Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-05-26

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

All times shown according to UTC.

Time Nick Message
00:00 noraatepernos joined #salt
00:07 Nahual joined #salt
00:10 nmadhok joined #salt
00:11 brianfeister joined #salt
00:13 bltmiller joined #salt
00:18 zmalone joined #salt
00:21 ElasticApps joined #salt
00:23 mosen joined #salt
00:24 aqua^c joined #salt
00:28 TheoSLC joined #salt
00:31 mosen hi saltines
00:37 dendazen joined #salt
00:38 subsignal joined #salt
00:39 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
00:40 UtahDave joined #salt
00:45 DEger joined #salt
00:49 peters-tx joined #salt
00:51 quasiben joined #salt
00:55 estahn joined #salt
00:59 armguy joined #salt
01:02 mpanetta joined #salt
01:09 racooper joined #salt
01:13 dabb joined #salt
01:17 k_sze[work] joined #salt
01:18 spuder joined #salt
01:20 subsignal joined #salt
01:24 nmadhok1 joined #salt
01:28 alexlist joined #salt
01:33 quasiben joined #salt
01:37 iceyao joined #salt
01:39 nmadhok joined #salt
01:40 schemanic_ joined #salt
01:40 schemanic_ Heya, I'm running the salt minion on a windows client and I'm trying to figure out where I put my state tree and other files considering I'm trying to go masterless
01:40 mowntan joined #salt
01:41 schemanic_ I've found C:/Program Files/salt, but from there I'm lost
01:41 mowntan joined #salt
01:46 UtahDave schemanic_: I haven't tried running masterless, but I would modify the minion config just like you would for a linux minion.  You could probably put the path anywhere you want
01:46 schemanic_ UtahDave, drinking from the firehose here - I've never done it on a linux machine aside from a vagrant box. Right now I'm just putting it in the salt 'root' so to speak, which I'm assuming is equivalent to /srv/salt. Can you tell me more about where to find that in the minion file?
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
01:49 UtahDave you need to set the file_roots option to the path you want and also set file_client: local
01:51 schemanic_ latter part I got, the former doesn't seem to be set by default
01:52 catpiggest joined #salt
01:53 schemanic_ If I just have the salt minion installed, can I run salt execution modules on it?
01:53 UtahDave Yeah, it's probably not in there
01:53 UtahDave yes, you can
01:53 UtahDave all those executions are always done minion-side
01:53 UtahDave even when you have a master
01:54 schemanic_ i'm wondering how to do that on windows. I'm trying to do something like salt '*' cmd.run 'blah' but powershell says salt isnt on the machine
01:57 tristianc joined #salt
01:57 UtahDave schemanic_: you run salt-call when you're running commands locally
01:58 schemanic_ hmm. I guess the installer didn't add c:\salt to my path
01:58 UtahDave no, probably not
01:59 barmaley joined #salt
02:00 schemanic_ Thank you UtahDave that helped alot
02:00 brianfeister joined #salt
02:01 aw110f joined #salt
02:03 drawsmcgraw joined #salt
02:05 barmaley joined #salt
02:05 nmadhok joined #salt
02:13 TheoSLC joined #salt
02:16 fracklen joined #salt
02:18 ageorgop joined #salt
02:23 devtea joined #salt
02:23 brianfeister joined #salt
02:23 rgrinberg joined #salt
02:42 dh__ joined #salt
02:42 dh__
02:46 tuxx joined #salt
02:55 quasiben joined #salt
02:56 yetAnotherZero joined #salt
03:02 estahn joined #salt
03:10 DEger joined #salt
03:12 DEger joined #salt
03:27 tristianc joined #salt
03:39 jcastle joined #salt
03:39 rnts_ joined #salt
03:40 muep_ joined #salt
03:42 MindfulMonk joined #salt
03:44 kidneb joined #salt
03:47 WildPikachu joined #salt
03:47 Twiglet joined #salt
03:47 systeem joined #salt
03:48 godlike joined #salt
03:48 godlike joined #salt
03:52 zmalone joined #salt
03:54 josuebrunel joined #salt
04:00 estahn joined #salt
04:00 riftman joined #salt
04:01 nmadhok joined #salt
04:08 josue joined #salt
04:20 drawsmcgraw left #salt
04:24 nmadhok joined #salt
04:25 rojem joined #salt
04:37 rdas joined #salt
04:42 alxchk joined #salt
04:54 josuebrunel joined #salt
04:56 hasues joined #salt
05:04 ivanjaros joined #salt
05:04 zmalone joined #salt
05:09 sauvin joined #salt
05:09 mohae_ joined #salt
05:12 edrocks joined #salt
05:20 jxm_ joined #salt
05:23 manji joined #salt
05:31 Ayo joined #salt
05:40 linjan__ joined #salt
05:40 barmaley joined #salt
05:44 onlyanegg joined #salt
05:44 fracklen joined #salt
05:44 evle2 joined #salt
06:03 kawa2014 joined #salt
06:06 hrumph hi everyone i'm having trouble with salt development
06:06 hrumph probably because i know (*& all about python
06:06 hrumph ok i followed the development install instructions (or thought I did) changed around a module, and did a sync_all
06:07 hrumph sync_all didn't report making any changes and it's as though I never made any changes to the module
06:16 hrumph help!!
