Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-07-27

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

All times shown according to UTC.

Time Nick Message
00:01 orion I am now editing /etc/hosts to have salt01-test point to 10.0.3.254.
00:01 tonybaloney joined #salt
00:01 krymzon joined #salt
00:01 orion I am also setting "master: 10.0.3.254" in the minion config.
00:03 tonybalo_ joined #salt
00:05 ninjada joined #salt
00:07 orion Now the minion won't connect at all.
00:07 hemebond Anything in its logs?
00:07 orion [salt.transport.zeromq][DEBUG   ][2073] Initializing new AsyncZeroMQReqChannel for ('/etc/salt/pki/minion', 'salt01-test', 'tcp://10.0.3.254:4506', 'clear') <-- this times out.
00:07 orion And yet, telnetting to 10.0.3.254 4506 succeeds.
00:08 orion "salt-call test.ping" fails for the same reason too.
00:08 infrmnt joined #salt
00:10 teryx510 joined #salt
00:13 hemebond orion: Which version of Salt?
00:13 hemebond salt-run manage.versions
00:13 hemebond salt-call test.versions
00:15 orion The former command timed out.
00:15 hemebond It didn't even show the master version?>
00:16 orion The latter command: https://gist.github.com/centromere/8662c62467532b26f5ba6cd24c6c8ea6
00:18 orion The former command: https://gist.github.com/centromere/53f634b78f93daedc49f69aefc3e2d09
00:22 hemebond The master is listening on 4505 and 4506, yeah?
00:22 hemebond Not something else?
00:29 west575 joined #salt
00:31 lorengordon joined #salt
00:34 orion Correct
00:34 orion 19:35:04 < orion> tcp        0      0 0.0.0.0:4505            0.0.0.0:*               LISTEN      29963/python
00:35 hemebond I'm afraid I'm at the limit of my ability to help here. I've not run a minion on my master and I'm not having any connection issues with my minions (even though my master is on my local home computer and behind a dynamic IP).
00:36 Brijesh1 joined #salt
00:36 west575_ joined #salt
00:37 orion Ok. I appreciate all the time you've given me today.
00:38 hemebond You're welcome. Hopefully someone else will pop their head in and be able to help. Good luck ????
00:40 nZac_ joined #salt
00:41 west575 joined #salt
00:42 nZac_ joined #salt
00:46 mapu joined #salt
00:48 hasues joined #salt
00:48 hasues left #salt
00:52 antpa joined #salt
00:53 racooper joined #salt
00:53 racooper joined #salt
01:01 Nahual joined #salt
01:02 edrocks joined #salt
01:04 tonybaloney joined #salt
01:08 catpigger joined #salt
01:09 Nahual left #salt
01:10 Nahual joined #salt
01:11 tonybaloney joined #salt
01:13 jhauser joined #salt
01:16 Brijesh1 joined #salt
01:19 AdamSewell joined #salt
01:20 PerilousApricot joined #salt
01:20 ZachLanich hemebond: So far, Salt Master didn't like installing onto Ubuntu 16. Ubuntu 15.04+ switched to systemd for their service framework and I tried switching it back to Upstart (which supposedly is as easy as 2 commands) and Salt still wouldn't work. So I went back to 14.04 for my master for now.
01:21 hemebond Hmm, that's odd.
01:21 flowstate joined #salt
01:21 hemebond I wonder if the master fails to run on Debian Jessie too.
01:21 hemebond Though I haven't had any trouble with minions.
01:26 ZachLanich Yea, minions seemed to work in vagrant, so I have high hopes. Just got this now too: http://dp.wecreate.com/13gfC
01:26 ZachLanich I can't catch a break with Salt AT ALL lol
01:26 hemebond How are you installing the master?
01:28 ZachLanich Bootstrap script
01:28 ZachLanich It said I could use apt-pkg but I also saw mentions that it is sometimes out of date, so I went for the script. Starting to wonder if I shouldn't have.
01:29 hemebond I think I installed using PIP.
01:29 ZachLanich Latest develop as well, also not sure that's a great idea, but I couldn't get it to recognize the latest tag either
01:29 hemebond https://github.com/saltstack/salt/issues/32761
01:29 saltstackbot [#32761][MERGED] Ubuntu 16.04 Xenial Xerus Support | Ubuntu 16.04 is out. http://releases.ubuntu.com/16.04/...
01:30 hemebond You downloaded the very latest version of the bootstrap script?
01:31 ZachLanich Yea
01:31 ZachLanich Thinking about going back to 16.04 and trying to apt-pkg
01:34 hemebond https://repo.saltstack.com/#ubuntu
01:34 hemebond Did you check that stuff?
01:41 ZachLanich hemebond: Boom, switched back to 16.04 and used apt and systemd and it works fine so far.
01:41 ZachLanich I would much rather use apt-pkg when possible
01:42 hemebond apt-pkg?
01:46 hemebond You mean APT?
01:46 tuxx joined #salt
01:46 hemebond Are you using the official Salt repo now?
01:46 Brijesh1 joined #salt
01:47 johnkeates joined #salt
01:48 johnkeates joined #salt
01:54 keldwud joined #salt
01:59 beardedeagle joined #salt
02:04 ninjada joined #salt
02:05 justanotheruser joined #salt
02:12 bastiand1 joined #salt
02:15 iceyao joined #salt
02:21 flowstate joined #salt
02:25 ninjada joined #salt
02:27 krymzon joined #salt
02:30 ZachLanich hemebond: I used apt to install it on the master, but I'm having trouble with this on the minions: script_args: -P git v2016.07.07 - I've tried it w/o the v too and I get a pathspec error. It obviously doesn't refer to real tags in the repo or something.
02:31 hemebond ZachLanich: The latest tag is v2016.3.1
02:32 hemebond Though you should be able to just let the bootstrap script do its thing and it'll install via APT and the official repos.
02:32 hemebond That's my understanding.
02:32 ZachLanich Really?: http://dp.wecreate.com/1hlVo
02:33 hemebond That's the bootstrap script./
02:33 hemebond The tag you specify is for the Salt repo tag.
02:33 ZachLanich Ahh
02:33 hemebond You're saying "install Salt via git and use the tag v2016.3.1" (or in your case v2016.07.07)
02:35 ninjada joined #salt
02:42 ZachLanich hemebond: Looks like the install worked :)
02:43 hemebond ????
02:49 sagerdearia joined #salt
02:53 ninjada joined #salt
02:54 antpa joined #salt
02:56 writtenoff joined #salt
03:01 iceyao joined #salt
03:05 edrocks joined #salt
03:05 PerilousApricot joined #salt
03:05 ZachLanich hemebond: So far, everything is working perfectly. All my states applied without issue :D
03:06 hemebond Nice :-)
03:06 ZachLanich I'm able to provision a web box with apache installed (default config for now), users created, ssh keys uploaded, mysql installed and configured for secure RDBMS, etc
03:07 ZachLanich I could have installed the master using the bootstrap, but I was messing up that git tag thing lol. Oh well.
03:08 ZachLanich Now I just have to figure out how to configure PHP7 + FPM and manage config files, create some dirs, etc and I have a full stack ready for tweaking
03:14 PerilousApricot joined #salt
03:15 sjmh joined #salt
03:20 flowstate joined #salt
03:21 ninjada joined #salt
03:21 ninjada joined #salt
03:22 ninjada joined #salt
03:27 krymzon joined #salt
03:47 antpa joined #salt
03:59 ZachLanich hemebond: How do I get a2enmod proxy_fcgi to fire after apache is installed in a state file?
03:59 ZachLanich I tried this and it errors out: http://dp.wecreate.com/14tv0
03:59 hemebond ZachLanich: Are you using the Apache formula or writing the states yourself?
03:59 ZachLanich Writing myself
04:00 hemebond Does service.running have a cmd parameter? Doesn't look right anyway.
04:00 ZachLanich I'd like to make all this work myself before starting to rely on other ppl's millions of lines of code that I don't fully understand the implications of. I'm not doing anything crazy complicated atm.
04:01 ZachLanich I saw it on StackOverflow, but it seems broken.
04:01 hemebond You want a state that uses cmd.run (if that's a state module)
04:01 hemebond I have to pop out. Back soon.
04:03 evle joined #salt
04:04 orion I think I found a bug in 2016.3.1: https://gist.github.com/centromere/f011c441641d17676033e12b220eb145
04:06 armguy joined #salt
04:12 orion hemebond: I also think I located the source of the issue we were discussing earlier.
04:15 rdas joined #salt
04:20 flowstate joined #salt
04:21 PerilousApricot joined #salt
04:24 barajasfab joined #salt
04:40 ZachLanich hemebond: All good so far. Full web box!: http://192.241.150.186/
04:41 hemebond ????
04:41 hemebond orion: Oh really?
04:42 hemebond orion: That definitely looks like a bug. I can see the try/except and that IP = IPDB() line doesn't use the HAS_PYROUTE2 flag.
04:43 orion hemebond: Shall I file a report?
04:43 hemebond Yes please, if there isn't one open already for it.
04:44 orion Other than a versions report and the contents of the gist, is there anything else I should include?
04:44 hemebond They should be able to find it with just that information.
04:44 orion ok
04:45 hemebond It _seems_ incorrect to me; though I'm not at all familiar with that code.
04:47 orion https://github.com/saltstack/salt/issues/34978
04:47 saltstackbot [#34978][OPEN] Beacon error being reported in logs | The following is appearing in `/var/log/salt/minion`:...
