetch  "salt 'saltm*' config.get wsf:customers:client3" which is OK, customer.sls is loaded. In master log I don't have any more error, but "salt '*' config.get mysql" return "database: None"
13:03 dyasny joined #salt
13:04 jad_jay joined #salt
13:04 jad_jay Hello
13:04 jad_jay Is it possible to comment out line in init.sls files ?
13:04 jad_jay with #
13:05 jad_jay or is there another comment system like /**/
13:05 abednarik joined #salt
13:05 Sylvain31 jad_jay: yes but sls are jinja processes before… be careful
13:05 jad_jay yes so HOW..
13:05 jad_jay Sylvain31, How could you comment lines ?
13:06 Sylvain31 jinja comment style: {# #}
13:06 jad_jay Sylvain31, thanks a lot
13:06 Sylvain31 yaml comment style # to end of line
13:07 Sylvain31 jad_jay: you may learn a better editor, if yours dont know how to comment a block with a prefix ;)
13:08 jad_jay Sylvain31, vim is in my soul !
13:10 babilen jad_jay: Embrace spacemacs! It's a better vim!
13:10 babilen ;)
13:12 ivanjaros joined #salt
13:15 Sylvain31 vim is good for visual block so instering bunch of # would be easy ;)
13:17 Ayo joined #salt
13:19 dh__ joined #salt
13:25 mschiff I have this in a state: {% for name in salt['pillar.get']('service:foo',{}) %} -- but when foo happens to be an empty list, I get a type error: "TypeError: 'NoneType' object is not iterable", why?
13:26 AndreasLutro are you sure foo is an empty list and not null? are you sure your type error is coming from that exact line of code?
13:28 mschiff AndreasLutro: yes. The problem sems to be that there is y key "foo", but withount any values (because the minion doe snot matxh any of the jinja if/then/else code)
13:29 mschiff so the pillar data is just "foo:"
13:29 mschiff (or "service:\n    foo:")
13:30 babilen That would make it None, not an empty list
13:30 babilen (the value that is)
13:30 AndreasLutro if you want an empty list you need to specify it as such
13:30 mschiff yes thats my problem, but how to add an empty list in a jinja else here?
13:30 babilen foo: []
13:30 mschiff yes but that does not fit the stuff with jinja
13:30 pseudonymous joined #salt
13:31 babilen But then .. why don't you make it an empty dictionary as that is what you default to in the pillar lookup (or the other way round)
13:31 mschiff where I have "foo:" and then "{% if blah ..." on subsequent lines
13:31 babilen Could you provide a little more information/context ?
13:31 mschiff I have:
13:31 mschiff foo:
13:31 babilen Such as the surrounding lines .. I am not sure why there is a need to define foo: at all in your pillar if it is emptyu
13:32 mschiff then "{% if grains.id == 'blah' %}- bar{% endif %} for example
13:33 mschiff I start "foo:", then have jinja if/then/else to have different lists for each minion
13:33 babilen jfindlay: 2016.3.0 has been uploaded to unstable (including the master fix)
13:33 mschiff and I wanted to avoid to repeat "foo:" for every miion so that visble for any minion
13:34 mschiff But I now understood whats happening, so I could rearrange everything a bit...
13:34 mschiff thanks AndreasLutroand babilen
13:34 babilen All the best! :)
13:40 abednarik joined #salt
13:49 abednarik joined #salt
13:51 dh__ Does anyone use salt with CoreOS? I am rather puzzled by whether this would work or if I am barking up the wrong tree since CoreOS relies so heavily on cloud-config. Maybe I install salt-minion in a priviledged container?
13:51 permalac What is the best / easiest way to document an infrastructure before making changes with salt?
13:52 XenophonF if you've got no documentation at all, maybe write salt states that reflects your existing config
13:52 XenophonF throw those into git
13:52 XenophonF or DVCS of your choice
13:53 XenophonF make sure the state applies without making changes e.g., salt minion-id state.sls sls-id-here test=True shouldn't show any changes
13:54 permalac XenophonF, understood. nice approach.