06:18 iceyao_ joined #salt
06:19 mosen ah okay
06:19 mosen then it isnt looking in the right place
06:20 mosen what I usually do is add my git dir to file_roots and run sync_all after changes
06:20 hrumph ok i started messing with an external module in _modules and everything was working fine
06:21 hrumph so i installed the development version in a devel environment using the instructions
06:21 mosen yep
06:21 mosen oh theres a development version? running out of the salt repo?
06:21 hrumph then i modified the module i wanted to "improve" (other can be the judge of that)
06:21 hrumph mosen, well the devel branch
06:22 mosen oh i see, i was thinking pure module development not hacking on the main repo :)
06:22 mosen lemme try to whiz through the tutorial
06:22 hrumph i changed reg.py
06:22 hrumph no change i make makes any difference afaics
06:24 mosen youre running the salt out of the virtualenv not the installed version?
06:24 hrumph yes
06:25 hrumph i did a pip install -e ~/salt
06:25 hrumph and ~/salt is straight from git
06:25 hrumph and it works
06:25 hrumph but the changes i make aren't doing anything at all
06:28 ivanjaros3916 joined #salt
06:29 sjmh joined #salt
06:30 babilen hrumph: You really don't have to install a devel version if all you want is to override a module by placing it in _modules
06:30 babilen You can just place it in there and sync it, that's all that's needed for it to override the default one
06:30 hrumph babilen, no i have done some worko and want to do a PR
06:31 hrumph i'm obviously doing something wrong
06:31 hrumph i just added a feature for volatile registry keys and i want the devels to merge this
06:40 hrumph it's the minion
06:41 hrumph it's not getting the new module
06:41 hrumph i don't know how to make it do it
06:41 hrumph other than a reinstall
06:42 cyborg-one joined #salt
06:42 babilen You would have to install the minion from the same source, won't you?
06:43 hrumph babilen, i don't really care if the minion is totally up to date
06:43 iggy /srv/salt/_modules (default location) is synced to minions automatically?
06:43 hrumph iggy, but the standard modules don't seem to be synced automatically
06:43 babilen The master won't magically sync anything, just because you changed the source locally... You essentially have two options: 1. Place the module in _modules and sync it 2. Install both, master and minion, from the same git HEAD
06:44 hrumph babilen, hehe ok...
06:44 hrumph fuck
06:44 hrumph i'll continue tommorrow
06:44 iggy I've overridden modules from _modules
06:44 babilen In the first case the master will just sync the module and the minion will use it. This is obviously only possible if you work on execution modules, states (_states), ...
06:44 iggy including upstream modules
06:45 babilen The second approach is necessary if you want to tinker with the salt code itself (e.g. minion.py, master.py, ...)
06:45 bantone hey iggy
06:45 hrumph iggy, yes i'm got a pr almost ready but i need to test the module as though it's part of the mainline
06:46 babilen But you can, naturally, also edit the modules that come preinstalled with the master, but then you would have to install master and minion from that git HEAD
06:46 babilen Modules placed in _modules behave *exactly* like those that ship with salt
06:46 iggy I wouldn't say it was a requirement to test that way
06:46 bantone I don't want anything just saying hi :) sitting up here working
06:47 babilen iggy: No, not at all .. but you can't expect changes in your master install to be synced to your standard minion install. That's, I think, the main misconception here
06:47 iggy hi bantone, how's things?
06:47 hrumph babilen, i understand. it was wishful thinking
06:49 hrumph babilen, it makes windows testing a bit difficult
06:49 hrumph since you can't make windows a salt-master
06:50 armyriad joined #salt
06:51 babilen hrumph: Why does that make it difficult? You make changes to your modules, place it in _modules, sync and then test it?
06:51 babilen Or you clone your git repository on both, Windows and Linux and install from git there.
06:52 estahn joined #salt
06:52 armyriad joined #salt
06:53 favadi joined #salt
06:53 hrumph babilen, i'll probably be doing that in future
06:54 subsignal joined #salt
06:54 hrumph until now i thought that the minions always fetched the modules from the master
06:54 hrumph shows how much i know
06:55 hrumph oh well the pr will wait for tommorrow
06:55 bantone iggy: i'm deploying a hotfix...using salt of course hah
06:57 babilen Hotfixing salt? ;)
06:57 Elsmorian joined #salt
06:57 bantone hehe performance hotfix for data striping
06:58 sjmh joined #salt
06:58 Sylvain31 joined #salt
06:59 mosen hrumph: i usually just develop with salt-call, at least when the platform machines my own
07:02 Sylvain31 hi, what is the c
07:03 Sylvain31 common way to organize pillar, such I can have single mysql root password, for a single host
07:03 Sylvain31 I mean…
07:05 Sylvain31 I've a mysql formula, which can embbed root's password, I want to share common config, charset, folders, stuff like that, but personalize root password at least…
07:06 Sylvain31 I may have a pillar/top.sls with a 'database*' selector, and a second one which match the minion excactly…
07:09 RandyT joined #salt
07:11 CeBe joined #salt
07:12 Sylvain31 http://paste.debian.net/704917/
07:15 hasues left #salt
07:16 babilen I'd keep a standard pillar in Python and then write a password one that uses a dictionary with entries for each box and a lookup on minion['id']. Or use hashicorp's vault in Salt.
07:17 toanju joined #salt
07:18 babilen You can, naturally, combine those two in a single SLS if you prefer
07:18 Sylvain31 I dont undersand all you said. Will default pillar sls files merge in order they are defined (loaded in top.sls)?
