Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2014-02-14

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

All times shown according to UTC.

Time Nick Message
00:00 jroweboy Hello everyone! I have a working (I think) salt master and minion setup where the master is sending messages every 5 seconds to the minion, and salt-minion -l debug shows that the minion is receiving the data just fine. (It prints out on the screen) But when I run this python code it never prints out that it got a message http://pastebin.com/ANq8aQtv
00:01 jroweboy Is there something wrong with my code? Or better yet, what is the proscribed method of receiving and handling events on the minion side from python?
00:02 Furao jroweboy: for data in event.iter_events():
00:02 Furao instead of your weird loop
00:03 jroweboy Ok lemme give it a shot
00:03 Furao you assume that you'll receive a msg on the bus under 15 sec
00:03 Furao which can be ok if master send each 5 seconds
00:04 Furao but that don't sound the best way to handle it
00:04 Furao oh wait
00:04 Gifflen joined #salt
00:04 jroweboy Well thats just for now. The master is just sending every 5 seconds just for testing purposes
00:04 Furao you're running that in a minion?
00:05 jroweboy Yeah... its a minoin
00:05 Furao but this is "MasterEvent"
00:05 Furao I'm running this code on the master side
00:05 jroweboy Oh... I just noticed there is a MinionEvent class
00:05 jroweboy is that what I should be using?
00:05 Furao You should try
00:05 Furao It's 8 am here and I just woke up and still sleepy :P
00:06 Furao brain is slow
00:06 jroweboy Hahaha its fine :)
00:10 jroweboy Now I'm getting 'NoneType' object has no attribute 'endswith'  when I changed it to salt.util.events.MinionEvent()
00:10 jroweboy .event*
00:11 jroweboy the path in posixpath.py ends up being None somehow
00:12 jtang1 joined #salt
00:15 joehoyle joined #salt
00:16 fllr joined #salt
00:19 n8n joined #salt
00:20 jroweboy_ joined #salt
00:21 forrest joined #salt
00:25 vbabiy_ joined #salt
00:27 frasergr_ joined #salt
00:27 Gordonz_ joined #salt
00:31 forrest mgw1, did you create an issue to see about the 'creates' stuff for cmd.script?
00:31 mgw1 no, not yet
00:31 forrest ok
00:31 mgw1 Do you agree it would be useful?
00:32 mgw1 It's a shortcut/more elegant way of doing "unless: [ -f /tmp/foo ]"
00:32 forrest To me it seems kinda pointless as you could just use file.managed, and just leave the options blank and it will be a blank file which you can then populate via the script, but I can see why it could come in handy for some users.
00:33 mgw1 How would you prevent the script from running every time?
00:33 mgw1 If it were a script out of your control?
00:33 mgw1 Obviously, you can wrap it
00:33 forrest yea very good point
00:33 forrest and something that should be mentioned in the issue
00:35 diegows joined #salt
00:35 Gordonz joined #salt
00:36 techhat joined #salt
00:39 Gordonz joined #salt
00:43 mgw1 forrest: did you see this: https://github.com/borgstrom/nacl ?
00:43 forrest I have not
00:43 nextdoorwarren joined #salt
00:44 forrest seems cool
00:44 mgw1 quite
00:45 mgw1 I extended it with a proper salt renderer: https://gist.github.com/mgwilliams/8986670
00:46 forrest nice
00:46 psyl0n joined #salt
00:46 kalloc joined #salt
00:47 joehillen that's awesome!
00:51 n8n joined #salt
00:52 bhosmer joined #salt
00:55 fllr joined #salt
00:55 gadams9991 left #salt
00:56 TJ1980 joined #salt
00:56 randeep joined #salt
00:57 randeep I was having a small problem getting schedule to run state.highstate .... it's just not doing it, I'm trying to run standalone salt and I can call state.highstate through salt-call just fine
00:57 randeep scheduler just seems like it's not running
00:59 randeep i put the stuff into my minion config and you would think it would run
01:02 Gordonz joined #salt
01:06 jtang1 joined #salt
01:14 andrej Does anyone have a working sample of how to use states.file.append in conjunction with module.run?
01:15 andrej I can't get the syntax from the two module docu pages together
01:18 jtang1 joined #salt
01:21 andrej should something like this http://pastebin.com/ZtCyua95 work?
01:22 xzarth_ joined #salt
01:22 Led_Zeppelin joined #salt
01:22 Led_Zeppelin i want to avoid using '-c' but it seems the environment variable isn't working either.
01:24 scoates andrej: I think this is closer to what you need/want: http://paste.roguecoders.com/p/e686167a79c1540d6cbec89c24c6203d.txt
01:25 Gifflen joined #salt
01:25 andrej scoates - I'll give it a look/try
01:27 andrej Hmmm ... I was told I need to use the file.append in conjunction w/ module.run ... you seem to be avoiding that - how come? :)
01:28 longdays joined #salt
01:28 n8n joined #salt
01:28 andrej yours would be so much simpler - I'll see what it does :)
01:28 scoates andrej: well, you're calling a state, not a module. there's a state for file.append, so you can just use that. https://salt.readthedocs.org/en/latest/ref/states/all/salt.states.file.html#salt.states.file.append
01:29 andrej can I run a state from within a state?
01:29 scoates not sure how you mean
01:29 andrej I'm trying to deploy a standard nrpe.conf and then modify it for a few specific hosts.
01:30 scoates andrej: I think I'd use a template for that
01:30 justBob joined #salt
01:30 scoates I have to go afk for a few mins. back later.
01:30 andrej I know what you mean. I have two templates already, one for ubuntu hosts, one for centos ...
01:31 andrej I don't want to maintain 2,3,4 almost identical files
01:31 andrej which is where file.append comes in
01:32 pdayton joined #salt
01:33 kohaku_river left #salt
01:33 andrej http://pastebin.com/MnLhC5pk ... this is what the deploy as such looks like atm
01:34 andrej I was wanting to do the append before the elseif section
01:34 andrej The reality is that I need a few specific lines for several machines,
01:34 scoates andrej: those are static files. I mean jinja-templated files. I bet you could use one template for all [4] scenarios
01:34 Led_Zeppelin when using hilight how do I point to the proper -c ?
01:34 andrej Not just one line appended to one template
01:35 n8n joined #salt
01:35 andrej scoates - what would that look like insde a state file (for a deployment)
01:38 scoates andrej: something like this: http://paste.roguecoders.com/p/fb487cad1a40c173759161948da82853.txt
01:38 scoates really need to run. back in a bit.
01:42 andrej k, ta
01:45 rojem joined #salt
01:46 kalloc joined #salt
01:55 Led_Zeppelin is SALT_MASTER_CONFIG the correct env variable for hilite ?
02:02 BenCoinanke joined #salt
02:04 higgs001 joined #salt
02:05 scoates andrej: did you get it?
02:10 TJ1980 joined #salt
02:10 higgs001 joined #salt
02:12 jtang1 joined #salt
02:13 andrej scoates - I think I understand where you're heading. I did some googling, and am about to have a play with a jinja based config file.  Thanks a ton for the tip, this looks far more sensible than my initial idea
02:17 patrek_ joined #salt
02:19 diegows joined #salt
02:28 ckao joined #salt
02:28 nextdoorwarren joined #salt
02:31 mgw1 joined #salt
02:36 yomilk joined #salt
02:37 dave_den joined #salt
02:39 toofer joined #salt
02:42 Gifflen joined #salt
02:45 krow joined #salt
02:46 kalloc joined #salt
02:49 xl1 joined #salt
02:50 Linz joined #salt
02:50 krow joined #salt
02:53 bhosmer joined #salt
02:56 krow joined #salt
02:56 joehoyle joined #salt
02:58 ravibhure joined #salt
02:59 favadi joined #salt
02:59 ravibhure1 joined #salt
02:59 Gifflen joined #salt
02:59 ravibhure1 left #salt
03:06 jtang1 joined #salt
03:08 jtang2 joined #salt
03:09 andrej Cool ... that worked
03:09 andrej Thanks scoates
03:11 Gifflen joined #salt
03:11 andrej Does my jinja file allow me to do matching of grains['fqdn'] where I match a substring rather than the whole string, e.g. a regex of infra[0-9]?