13:54 XenophonF make the desired changes to the salt state, commit to git (or whatevs) repo, push to servers, etc.
13:55 XenophonF that documents config as salt states, with change history provided by DVCS
13:59 Turl joined #salt
13:59 Turl moin
14:00 Turl is there any known issues re. state_verbose / state_output? I'm trying to reduce the amount of output generated by salt, but so far all the combinations of terse/mixed and False have had no effect
14:01 Turl running master 2016.3.0 on ubuntu xenial fwiw
14:03 Sylvain31 how to fetch my pillar value in jinja + pillarstack? line 65? http://hastebin.com/jezuvotaqe
14:04 XenophonF Sylvain31: can't read it b/c web site requires javascript
14:05 XenophonF but are you trying to access a pillar key via jinja from a pillar sls file?
14:05 XenophonF if so that's not going to work
14:05 Sylvain31 XenophonF: javascript?
14:05 XenophonF yeah hastebin requires javascript enabled to view the site
14:05 Sylvain31 XenophonF: shit!
14:05 Sylvain31 ;)
14:05 XenophonF call me paranoid but i usually browse without javascript enabled :)
14:06 Sylvain31 XenophonF: paste.debian.org fail with command line :( this morning
14:06 XenophonF that sucks :(
14:06 Sylvain31 which one are you able to see?
14:06 Turl Sylvain31: sprunge.us usually works fine
14:07 XenophonF hm, why don't you try http://ix.io/
14:07 XenophonF you can pipe to the web with that one, too
14:08 XenophonF or maybe gist.github.com?
14:08 XenophonF paste.debian.net seems to be working for me
14:08 abednarik joined #salt
14:08 Sylvain31 XenophonF: because I try to get all those Jinja yaml python stuff to fit in my brain, it seems hard to put into… ;) so I salted on command line paster. too bad.
14:08 Sylvain31 http://paste.debian.net/712533/
14:09 Sylvain31 copy pasted, with silly line number :~(
14:10 XenophonF it's ok
14:10 XenophonF i'm not sure you can normally put jinja into .yaml files
14:10 Sylvain31 Turl: we dont want pub, don't we? ;)
14:11 Turl Sylvain31: hm?
14:11 Sylvain31 XenophonF: it's pillarstack input, ./customers/db_database.yaml ?
14:12 Sylvain31 Turl: sprunge.us is for pastebin, right? there is pub on it, if I'm right.
14:13 Sylvain31 XenophonF: gist.github.com is using javascrit? right?
14:13 Turl Sylvain31: yes. What's pub?
14:13 XenophonF right but if you look at the pillarstack code, https://github.com/bbinet/pillarstack/blob/master/stack.py#L56
14:13 XenophonF you'll see that it's calling yaml.safe_load()
14:13 XenophonF as opposed to calling the jinja renderer first, and then calling yaml.safe_load on resulting data
14:14 Sylvain31 Turl: "BitBucket - Backup your code in the cloud! Host unlimited private projects, for free. SIGN UP takes 10 seconds, and it's free! " http://pastebin.com/
14:14 XenophonF let me guess you're getting the error "Can't parse as a valid YAML dictionary", right?
14:14 Turl Sylvain31: on sprunge.us? I don't get any of that there
14:15 mohae joined #salt
14:15 tristianc joined #salt
14:15 Sylvain31 Turl: OK, never used, I though it was a front end for pastebin.com, I'm lost, I may try, thanks. ;)
14:16 Turl Sylvain31: ah, no, it's a separate paste site
14:16 Sylvain31 XenophonF: no, I got nothing…
14:16 Sylvain31 I'm trying to configure pillarstack. I produced http://hastebin.com/jezuvotaqe How do I get at line 65, the sls data produced line 12? I can fetch  "salt 'saltm*' config.get wsf:customers:client3" which is OK, customer.sls is loaded. In master log I don't have any more error, but "salt '*' config.get mysql" return "database: None"
14:18 XenophonF well like i said, it's unlikely that yaml file is being rendered using jinja prior to loading
14:18 Sylvain31 XenophonF: OK, my value is may be in stack dictionary?