04:52 brent_ joined #salt
04:56 syndikate joined #salt
05:02 alexhayes joined #salt
05:04 impi joined #salt
05:05 JPT joined #salt
05:07 beardedeagle joined #salt
05:09 edrocks joined #salt
05:09 beardedeagle joined #salt
05:11 Brijesh1 joined #salt
05:12 ZachLanich hemebond: Any reason why this doesn't work, even though it was supposedly added to Salt in 2012?: service.running: - watch: - file: /etc/apache2/mods-enabled/*
05:12 hemebond ZachLanich: :-)
05:13 hemebond because watch doesn't watch files.
05:13 hemebond It watches states and modules.
05:13 hemebond Salt only knows about the files and things that it looks after.
05:14 hemebond So watch in this case would point to a state that manages a file, e.g., /etc/apache2/mods-enabled/mymod.conf
05:14 ZachLanich Ah, yea just found that. How can I make apache restart when mods are enabled?
05:14 hemebond You have to watch the state that enables the mod.
05:15 ZachLanich Ok, so how would I do that given this?: http://dp.wecreate.com/vbBA
05:15 ZachLanich Or if there's a better way to do what I'm doing, let me know.
05:15 hemebond watch cmd: 'sudo a2enmod rewrite'
05:16 hemebond Does the a2enmod do anything other than create symlinks?
05:17 ZachLanich Probably not
05:17 hemebond Then just create the symlink yourself :-)
05:17 ZachLanich So can I just replace my cmd calls completely like this?: http://dp.wecreate.com/1iVDS
05:17 hemebond No, you can't just watch a random file.
05:17 hemebond It has to be a file Salt manages.
05:18 ZachLanich Then explain how to use your "watch cmd: 'sudo a2enmod rewrite'" suggestion
05:18 hemebond Salt will see that it has updated or changed the file and notify the other states that are watching it.
05:18 hemebond I think the module is file.symlink
05:18 hemebond !file.symlink
05:18 hemebond !states.file.symlink
05:18 wm-bot4 http://docs.saltstack.com/en/latest/ref/states/all/salt.states.file.html#salt.states.file.symlink
05:18 ZachLanich No, I mean without making my own symlink
05:18 ZachLanich I don't see the benefit in that
05:18 ZachLanich I'm running a command either way
05:18 hemebond Well a2enmod is just a way to manually manage the Apache configs.
05:19 hemebond It's not really required when you're using your own configuration management.
05:19 hemebond Unless a2enmod installs the mod too or something.
05:19 ZachLanich Either way, I have to run a command to create the symlink lol
05:19 hemebond You're not running a command to create the symlink, you're writing a state that will create it for you.
05:19 ZachLanich So I need to make apache restart when those are enabled during state.apply
05:20 hemebond You can then watch that symlink state for changes.
05:20 ZachLanich Ah ok.
05:20 hemebond Yes, when the symlink is created or updated the state is identified as a change, this will trigger all watchers.
05:20 flowstate joined #salt
05:26 ZachLanich This seems to have worked: http://dp.wecreate.com/1hJYa - Thanks :)
05:26 felskrone joined #salt
05:26 hemebond ????
05:28 ZachLanich I'm seeing this issue every time I'm salt-cloud -d these DO boxes though: Error: There was an error destroying machines: An error occurred while querying DigitalOcean. HTTP Code: 404  Error: '{"id":"not_found","message":"The resource you were accessing could not be found."}'
05:28 irctc454 joined #salt
05:28 ZachLanich It destroys is successfully though
05:28 hemebond I'm not familiar with the DO module unfortunately.
05:30 Brijesh1 joined #salt
05:33 irctc454 Hi, I installed salt-minion 2015.8.10 through bootstrap script. When i checked the salt-minion process i noticed that there were two processes running.
05:33 irctc454 I see that amount of RSS memory consumed is (2.3 MB + 3.1MB) , total of 5.4 MB. Is this expected behaviour.
05:33 irctc454 root     14923  1.5  1.1 137952 23336 ?        Ss   21:21   0:00 /usr/bin/python /usr/bin/salt-minion root     14929  2.5  1.5 377012 31544 ?        Sl   21:21   0:00 /usr/bin/python /usr/bin/salt-minion
05:34 hemebond irctc454: Yes, expected.
05:34 hemebond One of my minions has four salt-minion processes
05:42 elias_ joined #salt
05:44 irctc454 sometimes we have seen that after deploying configuration a total of (24 + 37) = 61 MB is consumed.  Even after clearing the cache i dont dee any change in RSS memory.
05:45 irctc454 How can i keep the memory consumption by salt minion as low as possible. Please provide some suggestions
05:46 honestly You could check out salt-ssh, it works without persistent minions
05:48 honestly Definitely less comfortable though and not 100% feature parity
05:49 irctc454 Well.. thanks for the suggestion. I will try that out. We have 100's of minions connected in persistent manner. We have noticed sometimes that few saltminions consume 100% cpu
05:49 irctc454 when we delete the cache and restart salt-minion it comes back to normal
05:49 Miouge joined #salt
05:50 irctc454 can you provide some insight on , the possible reasons for few salt-minions to consume 100% CPU sometimes
05:51 armyriad joined #salt
05:52 honestly Python gnomes, CPU gnomes, gnomes in one of the libraries salt uses
05:52 honestly Hard to tell with the gnomes
05:53 honestly (this is not an enterprise support channel, consider paying for support if you want useful answers to questions like that)
05:54 honestly (or investigate the problem yourself and post to the issue tracker on github)
05:56 irctc454 Ok.. thanks.
05:59 Brijesh1 joined #salt
05:59 manji joined #salt
06:03 armyriad joined #salt
06:03 fracklen joined #salt
06:12 TyrfingMjolnir joined #salt
06:12 orion There should be a command in the inotify module which allows you to see which hooks are installed.
06:17 armyriad joined #salt
06:19 rdas joined #salt
06:20 flowstate joined #salt
06:20 jtang joined #salt
06:24 jrose1030 joined #salt
06:26 cmek joined #salt
06:30 krymzon joined #salt
06:32 ravenx joined #salt
06:32 ravenx is there a difference between #salt and #saltstack?
06:34 ZachLanich hemebond: I'm having issues with apache restarts failing on a first highstate. I've ran debug and collected all of the relevant info into a gist. Plz help!: https://gist.github.com/zlanich/1ddb65d3812c014c5a103457e3732f17
06:40 ZachLanich hemebond: Here's my sls: https://gist.github.com/zlanich/4e46af01c46e6c8ce18a6c275527523f
06:40 syndikate joined #salt
06:41 armyriad joined #salt
06:43 hemebond joined #salt
06:45 raspy_ joined #salt
06:45 ZachLanich_ joined #salt
06:45 raspy_ hi all, is there a way to have a state execute if any 2 other states trigger
06:46 raspy_ i want one state to execute if either a pkg is installed or if a file is modified
06:47 ZachLanich__ joined #salt
06:51 kawa2014 joined #salt
06:52 beardedeagle joined #salt
06:59 raspy_ onchanges is not an OR?
07:06 hemebond raspy_: watch?
07:06 hemebond onchanges?
07:06 hemebond Probably onchanges I think.
07:06 hemebond https://docs.saltstack.com/en/latest/ref/states/requisites.html#onchanges
07:07 raspy_ yeah im doing a onchanges: - file: jail-override
07:07 raspy_ but doesnt seem to work if I have to under onchanges:
07:08 hemebond That seems like a bug. The docs explicitly say "If a state has multiple onchanges requisites then the state will trigger if any of the watched states changes."
07:09 raspy_ hmm ok thx hemebond
07:12 edrocks joined #salt
07:18 ZachLanich__ hemebond: When you get a second, take a look at the last two things I posted to you. I don't know what I'm doing wrong with Apache.
07:19 hemebond ZachLanich__: I don't see any recent messages from you.
07:19 hemebond Appears I temporarily lost connectivity.
07:20 flowstate joined #salt
07:26 toanju joined #salt
07:26 ZachLanich__ joined #salt
07:27 POJO joined #salt
07:27 ZachLanich joined #salt
07:28 ZachLanich hemebond: I'm getting errors with apache starting when doing a provision, straight into a highstate: https://gist.github.com/zlanich/1ddb65d3812c014c5a103457e3732f17 - My SLS: https://gist.github.com/zlanich/4e46af01c46e6c8ce18a6c275527523f
07:30 hemebond So have you checked the resulting Apache config for errors?
07:30 hemebond Tried starting it manually to see what the error is>
07:30 hemebond ?
07:30 krymzon joined #salt
07:31 CeBe1 joined #salt
07:33 ZachLanich__ I can't remember what I did. It worked fine after installing it manually & running the state overtop if it though. I'm crawling into bed, so I'll check those things tomorrow :)
07:33 jrose1030 joined #salt
07:34 manji joined #salt
07:34 babilen ZachLanich__: What does "systemctl status apache2.service" give you?