03:12 Furao andrej: {% if grains['fqdn]
03:12 Furao oups
03:12 Furao andrej: {% if 'bleh' in grains['fqdn'] %}…
03:13 jaimed joined #salt
03:13 Furao In [4]: 'abc' in 'defabcegh'
03:13 Furao Out[4]: True
03:13 Furao In [5]: 'abzc' in 'defabcegh'
03:13 Furao Out[5]: False
03:13 Furao it's all python you can use python string logic in jinja2
03:13 andrej Furao: can those be regex?
03:13 Furao oh regex
03:13 andrej I'm not that good with python (yet)
03:14 Furao don't seem it's builtin jinja2
03:15 andrej how about having it check whether two different strings match? :)
03:15 Furao {% if grains['fqdn'].startswith('infra') %}{% if grains['fqdn'][5]|int in range(0,10) %}
03:15 Furao ugly
03:15 andrej heh
03:15 andrej I agree :)
03:16 Furao use pillar
03:16 andrej How? :)
03:16 Furao you can set pillar using pattern match such as 'infra*'
03:17 Furao and set a pillar key
03:17 Furao and have {% if salt['pillar.get']('your_pillar_key', False) %}
03:17 Furao you can't use salt without using pillar
03:17 Furao it's like buying a bike and not planing to use the wheels
03:21 scoates andrej: glad to hear it
03:22 rostam joined #salt
03:23 cachedout joined #salt
03:25 krow joined #salt
03:28 Eugene Sure you can. Use grains! :-p
03:29 krow1 joined #salt
03:29 Furao Eugene: grains is far less flexible as they need to be set in the minions
03:29 Eugene I'm well aware.
03:30 malinoff Well, pillars are not so flexible as the should be
03:30 malinoff they*
03:30 Furao malinoff: just move them out of .sls
03:30 malinoff Furao, did that
03:31 Furao I have pillars in a django app, in .csv files or http://reclass.pantsfullofunix.net
03:32 malinoff Furao, do you know this: https://groups.google.com/forum/#!topic/salt-users/nh8MqRiHVt4 ?
03:32 Furao yes I've seen this
03:32 malinoff Do you have any thoughts about?
03:33 Furao that is a big change, I wonder if they will ends with modular communication layer
03:33 Furao such as amqp
03:33 Poooogles joined #salt
03:34 Furao you pick the "driver" for your needs, such as unencrypted http in private cloud
03:35 malinoff Ye a modular layer is quite useful
03:35 malinoff But I can't imagine how long it will take
03:35 Furao rabbitmq is probably a lot faster and reliable than zmq
03:35 malinoff I see 0mq usage in almost every module
03:36 Furao yeah, that need big cleanup
03:37 dave_den wat
03:37 malinoff dave_den, wat wat? :)
03:39 dave_den http://mediamusings.dsc.rmit.edu.au/files/2011/05/wat.jpg
03:39 malinoff data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD/2wCEAAkGBxQSEhUUEhQWFRUXGBUVFRcUFRQUFxQVFxQXFhQVFRQYHCggGBolHBQUITEhJSkrLi4uFx8zODMsNygtLisBCgoKDg0OGhAQGiwmHyQsLCwsLDQsLCwsLCwsLCwsLCwsLCwsLCwsLCwsLCwsLCwsLCwsLCwsLDQsLCwsLDgsLP/AABEIALcBEwMBIgACEQEDEQH/xAAbAAABBQEBAAAAAAAAAAAAAAAFAAECBAYDB//EAEsQAAEDAgMEBgYECgoBBQEAAAEAAgMEERIhMQUTQVEGImFxgZEUMlKhscEHI5LRM0JDU2Jyc8Lh8BYkgpOUorLD0/GzNWODo6QV/8QAGgEAAgMBAQAAAAAAAAAAAAAAAAECAwUEBv/EACYRAAI
03:39 malinoff CAQQCAQQDAAAAAAAAAAABAhEDBBIhMRNBUQUUYXEiQrH/2gAMAwEAAhEDEQA/ANj0nLhR1BaSCInkFpLTkLmzhmMgcwjMfQiicAQyWxAI/rdZxF/zqqV8G8hkZ7THt+00j5o30Tqt7Q0sntwQu842psSMhtaCPZ1RNui9sfoMs+F8ssoxwyZuG8c4g2kaMuQRfZPQimMERlEzpDGwyO9KqxieWjGbCWwub6IV9KEQMsN/y0FRS9+9mphbyBXoTRbJIZh+kHR+CjEM0G9a4VEDHXqKiQFkr90WlskhafXHBPsbYEVZJVS1G8fad0UYbPPEGMjjjaQGxvA9cSG9uKKfSGy9C8+xLSydwjq4XuPkCpdAR/Uw8/lJamTwfUyFv+XCgAP0s6LU0FJLJCJmSgNEbvSqp2F7ntaw2dKQc3DIiy6Uex4qyqqd+ZHNh
03:39 malinoff 3MTA2aaIB273khtG9oJO8bmeQRHp7nDBGPylXSN7w2dsrh9mJyboKA5tVKPylXUf/UW0/8As+9AE/6D0fsTf4us/wCVCOk2woaKOOpg3rXMmga69RUSAslkELg5kkhaR9YDpwW1knDXNadX3w/2Rc+5BuntMZNnVbRruXub2OYMbT4FoKAAnRLopTT0NLNMJnSSQQySO9Kqxie+NrnGwlAFyToi39B6P2Jv8XWf8qzzKyuoqEOjlpXxQQNLAYJsTmMjGG7hNa5AGdvBegxOu0HmAfMIAz/9B6P2Zv8AF1n/ACoZsyibT19TDEX7oU9LIGvllls98lS1zgZHOIuGM05In0j2nUsqIIKYwt3jJpHOmjfJ+DMYAaGvbb8IfJVNm0MwnlqKiSN75I4ogIo3RtDYnSuBs57iSTMePAJoTC9091BPdMCV091C6dA
03:39 malinoff E7pKKV0hk7pwVzupXQ3QE7pXUAmab8tVDyQ+SW1/B0uldMAmKakn0xNNE7pXULpXUhErpXULpIAndMSopiUASJUSldMSmBBJIpIAHtXf6Pz/UYm/mzLF3buZ7APJoQWo23G3S7j2ZDzKv/RvVY4agWtgqpsuW8wzf7qTEip9I1HvKjZXIVzAe0BjpCD/de5bgoVtuh3klI781Ubz/APPOz/cVbpztb0ShmnvbAGZ/rSNb80hjdPYy7ZtWBruJCO8NJHwXToRAWbPpGnUQQ37TgBJ8yVc25DvKadvtRSDzYQuuyosMMTfZjYPJgCAMj9J0xBog0kETvlyyNmU8rdf/AJAiv0dsts6mdxkYZj2mZzpST2nHdZL6Xaotmjt+TpKyW36RMbWfBy9C2FSCGmgiGQjiiYP7LA35IAz/AEi2ng2tsyH84Ky/hC0t+
03:39 malinoff BWpqYQ9jmHRzXNPcRYrzzpcXHbdC7BIRGGWc2N7mDevkZJeQDC2ww6nivSEAeZGQnYZxG7m0jmP/XjYY3+9pXpNP6jf1R8F4vth74o9oQtcQ1klSCOFnnejLulC9op/Ub+qPgmBmtvf+oUv7Cq/106vIf0hdbaFL+wqv9cCvXQhMldPdQBTpiJ3TqAKe6AJXSxLk6Sy4unXBqNWocROnHhvllrGm3qp71LeLJyaqUn2daxJFvepb1U94lvFT9wyXjLm+UhN2qjvFISKcdS0J40XP
03:39 malinoff Oops
03:39 malinoff Sorry :(
03:39 malinoff Goddamn google
03:40 dave_den heh
03:40 malinoff http://stewd.io/javascript/media/watDuck.jpg
03:40 dave_den malinoff: are you saying you see 0mq usage in every module right now?
03:41 malinoff dave_den, not in each module, but almost, yes
03:41 dave_den can you point me to an example?
03:42 malinoff https://github.com/saltstack/salt/search?q=import+zmq&type=Code
03:42 malinoff I was talking about the core, of course
03:45 dave_den well yes, salt was built on 0mq
03:45 dave_den but of those 21 results, it is only in master, minion, payload, transport, event, and logstash
03:46 chuffpdx joined #salt
03:46 kalloc joined #salt
03:46 krow joined #salt
03:46 dave_den the others are tests
03:47 dave_den it's very do-able to make the transport layer pluggable, but as you can see from the mailing list thread, nothing is ever going to satisfy everyone
03:48 malinoff Users want some stability that salt can't offer right now
03:48 Jahkeup joined #salt
03:48 dave_den what stability issues are you having?