14:19 XenophonF possibly?
14:19 Sylvain31 https://github.com/bbinet/pillarstack#pillarstack
14:19 Sylvain31 multiple config files that are jinja2 templates with support for pillar, __grains__, __salt__, __opts__ objects ??
14:19 Sylvain31 is it not the purpose?
14:20 XenophonF this is the first i've heard of it, so i'm guessing here
14:20 Sylvain31 XenophonF: I gonna try to have the debug log above…
14:21 XenophonF just based on skimming the code and docs
14:21 babilen This is an external pillar that has little to do with the "normal" pillars and are simply yaml files that are rendered in a specific order and merged according to specific rules
14:21 babilen (pillarstack that is)
14:21 Sylvain31 XenophonF: you line of code L56 seem the missing point I was looking for, lets
14:21 Sylvain31 try.
14:24 rubenb joined #salt
14:25 Sylvain31 no, no output for "log.debug('YAML: basedir={0}, path={1}'.format(basedir, path))" :~(
14:25 abednarik joined #salt
14:25 pseudonymous left #salt
14:26 Sylvain31 with salt-call -lall should I have it?
14:28 tristianc joined #salt
14:31 Sylvain31 ok, enabled debug level in /etc/salt/master I got some more log…
14:31 Sylvain31 including "salt.loaded.ext.pillar.stack][DEBUG   ][3855] YAML: basedir=/srv/salt/pillar/customers, path="
14:32 DanSlimmon joined #salt
14:33 morissette joined #salt
14:34 Sylvain31 XenophonF: thanks for you link to the code, I parser don't seem to support comment "#" 8(
14:38 justanotheruser joined #salt
14:40 spuder joined #salt
14:42 impi joined #salt
14:43 fracklen_ joined #salt
14:47 pkruithof joined #salt
14:48 Sylvain31 XenophonF: pillar is passed as jenv on line https://github.com/bbinet/pillarstack/blob/master/stack.py#L46 it should be available in jinja as pillar, right?
14:48 pkruithof joined #salt
14:49 numkem joined #salt
14:50 pkruithof Anyone willing to help me with an issue I'm not able to figure out? For the past 4 hours I've been trying to upgrade to 2016.3 (from 2015.8), but I cannot get my custom modules to work
14:50 Sylvain31 found, "{%- for name, client in pillar['wsf']['customers'].items() %}" it wor
14:50 Sylvain31 works
14:56 babilen pkruithof: What do they do?
14:56 pkruithof a couple of utility functions that I needed in my states
14:56 pkruithof they've worked just fine for a couple of years, until now, they don't seem to load anymore
14:57 babilen Sylvain31: I'd make that salt['pillar.get']('wsf:customers', {}).items() as that won't break if 'wsf' or 'customers' is undefined
14:57 babilen pkruithof: Well, I have to leave very soon, but it is impossible to help without additional information (e.g. a simple reproducible test case, log entries, output that shows that are not being loaded, ...)
14:58 pkruithof I know, just wanted to ping someone first. Here goes:
14:59 pkruithof I'm running a masterless minion inside a Vagrant VM (previously Ubuntu 15.10, upgrading to 16.04 LTS now).
14:59 pkruithof The minion config has the following line to look for extra modules (note that the path is not standard): module_dirs: [- /srv/salt/stack/_modules]
15:00 babilen You might also want to consider filing a bug on https://github.com/saltstack/salt/issues that could serve as a point of reference
15:00 babilen Didn't the module loading paths change in 2016.3.0 ?