07:18 hrumph mosen, babilen, iggy, etc. i made a link to my source in _modules in my virtualenv root. That's working. thanks guys for telling me what i should have known
07:19 Nothos927 joined #salt
07:19 babilen hrumph: You know now, that is what is important
07:19 mosen whew
07:19 babilen And yay! :)
07:19 babilen Sylvain31: Read up on pillar merging
07:20 babilen There are a variety of settings that influence this behaviour
07:21 Sylvain31 babilen: OK. I seeking… thanks.
07:28 DEger joined #salt
07:28 sacosta joined #salt
07:29 slav0nic joined #salt
07:36 jhauser joined #salt
07:38 av_ joined #salt
07:38 jad_jay joined #salt
07:40 linjan__ joined #salt
07:41 Sylvain31 can I tell salt to produce debug output to explicitly show how it's processing pillar/*sls file? I do managed to produce a merge without particular config, but it's gone now. "salt-call -ldebug pillar.get 'mysql:server'" don't do that.
07:41 manji joined #salt
07:44 bmcorser hey guys
07:44 bmcorser i'm trying to get rest_cherrypy up and running on my salt master
07:44 bmcorser i've added the options to my config at /etc/salt/master and restarted the master
07:44 bmcorser but i'm getting connection refused on the port i configured
07:45 bmcorser i'm running salt-master -l debug but can't see any output relating to the netapi stuff ... ?
07:45 bmcorser any advices?
07:45 punkoivan joined #salt
07:46 impi joined #salt
07:46 punkoivan joined #salt
07:48 fracklen joined #salt
07:49 Sylvain31 nc -v -z salt.master.ip 4505
07:49 deniszh joined #salt
07:49 Sylvain31 nc -v -z salt.master.ip 4506.
07:49 Sylvain31 does it works ?
07:49 bmcorser Sylvain31: are you talking to me?
07:49 Sylvain31 bmcorser: yes for you. ;)
07:49 bmcorser :) ok let me try
07:50 bmcorser nc: getaddrinfo: Name or service not known
07:50 bmcorser okfuck
07:51 Sylvain31 bmcorser: salt.master.ip is an IP, right? ;)
07:51 bmcorser you want me to put the actual IP in lol just a sec
07:51 bmcorser :facepalm:
07:51 bmcorser succeeded on 4506
07:51 Rumbles joined #salt
07:52 fracklen joined #salt
07:52 bmcorser the master itself is working fine, but the REST API doesn't seem to be running
07:52 bmcorser ie if i curl localhost:8000 i get connection refused
07:54 Sylvain31 "netstat -an | less" gives you something listening?
07:54 bmcorser let me see
07:54 bmcorser nope, not on the port i have configured :(
07:55 babilen bmcorser: Connection refused either means that the service isn't running or that a firewall has gobbled up your SYN package.
07:55 dmaiocchi joined #salt
07:56 babilen I guess the former
07:56 babilen Is it, mayhaps, using a different port?
07:57 bmcorser i have explicitly asked it to run on 5001 :/
07:57 bmcorser rest_cherrypy: port: 5001
07:57 bmcorser etc
07:57 bmcorser i remember running the RESP API service "manually"
07:58 bmcorser but i am guessing if that stuff is configured in /etc/salt/master then it should just run
07:58 bmcorser or do i have to have some process running
07:58 babilen Do you see it running on a different port when you run "netstat -tulpen" ?
07:59 josuebrunel joined #salt
07:59 bmcorser i don't think so
08:00 bmcorser i guess the process would be called 1234/python
08:00 bmcorser like the salt master processes, i can see those on 4506 etc
08:01 bmcorser i'm migrating a master server
08:02 bmcorser and i can see from the .bash_history that (at one point) i ran a command salt-api
08:02 bmcorser do i need to install some extra thing on the new server?
08:02 bmcorser i have `salt-api: command not found ` here :(
08:03 dgutu joined #salt
08:04 bmcorser i also see https://github.com/saltstack/salt-bootstrap/issues/609
08:04 saltstackbot [#609]title: salt-api not installed on Oracle Linux | Is there a reason salt-api is not installed on Oracle Linux, I don't see an option to force its install
08:05 krymzon joined #salt
08:06 bmcorser i also installed salt with salt-bootstrap and notice that it doesn't install salt-api
08:06 bmcorser https://github.com/saltstack/salt-bootstrap/blob/76e9f02333940d8cdbf387f40b5c14f0cbfa695c/bootstrap-salt.sh#L2617
08:06 bmcorser perhaps there's my isssue
08:06 bmcorser now the question is how to install salt-api!
08:08 viq 'yum install salt-api'
08:09 viq or 'salt-call --local pkg.install salt-api'
08:09 devcorn joined #salt
08:09 dmaiocchi joined #salt
08:09 devcorn help with this http://stackoverflow.com/questions/37453549/saltstack-diff-in-state-using-python
08:09 Sylvain31 viq: +1 for the lastone. ;)
08:10 roksteady joined #salt
08:16 dmaiocchi joined #salt
08:25 keimlink joined #salt
08:29 GreatSnoopy joined #salt
08:31 Elsmorian joined #salt
08:35 s_kunk joined #salt
08:38 _mel_ joined #salt
08:42 _mel_ Morning salt-users :-) i've a general question about using more then one linux release like wheezy and jessie. by now i have a lot of if the else logic inside the states to make sure it works on every release. but its getting very complicated over time. so is there a better way?  I thought about something like a tree structure like /apps/squeeze/php /apps/wheezy/php, but maybe there is a better way?