03:49 dave_den as long as you have zmq >= 3 then the transport layer of salt is very reliable.
03:49 codekoala joined #salt
03:49 codekoala joined #salt
03:49 EWDurbin_ joined #salt
03:49 malinoff I don't talk about zmq, but about salt in common
03:49 Xiao joined #salt
03:50 malinoff 1/4 issues are bugs on github
03:50 [vaelen] joined #salt
03:50 carmony joined #salt
03:51 mpoole joined #salt
03:51 joonas joined #salt
03:51 malinoff I mean, developing a custom transport does not seem to make salt more stable, it brings up new bugs and issues
03:51 dave_den indeed. salt is a young product that must handle a huge variability in usage. that is why the saltstack coders have emphasized test coverage and regression testing instead of focusing on new feature implementation
03:52 dave_den malinoff: not exaclty. if the transport is made pluggable and 0mq is teh default, it should be just as stable as it is now (on the transport layer)
03:53 malinoff dave_den, but now it's not stable at all
03:54 malinoff And implementing a new protocol won't help
03:55 dave_den malinoff: what's not stable? the transport layer? module functions? backwards compatibility? new features explicitly noted as new and alpha?
03:55 Gifflen joined #salt
03:55 dave_den i would argue that the transport layer of 0mq is very stable.
03:56 gw_ joined #salt
03:56 gw_ joined #salt
03:56 dave_den there has been a higher incident of regression bugs before 0.17.2 with regards to module functionality
03:56 Gifflen joined #salt
03:57 dave_den which i think was due to a severe lack of test coverage
03:57 dave_den but in the last 3-4 months the focus has shifted from feature implementation to test and regression covage
03:57 dave_den coverage
03:57 dave_den which is very evidently increasing the bug issues there
03:58 dave_den err, reducing, not increasing ;)
04:01 whiteinge i'm skimming the chatlog, but the end-goal is to have pluggable transport and pluggable encryption
04:01 dave_den and i've got to hand it to the salt project maintaners and coders - for such a young project that people are applying to *extremely* different use cases, they are handling the growth and adoption very well
04:01 whiteinge amqp (if someone wants to write a backend), tls (if someone wants to write a backend), etc
04:02 jtang1 joined #salt
04:02 whiteinge there was a question in here a couple weeks ago about transport over a vendor-specific lights-out tech (hp perhaps? ibm? i forget)
04:02 whiteinge that's the goal anyway
04:04 dave_den right. i think malinoff and i were debating the instability that would cause, but i think as long as the 0mq plug transport is the default then it would be pretty much seamless and transparent to implement
04:04 whiteinge ah, sorry i missed that distinction. in any case, i agree with your assertion :)
04:05 whiteinge zmq is likely to remain the default for a good long while
04:05 krow joined #salt
04:06 dave_den i'd like to work on making the data minion cache pluggable when i have time, which would make multi-master as fully functional as single master as long as the cache backend was shared
04:06 dave_den get away from data.p and mine.p file stores
04:06 whiteinge ooh, nice
04:07 dave_den the code for accessing those could be DRYed by about 10x
04:07 whiteinge yeah....
04:07 * whiteinge shudders
04:08 dave_den we'll get 'er there
04:08 dave_den anyway - i'm off. see ya tomorrow
04:16 tomtomtomtom joined #salt
04:21 Gifflen_ joined #salt
04:23 UtahDave joined #salt
04:31 yomilk joined #salt
04:33 Gifflen joined #salt
04:35 mgw1 joined #salt
04:36 oz_akan__ joined #salt
04:37 frasergraham joined #salt
04:41 ausmarton joined #salt
04:41 rojem joined #salt
04:42 kermit joined #salt
04:44 krow joined #salt
04:46 kalloc joined #salt
04:47 forrest woah woah woah, whiteinge and UtahDave in here? What is this madness!
04:47 whiteinge what? UtahDave is here?!
04:47 * whiteinge leaves
04:47 forrest at the very least he is here in spirit form to haunt the irc.
04:47 UtahDave no, don't leave!!!!!
04:48 UtahDave Yeah, my wife and kids are sleeping so I sneaked (snuck??) down to the hotel lobby to get some work done
04:48 forrest oh are you on vacation?
04:48 Eugene snuck, I believe
04:49 jesusaurus i concur, snuck
04:49 jesusaurus <- linguist
04:49 Eugene How cunning.
04:49 krow joined #salt
04:50 UtahDave thanks!
04:51 joehoyle joined #salt
04:53 whiteinge lol
04:54 basepi joined #salt
04:54 bhosmer joined #salt
04:56 whiteinge and now basepi joins!
04:56 forrest he's just pretending and on his phone :P
04:56 jtang1 joined #salt
04:57 jesusaurus wait, you guys arent all the same person?
04:57 forrest jesusaurus, I saw all 3 of them in the same room
04:57 forrest so I can confirm they are indeed not the same person
04:57 malinoff Ye, and his name is Thomas Hatch
04:57 bdf joined #salt
04:57 __number5__ lol
05:04 higgs001 joined #salt
05:05 UtahDave lol
05:06 Eugene His name was Robert Paulson
05:06 jimallman joined #salt
05:07 Regek joined #salt
05:09 faldridge joined #salt
05:10 anuvrat joined #salt
05:11 Regek left #salt
05:12 gw joined #salt
05:17 elfixit joined #salt
05:20 mgw1 joined #salt
05:20 krow1 joined #salt
05:25 danielbachhuber joined #salt
05:26 krow joined #salt
05:29 krow1 joined #salt
05:29 joehoyle joined #salt
05:33 jidar joined #salt
05:40 l0x3py joined #salt
05:46 frasergraham joined #salt
05:46 kalloc joined #salt
05:51 jtang1 joined #salt
05:54 user_specific joined #salt
05:55 user_specific i need to install a pkg to the specific user in the salt_minion. how to configure this?
05:59 UtahDave user_specific: Can you explain what you're trying to do?
06:02 gw joined #salt
06:02 user_specific UtahDave: If i install a pkg using salt '*' state.sls svn. it installs in the system. If the system has 3 users say AB,C. i need to install only one user of the system say for A. how should i do this?
06:04 user_specific UtahDave: i don't want to install for users B,C. I hope you got it
06:05 UtahDave How would you do that without Salt?
06:06 joehoyle joined #salt
06:07 krow joined #salt
06:10 user_specific i want to do with salt
06:10 user_specific UtahDave: i want to do with salt
06:10 rojem joined #salt
06:11 UtahDave user_specific: what OS?
06:11 daneroo joined #salt
06:12 daneroo joined #salt
06:14 user_specific UtahDave: Ubuntu
06:15 user_specific UtahDave: i have installed many packages like mongodb, svn , pidgin etc.. in system specific. but for now i want to install in user specific.
06:16 oz_akan_ joined #salt
06:17 UtahDave Hm. I'm not 100% sure, but I don't think Salt can do that. I'm not sure how to do that manually
06:18 oz_akan__ joined #salt
06:19 sroegner_ joined #salt
06:19 user_specific UtahDave: oh thanks
06:21 malinoff user_specific, So basically you want to restrict access for users B and C. Thats how I would do that: add user A to the group svn, then chmod 0770 -R /path/to/svn or something like that
06:22 user_specific malinoff: can you let me know more exactly.