15:00 babilen https://docs.saltstack.com/en/develop/topics/releases/2016.3.0.html#backwards-incompatible-changes
15:01 pkruithof That mentions extension_modules, I'm using module_dirs. I'm not sure what the difference is, but module_dirs does not seem deprecated or anything
15:01 babilen right
15:01 babilen It was more a "I read something related to changed module paths in the release notes ..." though
15:01 babilen t
15:02 pkruithof I know, that was the first place I looked
15:02 babilen Cool
15:02 pkruithof I've consulted the docs and GH issues, but can't find anything related so far..
15:03 pkruithof and I don't know how to debug this: I've set the log level to debug and inspected the log files, but I don't see any lines about loading modules
15:04 Sylvain31 babilen: tested that previously "__salt__['pillar.get']('wsf:customers', {}).items()"  doesn't return anything
15:08 Sylvain31 pkruithof: log debug in which config?
15:08 pkruithof in /etc/salt/minion
15:09 Sylvain31 pkruithof: OK
15:09 abednarik joined #salt
15:10 Sylvain31 is it possible to trace salt evaluation with a debugger?
15:10 pkruithof I have no idea how I would do that
15:11 Sylvain31 this multiple evaluation chain make me crazy… :\
15:13 pkruithof I wanted to list all the loaded modules by doing "salt-call sys.list_modules", but it seems like a couple of builtin modules are not listed in the output from that, so I'm wondering if that's the right command to check it.
15:14 babilen Sylvain31: Note my usage of salt
15:14 babilen pkruithof: You could stop the master service and then run "salt-master -ldebug"
15:14 babilen So .. I'm off now. All the best!
15:14 pkruithof thank you
15:15 Sylvain31 babilen: sorry I dont see the point, are you speaking about sal['pillar.get'] ?
15:15 babilen I am
15:15 babilen salt, not __salt__
15:16 Sylvain31 babilen: tested, it doesn't work, I don't know why… :~(
15:16 babilen It would be equivalent
15:16 babilen (but safe)
15:16 Sylvain31 babilen: not looking at : https://github.com/bbinet/pillarstack/blob/master/stack.py#L46
15:16 babilen As I said: I'm off now ;)
15:16 DEger joined #salt
15:16 Sylvain31 do speak about that? I got an error for salt[]
15:16 babilen Oh, and I wouldn't expect salt execution modules to be available in external pillars (of which pillar.get is one)
15:17 babilen Anyway .. ciao
15:17 Sylvain31 YAML: basedir=/srv/salt/pillar/customers, path=db_database.yaml
15:17 Sylvain31 2016-05-30 17:17:09,605 [salt.pillar      ][CRITICAL][3858] Pillar render error: Failed to load ext_pillar stack: 'salt' is undefined2
15:17 Sylvain31 bye.
15:20 Sylvain31 from the doc: __salt__: a dictionary of Salt module functions, useful so you don't have to duplicate functions that already exist (note: runs on the master)
15:43 DanSlimmon joined #salt
15:46 Nahual joined #salt
15:55 Muchoz joined #salt
16:06 armyriad joined #salt
16:07 Sylvain31 found here, how to attach ipython https://docs.saltstack.com/en/latest/topics/development/modular_systems.html#interactive-debugging and it works… ;)
16:08 danielcb joined #salt
16:11 armyriad joined #salt
16:27 writtenoff joined #salt
16:28 danielcb joined #salt
16:41 aagbds joined #salt
16:46 MTecknology c'mon saltstack! We need proper sls highlighting!! (for pygments at least)
16:47 impi joined #salt
16:48 Sylvain31 MTecknology: sounds, cool. jinja included?