08:49 bmcorser cheers viq
08:49 bmcorser apt-get did it for me
08:49 babilen _mel_: Look into https://docs.saltstack.com/en/latest/topics/development/conventions/formulas.html and the scheme adopted in https://github.com/saltstack-formulas/template-formula
08:49 babilen _mel_: You can, additionally, also filter on osrelease grain if you want to incorporate additional differences
08:50 manji joined #salt
08:51 babilen _mel_: Really depends on what kind of logic/differences you want to account for
08:52 _mel_ babilen: i've a example. squeeze and wheezy have php with apc, but jessie uses php with apcu and zend-optimizer. how would i write a state for that?
08:54 thehaven joined #salt
08:54 AndreasLutro just put different states in an if/else statement where you check the os version number
09:04 lero joined #salt
09:04 _mel_ AndreasLutro: ok, thats what i thought too. seems to be the best compromise
09:05 hellosalt joined #salt
09:08 dmaiocchi joined #salt
09:15 hrumph new pull request now ready
09:15 hrumph https://github.com/saltstack/salt/pull/33527
09:15 saltstackbot [#33527]title: Allow creation of volatile registry keys in reg.py module | ### What does this PR do?...
09:17 barmaley joined #salt
09:18 barmaley joined #salt
09:21 Lionel_Debroux joined #salt
09:25 impi joined #salt
09:32 bmcorser anyone have any experience running salt on GCP
09:33 bmcorser (that is google compute platform)
09:33 bmcorser minions don't appear visible to master
09:41 hellosalt sounds like tagging on the network (security group)?
09:42 hellosalt bmcorser: did you try to telnet the master from minion - see if there's firewall issue?
09:49 fredvd joined #salt
09:52 Herugrim joined #salt
09:54 bmcorser hellosalt: that sounds about right
09:55 bmcorser what port does the salt-minion service run on?
09:55 hellosalt 4505 - 4506
09:56 bmcorser ok
09:56 Herugrim hi all, i'm running 2015.8.10 and it seems file.blockreplace does not accept a list for source: , if i use - source: salt://files/etc/environment then it works but if i use a list it borks, check https://gist.github.com/anonymous/c7d932fb5791ab4b94d5419d03d41eb7 for an example and error
09:57 bmcorser hellosalt: so i run 'telnet minion.ip 4506' and get connection refused (which feels like firewall issue) whereas 'telnet minion.ip 4505' hangs
09:58 hellosalt try telnet master host or ip
09:58 hellosalt from minion
10:00 bmcorser ok
10:01 bmcorser hellosalt: just get 'Trying minion.ip ...'
10:01 bmcorser not sure what this indicates
10:04 bmcorser would there by anything else i would need to do apart from open 4505-4506 on master and minions
10:04 bmcorser ?
10:09 Ayo joined #salt
10:12 viq joined #salt
10:14 cyborg-one joined #salt
10:16 estahn joined #salt
10:35 devcorn joined #salt
10:35 devcorn how can I run following using Python api.. client.cmd
10:35 devcorn salt <minion_id> state.highstate test=true --state-verbose=False
10:35 devcorn client.cmd(minion_id, 'state.highstate', [--state-verbose=False], test=True)
10:36 devcorn is this correct? worried that if don't take test then it will apply the changes
10:36 devcorn as of now I don't have any test instances to try it
10:46 quasiben joined #salt
10:49 AndreasLutro devcorn: no, client.cmd doesn't take command line arguments, but module function arguments. look up the highstate function in salt.modules.state and pass the appropriate dict of kwargs as "kwarg"
10:49 AndreasLutro also state-verbose might not even be a valid kwarg for non-cli usage
10:49 amcorreia joined #salt
11:01 DEger joined #salt
11:08 jakshi joined #salt
11:12 devcorn AndreasLutro: Thanks, can you suggest any python api way to set verbose or using like --state-output=mixed.. The doc don't speak much :(
11:13 AndreasLutro there is no way to do that through the the python api, since it's part of the console outputter module
11:15 dmaiocchi joined #salt
11:17 keimlink joined #salt
11:22 irctc478 joined #salt
11:23 irctc478 Hi all, I'm using Salt version 2015.8.8 on master and minions, and the "pkg" module fails to detect that "python-software-properties" is installed even though it is... this is causing a cascade of failures in our setup instructions... not seeing any exceptions in the logs. Any ideas?
11:26 viq hm, seems the peer publish system doesn't work for me. I have in /etc/salt/master https://pbot.rmdir.de/L4AAbPDSTgrIBcQ-PJHjrg but 'salt minion1 publish.publish \* grains.get os' returns nothing
11:26 viq irctc478: did you restart minions since "python-software-properties" was installed?
11:27 irctc478 Yes, and we also applied system upgrades and rebooted after that, as well.
11:28 viq irctc478: try running one of those states locally with 'salt-call -l debug' and see what it shows you?
11:29 VSpike irctc478: what about "sudo salt-call pkg.info_installed python-software-properties"
11:30 VSpike although on 2015.8.7 that throws exceptions for me
11:30 VSpike on latest, it doesn't
11:30 VSpike irctc478: alternatively, sudo salt-call pkg.list_pkgs | grep python-software
11:31 irctc478 @VSpike it does actually come up on that latest command of yours
11:33 viq any ideas how to debug non-working publish.publish?