06:22 malinoff user_specific, You can't "install a package for a specific user", you can restrict access to package's files
06:23 user_specific malinoff: thanks
06:23 malinoff Linux is not Windows :) np
06:26 higgs001 joined #salt
06:28 ram_ joined #salt
06:28 ram_ Hi All
06:28 ram_ How to use Halite GUI to install apecific package in linux
06:28 ram_ ex: yum install screen. how do we use this one in Halite GUI
06:31 Psyke joined #salt
06:36 ml_1 joined #salt
06:37 mgw1 joined #salt
06:38 frasergraham joined #salt
06:45 jtang1 joined #salt
06:46 kalloc joined #salt
06:50 Ryan_Lane1 joined #salt
06:55 bhosmer joined #salt
07:12 Ryan_Lane1 joined #salt
07:19 oz_akan_ joined #salt
07:20 jtang1 joined #salt
07:29 Shenril joined #salt
07:30 SaveTheRbtz joined #salt
07:38 Iwirada joined #salt
07:39 ml_1 joined #salt
07:40 Nexpro joined #salt
07:40 fllr joined #salt
07:44 lessismore joined #salt
07:46 kalloc joined #salt
07:49 faldridge joined #salt
07:50 n8n joined #salt
08:00 Kenzor joined #salt
08:07 Psyke joined #salt
08:07 balboah joined #salt
08:07 slav0nic joined #salt
08:08 sroegner_ joined #salt
08:09 linjan_ joined #salt
08:10 debaser joined #salt
08:13 anuvrat joined #salt
08:14 kalloc joined #salt
08:15 Shenril joined #salt
08:17 fllr joined #salt
08:18 Poooogles joined #salt
08:20 oz_akan_ joined #salt
08:26 Poooogles joined #salt
08:27 rbtz joined #salt
08:27 harobed_ joined #salt
08:28 zooz joined #salt
08:35 yomilk joined #salt
08:41 jeff joined #salt
08:51 user_specific how to install jboss package
08:54 user_specific what is salt pillar? what is the main use of it?
08:56 bhosmer joined #salt
08:56 MrTango joined #salt
08:57 d1rk joined #salt
08:59 Poooogles joined #salt
09:00 Poooogles joined #salt
09:02 giantlock_ joined #salt
09:03 kalloc joined #salt
09:04 y0j joined #salt
09:05 astol joined #salt
09:08 user_specific i want to install rvm and ruby to the specific user in the salt-minion.
09:09 astol joined #salt
09:09 krak3n` joined #salt
09:10 TJ1980 joined #salt
09:12 astol joined #salt
09:14 user_specific malinoff: i want to install rvm and ruby to the specific user in the salt-minion. how do i do this?
09:17 fllr joined #salt
09:21 oz_akan_ joined #salt
09:28 Schmidt joined #salt
09:29 miguitas joined #salt
09:33 martoss joined #salt
09:33 martoss1 joined #salt
09:39 babilen Hi all - I am a bit confused when to use 1. pillar.get(...) 2. pillar[..][..] 3. salt['pillar.get'](..) -- could somebody elaborate on these and when each is most idiomatic?
09:41 Shenril joined #salt
09:42 Psyke joined #salt
09:43 d1rk babilen: did you read http://docs.saltstack.com/topics/tutorials/pillar.html ?
09:44 babilen d1rk: I did, but I read a lot and am not entirely sure what you refer to now.
09:45 d1rk babilen: to be honest, i am not able to answer your question, still need to get my head around it. But from what i know now, it is depending on 'where' you want to use your pillar data, it can be accessed in different ways.
09:47 kalloc joined #salt
09:47 babilen Fair enough - I simply lack the experience to judge that. It seems as if 2 is used quite frequently in the formulars and always "wrapped" into guards such as "if FOO in pillar and BAR in pillar[foo]"
09:53 babilen Also .. is there a good way to deal with the "no states for minion" if there simply might not be a state? I generate most stated dynamically from pillar data and sometimes I just don't want to do anything yet. Is there a good "do nothing" default state?
09:57 sroegner_ joined #salt
10:03 TJ1980 joined #salt
10:17 joehh joined #salt
10:17 fllr joined #salt
10:21 oz_akan_ joined #salt
10:23 oz_akan__ joined #salt
10:25 tomspur joined #salt
10:25 tomspur joined #salt
10:29 diegows joined #salt
10:34 joehoyle joined #salt
10:35 ravibhure joined #salt
10:37 ninkotech joined #salt
10:37 EWDurbin joined #salt
10:40 rofl____ joined #salt
10:54 joehoyle joined #salt
10:57 bhosmer joined #salt
10:57 astol joined #salt
10:58 kalloc joined #salt
10:58 higgs001 joined #salt
11:07 vbabiy_ joined #salt
11:08 sroegner_ joined #salt
11:17 fllr joined #salt
11:17 ml_1 joined #salt
11:20 anuvrat joined #salt
11:24 oz_akan_ joined #salt
11:34 david_a joined #salt
11:47 vbabiy_ joined #salt
11:47 techdragon joined #salt
11:49 istram joined #salt
11:51 ml_1 joined #salt
11:53 gnugnu2 joined #salt
11:57 gnugnu4 joined #salt
11:58 johnny_bravo joined #salt
11:58 gadams999 joined #salt
11:58 zatricky joined #salt
12:02 mgw2 joined #salt
12:02 sroegner joined #salt
12:04 giantlock joined #salt
12:04 krak3n` joined #salt
12:10 mgw1 joined #salt
12:12 zatricky Hi, all. My apologies if asking silly questions - my google fu is leading me to unhelpful. Can I use Salt to do configuration management (network/snmpd for example) without having to install the salt minion on the servers being managed?
12:14 astol joined #salt
12:17 fllr joined #salt
12:17 Schmidt (still new to this myself) but yes, you can use ssh
12:18 zatricky So ... when I update a config, it will upload the changes via ssh/rsync?
12:19 zatricky Thanks. I'll do more digging.
12:19 Schmidt zatricky: as long as the config file is specified in a .sls file applied to the minion i would assume yes
12:22 EnTeQuAk joined #salt
12:25 oz_akan_ joined #salt
12:27 astol joined #salt
12:30 joehoyle joined #salt
12:37 miguitas_ joined #salt
12:50 ml_1 joined #salt
12:56 toastedpenguin joined #salt
12:58 bhosmer joined #salt
13:05 vbabiy_ joined #salt
13:08 ml_1 joined #salt
13:09 vbabiy_ joined #salt
13:17 Pate_ joined #salt
13:17 fllr joined #salt
13:18 blee joined #salt
13:18 Pate_ I'm having some trouble installing packages on my Windows minions.
13:18 Pate_ Lodged issue here: https://github.com/saltstack/salt/issues/10444
13:19 _fllr_ joined #salt
13:20 Pate_ if I run salt '*' pkg.install chrome from the saltmaster, the Chrome MSI is downloaded from the saltmaster to each client's cache, but no install takes place and no messages or errors are displayed.
13:21 Pate_ minion logs at c:\salt\var\log\salt\minion are empty.
13:23 astol joined #salt
13:24 Pate_ Windows Event Viewer logs alo show no sign of an attempted Chorme install.
13:27 elfixit joined #salt
13:28 Led_Zeppelin i would like to group my servers. what should I look into in the docs? ie, servers that are a different OS.
13:30 gnugnu4 joined #salt
13:35 astol joined #salt
13:35 gnugnu4 joined #salt
13:39 lessismore I am confused about latest salt release, in https://github.com/saltstack/salt-cloud there is a link to the Hydrogen Release: https://github.com/saltstack/salt/issues?milestone=39 that seems complete but not yet released, even in the page http://docs.saltstack.com/topics/releases/2014.01.0.html there is the string "release:2013-12-30". Is there a place where I can track the progress about the new release?
13:40 Narcolapser joined #salt
13:43 farra joined #salt
13:46 N-Mi joined #salt
13:46 N-Mi joined #salt
13:48 Whissi joined #salt
13:50 jfzhu_us left #salt
13:54 jankowiak joined #salt
13:56 cpenner461 joined #salt
14:01 brianhicks joined #salt
14:01 sroegner joined #salt
14:05 ml_1 joined #salt
14:07 oz_akan_ joined #salt
14:09 scott_w joined #salt
14:10 zzzirk joined #salt
14:17 fllr joined #salt
14:17 bhosmer joined #salt
14:19 jaimed joined #salt
14:20 astol joined #salt
14:23 TJ1980 joined #salt
14:25 faldridge joined #salt
14:25 faldridge joined #salt
14:30 juicer2 joined #salt
14:32 ccase joined #salt
14:39 astol joined #salt
14:39 quickdry21 joined #salt
14:46 vejdmn joined #salt
14:47 Damoun joined #salt
14:51 Gifflen joined #salt
14:53 d1rk_ joined #salt
14:55 bstr joined #salt
14:56 JasonSwindle joined #salt
14:57 s8weber joined #salt
15:00 mpanetta joined #salt
15:01 steverw can i have modules in subdirs like /srv/salt/_modules/example/infoblox.py ?
15:01 steverw the seem to sync with minions but I dont know how to salt-call them
15:02 MrTango joined #salt
15:02 steverw or rather salt '*' example.infoblox.gethost ip=123.123.123.123
15:03 steverw im developing salt _modules.. and have each module in a git repo.. I want to include these modules on my salt master...