07:34 kshlm joined #salt
07:35 babilen Ah .. good night
07:37 dh__ joined #salt
07:37 cmek joined #salt
07:37 karlthane joined #salt
07:38 stack joined #salt
07:39 saltsa joined #salt
07:39 ronnix joined #salt
07:48 antpa joined #salt
07:51 antpa joined #salt
07:51 ozux joined #salt
07:55 stooj joined #salt
07:56 mohae_ joined #salt
07:56 mikecmpbll joined #salt
07:56 Brijesh1 joined #salt
08:01 ribx joined #salt
08:04 Rumbles joined #salt
08:06 cyborg-one joined #salt
08:07 snaggleb joined #salt
08:07 saltsa_ joined #salt
08:09 tawm04 joined #salt
08:10 whytewolf joined #salt
08:10 renoirb joined #salt
08:11 thraxil joined #salt
08:14 ivanjaros3916 joined #salt
08:16 tho joined #salt
08:17 hoonetorg joined #salt
08:18 Brijesh1 joined #salt
08:19 flowstate joined #salt
08:20 lero joined #salt
08:20 Rumbles joined #salt
08:23 kaushal_ joined #salt
08:25 AvengerMoJo joined #salt
08:27 GreatSnoopy joined #salt
08:31 s_kunk joined #salt
08:32 kawa2014 joined #salt
08:34 impi joined #salt
08:41 sagerdearia joined #salt
08:43 Cadmus joined #salt
08:46 iceyao joined #salt
08:52 JPT joined #salt
08:53 woodtablet joined #salt
08:54 MTecknology joined #salt
08:55 ivanjaros joined #salt
09:00 renoirb joined #salt
09:02 CeBe1 joined #salt
09:03 TyrfingMjolnir joined #salt
09:06 TyrfingMjolnir joined #salt
09:06 MTecknology joined #salt
09:09 kbaikov joined #salt
09:11 krymzon joined #salt
09:14 edrocks joined #salt
09:14 Rumbles joined #salt
09:15 Cadmus Hi, I'm using the jenkins-saltstack plugin which uses the cherrypy local_batch function, but I can't figure out where to put in the equivalent of --batch-wait .
09:17 west575 joined #salt
09:17 kbaikov joined #salt
09:20 flowstate joined #salt
09:23 kaushal_ joined #salt
09:27 manji joined #salt
09:28 iceyao joined #salt
09:34 willh joined #salt
09:39 stopbyte joined #salt
09:42 iceyao joined #salt
09:50 alexhayes joined #salt
09:51 Rumbles joined #salt
09:52 ronnix joined #salt
09:55 fracklen joined #salt
09:55 jhauser joined #salt
10:07 fracklen_ joined #salt
10:09 N-Mi joined #salt
10:09 N-Mi joined #salt
10:13 fracklen joined #salt
10:20 flowstate joined #salt
10:22 fracklen_ joined #salt
10:24 kaushal_ joined #salt
10:29 ajw0100 joined #salt
10:35 rylnd joined #salt
10:38 raspy_ joined #salt
10:44 JohnnyRun joined #salt
10:45 ingslovak joined #salt
10:48 flowstate joined #salt
10:54 badon_ joined #salt
10:58 teryx510 joined #salt
11:02 p3rror joined #salt
11:03 west575 joined #salt
11:15 amcorreia joined #salt
11:18 edrocks joined #salt
11:18 Kelsar joined #salt
11:19 Brijesh1 joined #salt
11:21 alexhayes joined #salt
11:24 kaushal_ joined #salt
11:25 flowstate joined #salt
11:33 ribx joined #salt
11:46 viq https://pbot.rmdir.de/lvJsafxhx6LsHa3GiUsW-g  yet salt claims "[DEBUG   ] Error loading module.augeas_cfg: Cannot load augeas_cfg module: augeas python module not installed"  - any ideas?
11:47 viq oh, right, versions-report, here we go https://pbot.rmdir.de/U9TwWQrAIPm7gOBdR_mJIQ
11:52 JohnnyRun joined #salt
11:53 babilen viq: Could you open a Python REPL and run "from augeas import Augeas as _Augeas" -- Does that work?
12:15 badon_ joined #salt
12:18 berto- joined #salt
12:23 JohnnyRun joined #salt
12:23 Sarphram joined #salt
12:24 kaushal_ joined #salt
12:32 s_kunk joined #salt
12:39 datamann joined #salt
12:42 numkem joined #salt
12:45 johnkeates joined #salt
12:48 mapu joined #salt
12:51 infrmnt joined #salt
12:56 west575 joined #salt
12:56 ronnix joined #salt
13:04 TooLmaN joined #salt
13:04 johnkeates joined #salt
13:08 racooper joined #salt
13:09 ssplatt joined #salt
13:09 POJO joined #salt
13:10 dyasny joined #salt
13:11 antpa joined #salt
13:12 viq babilen: https://pbot.rmdir.de/mRPw8U-5S9ACNeE7hep1Mw
13:12 babilen And therein lies your error
13:12 viq which doesn't sound good...
13:13 mikecmpb_ joined #salt
13:13 babilen Yeah, that's a nice one :)
13:13 flowstate joined #salt
13:13 babilen Could you do me a favour and ask about that error on #python?
13:14 babilen Do you have libpython2.7 installed?
13:15 viq Apparently I don't, let me see if installing will fix this
13:15 viq yeah, now it imports fine
13:15 babilen It should .. you will have to restart the minion (or install with reload_modules: True)
13:15 babilen Cool, you should also have augeas_cfg now
13:16 viq yeah, trying
13:16 viq As for restarting minion, that's a test-kitchen environment, so AFAIK it doesn't even really start the salt-minion, and I'm testing through salt-call anyway
13:16 babilen *nod*
13:17 babilen Just wanted to make sure that your minion picks up on it
13:17 ivanjaros joined #salt
13:18 viq yeah, thank you
13:20 raspy_ joined #salt
13:21 viq OK, seems it was the libpython2.7 missing, I wonder how that happened o.O
13:22 datamann joined #salt
13:23 babilen Debian?
13:24 kaushal_ joined #salt
13:25 infrmnt joined #salt
13:31 viq yeah, debian 7
13:32 babilen https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=734557
13:33 viq ah, thank you
13:35 mikecmpbll joined #salt
13:39 viq huh, now it's saying "Error: Unable to save to file!"
13:40 viq via state, via augeas.setvalue it returns true but doesn't do any changes...
13:41 nicola_pav joined #salt
13:41 babilen That fucker
13:41 babilen More details in the minion log?
13:42 viq Ah, now managed to make the augeas.execute give same error, though with no explanation
13:42 nicola_pav Hi all. I am trying to use salt-cloud against openstack. I have prepared a profile and a provider.
13:42 Tyrm joined #salt
13:42 nicola_pav I am able to create instance but salt-cloud cannot login to the machine. I am using a password
13:42 nicola_pav any hints please?
13:43 perfectsine joined #salt
13:43 viq https://pbot.rmdir.de/7QAEw9lJQP54k3BL6mb22g
13:44 viq nicola_pav: maybe try stuff from https://docs.saltstack.com/en/latest/topics/cloud/saltify.html  ?
13:45 nicola_pav viq: I will try now - thanks
13:49 tapoxi joined #salt
13:51 Ryndaxon joined #salt
13:51 pfallenop joined #salt
13:51 tapoxi anyone using salt for monitoring? thoughts on how to only alert after x number of failures?
13:52 nicola_pav viq: I tried it - I create the instance but it seems it does not use the password in the profile
13:52 nicola_pav viq: it keeps returning Permission denied after the instance is created
13:53 PerilousApricot joined #salt
13:53 viq nicola_pav: 'ssh_password' in openstack settings maybe?
13:54 antpa joined #salt
13:56 west575_ joined #salt
13:56 Pulp joined #salt
13:58 nicola_pav viq: tried it but no luck :(
13:59 nicola_pav viq: in the password I have ! - should I escape it?
13:59 viq Or put the whole thing in ''
14:00 * viq is mostly guessing here
14:00 ronnix_ joined #salt
14:01 protoz joined #salt
14:02 mpanetta joined #salt
14:02 viq https://github.com/saltstack/salt/blob/develop/salt/cloud/clouds/openstack.py mentions some 'data.extra' but I have no idea what that is and how to get at it
14:04 viq meh, I guess time to throw out the augeas idea and start templating files
14:04 Tyrm joined #salt
14:05 Brew joined #salt
14:05 nicola_pav viq: put the whole thing in '' but still no luck :(
14:06 viq nicola_pav: sorry, time I spent (zero) with cloud in general and openstack in particular is not enough to have any further advice
14:07 viq nicola_pav: any way you could put your master's ssh key into the VMs ahead of time, or on provisioning?
14:07 nicola_pav viq: thanks anyway - I will continue looking
14:07 viq I think OpenStack actually supports that upon machine creation?
14:07 nicola_pav viq: I hoped I could do this via salt-cloud
14:07 Tanta joined #salt
14:08 Andrew joined #salt
14:09 viq tapoxi: https://github.com/lincolnloop/salmon
14:10 hasues joined #salt
14:10 hasues left #salt
14:13 barajasfab joined #salt
14:13 Andrew joined #salt
14:14 teryx510 joined #salt
14:14 PerilousApricot Hello all - does anyone know if there is an example of a file_tree ext_pillar somewhere? I'm struggling to get it working locally :(
14:14 PerilousApricot And learning-by-example is always great
14:14 _JZ_ joined #salt
14:17 viq PerilousApricot: https://docs.saltstack.com/en/latest/ref/pillar/all/salt.pillar.file_tree.html  ?
14:17 CTaylor joined #salt
14:17 PerilousApricot viq: I tried to follow that back and forth, my minions don't see any sign of the pillar and I'm not sure why
14:18 viq show your config?
14:18 pcdummy PerilousApricot: you work with the saltenv "base" ?
14:19 PerilousApricot viq: absolutely, let me collect the different little snippets into a gist
14:19 fredvd joined #salt
14:19 pcdummy I don't see an option to change the saltenv with that ext_pillar.