16:49 abednarik joined #salt
16:51 manji joined #salt
16:55 duncanmv joined #salt
16:56 writtenoff joined #salt
16:57 manji joined #salt
16:59 Muchoz joined #salt
17:02 danielcb joined #salt
17:04 MTecknology Sylvain31: jinja+yaml ... both which seem to be lacking on their own
17:05 mdasilva joined #salt
17:05 aagbds joined #salt
17:06 duncanmv joined #salt
17:06 viq MTecknology: stuff pulled in by http://vim.spf13.com/ seems to do a pretty good job for me
17:07 mdasilva hey guys im running 2015.8.7 and some of my minions are all of a sudden throwing [ERROR] Minion failed to start
17:07 DEger joined #salt
17:07 mdasilva any idea what could be causing this?
17:14 MTecknology viq: this seems like a scary script to run
17:15 mdasilva upgrading to 2015.8.10 and upgrading python-tornado > 3.3 fixed my issue :)
17:15 MTecknology viq: anything for pygments?! :)
17:16 MTecknology ooooh!!!
17:20 MTecknology viq: I take back what I said! I found love again!
17:20 Sammichmaker joined #salt
17:25 mdasilva left #salt
17:27 numkem joined #salt
17:30 punkoivan joined #salt
17:31 fracklen joined #salt
17:33 pmcnabb joined #salt
17:38 pkruithof joined #salt
17:42 devster31 if I understand correctly I can see the order in which the formulas are run using salt '*' state.show_highstate, what's the best way to change this order?
17:47 linjan_ joined #salt
17:48 Eugene devster31 - https://docs.saltstack.com/en/latest/ref/states/ordering.html
17:50 subsignal joined #salt
17:51 mowntan joined #salt
17:51 mowntan joined #salt
17:55 s_kunk joined #salt
17:58 impi joined #salt
18:02 onlyanegg joined #salt
18:05 s_kunk joined #salt
18:08 GreatSnoopy joined #salt
18:18 babilen devster31: The best way is to declare explicit requisites between states. Why do you want to change the order to begin with?
18:21 devster31 to avoid requiring states from other formulas, but it seems the best way to do it is extending them, I didn't read about that
18:23 babilen What's the problem with requiring states from other formula?
18:24 * babilen does that all the time
18:24 deniszh joined #salt
18:24 alvinstarr joined #salt
18:24 babilen 2016.3.0+ds-1 is in Debian unstable :)
18:24 Sammichmaker joined #salt
18:29 deniszh joined #salt
18:29 Muchoz joined #salt
18:30 khaije1 joined #salt
18:33 khaije1 does the Pillar system support use of a default init.sls inside a directory in the way that it works in the State system?
18:36 komputes joined #salt
18:37 dmaiocchi joined #salt
18:41 martoss joined #salt
18:44 DanSlimmon joined #salt
19:04 komputes joined #salt
19:17 alvinstarr joined #salt
19:19 cyborg-one joined #salt
19:20 noraatepernos joined #salt
19:21 devster31 khaije1: I believe it does, I have folder/init.sls for pillars, is that what you meant?
19:23 dmaiocchi joined #salt
19:23 khaije1 devster31: yep!
19:24 devster31 reference here: https://docs.saltstack.com/en/latest/topics/tutorials/pillar.html#more-complex-data
19:24 khaije1 so you can evaluate folder/init.sls in your pillar's top file as '   - folder'
19:24 * khaije1 goes to read
19:25 spuder joined #salt
19:25 khaije1 brilliant, devster31 thats exactly what I wanted to know thanks!
19:26 martoss joined #salt
19:31 spuder_ joined #salt
19:36 DEger joined #salt
19:43 amcorreia joined #salt
19:46 keimlink joined #salt
19:46 permalac Hello guys, does anyone here manage cisco switches?   Catalyst and such ???
19:51 mrud joined #salt
19:51 mrud joined #salt
20:01 devster31 you can do that?