15:05 steverw while keeping them under version control.
15:07 steverw should i put each git salt module under a subfolder like _modules then use gitfs_remotes: for each module i want to include?
15:08 GradysGhost joined #salt
15:09 debaser have you tried without specifying subfolder, e.g. salt '*' infoblox.gethost ...
15:09 steverw yes and it never seemed to work... should it work?
15:10 debaser look for module_dirs in minion conf
15:11 debaser it should work if you add /srv/salt/_modules there
15:11 tyler-baker joined #salt
15:11 psyl0n joined #salt
15:11 debaser at least it works for me, as im doing something similar
15:12 Gordonz joined #salt
15:13 steverw ok i see module_dirs...
15:14 seanz joined #salt
15:14 steverw this is not a good solution because it requires each minion to have this set.  I tent to only want to set one value and thats the minion id.
15:14 steverw ill try some more things... I think the gitfs_remotes might work.
15:14 ipmb joined #salt
15:15 steverw thanks
15:19 debaser steverw: have a look at dynamic_modules then, ive not used it but it might work for you
15:22 gadams999 left #salt
15:22 rostam joined #salt
15:23 pass_by_value joined #salt
15:23 debaser steverw: using autoloading of dynamic modules is turned on by default and you only need to keey your custom modules on master in /srv/salt/_modules/
15:24 debaser calling state.highstate should fetch them
15:24 fllr joined #salt
15:25 steverw thanks/
15:33 Gifflen joined #salt
15:34 higgs001 joined #salt
15:36 vejdmn joined #salt
15:41 kalloc joined #salt
15:42 psyl0n joined #salt
15:42 Ahlee Can anybody tell what i'm doing wrong here?  trying to set pillar data up based on the lower() of a grain
15:42 Ahlee https://gist.github.com/jalons/0af35614e1ec2511268d
15:43 Ahlee grain is either Foo or foo
15:44 cachedout joined #salt
15:44 forrest joined #salt
15:46 rojem joined #salt
15:46 debaser left #salt
15:49 kalloc joined #salt
15:51 scoates Ahlee: looks right to me in Real Python™ (not Jinja… unsure about Jinja…) https://gist.github.com/jalons/0af35614e1ec2511268d
15:52 alunduil joined #salt
15:55 Ahlee thanks scoates
15:57 ml_1 joined #salt
16:02 dave_den grains['foo'].lower() and grains['foo']|lower() should both work. and with grains.get(), too
16:03 moped joined #salt
16:05 MTecknology so... saltstack: I HEART YOU ALL THIS MUCH!!!   |-[earth]---[sun]-----[galaxy]-----------[observable universe]-----------------------------[saltstack & community]--------|
16:07 dave_den feel the love
16:07 JasonSwindle Would you guys like a New Relic formula for Salt?
16:09 MTecknology When was saltstack formed?
16:10 JasonSwindle MTecknology:  Not too long ago
16:10 JasonSwindle Initial release
16:10 JasonSwindle 19 March 2011
16:10 JasonSwindle I think Salt got started in late Feb in 2011…… I think.
16:10 drawsmcgraw joined #salt
16:11 Ahlee i think i done found another bug
16:11 jeremyBass joined #salt
16:11 drawsmcgraw Salt Dudes - There's some curiosity re: if/when the Saltconf videos will be available.
16:11 drawsmcgraw http://www.reddit.com/r/saltstack/comments/1xtw86/will_the_saltconf_talks_be_made_available_online/
16:11 abe_music joined #salt
16:12 forrest yea I saw that last night drawsmcgraw
16:12 forrest I spoke with basepi about it last week, he said they aren't sure yet, apparently a lot of the breaks were recorded so they have to edit out all that.
16:12 drawsmcgraw forrest:  I figured one the important guys saw it but wanted to be sure.
16:13 drawsmcgraw Ah. So it's alotta work is what you mean to say :)
16:13 forrest drawsmcgraw, basically yea
16:13 forrest from what I understand the plan was to get all the presentations up first, then do the videos
16:13 forrest I have no idea who on the team is doing the editing.
16:13 drawsmcgraw Makes sense. Thanks
16:13 forrest yup
16:13 drawsmcgraw You mind if I mention that in the thread?
16:14 forrest I was gonna let timoguin know last night, but he wasn't around
16:14 drawsmcgraw (that it's in the editing process and will be up... uh, when it's up)
16:14 forrest drawsmcgraw, I don't even know if the editing process has started
16:14 drawsmcgraw ah, fair enough
16:14 forrest yea, don't want to make any assumptions
16:14 drawsmcgraw forrest: I'll let it be, then. timoguin's message is sufficient
16:14 drawsmcgraw Thanks!
16:15 forrest yea np
16:15 MTecknology JasonSwindle: thanks :)
16:16 nineteeneightd joined #salt
16:17 Ahlee hrm.
16:18 Ahlee even without the .lower i can't get that to work
16:18 xmltok joined #salt
16:18 dave_den Ahlee: can you gist it again?
16:18 Ahlee yeah
16:18 Ahlee one second
16:18 dave_den the other gist looked incomplete
16:18 Ahlee that was the entirety of my pillar file
16:18 Ahlee so if i'm missing something, that could be why
16:19 Ahlee one second
16:20 Iwirada left #salt
16:20 Ahlee https://gist.github.com/jalons/aa1ee0a57d0baaaae977
16:21 Ahlee even if i make grains['foo'] foo on hte minions, it's still not setting baz, and my yaml return is {} according to the logs
16:21 chrisjones joined #salt
16:21 krow joined #salt
16:21 jalbretsen joined #salt
16:22 basepi forrest: drawsmcgraw: pretty sure it's just Rhett working on it.  we did it all in-house rather than hiring a company to record, so we have to find time to edit them ourselves, which is easier said than done.  =)
16:22 dave_den so you want to set baz equal to either 'bar' or 'bar2' dict, depending on if grains['foo'] is either 'bar' or 'bar2'?
16:23 forrest basepi, jeez that Rhett guy, what a slacker!
16:23 drawsmcgraw basepi: Goodness. You guys don't know anything but hard work!
16:25 TheRealBill joined #salt
16:25 basepi forrest: last word I would use to describe Rhett.  =)
16:25 basepi drawsmcgraw: true enough.  ;)
16:25 jslatts joined #salt
16:25 forrest Ahlee, why can't you just do {% if 'bar' in grains['foo'] %} \n {% set baz = 'abc' %} {% else %} {% set baz = 'xyz' %}
16:25 forrest basepi, yea I know, I'm joking
16:25 drawsmcgraw basepi: Thanks for all the awesome, btw
16:25 forrest should have added a smiley face of some sort
16:25 frasergraham joined #salt
16:25 jankowiak joined #salt
16:26 rallytime joined #salt
16:26 dwiden joined #salt
16:27 dwiden is there good documentation on using salt to install software, python packages, etc on windows systems?
16:28 elfixit joined #salt
16:29 forrest dwiden, check out http://docs.saltstack.com/ref/windows-package-manager.html that will get you started on the package manager stuff, from there it's as easy as a normal pkg.installed
16:31 Ahlee forrest: is there an elif?
16:31 Ahlee as i technically have 3 items
16:31 viq Ahlee: yes, there is
16:31 dwiden forrest: thank you, I started reading that.  This may seem like a terrible, horrible question to ask, but would that solution work in a non-networked environment?
16:31 anuvrat joined #salt
16:32 forrest Ahlee, yea just do elif instead of else
16:32 viq dwiden: you can run salt in local mode... but I have no idea whether it would work on windows
16:32 forrest dwiden, do you mean one where you aren't connected to the internet and only have local stuff available?
16:33 dwiden let me restate my question: I will have my master and minions on a local intranet so that they can communicate with each other, but they will not be able to talk to the outside world
16:33 forrest dwiden, yea that's fine as long as you're ok with getting whatever software you need to your master.
16:33 dwiden I can do that
16:34 forrest the example that is up is pretty great, shows how to use the installer from all local locations in this section: http://docs.saltstack.com/ref/windows-package-manager.html
16:34 dave_den Ahlee: you know you're only setting a jinja variable inside the {% %} - you're not actually outputting that data anywhere
16:35 dwiden forrest: thanks again, I will definitely be back within the next day/week/months asking more questions that get increasingly more terrible
16:35 Ahlee dave_den: No, I did not know that.