14:21 mikecmpb_ joined #salt
14:23 kshlm joined #salt
14:23 tapoxi viq - I saw that, it looks awesome
14:23 viq tapoxi: except it's been dead for over a year, IIRC
14:24 tapoxi yeah I've been debating forking it. I don't really need a UI though
14:24 kaushal__ joined #salt
14:25 tapoxi thinking of a salty way to do this but I'm thinking having beacons fire when stuff goes bad and having a reactor see if its met alarm conditions and send off to pagerduty
14:26 jenastar joined #salt
14:26 pcdummy tapoxi, viq: i have big interests in a salt based monitoring solution, i'll help on a fork.
14:28 viq tapoxi: look up cfgmgmt and sensu (and maybe stackstorm)
14:28 viq I want to use salt to deploy icinga
14:28 tapoxi yeah I have sensu now
14:28 tapoxi but if salt can do the same thing I can 1) eliminate ruby and 2) save money on the sensu vm
14:28 * viq is programming-illiterate, therefore rather unsuitable for such projects
14:29 cmarzullo At saltconf the talk about what thorium can bring to the table might help with the monitoring.
14:29 wwalker we're using gitfs for everything, been working fine.  I have some data we want kept in a pillar, but I need access to the data from a program that a state will run.  Is there a way to get a path to the cached pillar? (or a better way to do it?)
14:30 tapoxi I set up a big icinga 1.x install at my last place, it was easy enough. I can't imagine automating its setup would be that hard
14:30 cmarzullo wwalker: take your pillar your interested in and stick it in a file on a host. just do {{ myvalue | yaml }}
14:30 tapoxi cmarzullo thorium is in 2016.3 right?
14:30 gh34 joined #salt
14:31 wwalker cmarzullo: thank you.
14:32 PerilousApricot viq: sorry, it appears comcast is wanting to make my life difficult. nearly done
14:32 pcdummy tapoxi: i use OMD for that http://omdistro.org/
14:33 viq pcdummy: last I looked OMD had ancient versions of stuff
14:33 pcdummy viq: for me its great
14:33 tapoxi I've heard OMD was okay, I set them up on wheezy with a few puppet scripts for management
14:33 ageorgop joined #salt
14:33 perfectsine joined #salt
14:33 tapoxi I ran icinga-web on a separate box so they could share a ui
14:34 pcdummy I use it with check_mk and wato and will automaticaly configure wato in the near future.
14:34 Sketch if you don't need everything in OMD, you can use check_mk raw edition
14:34 Sketch https://mathias-kettner.de/check_mk_news.html
14:34 PerilousApricot viq: this is my config https://gist.github.com/PerilousApricot/ae93fbd955bcf7d16f70a3c9aa6e23af
14:35 pcdummy Sketch: its with nagios, i like icinga more
14:35 ozux joined #salt
14:35 tapoxi you can run check_mk with icinga 1.x
14:36 cmarzullo tapoxi: Yeah it's still early dev. Has promise although I don't have a need for it yet.
14:36 Sketch i haven't used icinga, but it doesn't seem like it would matter much either way if you're using CMK
14:36 tapoxi icinga 2 is a rewrite so it's not completely nagios compatible
14:36 Sketch if you're not, then yeah, cmk raw wouldn't make much sense
14:37 pcdummy I want the IPv6 support of the newer check_mk versions in OMD, need to check versions agains.
14:37 bluenemo joined #salt
14:37 viq PerilousApricot: anything in master logs?
14:38 viq Or minion logs for that matter
14:38 mpanetta joined #salt
14:39 mpanetta joined #salt
14:39 PerilousApricot viq: let me bounce both of them and see if anything interesting comes by
14:39 pcdummy OMD doesn't deliver Check_MK 1.2.7i3 but 1.2.6p12 :/
14:39 viq pcdummy: as I said :P
14:40 Sketch 1.2.7 is a beta release.. 1.2.8pX is the new stable release
14:40 pcdummy ahh ok, still its not in OMD :/
14:40 Sketch yeah, omd is usually quite a bit behind
14:40 garphy joined #salt
14:41 mikecmpbll joined #salt
14:41 bowhunter joined #salt
14:42 PerilousApricot viq: Minion looks OK, master is complaining: [CRITICAL] Specified ext_pillar interface file_tree is unavailable ...
14:42 ALLmightySPIFF joined #salt
14:42 PerilousApricot viq: Not sure what I could've done wrong, I installed the master via a predefined docker image
14:42 west575 joined #salt
14:43 yuhlw_ Hi, we are mixing debian as a master (version: 2015.5.3) with ubuntu 16.04 (version 2016.3.1). And we get on the minion:"[ERROR   ] Got a bad pillar from master, type str, expecting dict:"
14:43 viq PerilousApricot: try starting master with "-l debug" ?
14:43 Xopher joined #salt
14:44 viq yuhlw_: master needs to be same or newer than minions or Bad Things Will Happen
14:45 yuhlw_ viq: so in our case, as there is no lithium version for ubuntu 16.04, this will forces us to migrate to Boron ?
14:46 yuhlw_ viq: I know that I can install it myself. But I was talking about version prepackaged.
14:46 viq yuhlw_: ...I never was good with codenames, numbers any day please ;)
14:47 _mel_ joined #salt
14:47 Brijesh1 joined #salt
14:47 viq yuhlw_: and, uhm, you're already on 2016.3 on your minions, so it isn't much of "migration"
14:47 shawnbutts joined #salt
14:48 babilen And your master should be at least as recent as your minion
14:48 babilen yuhlw_: Which Debian version is that?
14:48 yuhlw_ babilen: Debian GNU/Linux 8.5
14:49 jgarr is there a way to watch state.event for only specific events? Ideally I can filter out all the minion pings and only watch for my custom reactor event being fired
14:49 aswini joined #salt
14:49 babilen yuhlw_: You'll find 2016.3.0 in jessie-backports (with some patches) or 2016.3.1 on http://repo.saltstack.com/#debian
14:49 yuhlw_ babilen, viq: thanks for your help.
14:50 yuhlw_ yes, we will use it. I'll ugrade my master.
14:50 babilen Ah, .1 hit testing so I could backport that now
14:51 poiuasdf joined #salt
14:51 raspy_ joined #salt
14:51 flowstate joined #salt
14:52 catpig joined #salt
14:54 __mel__ joined #salt
14:55 _mel_ joined #salt
14:56 _mel_ joined #salt
14:56 flowstat_ joined #salt
14:56 lungaro joined #salt
14:58 _mel_ joined #salt
14:58 __mel__ joined #salt
14:59 __mel__ left #salt
15:00 flowstate joined #salt
15:01 yuhlw_ babilen,viq : many thanks it works again like a charm.
15:04 flowstate joined #salt
15:05 melvin joined #salt
15:06 PerilousApricot viq: Hmm, I don't see anything useful in there. The line before the one I sent before is simply: [INFO    ] AES payload received with command _pillar
15:08 flowstate joined #salt
15:10 flowstat_ joined #salt
15:12 fracklen joined #salt
15:12 perfectsine joined #salt
15:18 corichar joined #salt
15:19 rdas joined #salt
15:21 antpa joined #salt
15:22 ZiLi0n joined #salt
15:22 iceyao joined #salt
15:23 flowstate joined #salt
15:25 ZiLi0n Hello everyone. I would like to collect some data from few servers where there are minions running. The idea is to collect the data using the minions, then store (like in a database), and perform some work on it. I have read about Salt returners, as a way to store job results in external systems. My question is... if I configure an external system, would the result of all jobs be sent to the external system, or only the data collected by certain so
15:25 ZiLi0n me Salt custom modules?
15:26 brent_ joined #salt
15:26 armin hi. in this example: https://docs.saltstack.com/en/getstarted/config/jinja.html (first text box on that page), what does that "apache: httpd" mean? what's the key and what's the value representing in that example?
15:27 raspy_ Hi all, in salt is there a way to run a command as a certain user?   I want to run "google-authenticator -tfud -w 10" but it needs to run on each individual user that will require 2factor authentication
15:27 corichar joined #salt
15:28 racooper armin,  if you keep reading that page will explain it
15:28 antpa joined #salt
15:29 armin racooper: i read the entire page already.
15:32 armin racooper: in that example, the apache: httpd statement is not indented, at all, so i'm asking myself what could be meant. if i literally use that example i'll get an error on a salt-call, hence i'm asking...
15:32 perfectsine_ joined #salt
15:35 raspy_ joined #salt
15:35 asdf__ joined #salt
15:35 tapoxi joined #salt
15:35 tapoxi anyone know of good documentation on the thorium reactor?
15:36 Tanta tapoxi: https://whatisnuclear.com/articles/thorium.html
15:36 tapoxi this looks like it'll make my monitoring easier
15:36 tapoxi Tanta
15:36 tapoxi https://www.youtube.com/watch?v=9CdVTCDdEwI
15:37 Tanta sorry, couldn't resist
15:37 mpanetta_ joined #salt
15:38 dfinn joined #salt
15:41 krymzon joined #salt
15:42 ALLmightySPIFF joined #salt
15:43 ingslovak joined #salt
15:44 POJO joined #salt
15:46 beardedeagle joined #salt
15:46 mpanetta joined #salt
15:47 cmarzullo armin: what they are showing is how you can use jinja templating. What isn't clear is that they are templating their pillar data.
15:48 cmarzullo But you can use jinja everywhere. You can template your pillar, your states, and the files that your states call.
15:49 racooper it does state that it's pillar data. "This example checks Salt grain values to set OS specific Salt pillar keys."
15:49 flowstate joined #salt
15:49 armin cmarzullo: to me it was unclear whether the apache word there is the id declaration, and whether or not (or in what case) an id declaration may have a 2nd word on the same line, after the ":".