11:34 irctc478 @VSpike also, not sure if it's relevant, when issuing the above salt-call command the minion actually disconnects from the master, and I have to restart it
11:35 viq that's not an expected behaviour, AFAIK. You're using the default zmq transport?
11:35 skrobul joined #salt
11:36 pjs joined #salt
11:37 irctc478 @Viq I assume so, I haven't changed that
11:37 babilen irctc478: Could you paste the debug log to one of http://refheap.com, http://paste.debian.net, https://gist.github.com, http://sprunge.us, … ? (master and minion if possible)
11:38 babilen (and that of the state run)
11:42 quasiben1 joined #salt
11:51 irctc478 @VSpike haven't got anywhere yet. Salt "pkg" module stills reports the expected packages to be there when doing "pkg.list_pkgs" *and* even when instructing "pkg.installed" it succceeds. But when using a "require: pkg: python-software-properties" or "require: pkg: python-pip" it just fails
11:51 irctc478 I've had to remove those "require" sections in order for the rest of the instructions to run
11:51 irctc478 Which did run successfully :)
11:52 viq irctc478: 'require' isn't for arbitrary stuff, but for state IDs
11:53 viq Which may happen to be installing the packages in question, but they still need to be state IDs
11:53 babilen irctc478: Mind pasting your SLS? Are you sure that you have a state with "python-software-properties" as either state ID or name argument?
11:53 babilen Ah, viq already pointed that out :)
11:54 viq In a less polite way ;)
11:59 * viq remembers about theforeman and sets that up to see state of salt integration
12:05 evle joined #salt
12:16 Ayo_ joined #salt
12:26 justanotheruser joined #salt
12:30 scoates joined #salt
12:33 tracphil joined #salt
12:39 ssplatt joined #salt
12:40 toastedpenguin joined #salt
12:45 zmalone1 joined #salt
12:45 irctc478 @VSpike @Viq good points there, didn't think of that. But I haven't changed my SLS around this area in ages, and it was working fine until now. However maybe Salt's behavior changed around detecting that prerequisite? although pretty sure I didn't upgrade Salt either
12:48 irctc478 the error is indeed "The following requisites were not found:   require:  pkg: python-software-properties" when the declaration of that is "python-software-properties:   pkg.installed"
12:48 west575 joined #salt
12:50 gh34 joined #salt
12:52 irctc478 @VSpike @Vieq http://pastebin.com/Mie0v4Hy
12:58 irctc478 OMG.... it was because I was issuing "state.apply docker" and not "state.apply python-packages,docker", sorry to have bothered
13:02 subsignal joined #salt
13:02 babilen viq: Politeness obviously doesn't pay off ;)
13:04 estahn joined #salt
13:07 numkem joined #salt
13:13 subsignal joined #salt
13:14 metalseargolid joined #salt
13:15 dgutu_ joined #salt
13:20 DEger joined #salt
13:22 west575_ joined #salt
13:22 rgrinberg joined #salt
13:27 toastedpenguin joined #salt
13:27 mapu joined #salt
13:28 rbjorklin joined #salt
13:35 teryx510 joined #salt
13:37 perfectsine joined #salt
13:39 teryx5101 joined #salt
13:39 racooper joined #salt
13:43 rem5 joined #salt
13:45 scoates joined #salt
13:52 ekristen joined #salt
13:54 sjorge joined #salt
13:55 DEger joined #salt
13:56 jerredbell joined #salt
13:58 * babilen contemplates upgrading to salt 2016.3
14:00 quix joined #salt
14:03 dendazen joined #salt
14:04 Sylvain31 joined #salt
14:05 mpanetta joined #salt
14:06 mpanetta joined #salt
14:06 zmalone joined #salt
14:07 estahn joined #salt
14:21 * viq ponders why publish doesn't work
14:24 mage_ is it possible to manage require,watch, ... out of a module?
14:27 DEger joined #salt
14:31 glickind joined #salt
14:31 glickind hi
14:31 spuder_ joined #salt
14:32 glickind FYI there is a 404 Not found on https://docs.saltstack.com/en/pdf/Salt-2016.3.0.pdf
14:34 mohae joined #salt
14:34 rawzone joined #salt
14:34 DEger joined #salt
14:39 Nebraskka joined #salt
14:42 brew joined #salt
14:42 babilen glickind: Has it been generated yet?
14:43 tapoxi joined #salt
14:43 glickind I don't know
14:44 manji joined #salt
14:44 babilen Okay, I guess not :)
14:44 tapoxi hey #salt, so I'm new and trying to figure out if I should use the orchestration runner or write my own using the python api
14:44 glickind but I could get https://docs.saltstack.com/en/pdf/Salt-2015.8.10.pdf
14:45 tapoxi what's a 'failure' in salt? can I look for a specific output from a command and say "nope this blew up lets stop" ?
14:45 Lee_ joined #salt
14:47 west575 joined #salt
14:47 rojem joined #salt
14:50 morissette joined #salt
14:51 hasues joined #salt
15:00 tapoxi 2016.3 come out today?
15:02 hasues left #salt
15:03 fracklen joined #salt
15:03 brew joined #salt
15:03 Sylvain31 can I apply a particular state_id: from a state.sls, does it make sens?