16:35 dave_den Ahlee: so you need to actually put {{ baz }} in there to get it into pillar
16:35 Ahlee i'm not following
16:35 dave_den sec
16:35 forrest dwiden, yea np! I'm no windows expert but there is usually someone around!
16:36 dave_den Ahlee: https://gist.github.com/jalons/aa1ee0a57d0baaaae977
16:36 forrest Ahh sorry Ahlee, came into the conversation kinda late and thought you were trying to set stuff local to a state
16:36 forrest dave_den, to the rescue!
16:36 chrisjones joined #salt
16:37 * dave_den makes silly police siren noises
16:37 dave_den weeee-oooo-weeee-ooo
16:39 Ahlee dave_den: thanks
16:40 Ahlee still poking it
16:40 zooz joined #salt
16:43 thayne joined #salt
16:46 kossy joined #salt
16:48 meteorfox joined #salt
16:50 meteorfox DeprecationWarning: This minion recieved a job where kwargs were passed as string'd args, which has been deprecated. This functionality will be removed in Salt Boron.
16:50 meteorfox 'This minion recieved a job where kwargs were passed as '
16:50 meteorfox Hi, after firing an event using salt-call event.fire_master '{"minion":{"id":"foo"}}' 'salt/cloud/foo/bar/complete', I received this deprecation warning
16:50 meteorfox what is the 'new' way to achieve the same functionality?
16:52 bemehow joined #salt
16:58 yogi just finished a salt recipie for installing halite / PythonGit   using pip in an virtualenvironment is it possible to submit it somewhere ...
16:58 jeremyfelt joined #salt
17:00 djaime joined #salt
17:01 jeremyfelt joined #salt
17:01 frasergr_ joined #salt
17:01 viq yogi: https://github.com/saltstack-formulas/ is the top level, but I guess someone would need to set up project first for you to publish a pull request
17:03 yogi viq: whom should I approach for that
17:05 viq yogi: I guess any of the people listed as members there
17:05 viq For example forrest ;)
17:05 forrest who what?
17:05 forrest :P
17:06 viq forrest: yogi formula pull request ;)
17:06 forrest yea  Isaw, just joking
17:06 viq :P
17:06 forrest yogi, does it provide support for installations outside of a virtualenv as well?
17:08 yogi forrest: If you drop the require line and comment out the env_bin option it should install system wide
17:08 joehillen joined #salt
17:08 forrest yogi, https://github.com/saltstack-formulas/halite-formula
17:08 yogi forrest: If you are running master as root. As I'm not running master as root hence the venv option
17:09 forrest yogi, gotcha
17:09 forrest fork that repo, push your content in and make a pull request, I'll approve it
17:09 vejdmn joined #salt
17:09 viq yogi: would be great if you documented what you just said, and maybe even made it tunable via a pillar for example
17:09 forrest viq, yea agreed. I was gonna open an issue once he made the pull rq
17:10 forrest *req
17:10 forrest but I'll take a formula with some support, over no formula any day!
17:10 yogi sure
17:10 viq ;)
17:10 yogi I have towrap my head around pillar
17:10 viq I need to finally clean up and push out the progress I made on my gitlab formula
17:10 forrest no worries.
17:11 forrest viq, https://github.com/saltstack-formulas/gitlab-formula
17:11 forrest have you finished making it os agnostic?
17:11 forrest or is it still only for ubuntu
17:11 forrest either way, I'd still like to see a pull request on that :P
17:11 viq forrest: mine currently is tried on debian, want to make it work on CentOS as well, haven't tried anything else yet
17:11 frasergraham joined #salt
17:12 forrest ahh right debian, I couldn't remember
17:12 forrest viq, yea we talked about it a bit before, I just wanna see it in the formula repo ;P
17:12 viq I started with centos and rvm, but was shown the error of my ways ;)
17:12 JasonSwindle joined #salt
17:12 forrest hah
17:13 anuvrat joined #salt
17:13 viq ie: rvm is a huge pain and doesn't really work, and ruby 1.9.3 that you can possibly get in most places is enough, and possibly even the git 1.7 that's in centos is enough, though it's easy enough to get a newer one
17:14 TheRealBill joined #salt
17:14 forrest yea when I worked at my last company I tried to make RVM available to customers so they could use it
17:14 viq Though I have yet to try it properly on CentOS, or get around the strange place that ruby ends in
17:14 forrest It didn't go well.
17:14 forrest virtualenv on the other hand was done in a week
17:15 viq Also right now I'm using postgresql, "because", but since there were some comments here I'd like to make that tunable.
17:15 viq But at the very least "gitlab with postgres on debian" "works for me", just deployed it at work ;)
17:15 forrest nice
17:16 mikeywaites joined #salt
17:16 majoh joined #salt
17:16 viq With a bunch of settings I found in config files settable via pillar ;)
17:16 Guest61276 joined #salt
17:17 mikeywaites Hey peeps! hoping someone can help.  Just trying out salt for the first time with vagrant and i think im having some permissions issues.
17:17 mikeywaites ill just gist my setup
17:17 troyready joined #salt
17:17 viq mikeywaites: https://github.com/viq/cm-lab-salt ;)
17:18 AdamSewell joined #salt
17:18 KyleG joined #salt
17:18 KyleG joined #salt
17:18 [diecast] joined #salt
17:18 mikeywaites viq:  Im actually hoping for a masterless config at the moment
17:19 viq ah, ok
17:22 mikeywaites https://gist.github.com/mikeywaites/cc97bd24637e141f40b9
17:23 viq mikeywaites: so what's the problem?
17:24 mikeywaites output.txt
17:24 mikeywaites im assuming i should get the same error i see when i run it manually
17:24 mikeywaites it seems like its not running the states at all when running vagrant provision
17:24 viq I'm not sure you would get it to terminal
17:25 viq check salt logs
17:25 mikeywaites ah ok maybe this is a -v verbose output jobby
17:26 kalloc joined #salt
17:27 adizam joined #salt
17:27 adizam left #salt
17:27 kalloc_ joined #salt
17:29 kalloc joined #salt
17:33 kalloc_ joined #salt
17:34 pirateduck81 joined #salt
17:34 TJ1980 joined #salt
17:34 joehoyle joined #salt
17:36 joehoyle joined #salt
17:37 joehillen is is possible to have multiple unless statements for a cmd.run?
17:39 forrest joehillen, I haven't seen someone try it, however cmd.run does support kwargs so technically it might work
17:39 forrest if you try it and confirm can you create an issue to get the docs updated?
17:40 forrest not sure if this might be a problem though: https://github.com/saltstack/salt/blob/develop/salt/states/cmd.py#L512
17:40 JasonSwindle forrest:  Salt Formula question
17:40 forrest JasonSwindle, what's up
17:40 JasonSwindle I have a GPG Public key block I need to put somewhere……. PIllar or JINJA MAP?
17:41 JasonSwindle Or in a file that I just manage?
17:41 forrest can you get by without the key?
17:41 forrest or does the key change based on OS?
17:41 vbabiy__ joined #salt
17:41 JasonSwindle Right now it looks to be for RedHat base systems, but that is a great question
17:42 viq FWIW in my gitlab formula I'm putting SSL keys for it in pillar, but our CA cert that the cert is signed with I have as simply a file
17:42 cro joined #salt
17:42 forrest viq, as a managed file you drop in place?
17:43 viq forrest: precisely
17:43 forrest viq, gotcha
17:43 mikeywaites viq:  salt.verbose = true is the ticket!
17:43 cb [M ïS
17:43 viq So CA is simply file.managed, while cert and key are file.managed with contents_pillar
17:43 forrest JasonSwindle, looking at Dave's old example: https://gist.github.com/UtahDave/3785738 you can see he's using his pub keys as a managed file.
17:44 forrest viq, gotcha
17:44 forrest forgot the lines: https://gist.github.com/UtahDave/3785738#file-saltsample_09252012_01-L104
17:45 JasonSwindle OK
17:45 JasonSwindle That is what I was thinking
17:45 forrest managed file?
17:45 JasonSwindle Yep
17:45 JasonSwindle Just wanted to see if there was a "better" way or a Salt-Formula way
17:45 forrest I don't think there is a standard when it comes to pub keys
17:45 rallytime joined #salt
17:46 forrest the formulas really aren't 'mature' enough yet to where we're having some kind of puppet/example42 style split in thinking.
17:46 forrest right now it's more 'can we get everything in a formula?'