15:50 cmarzullo it's in pillar, not states
15:50 cmarzullo it's just a variable.
15:50 armin ah. gotcha.
15:51 cmarzullo raspy_: you can do that with cmd.run
15:51 cmarzullo https://docs.saltstack.com/en/latest/ref/states/all/salt.states.cmd.html#salt.states.cmd.run
15:52 ronnix joined #salt
15:52 armin i have major problems understanding the terminology of elements in a typical .sls file. is there any good explanation of this? the docs is full of examples, but searching for a specific TYPE of elements that may occur there seems impossible to me.
15:53 cmarzullo in a state sls file you have a collection of state ids. then the state module declaration. Then the arguments for the state module delcaration
15:55 armin ok, so i just came up with the following .sls file here that i wrote: https://paste.fedoraproject.org/396455/34883146/raw/
15:55 armin how do i describe in english language and with human speech that he httpd keyword is one level more indented ("below"?) the pkgs keyword there?
15:56 armin cmarzullo: thanks for your explanations!
15:56 antpa joined #salt
15:56 flowstate joined #salt
15:57 cmarzullo One thing to keep in mind is that when you doing jinja, 90% of the time you are just using it to generate yaml.
15:57 armin cmarzullo: my understanding was 100%, because it's to be understood as a pre-processor.
15:58 cmarzullo great. so you sorted/
15:58 cmarzullo ?
15:58 armin cmarzullo: i don't know. "pkgs" would be an argument for the module delcaration "pkg.latest" here. correct?
15:59 cmarzullo correct
15:59 armin cmarzullo: is "- httpd" an argument for an argument for the module declaration then? :)
15:59 cmarzullo it's just the value pkgs
15:59 cmarzullo in this case it's an array.
16:00 cmarzullo You could also write it like - pkgs: ['http','mariadb']
16:00 cmarzullo but that'd be tough to tempalte
16:00 armin so my additional indentation is just plain wrong?
16:00 cmarzullo not it's right.
16:00 armin :)
16:00 cmarzullo err. no it looks right to me.
16:01 PerilousApricot viq: Ohhhhhhh, it looks like this docker image is using "salt-master=2014.1.11+ds-2trusty1" ... that's not right :/
16:01 viq :P
16:01 viq PerilousApricot: any particular reason why docker?
16:02 viq Also, trusty? Isn't that long dead?
16:02 PerilousApricot viq: It's my dev machine, and I try and use docker for "playing around" things so it doesn't get all cluttered
16:02 armin cmarzullo: i just tried. if i put the "- httpd" line on the same level of indentation as the "- pkgs" line, i get "Too many functions declared in state 'pkg' in SLS 'apache'"
16:02 armin oh wow, this reminds me why i hate yaml so much.
16:02 viq PerilousApricot: vagrant ;)
16:03 cmarzullo when you do a salt-call -l debug you should see the rendered yaml output
16:03 PerilousApricot Well, I saw the image in docker hub, and it was recently updated. I assumed it was a recent build
16:03 viq PerilousApricot: and/or lxc / vagrant-lxc
16:03 PerilousApricot viq: Eh, it's 6 of one half-dozen of another. The isolation/VM software wasn't the problem. I was just installing an ancient version like a dummy
16:04 cmarzullo yeah you couldn't do that. since pkgs is a hash
16:04 viq yah, I just don't think I agree docker is a good place to test things that normally run on full OS
16:04 armin hash as the perl style "hash" word? like a dictionary?
16:05 cmarzullo sorry yeah it dict
16:05 armin cmarzullo: ahhh, now i get it. each of the elements following then are members of the pkgs dict?
16:06 cmarzullo so the whole thing is a dict. with the first key being apache. which gets mapped into the state id. The value of the the apache key is another dict. With pkg.latest as it's key
16:06 armin cmarzullo: ok, good. thank you!!!
16:07 cmarzullo you are welcome
16:08 armin 2 further dumb noob questions: a) why am i able to specify any state id identifier for the /tmp/foobar state id there? b) what is that "[]" i keep seeing after "pkg.latest: " statements often?
16:08 ssplatt [] makes an empty list
16:08 ssplatt its like initializing a variable, like myvar = “”
16:08 ssplatt ensures it’s there.
16:08 flowstate joined #salt
16:10 KingOfFools_ joined #salt
16:11 west575 joined #salt
16:11 cmarzullo When writing states I try to make the state ids more descriptive and ensure that we don't have collisions. By default it'll take the state id as the 'name' for the module. I don't like that.
16:11 cmarzullo All our stuff is like apache_config_sites_enabeld_mysite
16:12 flowstate joined #salt
16:12 cmarzullo Which is helpful when visually looking at the state output and allows better log analisys
16:12 KingOfFools_ Can someone tell me why lxc_conf_unset is mandatory argument in salt.modules.lxc.update_lxc_conf? :/
16:13 armin cmarzullo: fair enough. thanks for your valuable help again.
16:14 armin like, what i really noticed is that speaking with other co-workers about .sls files reveals that 3 people have 5 different terms for 1 thing...
16:14 cmarzullo heheh. it's important to have a common vocabulary. It's one of the charecterics of a tribe. (that's what my anthro prof wife tells me)
16:14 ssplatt takes time to get everyone on the same page, depending on how entrenched others are in competing tech
16:15 syndikate joined #salt
16:15 irctc170 joined #salt
16:17 armin cmarzullo: how many vitruvian man posters do you have in your home? :)
16:18 cmarzullo hehe.
16:19 antpa joined #salt
16:20 armin cmarzullo: scnr. greet her, and again thanks for your help. :-)
16:26 raspy_ joined #salt
16:34 lero joined #salt
16:41 impi joined #salt
16:43 krymzon joined #salt
16:50 Ryndaxon joined #salt
16:53 Ryndaxon Hi. I have a question regarding the REST API that Salt provides. How do I find out the structure of the data that is returned by a certain module if it is executed through the API? Do I need to query the API or is it documented somewhere?
16:53 whitenoise joined #salt
16:56 brotatochip joined #salt
16:57 tpaul joined #salt
17:00 cmarzullo could just do a curl and observe the result
17:01 Ryndaxon True, but I would have to set up all the software to get it working
17:01 cmarzullo it returns a dict.
17:01 Ryndaxon Just wondering if there might be an easier way
17:03 writtenoff joined #salt
17:04 LotR joined #salt
17:05 LotR joined #salt
17:10 flowstate joined #salt
17:12 ssplatt joined #salt
17:16 amcorreia joined #salt
17:17 edrocks joined #salt
17:18 aw110f_ joined #salt
17:20 west575_ joined #salt
17:21 antpa joined #salt
17:22 Brijesh1 joined #salt
17:22 Brijesh1 joined #salt
17:23 teryx510 joined #salt
17:24 antpa joined #salt
17:24 west575 joined #salt
17:28 west575 joined #salt
17:30 raspy_ hi all, how can trigger an sls manually?
17:30 raspy_ and not through the top file
17:31 raspy_ basically, I want to create an sls file that can be ran adhoc as the user needs to read the output from the commands this sls file will run
17:34 lero joined #salt
17:34 babilen raspy_: You can use state.sls or state.sls with the name of the sls you want to run
17:34 ivanjaros joined #salt
17:34 babilen state.sls or state.apply that is
17:35 raspy_ babilen: how exactly?
17:35 raspy_ do like a salt <state_file_name>.apply ?
17:35 flowstate joined #salt
17:35 lero joined #salt
17:36 babilen raspy_: salt 'theminion' state.sls path.to.the.sls
17:36 babilen (or state.apply .. doesn't make a difference)
17:39 raspy_ babilen: awesome thx!
17:39 babilen yw
17:41 LessSneaky joined #salt
17:44 west575_ joined #salt
17:44 ktosiek joined #salt
17:44 s_kunk joined #salt
17:53 Andrew joined #salt
17:56 Pulp joined #salt
17:57 manji joined #salt
17:58 dean joined #salt
18:00 cwyse joined #salt
18:01 bowhunter joined #salt
18:02 _beardedeagle joined #salt
18:03 ponyofdeath is there a way to detect if cmd.run fails and then do another action?
18:04 Tyrm joined #salt
18:06 west575 joined #salt
18:07 darvon joined #salt
18:07 mikecmpbll joined #salt
18:07 GnuLxUsr joined #salt
18:08 antpa joined #salt
18:09 Sketch you could make cmd.run run a script which tries to do something and if that fails does something else
18:14 tapoxi joined #salt
18:17 pipps joined #salt
18:17 raspy_ is there a way to pass a varaible into an sls ? for example  if I want to do "salt 'theminion' username=cooluser state.sls path.to.the.sls" can the sls file pick up that username var?
18:17 ponyofdeath Sketch: i want to do detect if a state fails or succeeds how is that done? ie if my state succeeds i want to touch a file if it fails i want to do something else
18:18 Sketch i don't know if there's any way to do that, though i'm far from an expert in salt
18:18 ponyofdeath Sketch: thanks
18:19 ALLmightySPIFF joined #salt
18:20 Miouge joined #salt
18:21 antpa joined #salt
18:23 tapoxi raspy_ not sure about sls but you can with a jinja template
18:23 raspy_ tapoxi: ok ill try thx
18:23 cmarzullo raspy_: you can pass pillar in on the cli. but it's awkward.