15:07 DEger joined #salt
15:10 blw0rm joined #salt
15:14 dh__ joined #salt
15:14 salty_solution joined #salt
15:15 ALLmightySPIFF joined #salt
15:16 ALLmightySPIFF joined #salt
15:24 west575_ joined #salt
15:26 traph joined #salt
15:29 west575 joined #salt
15:29 DammitJim joined #salt
15:33 s_kunk joined #salt
15:33 noraatepernos joined #salt
15:34 west575_ joined #salt
15:35 Heartsbane joined #salt
15:35 Heartsbane joined #salt
15:38 punkoivan joined #salt
15:38 toanju joined #salt
15:40 punkoivan joined #salt
15:41 punkoivan joined #salt
15:43 scottk_ joined #salt
15:45 scottk_ To join a channel type: /JOIN #Does anyone know how to check in jinja if a minion is part of a nodegroup? I tried following: "https://docs.saltstack.com/en/latest/topics/targeting/nodegroups.html" but when i run 'salt 'minion' pillar.get state'  I don't get any results. The minion is in the correct nodegroup, and i have restarted the master and refreshed the pillar items.
15:45 scottk_ Does anyone know how to check in jinja if a minion is part of a nodegroup? I tried following: "https://docs.saltstack.com/en/latest/topics/targeting/nodegroups.html" but when i run 'salt 'minion' pillar.get state'  I don't get any results. The minion is in the correct nodegroup, and i have restarted the master and refreshed the pillar items.
15:45 scottk_ sorry for the repost.
15:47 pcn Has anyone heard of any follow-up on https://github.com/saltstack/salt/issues/26467 ?
15:47 saltstackbot [#26467]title: [salt-ssh] aborting a job leads to corrupted salt-thin.tgz in the next run | Aborting a job leads to corrupted salt-thin.tgz in the next run...
15:47 fracklen joined #salt
15:48 ivanjaros joined #salt
15:49 pcn scottk_: did you enable pillar_opts per https://docs.saltstack.com/en/latest/topics/targeting/nodegroups.html#using-nodegroups-in-sls-files ?
15:50 TheoSLC joined #salt
15:50 scottk_ @pcn yes i di. it is set to True.
15:50 scottk_ did*
15:54 pcn If you do a salt-call pillar.items, do you see any related keys/values present?
15:56 pcn The documentation there doesn't refer to "state" as a pillar that's provided by enabling nodegroups, it looks like it'd be via pillar.get master:nodegroups:<whatever>
15:56 brew joined #salt
15:57 salty_solution anyone here a rbenv guru?
16:00 pcn I've been told to use chruby instead?
16:00 brianfeister joined #salt
16:00 pcn I'm sorry I know that's not as helpful, I haven't tried using rbenv with salt, chruby seems to provide some flexibility is all.
16:01 salty_solution never heard of it. i guess i had better do my research.
16:04 cyborg-one joined #salt
16:07 scottk_ sory pcn had something come up. tryint it now.
16:11 scottk_ <pcn> I did salt-call pillar.items and did not see anything relevant. Are you saying I should put pillar.get master:nodegroups:<whatever> into a pillar file? or should it work from a state.sls?
16:11 fas3r joined #salt
16:11 fas3r Hello
16:12 fas3r does  "noifupdown = True"  is suppose to prevent the network.managed states to reload the network configuration ?
16:15 sjmh joined #salt
16:19 onlyanegg joined #salt
16:29 amcorreia joined #salt
16:33 bltmiller joined #salt
16:37 yetAnotherZero joined #salt
16:39 writtenoff joined #salt
16:50 rem5 joined #salt
16:52 viq Ah, found why publish was not working for me - I had 'peer' defined twice in the master config
16:52 lero joined #salt
17:02 Trauma joined #salt
17:05 tharkun joined #salt
17:08 toastedpenguin joined #salt
17:11 ssplatt joined #salt
17:16 bltmiller joined #salt
17:23 writtenoff joined #salt
17:25 west575 joined #salt
17:27 toastedpenguin joined #salt
17:28 woodtablet joined #salt
17:30 punkoivan joined #salt
17:31 punkoivan joined #salt
17:34 bltmiller joined #salt
17:41 rem5 joined #salt
17:53 gmeno joined #salt
17:53 brianfeister joined #salt
17:54 ajw0100 joined #salt
17:55 brianfeister joined #salt
18:05 brew joined #salt
18:10 west575_ joined #salt
18:11 hasues joined #salt
18:11 hasues left #salt
18:12 lero joined #salt
18:15 s_kunk joined #salt
18:21 ajw0100 joined #salt
18:23 glickind joined #salt
18:23 DEger joined #salt
18:25 snergster joined #salt
18:25 fullstop hi all.  I added a new salt-minion, same version, but now my output looks a little different than it previously did.
18:25 fullstop http://pastebin.com/raw/s43FicSW
18:25 rgrinberg joined #salt
18:25 fullstop version 2015.5.3+ds-1trusty1 on all nodes
18:27 fullstop wait.  that's not the same version at all.
18:27 GreatSnoopy joined #salt
18:29 DEger joined #salt
18:31 west575 joined #salt
18:33 devster31 how is it that salt-bootstrap doesn't support xenial but cloud-init does?
18:38 babilen devster31: Are those projects that closely related?
18:39 viq seblu: sorry to mark your package out of date pretty much as soon as you published an update ;)
18:39 babilen And given that cloud-init is used in the Ubuntu cloud images it is not that surprising, isn't it?