17:46 JasonSwindle Maybe that would be a good doc to have
17:47 JasonSwindle like for example, don't use short decks
17:47 forrest well, the formulas docs are pretty extensive
17:47 forrest but they don't cover some of the more mechanical stuff like that I don't think
17:50 anuvrat joined #salt
17:51 hunter joined #salt
17:52 schimmy joined #salt
17:55 bemehow_ joined #salt
17:57 QuantumRiff joined #salt
17:59 dleedyna joined #salt
17:59 QuantumRiff I'm trying to setup a base and dev environment.  in my "top.sls" I have my test server listed below "dev:" however, below base:, i have a '*' statement.  the highstate is trying to apply "/srv/salt/common/packages/ and /srv/salt/dev/common/packages/" and is seeing a conflict
17:59 QuantumRiff is there a good way to 'exempt' this server from the base: * match?
18:01 QuantumRiff or do I need to rename the element in my dev state files that is conflicting? that would seem to make it hard to copy state files back and forth
18:02 forrest QuantumRiff, did you already try '* and not L@devminion' ?
18:02 forrest QuantumRiff, or even better '* and not minion_name'
18:04 QuantumRiff forrest: I didn't know I could do that.. nice
18:04 mikeywaites left #salt
18:04 forrest QuantumRiff, it should work in the top
18:04 dwiden joined #salt
18:04 QuantumRiff my dev systems start with 'test-' so would it be something like '* and not test-*' ?
18:05 forrest QuantumRiff, yea that should be the right syntax
18:05 psyl0n joined #salt
18:05 forrest I'm just comparing it off the existing examples that I see.
18:05 dwiden how do you stop the salt-master daemon?  is it something as simple as "sudo salt-master stop"
18:05 kalloc_ joined #salt
18:05 Poooogles joined #salt
18:06 pedrovc joined #salt
18:06 nkuttler dwiden: on which os? use service?
18:06 dwiden running on ubuntu
18:06 nkuttler try /etc/init.d/salt-master stop
18:06 meteorfox joined #salt
18:08 joehoyle joined #salt
18:08 dwiden nkuttler: thank you!
18:09 nkuttler yw
18:09 frasergraham joined #salt
18:12 djaime joined #salt
18:13 kballou joined #salt
18:14 zzzirk joined #salt
18:17 meteorfox joined #salt
18:18 jaimed joined #salt
18:18 hunter joined #salt
18:20 ml_1 joined #salt
18:20 yogi forrest: I get a error message when I try to do a commit, The requested URL returned error: 403 while accessing https://github.com/saltstack-formulas/halite-formula.git/info/refs
18:23 Ryan_Lane1 joined #salt
18:23 forrest yogi, did you fork it?
18:25 forrest you need to fork it first so you have a copy in your own personal repo, then you push back to your copy, and then make the pull request to merge back into the main repo
18:27 Gareth morning
18:27 forrest hey Gareth
18:27 Gareth hey forrest, hows it going?
18:27 forrest It's friday! What about you?
18:27 Gareth Friday :)
18:28 forrest exactly
18:30 forrest Is anyone in here going to pycon?
18:30 Gareth PyCon is in the Phillipines right?
18:31 forrest it might as well be
18:31 forrest montreal this year
18:31 forrest could probably fly to the Phillipines for less.
18:32 TheRealBill joined #salt
18:32 Ryan_Lane1 do returners work when using masterless salt?
18:32 MrTango joined #salt
18:33 Ryan_Lane1 I guess I could write a simple returner to test
18:34 pass_by_value salt-call has a —return option
18:34 Ryan_Lane1 sweet
18:35 xmltok joined #salt
18:35 Ryan_Lane1 I'm working a path towards master/minion salt, but getting people onboard first is key
18:35 Ryan_Lane1 which means masterless first
18:36 jeremyfelt joined #salt
18:40 JasonSwindle joined #salt
18:41 zooz joined #salt
18:43 pedrovc Hello everyone! Is it possible to use the salt-bootstrap script to install multiple minions in one machine?
18:43 pedrovc "export BS_SALT_ETC_DIR=/etc/salt-2 && sudo sh install_salt.sh" after a clean install didn't work as I was expecting, there is only one minion running and using the /etc/salt configs
18:43 seanz left #salt
18:44 meteorfox joined #salt
18:44 forrest pedrovc, http://docs.saltstack.com/topics/tutorials/salt_bootstrap.html maybe try using the -c ,-k and -K options?
18:44 forrest That might only be for the install files however, I'm not sure.
18:45 pedrovc I've tried with the -c flag, no success
18:46 n8n joined #salt
18:47 pass_by_value Ryan_Lane1 that sounds like a good idea. It might be easier to get started that way.
18:47 Ryan_Lane1 yeah, if there's no need to manage another cert system and deal with another set of masters/minions it helps
18:48 Ryan_Lane1 I'm going to have to tackle the cert system issue at some point
18:48 hunter joined #salt
18:48 Ryan_Lane1 we use puppet and no one is going to be sold on having two cert systems
18:48 Ryan_Lane1 it's like that at wikimedia right now and people are not pleased
18:49 kalloc_ joined #salt
18:49 thayne joined #salt
18:52 pass_by_value pedrovc we have something called the minionswarm (thought I am not sure if that is exactly what you are looking for). The swarm allows us to run multiple minion processes on one machine. But they are essentially the same minion (connect back to the same master, have the same grains). Their keys are different though.
18:53 pass_by_value s/thought/though
18:54 pedrovc Humm, not exactly what I'm looking for. The masters are different for each minion. :/
18:55 forrest pedrovc, can you flesh out your idea, and create an issue over here for the bootstrap: https://github.com/saltstack/salt-bootstrap/issues/new
18:55 forrest or make a pull request fixing it :P
18:55 forrest well, adding that functionality
18:56 pedrovc I'll try!
18:56 joehoyle joined #salt
18:56 forrest pedrovc, awesome thanks! Even if you don't want to write the code, an issue with the functionality would be great!
18:57 linjan__ joined #salt
18:58 racooper joined #salt
18:59 Gifflen joined #salt
19:00 amckinley joined #salt
19:02 JordanRinke Morning
19:02 dwiden how do you specify multiple grains to match in a sls file?  For example, I want all minions with an id of 'dev*' and cpuarch of 64AMD to install 1 version, but cpuarch of x86 to install another version
19:03 redondos joined #salt
19:04 hunter joined #salt
19:05 jalbretsen joined #salt
19:09 Ryan_Lane1 joined #salt
19:10 jidar is there a .erb like configuration template? if so what's it called?
19:11 ipmb dwiden: a top file or a state file?
19:12 aboe joined #salt
19:12 faef joined #salt
19:12 ipmb dwiden: for top file, you can use a compound matcher http://docs.saltstack.com/ref/states/top.html#other-ways-of-targeting-minions
19:13 ipmb for a state file you can use a jinja {% if %} statement
19:13 dwiden I would like to do it in a top file, but I can put it into a state file.  My top is currently set up as this http://pastebin.com/U4s5La4f but this does not work
19:14 krow joined #salt
19:14 ipmb cpuarch needs to be on the same level as windows*
19:14 ipmb so you'd do something like this...
19:14 MrTango joined #salt
19:15 ipmb dwiden: http://pastebin.com/pRJUf2n4
19:16 dwiden ipmb: thank you!  I'm assuming that I can change the or to an and?
19:16 ipmb yep
19:16 ml_11 joined #salt
19:16 faef we're supposed to do pull requests against develop branch right? not any of the releases?
19:16 ipmb dwiden: http://docs.saltstack.com/topics/targeting/compound.html
19:16 sgviking joined #salt
19:16 hunter joined #salt
19:17 dwiden ipmb: thank you for the help!
19:17 ipmb np
19:17 psyl0n joined #salt
19:18 ipmb jidar: I don't understand your question. By default, salt uses Jinja for a template language.
19:22 jeremyfelt joined #salt
19:24 dwiden ipmb: I have another question if you're willing.  Is there a page that has all possible values for grains such as cpuarch?  I'm trying to match all 64-bit machines to a certain set of software, and I've already seen 3 different grains values for this processor type
19:25 gamingrobot joined #salt
19:27 racooper dwiden,  why not run salt '*' grains.item cuparch and get everything in your environment?