20:02 Lionel_Debroux joined #salt
20:05 justanotheruser joined #salt
20:09 lero joined #salt
20:19 Ayo joined #salt
20:21 deniszh1 joined #salt
20:22 jhauser joined #salt
20:23 Eugene permalac - https://docs.saltstack.com/en/latest/topics/proxyminion/index.html is what you're looking for; AFAIK there is not a module for Cisco
20:25 kus joined #salt
20:28 CeBe joined #salt
20:33 babilen There are also the napalm states and modules
20:34 babilen https://github.com/napalm-automation/napalm-salt
20:34 babilen permalac: ^
20:36 bittler joined #salt
20:38 bittler hello! I have been running 2015.5.x for a while (based on the community supported PPA), but I recently updated my build process to use the official deb repo, and I'm now running 2016.3.0 - having made that upgrade, I now see the consul ext_pillar failing hard and killing state.sls or highstate, whereas previously, the consul ext_pillar could fail, but it would not kill the state.sls or highstate that was running. Is there a workaround
20:38 bittler Is there a workaround for this, or do I need to disable that ext_pillar module if consul is not (yet) available?
20:40 titilambert joined #salt
20:41 martoss joined #salt
20:44 abednarik joined #salt
21:01 amcorreia joined #salt
21:03 babilen bittler: Might want to report a bug
21:06 bittler babilen: I'm still investigating, but it seems like some folks _want_ that behaviour (eg, if the ext_pillar module fails, the state run should fail hard) - https://github.com/saltstack/salt/issues/31541 and https://github.com/saltstack/salt/issues/23402, but neither of those are closed/resolved, and so I'm confused as to the expected behavior.
21:06 saltstackbot [#23402]title: State runs should fail if external pillar retrieval fails | If you throw an exception in a custom ext_pillar, highstate still runs.  Is it possible to control this behavior somehow, or by default, fail the state run due to this?...
21:07 bittler I'm also confused as to what version these details changed in, I'm going to test with a few different versions to see when this changed
21:09 bittler bah, I'm not sure if the official deb repo has anything other than 2016.x
21:12 babilen There's an archive bit
21:12 DanSlimmon joined #salt
21:14 babilen http://repo.saltstack.com/apt/debian/8/amd64/archive/ (for example)
21:14 babilen bittler: It's also pretty easy to just install different tags from git using the bootstrap script
21:15 jhauser joined #salt
21:16 bittler ah, yes - http://repo.saltstack.com/apt/ubuntu/14.04/amd64/
21:21 deniszh joined #salt
21:23 CeBe joined #salt
21:33 mowntan joined #salt
21:34 martoss joined #salt
21:35 lero joined #salt
21:40 deniszh1 joined #salt
21:45 scottcrooks joined #salt
21:46 sc250024 Quick question. The latest release documentation (2016.03) for the `cmd` state says that `cmd.wait` will eventually be deprecated
21:46 sc250024 https://docs.saltstack.com/en/latest/ref/states/all/salt.states.cmd.html
21:46 sc250024 Will `onchanges`, which is the recommended way now, work if it's waiting for another command to be run?
21:49 deniszh joined #salt
22:00 hasues joined #salt
22:00 hasues left #salt
22:10 oida joined #salt
22:13 oida joined #salt
22:37 west575 joined #salt
22:47 west575 joined #salt
22:50 kiorky joined #salt
22:59 subsignal joined #salt
22:59 majikman joined #salt
22:59 punkoivan left #salt
23:00 west575 joined #salt
23:04 abednarik joined #salt
23:10 keimlink_ joined #salt
23:14 west575 joined #salt
23:19 al joined #salt
23:20 justanotheruser joined #salt
23:24 west575 joined #salt
23:25 DanSlimmon joined #salt
23:26 keimlink joined #salt
23:27 jeddi joined #salt
23:27 Muchoz joined #salt
23:29 DEger joined #salt
23:32 majikman joined #salt
23:34 subsignal joined #salt
23:39 rgrinberg joined #salt
23:46 fracklen joined #salt

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