18:42 djgerm joined #salt
18:42 djgerm is there a state to make sure a debian based system is up to date? (i.e. something that does apt-get update, apt-get upgrade)
18:43 tapoxi joined #salt
18:43 viq djgerm: https://docs.saltstack.com/en/latest/ref/states/all/salt.states.pkg.html#salt.states.pkg.uptodate
18:43 deniszh joined #salt
18:45 djgerm thanks! I should have read all the way to the bottom >_<
18:48 rojem joined #salt
18:50 toanju joined #salt
18:54 beardedeagle joined #salt
18:55 edrocks joined #salt
18:59 josuebrunel joined #salt
19:02 ajw0100 joined #salt
19:04 ajw0100 joined #salt
19:05 DEger joined #salt
19:06 teryx510 joined #salt
19:06 zmalone joined #salt
19:07 tapoxi hey everyone, http://hastebin.com/nisizevina.vhdl
19:07 tapoxi so in the second case tmux was already installed. in the third case cowsay wasn't found. but I get the same data back from each
19:07 tapoxi how can I tell if I had a failure?
19:08 teryx5101 joined #salt
19:09 edrocks joined #salt
19:19 devster31 babilen: no, they're completely unrelated, but in the issues for salt-bootstrap the step failing is install_ubuntu_stable_deps() but all the required dependencies are already packaged in the xenial repos
19:19 fracklen joined #salt
19:20 subsignal joined #salt
19:23 Elsmorian joined #salt
19:24 TheoSLC joined #salt
19:24 fracklen joined #salt
19:26 ldelossa joined #salt
19:26 ldelossa Hey guys, quick question. How can I say 'if this grain exist, do this stuff, but if not, exit state and do not fail the state'
19:27 babilen devster31: You could just install the packages in xenial for now
19:27 babilen (they are reasonably up to date)
19:27 viq ldelossa: wrap state in {%if%} {%endif%}
19:27 babilen ldelossa: if block with test on the grain, state inside
19:28 ldelossa {% if grains.get some:value %} state stuff {% endif %}
19:28 ldelossa just like that?
19:28 ldelossa And the state will still pass if it just his the end if and does nothing ?
19:28 ldelossa (if values are not set)
19:29 viq ldelossa: not quite right syntax, and if the if is false then the contents just "don't exist", so it will be like running an empty state
19:29 ldelossa Which passes?
19:29 viq I believe so
19:29 ldelossa I can figure out the syntax, it's more of whether an empty state fails or not
19:29 babilen We might want to refer to SLS and state :)
19:29 ldelossa that I'm not sure
19:29 zmalone joined #salt
19:29 babilen It gets confusing otherwise
19:30 ldelossa sls could be a pillar also tho?
19:30 viq Empty SLS is fine, I have some that do nothing unless on a specific OS, because they don't make sense elsewhere
19:30 babilen It could .. just to make it even more confusing
19:30 babilen If you have an empty SLS that isn't a problem at all
19:31 babilen SLS do not pass themselves, individual states in them do (or don't)
19:32 ldelossa Okay awesome
19:32 ldelossa one other question then...
19:32 ldelossa So I have a pillar file that looks like this: ephemeral: [ drive1, drive2]
19:32 ldelossa I believe ephemeral will always be set
19:32 ldelossa but how can  I test to see if there's any items within ephemeral
19:32 ldelossa if iteritems() fails?
19:33 babilen You could use {% if salt['pillar.get']('ephemeral') %}
19:34 s_kunk joined #salt
19:34 s_kunk joined #salt
19:34 babilen That assumed that it is set and that you don't need the value later on .. if it is not always set you might want to define an empty/falsy default value and if you need the value you might want to assign it to a variable and test that variable
19:34 babilen Just the general idea
19:35 babilen But if you want to iterate over that list anyway, I'd just use {% for item in salt['pillar.get']('ephemeral') %} $DO_STUFF {% endfor %}
19:35 babilen That way nothing will be generated if the list is empty
19:36 babilen Ah, you might want {% for item in salt['pillar.get']('ephemeral', []) %} in that case (same case as before, if you can't be 100% sure that it isn't set .. which you can't typically)
19:36 babilen Sometimes it is nice though if stuff likes this fails explicitly rather than result in a "no-op"
19:38 deniszh joined #salt
19:40 subsignal joined #salt
19:40 onlyanegg joined #salt
19:41 bltmiller joined #salt
19:44 toastedpenguin joined #salt
19:46 DEger joined #salt
19:55 edrocks joined #salt
19:55 ajw0100 joined #salt
20:01 TyrfingMjolnir joined #salt
20:04 quix joined #salt
20:04 lero joined #salt
20:05 ldelossa Hmm
20:05 ldelossa let me read this
20:05 viq I have following states - https://pbot.rmdir.de/pJ7LfikSvnRHBs4xOcvRow - and x509.pem_managed nicely generates a well formatted multi-line pem, but it overwrites the file. Below file.blockreplace puts contents of the pem into the existing file, but not properly formatted. Any idea how I could chain things to have it formatted properly?
20:09 jfindlay viq: how is it misformatting the file block?
20:10 ldelossa @babilen, I don't believe I want a default value
20:11 ldelossa I just want to say 'hey is it set' if not, 'do nothing and exit state'
20:12 viq jfindlay: https://pbot.rmdir.de/WDM1jzob3i5vlhNG6a92TA
20:14 noraatepernos joined #salt
20:14 jfindlay viq: is the source file properly formatted?