18:24 tapoxi raspy_ - defaults key : value
18:24 jenastar joined #salt
18:28 lero joined #salt
18:29 schemanic_ joined #salt
18:30 west575 joined #salt
18:35 lero joined #salt
18:35 west575_ joined #salt
18:35 toanju joined #salt
18:38 brotatochip joined #salt
18:38 Miouge joined #salt
18:40 catpig joined #salt
18:40 west575 joined #salt
18:43 pipps joined #salt
18:43 elias_ joined #salt
18:45 krymzon joined #salt
18:45 majuscule I'm trying to use https://github.com/saltstack/salt-contrib/blob/master/grains/ec2_info.py. I've added it to _grains and run salt '*' saltutil.sync_grains on the master, but i'm not seeing any new attributes on the minions. what might i be missing?
18:45 west575_ joined #salt
18:46 majuscule running salt-call grains.item ec2 -ltrace on a minion shows [TRACE   ] data = {'local': {'ec2': ''}}, which makes no sense
18:46 Miouge joined #salt
18:46 pipps joined #salt
18:49 pipps99 joined #salt
18:50 west575 joined #salt
18:51 pipps99 joined #salt
18:51 majuscule running the script on the master with python directly does show all the attributes
18:52 ZiLi0n Hello, where are retuners configured for master job cache? I just see retuner=mysql in master config file, but where is the user,pass,db,port to use filled out?
18:53 cmarzullo https://docs.saltstack.com/en/2015.8/ref/returners/all/salt.returners.mysql.html
18:53 cmarzullo gotta put them in a conf file also
18:53 brotatochip joined #salt
18:54 pipps99 joined #salt
18:56 majuscule is there an equivalent to runner_dirs for grain modules?
18:56 pipps_ joined #salt
18:56 mpanetta_ joined #salt
18:56 mikecmpbll joined #salt
18:57 mpanetta joined #salt
18:57 pipps__ joined #salt
19:02 dh joined #salt
19:04 ZiLi0n cmarzullo thanks, but is that conf file /etc/salt/master
19:04 ZiLi0n ?
19:04 Tyrm joined #salt
19:05 antpa joined #salt
19:06 brent_ joined #salt
19:07 cmarzullo Yeah it can be. or in the minion or in /etc/salt/master.d/extraconfigsireallywanttohave
19:09 antpa joined #salt
19:11 majuscule I've restarted the salt-master process. I don't think the minions are running the _grains/ec2_info.py code at all.
19:12 ZiLi0n cmarzullo thanks! do I need to configure 'event_return: postgres', 'return: postgres' as well or only 'return'. I am trying, but unless event_return is not commented out I don't see anything in the logs trying to connect to postgresql
19:12 majuscule Trying to peek inside with salt-call grains.item ec2 -ltrace 2>&1 | grep Loading, i only see core.* grains getting loaded
19:12 Tyrm joined #salt
19:13 cmarzullo I haven't don't much work with returners. Only so much to realize the elasticsearch returner is broken.
19:13 mikecmpbll joined #salt
19:15 ZiLi0n cmarzullo, thanks very much. I am getting 'error coudl not store return for events returner postgres not found'
19:15 babilen majuscule: Check it's virtual function and see if you have the required modules installed
19:16 babilen (Python modules that is)
19:16 ZiLi0n I don't really know if I need to install it, maybe it does not come with the default install from salt repo, or that the name I am referring too is wrong...
19:16 impi joined #salt
19:16 babilen majuscule: Ah, nvm .. it's a grain
19:17 ssplatt ZiLi0n: iirc returners are generally defined in the minion config
19:18 ssplatt not 100% cuaght up on what you were trying to accomplish, but maybe check that
19:18 pipps99 joined #salt
19:19 ZiLi0n ssplat, thanks. I was thinking on using master job cache return, so that the minions return to the master, and then the master redirect the a database
19:19 ssplatt ah
19:22 tedbot joined #salt
19:22 tedbot quick q: any preference to nodename vs localhost when referencing a grain??
19:23 majuscule babilen: thanks for thinking about it anyway :-), this is driving me insane
19:23 tedbot I'm about to commit some changes
19:23 majuscule i see this in the -ldebug: [DEBUG   ] Local cache dir: '/var/cache/salt/minion/files/base/_grains'
19:24 majuscule that dir didn't exist, i tried wiping all of /var/cache/salt, but it didn't seem to make any difference
19:24 majuscule (tried salt-call state.clear_cache as well but no difference)
19:28 ZiLi0n ssplatt you are right Sir, I see the minion trying to connect to the database as well
19:28 babilen majuscule: Is it executable?
19:28 fracklen joined #salt
19:28 ZiLi0n I though there is a way to do it only from the master, I haven't configured anything on the minion
19:29 babilen majuscule: Just saw that my custom grains all have the executable bit set, but I'm not sure if that's actually needed.
19:30 majuscule babilen: it is not, but i'll try that, thanks
19:30 majuscule i don't expect them to need it as runners aren't exec either?
19:31 majuscule yeah no difference :-/
19:31 babilen (and you synced again after cleaning caches?)
19:31 babilen So the copy on the minion is executable and you restarted the minion?
19:31 babilen I didn't really expect that to work, just something I noticed
19:34 alexhayes joined #salt
19:34 babilen majuscule: And you confimed that the grain itself works if you run it directly on the minion, didn't you?
19:39 xist joined #salt
19:39 pipps99 joined #salt
19:41 majuscule babilen: yes, it does
19:41 majuscule erm wait
19:41 majuscule it works when i run it on the master
19:41 majuscule there is no copy on the minion
19:41 majuscule it's not getting synced over when i run sync
19:41 majuscule it's just getting ignored
19:42 majuscule i have restarted both the master and minion processes
19:43 hasues joined #salt
19:43 hasues left #salt
19:44 babilen majuscule: How do you sync it and where does it live on the master?
19:52 mapu joined #salt
19:54 sagerdearia joined #salt
19:57 majuscule it lives in /etc/salt/config/_grains, i am following the documentation here: https://docs.saltstack.com/en/latest/topics/grains/#loading-custom-grains, and attemping to sync with salt '*' saltutil.sync_grains
19:58 flowstate joined #salt
19:58 majuscule and for the record, if i just scp the grainfile over to the minion and call it with python, it works
19:58 majuscule but that's not helpful of course :-)
20:02 tapoxi joined #salt
20:05 fredvd joined #salt
20:05 johnkeates joined #salt
20:05 ZiLi0n Hello I am trying to configure a external master job cache. As soon as I configure master_job_cache: postgres under master conf, and send a command to a minion, excetpions about could not serialize msgpack appears!
20:06 zmalone_ joined #salt
20:06 ZiLi0n someone using a master_job_cache with a database? maybe mysql?
20:06 babilen majuscule: Shoudln't that be /srv/salt/_grains/ ? (i.e. file_roots)
20:06 pipps99 joined #salt
20:07 pipps99 joined #salt
20:07 babilen majuscule: The master syncs grains from salt://_grains/ to the minions, so you have to put them somewhere in file_roots
20:09 babilen (oh, and highlight me or I might miss replies for a while)
20:10 majuscule babilen: yes, /etc/salt/config/ is listed as a file_root in my /etc/salt/master config file
20:10 babilen what?
20:10 babilen Okay .. well .. fair enough
20:10 yidhra joined #salt
20:11 babilen Would you mind, just for the sake of trying it, just use a more standard setup?
20:11 pipps_ joined #salt
20:11 tedbot left #salt
20:12 majuscule babilen: i did try putting it directly in /etc/salt/_grains, nothing in my configuration is in /src, and i'm unclear on what you're suggesting that from
20:12 majuscule the documentation i linked above says /etc/salt/_grains
20:12 babilen majuscule: Where do you keep your states and pillars?
20:13 majuscule babilen: /etc/salt/config/pillars
20:14 majuscule not sure about states (sorry, i am new to this system)
20:14 babilen Mind pasting you master config?
20:15 babilen But it is not customary to keep them in /etc .. but it should™ work I guess :)
20:15 brent_ joined #salt
20:15 babilen I mean the SLS files that actually do things (the file.managed, pkg.installed, ....)
20:17 babilen What does "salt 'someminion' cp.list_master" give you?
20:17 majuscule hrmm, not sure. i'm running find / -name file.managed etc, and not seeing anything at all, heh
20:17 majuscule i'll check that, one second
20:17 cmarzullo so I see that in the master config you can specifiy state_top: but I don't see a similar pillar_top: declaration.
20:17 Roelt joined #salt
20:17 cmarzullo anyone know about that? Maybe just not in the docs?
20:18 GreatSnoopy joined #salt
20:19 majuscule babilen: woah i'm getting some weird errors now, Failed to authenticate!, from root@saltmaster
20:19 jenastar1 joined #salt
20:19 babilen majuscule: Did you work through the tutorials already? And why did you decide to keep pillars in /etc ? I am referring to states as detailed in, for example, https://docs.saltstack.com/en/latest/topics/tutorials/states_pt1.html
20:19 babilen majuscule: Does "salt '*' test.ping" work?
20:20 majuscule nope, same error, it worked before, give me a minute to try to figure out wtf is going on :-)
20:20 majuscule thinking/hoping this is something unrelated
20:20 babilen Did you install and setup this system?
20:20 cyborg-one joined #salt
20:21 majuscule yes & no, haha. i provisioned and bootstrapped the system with another salt master and then switched it over to talk to itself.
20:22 majuscule that may sound a bit insane but i can explain ...
20:22 Trauma joined #salt
20:22 babilen No, I meant the highly opinionated setup
20:23 babilen (the /etc bit .. not that its wrong, but I simply wouldn'y expect someone completely new to salt to change that first)
20:23 majuscule the highly opinionated setup was configured by the original salt master
20:23 babilen And who configured that?