20:14 viq jfindlay: yes, https://pbot.rmdir.de/SPlhsjcu6bhAi58ioEe6wA
20:16 jfindlay I wonder if yaml is doing that
20:16 viq Actually, I have an idea...
20:16 AirOnSkin joined #salt
20:17 babilen ldelossa: Well, I'm sure that I provided enough examples for you to take it from there .. :)
20:18 jfindlay viq: could you try something like this: https://github.com/saltstack/salt-pkg-tests/blob/master/test_run/init.sls#L53?
20:19 ldelossa Yeah I think I'm good!
20:19 ldelossa thanks man
20:19 ldelossa man/women
20:19 ldelossa man-women ? O_o
20:19 viq jfindlay: yeah, that was my idea ;) Fails, preparing paste
20:19 jfindlay :(
20:20 viq https://pbot.rmdir.de/-LqLU4XlsCA90cwzvFBkaQ
20:20 deniszh joined #salt
20:20 viq actually...
20:20 jfindlay viq: can you eliminate the jinja lookup and blockreplace it with lines directly from the sls file?
20:21 babilen ldelossa: man :)
20:21 viq jfindlay: I could, but then I lose the wonderful "everything happens by itself" of the x509 modules
20:22 jfindlay right, I'm just wondering if that would help figure out where the newlines are getting stripped
20:22 viq sure, give me a sec then
20:24 viq https://pbot.rmdir.de/aEOcSE_4XuPon6dgrWrUYg
20:24 jfindlay I think those lines need to be indented
20:25 viq Oh, more, ok
20:26 TheoSLC joined #salt
20:26 jfindlay I wonder if you template in the jinja in a similar manner if it would work
20:27 jfindlay - content:
20:27 jfindlay {{ my_pk }}
20:27 jfindlay well, with the |, of course
20:28 lero joined #salt
20:28 viq With manually put content indented properly it works. With {{}} indented more it doesn't https://pbot.rmdir.de/IhCw6IDrlRY_lz6vM8U2-g
20:29 jfindlay hm
20:29 keimlink joined #salt
20:29 Elsmorian joined #salt
20:30 viq so yeah, {{}} indented makes only the first line indented
20:30 jfindlay right
20:32 jfindlay viq: are you able to reference the source file with `- source:`?
20:32 jfindlay rather than the file contents?
20:33 viq the cert I want to add to the file you mean?
20:33 jfindlay right
20:36 viq I could probably use MinionFS instead of mine for it, I guess. Mine feels cleaner ;)
20:37 * viq is trying to replicate https://docs.saltstack.com/en/latest/ref/states/all/salt.states.x509.html on a system that has all certs in a single file instead of a dir
20:38 Knuta viq: put them in a dir, have a file.waiting which concatenates all of them in the end? ;-)
20:39 viq Knuta: except the majority of the file comes from upstream. And managing it where I try to have as much as possible in git makes this awkward ;)
20:40 Knuta viq: no, I mean put them in a dir on the minion, then cat them together into a file ;-)
20:41 Knuta viq: alternately, make a crazy template which concatenates it all by itself. I abused the py template language to do lots of stuff before.
20:41 viq Knuta: I guess it would need to be "mv cert.pem cert.pem.orig && cat cert.pem.orig intca.pem >> cert.pem"
20:41 onlyanegg joined #salt
20:41 viq I'd need to learn python first ;P
20:44 viq Time to call it a day. I guess I'll try with MinionFS tomorrow
20:47 brianfeister joined #salt
20:49 quix joined #salt
20:55 subsignal joined #salt
21:00 teryx510 joined #salt
21:08 sjorge joined #salt
21:11 onlyanegg joined #salt
21:16 tracphil joined #salt
21:19 DEger joined #salt
21:24 sjorge joined #salt
21:26 kapra joined #salt
21:29 kapra left #salt
21:31 sjorge joined #salt
21:33 yetAnotherZero ok viq, why can't I get mines to work?
21:33 kapra joined #salt
21:34 yetAnotherZero no errors or anything, but no data from the saltmine
21:34 brianfeister joined #salt
21:37 teryx510 joined #salt
21:38 cyteen_ joined #salt
21:40 yetAnotherZero anybody know how to debug my saltmine attempt?
21:41 yetAnotherZero I've devined the mine_functions in a pillar
21:45 yetAnotherZero *defined
21:46 teryx510 joined #salt
21:47 yetAnotherZero ^ sync_grains
21:48 cpowell joined #salt
21:48 west575 joined #salt
21:49 Laogeodritt joined #salt
21:51 rocketnova joined #salt
21:51 yetAnotherZero still can't figure out how to get any mine data to come back
21:53 west575 joined #salt
21:58 GreatSnoopy joined #salt
22:01 rojem joined #salt
22:06 DEger joined #salt
22:33 brianfeister joined #salt
22:38 zmalone joined #salt
22:42 TheoSLC joined #salt
23:09 onlyanegg joined #salt
23:10 linjan__ joined #salt
23:15 keimlink_ joined #salt
23:15 debian112 joined #salt
23:27 mowntan joined #salt
23:27 amcorreia joined #salt
23:28 keimlink joined #salt
23:30 brotatochip joined #salt
23:37 brianfeister joined #salt
23:37 zmalone joined #salt
23:38 djgerm left #salt
23:38 mosen joined #salt
23:39 cpowell joined #salt
23:45 TheoSLC joined #salt
23:53 rgrinberg joined #salt
23:54 iceyao joined #salt

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