20:23 yidhra joined #salt
20:23 majuscule this machine was originally pointing at another salt master, and configured itself with a "salt_master" role
20:23 majuscule not me
20:23 babilen Okay
20:24 babilen So this might be a *very unusual* system in that it is likely that many bits and pieced have been changed in similarly opinionated ways?
20:24 babilen *pieces
20:24 majuscule however, i'm still a bit confused by your comments about it, as the documentation i linked *explicitly* states /etc/salt/grains
20:24 babilen Either way, did you get that cp.list_master running?
20:24 majuscule babilen: https://docs.saltstack.com/en/latest/ref/configuration/master.html
20:25 majuscule explicitly notes /etc/salt/master
20:25 babilen majuscule: That's for keeping grains directly on the minion.. The salt master synchronises grains to minion from salt://_grains
20:25 babilen (which is relative to file_roots)
20:25 majuscule hmm, okay
20:25 babilen And file_roots default to /srv/salt
20:25 majuscule and no, i've been distracted explaining this :-)
20:25 majuscule let me go work on it for a minute
20:29 alexhayes joined #salt
20:29 lero joined #salt
20:30 puresalt joined #salt
20:33 pipps99 joined #salt
20:33 puresalt Hi guys need some help with salt-cloud maps
20:36 pipps_ joined #salt
20:37 yidhra joined #salt
20:42 puresalt I have a template in vmware that already has grain values set. What im trying to do is clone from this template and in the map file specify the new grain values. The hosts build new hostnames are applied, ips, etc but the values in /etc/salt/grains are not updated.
20:46 krymzon joined #salt
20:47 west575 joined #salt
20:48 pcdummy puresalt: you wanna change grains in /etc/salt/grains with salt-cloud, right?
20:48 alexhayes joined #salt
20:48 antpa joined #salt
20:48 antpa joined #salt
20:48 puresalt yes
20:49 pcdummy need to read salt-cloud docs for that
20:49 pcdummy puresalt: you have it set like that: https://docs.saltstack.com/en/latest/topics/cloud/profiles.html ?
20:51 saltycharles joined #salt
20:53 bowhunter joined #salt
20:53 puresalt yea i have it working in maps that work on a clean install. Where i run into issues is if i provision a server and set grains via a map file, then clone that server to a template, then try to redeploy from the template and change map grains.
20:54 sfxandy joined #salt
20:54 sfxandy hi
20:54 puresalt seems to keep the original grain and not overwrite them
20:55 puresalt grains*
20:55 pcdummy puresalt: can you remove /etc/salt/grains in the template?
20:55 pcdummy puresalt: i'm reading through the code to find that part...
20:56 sfxandy quick question, if I raise an event with salt-call i.e. salt-call 'foo/bar/tag' '{param1: foo, param2: bar}'  and I lsiten for this event with a reactor and fire off an SLS....... how do I access the event payload inside the SLS?
20:56 ajw0100 joined #salt
20:59 DammitJim joined #salt
20:59 puresalt i can remove the /etc/salt/grains, just wanted to check in and make sure that is a desired way it should work
20:59 pcdummy puresalt: think its best to ask the ML salt-users for that, it looks like "bootstrap-salt.sh" would overwrite the grains file, but not sure it gets generated, i'm still reading the code for that.
21:02 puresalt thanks appreciate it
21:06 pcdummy puresalt: you don't get current roles with "salt <machine> grains.items" right?
21:07 * viq prefers to keep roles in pillars
21:07 pcdummy i'm also not a fan of /etc/salt/grains
21:08 pcdummy puresalt: sorry have to go, ask the ML salt-users please, maybe you found a bug maybe its intentional.
21:09 puresalt pcdummy i will try in a sec and report
21:13 majuscule babilen: figured it out. it was a pathing issue. it needs to be in <file_root>/salt/_grains, not <file_root>/_grains
21:13 majuscule thank you for your help earlier!
21:20 flowstate joined #salt
21:21 majuscule is it possible to include a recipe conditionally?
21:22 majuscule (i.e. i'd like to include dns management if an 'ec2' attribute is found, but skip it otherwise)
21:23 viq majuscule: jinja to the rescue! (ie, yes)
21:24 viq Or, erm, I would expect it to work
21:25 puresalt pcdummy  grains.items returns the old values
21:26 brotatochip joined #salt
21:28 MindDrive Running a basic 'salt ... cmd.run ...' over quite a few hosts and I'm getting "ERROR: Minions returned with non-zero exit code" in the output - is there a simple way to find out which minions are having the issue?
21:31 antpa joined #salt
21:33 flowstate joined #salt
21:35 majuscule ahh right, ofc, thank you viq
21:35 teryx510 joined #salt
21:37 antpa joined #salt
21:37 hemebond MindDrive: I'm surprised it didn't show each minion result.
21:38 hemebond MindDrive: http://stackoverflow.com/questions/26098365/how-to-know-which-minions-failed-to-execute-a-salt-state
21:42 MindDrive hemebond: Okay, I think the 'failure' in this case is the response being the empty string, since the command I'm running is a grep.
21:42 hemebond MindDrive: Ah, yes, if it returns an error code then I guess that's the message you'll get.
21:42 MindDrive (Although I do have non-responsive minions as well, some of which shouldn't be running Salt at all... *sigh* someone's Puppet change wasn't complete enough.)
21:44 hemebond lol
21:46 antpa joined #salt
21:46 jacksontj joined #salt
21:51 mikecmpbll joined #salt
21:51 racooper howdy folks.  I'm getting the following error when trying to run a highstate with a new config: "Too many functions declared in state 'file' in SLS 'idp-dev.files'"
21:51 racooper the referenced file is at https://gist.github.com/racooper/b6ccc4dde7d937ad479b854608465798
21:52 racooper debug isn't giving me any line numbers, and I'm not seeing any problems, so it's obviously something I'm completely missing.
21:54 KajiMaster joined #salt
21:54 hemebond racooper: Line 140 have a problem?
21:57 GnuLxUsr joined #salt
21:57 hemebond racooper: Try putting your state IDs into single quotes.
21:57 hemebond No wait, that's not required either here. Hmm.
21:58 racooper I was almost wondering if it's too many statements in one file, but that seems a bit absurd.
21:58 hemebond That's what the message suggests but I haven't heard of any limit like that.
21:59 racooper I don't see anything wrong with line 140; it was really just copied from another file.managed in that sls and the filenames changed.
21:59 hemebond Yeah, there's no problem there, it was a temporary parsing issue.
21:59 hemebond I guess you'll just have to try removing/commenting out entries until it works without error.
22:00 racooper yeah...project for tonight I suppose.  it's time to go home.
22:02 west575 joined #salt
22:04 hemebond Oh, I found a possible indentation error.
22:05 TyrfingMjolnir joined #salt
22:05 hemebond # Shibboleth-IdP
22:05 hemebond Oh well.
22:05 TyrfingMjolnir joined #salt
22:08 MindDrive Well, I don't get it.  I have two minions that keep getting denied (rejected key) by the master, I delete the key from the master, restart the minion and it fails again.  Ideas?
22:09 flowstate joined #salt
22:12 pipps99 joined #salt
22:14 hemebond MindDrive: Stop salt-minion. Delete the minion key from the master and the minion. Delete master key (pub) from the minion. Start salt-minion.
22:14 hemebond Accept key.
22:14 nmadhok joined #salt
22:15 MindDrive You can't accept when it's denied, you can only delete.
22:16 MindDrive But I will try again... of course, now I'm having performance issues for my Salt masters in production.
22:16 hemebond It gets denied immediately even after you've created new minion keys?
22:16 hemebond Oh dear. What kind of performance issues? How many minions do you have?
22:18 MindDrive I switched hash_key from 'md5' to 'sha256', what I thought would be an okay change.  Now running a simple grep command against all our prod hosts (around 800) is causing about 70 at any given time to be unable to respond - seeing on the failed hosts "The minion failed to return the job information for job 20160727150518732342. This is often due to the master being shut down or overloaded. If the master is running consider increasing the wor
22:18 MindDrive ker_threads value."
22:19 hemebond Ah
22:19 MindDrive We're getting major flack from execs now about "why are we using this thing in our environments" as it's been a nightmare getting it to be stable and consistent.
22:20 MindDrive And I don't have answers for them.
22:20 edrocks joined #salt
22:20 hemebond Oh dear. Do you have a support plan from Saltstack?
22:21 hemebond btw, do you mean hash_type instead of hash_key?
22:21 MindDrive Yes, sorry, hash_type.
22:22 MindDrive No I don't.  I would like such a thing, but I'm not sure the execs will go for it.
22:22 hemebond I could see that increasing the load on the master, yeah.
22:22 hemebond What do they want you to use?
22:23 hemebond Were you getting collisions or something?
22:23 hemebond ^ regarding the hash_type.
22:24 krymzon joined #salt
22:25 coval3nce joined #salt
22:25 coval3nce Anyone see their pillars between environments getting merged all of a sudden?
22:25 MindDrive hemebond: How would I check that?
22:25 MindDrive Looking at the CPU and memory usage on the box, I'm not seeing any real issues.  Maybe I can just bump the threads...
22:26 hemebond coval3nce: All of a sudden? Only when I bork the targeting.
22:27 hemebond MindDrive: I'm trying to understand why you changed hash_type.
22:27 coval3nce lol, well im using “{{ saltenv}}” as includes in top fiels
22:27 antpa joined #salt
22:27 coval3nce something had to change with that variable in pillar facility
22:27 MindDrive hemebond: 'md5' is considered highly insecure.  It was warning in the salt minion logs every time I restarted the minions.
22:27 coval3nce now i guess it doesn’t resolve, and default strategy is merging all my top files
22:27 coval3nce including base.sls and stage.sls
22:28 MindDrive Also getting a ton of these messages in the master: 2016-07-27 15:27:10,322 [salt.transport.zeromq][ERROR   ][28234] Bad load from minion: AuthenticationError: message authentication failed
22:28 hemebond MindDrive: "The hash_type is the hash to use when discovering the hash of a file on the master server." I didn'
22:28 hemebond didn't think it had anything to do with security/
22:28 hemebond I could be wrong though.
22:28 mikecmpbll joined #salt
22:29 tduerr joined #salt
22:30 hemebond coval3nce: The default is always to merge top files.
22:31 coval3nce hemebond: if my init.sls has an include with a “{{ saltenv}}” and if the saltenv is “stage” should it not only use “stage.sls”?  Its worked this way for 1+ years
22:31 hemebond MindDrive: Have you customised the minions at all?
22:31 hemebond coval3nce: It sounds like you only have one top.sls, is that correct?
22:32 coval3nce hemebond: correct, in it, simply includes a saltenv.sls file
22:32 hemebond Then I don't think there's any merging going on there.
22:32 hemebond The merging is only for the top.sls files.
22:32 hemebond If you're actually using environments.
22:32 ribx joined #salt
22:33 aswini1 joined #salt
22:33 MindDrive hemebond: No.  I *think* bumping the threads from 32 to 64 is helping, though.
22:33 coval3nce hemebond: deifnitely have “environment” set on the minion, and confirmed all the .sls files are being merged
22:33 MindDrive The number of 'no response' systems appears to be going down.
22:33 hemebond MindDrive: ????
22:34 hemebond coval3nce: Well environments (actual Salt environments) are not really separate. So the top.sls are still merged. And your base/top.sls will still be used. But in this case it doesn't matter because you're just referencing another SLS file depending on a grain/variable.
22:35 hemebond Can I see your top.sls?
22:35 brotatochip joined #salt
22:35 coval3nce hemebond: https://github.secureserver.net/gist/dkiser/1ecee1445f77ef84954a2a40ff9a7fea
22:35 antpa joined #salt
22:36 hemebond I'm not sure I can reach that host.
22:36 coval3nce hemebond: just updated it, had a hacked in variable there
22:36 coval3nce oh yueah my bad, one sec
22:36 hemebond It says it's connecting but not getting far.
22:36 coval3nce https://gist.github.com/dkiser/e270dc22374461b8efd48c23e7f4751d
22:37 hemebond That'
22:37 hemebond That's your top.sls?
22:37 hemebond That... I would not have expected that to work.
22:37 coval3nce ha, been working for 1+ years
22:37 coval3nce it would include either base.sls or stage.sls depending on environment
22:37 coval3nce that would do all the traget matching
22:38 hemebond What happens if you do a 'salt "minion" state.show_top'?
22:38 hemebond and state.show_sls
22:38 hemebond Oh this is pillar stuff...
22:39 coval3nce hemebond: pillar correct
22:41 hemebond Does the environment setting show up in the grains?
22:41 hemebond Or the config?
22:41 hemebond salt 'minion' config.get environment
22:42 coval3nce its in config, let me see if i can drum up the command
22:42 coval3nce ah there you go, thx
22:42 hemebond (I guess)
22:42 coval3nce Shows up in config.
22:43 hemebond https://github.com/saltstack/salt/issues/22048 ?
22:43 saltstackbot [#22048][OPEN] state.sls: saltenv is being ignored | salt-call --versions-report:...
22:43 hemebond Oh, probably not relevant.
22:43 TyrfingM1olnir joined #salt
22:43 coval3nce Unless it also applies to pillar, probably not
22:45 TyrfingMjolnir joined #salt
22:47 jtang joined #salt
22:47 hemebond coval3nce: What happens if you explicitly provide the saltenv on the command line?
22:47 mohae joined #salt
22:47 coval3nce can’t via a salt-call command as far as i can tell
22:47 hemebond Also, does the environment show up in the grains?
22:47 coval3nce at least fora pillar.get, invalid keyword arg
22:48 coval3nce Its not in the grains at all.
22:48 hemebond What does pillar.items show? All pillars from all "environments"?
22:48 coval3nce hemebond: precicely lol
22:49 pcdummy Hihi
22:49 TyrfingMjolnir joined #salt
22:49 pipps joined #salt
22:49 pcdummy pillar.items Shows the current pillar for the current Host on the given saltenv
22:50 coval3nce pcdummy: in my particular case right now, its returning pillar for all saltenvs
22:50 coval3nce e.g. merging *.sls in top directory of pillar
22:51 hemebond See, that's a point, if saltenv was blank, then nothing would get included.
22:51 hemebond Only top.sls files get merged as far as I understand it.
22:51 hemebond It doesn't just merge all files in the root of the directory.
22:52 hemebond You only have a base directory, correct?
22:52 hemebond No actual environments?
22:52 pcdummy Hmm thought there's no way to have multiple saltenv for a single Host, there is?
22:52 hemebond You have only the "base" entry in your file_roots?
22:52 coval3nce nah, using pillar backend
22:52 coval3nce hemebond: so i have environments explicitly set to branches in git
22:52 hemebond Aaaaa
22:53 hemebond OKay, that's completely different.
22:53 hemebond Then you do have separate environments and it will be merging top.sls files it finds.
22:53 hemebond And saltenv will change depending on which environment Salt is currently checking.
22:54 hemebond "The saltenv variable is available in only in sls files when gathering the sls from an environment."
22:54 hemebond Though using the "environment" config property should fix the minion to a particular environment.
22:54 coval3nce Right, so should be working as these sls files are in enviornments created by gitfs
22:54 coval3nce right, been working a while like this
22:54 coval3nce I think upgrading to 2016 got me in this weird state
22:55 coval3nce BRB
22:57 hemebond pcdummy: You can provide multiple environments with some commands apparently.
22:57 hemebond And you can also configure the environments with fallbacks.
22:57 hemebond coval3nce: If you've just upgraded and hit this issue then yeah, possibly a bug.
22:58 hemebond Are you able to post your pillar_roots?>
22:58 west575_ joined #salt
23:01 TyrfingM1olnir joined #salt
23:06 orion left #salt
23:07 ninjada joined #salt
23:08 amcorreia joined #salt
23:09 antpa joined #salt
23:10 GothAck joined #salt
23:10 coval3nce hemebond: not really
23:11 coval3nce hemebond: i have to jet, i’ll keep poking around later, thx man
23:12 hemebond Well, according to the docs, I would expect each of your top.sls to get the saltenv that they are in. Which means they will match in every environment.
23:12 hemebond I'm not entirely sure if Salt still merges all the top.sls even if you specify "environment" config variable.
23:12 hemebond I wouldn't have expected it to.
23:13 hemebond Good luck ????
23:17 orion joined #salt
23:21 orion Hi. On salt 2016.3.1 I am getting weird jinja output: https://gist.github.com/centromere/e2ff7822db80cb6ba230751b01eaef1f
23:21 orion Does anyone know why I would be getting odd strings of text such as "&id001"?
23:22 hemebond orion: I think that's a YAML anchor.
23:23 hemebond I'm struggling to understand what you're doing here.
23:24 hemebond You're trying to import a YAML file using Jinja?
23:25 orion hemebond: This is the config file for filebeat. It's a program which monitors log files and ships them elsewhere. I have a pillar dict called "filebeat". All hosts are configured to monitor /var/log/auth.log and /var/log/syslog. But on some hosts I also want to log extra files, such as /var/log/foo/requests.log.
23:26 ageorgop joined #salt
23:26 hemebond Sure, it's the import part that's confusing me.
23:26 hemebond I can't tell how the pillar file is read.
23:26 orion I'm importing map.jinja.
23:26 cyborg-one joined #salt
23:27 orion Its contents are as follows: {% set filebeat_settings = salt['pillar.get']('filebeat', default={}, merge=True) %}
23:29 antpa joined #salt
23:33 orion *In fact* the output from salt is wrong.
23:34 orion If you paste the rendered output in to http://www.yamllint.com/ you'll find that the web-request and web-exception log entries are listed twice.
23:34 orion This is a bug.
23:34 hemebond Yeah, the anchor and references will cause that.
23:34 hemebond &blah is the anchor and *blah is the reference/pointer.
23:35 orion What I don't understand though is why it's even being formatted using references and pointers.
23:35 orion Something seems wrong with the .extend function
23:36 hemebond Well that's a Jinja thing.
23:36 hemebond (I think)
23:42 orion hemebond: If you wanted to merge pillar lists "foo" and "bar", how would you do it?
23:43 hemebond Lists would just be concatenated.
23:43 hemebond You can only merge dicts.
23:44 orion How would you concatenate them?
23:44 hemebond Oh, right, your file_extends is a list, not a dict.
23:44 orion Both "files" and "files_extend" are lists.
23:45 hemebond I would try +
23:46 orion hmm, well in python I would naturally do: x = [1, 2, 3]; x.extend([4, 5])
23:48 hemebond Oh, maybe that should work then.
23:49 orion But, that's what I'm doing and I'm getting the duplicate entries.
23:49 hemebond Try +
23:49 hemebond See how you get on.
23:50 west575 joined #salt
23:53 antpa joined #salt
23:53 hemebond Seems to work for me.
23:54 orion Does .extend() work for you?
23:54 hemebond No, I used + to join the lists.

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