19:30 ipmb joined #salt
19:31 nextdoorwarren joined #salt
19:33 dwiden racooper: I only have one test minion to work with with certain attributes, but those attributes could be different when we deploy with many different minions.  The only real constant is that for this specific set of dependencies, all minions requiring them will be Windows
19:33 mgw joined #salt
19:34 dwiden So for my testing purposes, I ran that command and saw cpuarch=AMD64, but I've also seen amd64 (not sure if case sensitivity matters) and x86_64.  that should cover everything, I was just wondering if there was some master list that had them all
19:34 racooper the value for a particular grain is going to vary depending on your environment. If you don't have access to your production to run that query, I'd suggest finding someone who can. you can't test unless you have the correct data.
19:34 racooper when it comes to what Windows reports for cpuarch? no telling. I doubt any list would be complete.
19:35 dwiden racooper: Thank you.  Thinking about my question, it makes sense that such a list would not exist because I'm guessing salt asks the OS for an answer and just passes it through instead of normalizing it
19:37 bhosmer joined #salt
19:38 Gifflen joined #salt
19:39 Gifflen_ joined #salt
19:39 ml_1 joined #salt
19:40 diegows joined #salt
19:41 rojem joined #salt
19:41 mgw1 joined #salt
19:50 krow joined #salt
19:51 druonysus joined #salt
19:54 krow1 joined #salt
19:54 Hefeweizen joined #salt
19:55 ajw0100 joined #salt
19:55 Ryan_Lane1 joined #salt
19:58 jidar ipmb: I hadn't found this until you mentioned that: http://docs.saltstack.com/ref/renderers/all/salt.renderers.jinja.html#jinja-in-files
19:58 jidar thanks
19:58 ipmb np :)
19:59 hunter I'm attempting to duplicate the pillar seetup in http://docs.saltstack.com/topics/tutorials/states_pt4.html
19:59 hunter http://pastebin.com/KbP04QbT
20:00 faef is there anywhere posted why saltstack decided to go with global dunder variables created by the loader?
20:00 faef instead of passing opts along ?
20:00 hunter I'm getting that - The minion never replies when I invoke salt through the pillar matching - but it DOES if I specify the minion name directly.
20:00 hunter I _think_ this is a bug?
20:00 thayne joined #salt
20:01 hunter forrest: more pillar questions I'm afraid
20:01 scoates faef: I'm guessing to make it "easy"… but I'd prefer it to not do that (-:
20:01 faef scoates: rewriting fileserver backend right now, making it an ABC and forcing fileserver to pass along opts as a variable
20:02 faef and at the same time, abstracting out fileserver so it can be used for pillars too
20:02 ipmb hunter: try this...
20:02 ipmb salt leipzig.mcs.anl.gov pillar.items
20:02 ipmb does the pillar you are targeting show up there?
20:03 hunter Yes
20:03 hunter salt-call on the minion in question also returns the expected data.
20:03 hunter Do you want a pastebin?
20:03 jidar also, is there example salt states / sls modules? like a full apache2/httpd configuration I can pull down as an example
20:03 ipmb hmm, I'm not sure then
20:04 joehoyle joined #salt
20:04 ipmb jidar: https://github.com/saltstack-formulas/apache-formula
20:04 jidar you have that bookmarked ? :P
20:04 ipmb google is fast
20:04 jidar ah, then formula was the term
20:04 Ryan_Lane1 joined #salt
20:05 ipmb the formula is probably overkill for what you'd want
20:05 jidar still trying to get all the terms right, everybody has their own names for everything :/
20:05 ipmb jidar: yep, it sucks
20:05 hunter yeah - the nomenclature doesn't quite match up sometimes
20:05 jidar ipmb: looking for more real world examples than anything
20:05 ipmb http://saltstarters.org/
20:05 jidar truth be told, I'm hoping to find something more complicated so I get a better idea of where things work out
20:05 hunter Ok, this is a real deal-breaker for me - if I can't use pillar data to target minions, I'm sunk
20:06 ipmb jidar: there's more than one way to do it
20:06 ipmb hunter: I didn't even know that pillar matching existed, but have used grains to do what you are trying to do in the past
20:07 fllr joined #salt
20:07 hunter ipmb: yes, many do - but like a few others here (madduck, etal) I dislike asking the minion what they want to be - I want the master to TELL them.
20:08 ipmb that makes sense
20:08 hunter Which is why I was following that example from the docs - that does pretty much exactly what I want.
20:08 ipmb hunter: probably worth filing a bug in GitHub
20:08 hunter CRUD.
20:08 ipmb they are usually really responsive
20:09 _fllr_ joined #salt
20:09 ipmb like I said, I've never used that part before, so it might just be that the docs are off
20:09 ipmb hunter: instead of using --pillar, can you use I?
20:10 hunter I get the same result - tried that.
20:10 ipmb I@webserver_role:dev
20:10 ipmb darn
20:10 patrek_ joined #salt
20:11 ytrewq joined #salt
20:11 ytrewq left #salt
20:12 jimallman joined #salt
20:12 ipmb hunter: are you on the current release? `salt --versions`
20:12 hunter 0.17.4
20:12 ipmb and using ZMQ 3?
20:13 hunter Its 3 -- this is a bone stock install from the packages in EPEL, which is what i need for rollout.
20:13 hunter I'm searching the github ticket repo (or trying)
20:15 ipmb hunter: fwiw, this works for me
20:15 krow joined #salt
20:16 ipmb hunter: https://dpaste.de/VnqE
20:16 hunter what version are you running?
20:16 ipmb see the paste :)
20:17 HeadAIX joined #salt
20:17 hunter I found bug 6126 but thats not the same problem.
20:17 fllr joined #salt
20:17 hunter let me find the .5 release notes
20:18 Mua joined #salt
20:19 hunter I don't see anything in the release notes.
20:19 hunter Well spit.
20:19 hunter Let me file a bug in github.
20:20 joehoyle joined #salt
20:20 dave_den faef: fyi terminalmage is working on the gitfs code right for mountpoints right now ( https://github.com/terminalmage/salt/tree/gitfs-mountpoints )
20:21 terminalmage that is WIP
20:21 terminalmage it will not work
20:21 dave_den seems like you'll be duplicating a lot of functionality that can be used between the gitfs and git_pillar stuff.. i mentioned it to terminalmage eariler. you might want to just ping him about what you're working on
20:21 forrest lol
20:21 faef dave_den: not interested in mountpoints
20:22 faef honestly, i'm not too happy with the loader either
20:22 dave_den just trying to point out that overlap in effort and how it could be easily combined
20:22 jidar ipmb: how is the multi-os support done with the saltstack-formulas? I see thie apache-formula, but its debian/ubuntu specific
20:23 faef oh i'm just going to throw out git _pillar, it seems like a subset of what gitfs can do
20:23 ipmb jidar: they use that map file
20:23 jidar ah ok, there is no map file for that specific formula
20:23 ipmb jidar: https://github.com/saltstack-formulas/apache-formula/blob/master/apache/map.jinja
20:24 ipmb that is in the same repo
20:24 jidar well color me an idiot
20:24 jidar thanks again!
20:26 meteorfox joined #salt
20:26 patrek_ joined #salt
20:31 hunter forrest: any ideas on my latest pillar problem?
20:31 fllr joined #salt
20:32 patrek__ joined #salt
20:32 patrek_ joined #salt
20:32 DaveQB joined #salt
20:34 hunter Attempting to rebuild 0.17.5 for centos-6
20:35 hunter I'll be annoyed if thats it
20:39 rawtaz joined #salt
20:39 rawtaz o hay
20:39 forrest hunter, sorry a bit busy here at work, haven't looked
20:40 hunter I understand
20:46 hunter stab me with a spoon I just hit the rpm file format change again
20:47 bhosmer_ joined #salt
20:47 rawtaz who are they guys in the vid at https://www.youtube.com/watch?v=uWGDC1PdySQ ? amirite that the guy talking the most is Thomas? who's the other one?
20:49 TheRealBill joined #salt
20:50 hunter Upgraded to 0.17.5 on both master and minions - now I get No minions matched the target. No command was sent, no jid was assigned.
20:54 alexthegraham joined #salt
20:55 alexthegraham What's the flag to quiet output from salt commands? As in "only show me summary and errors"
20:56 hunter I'm gonna keep going - targetting is important but I can live without it.
20:56 hunter The minions are getting the right pillar data and thats all that is critical for now.
20:56 hellome joined #salt
20:56 dave_den alexthegraham: state_verbose
20:57 forrest that reminds me I need to close that issue I opened this week and modify it so that the note about the default status is correct.
21:02 schimmy joined #salt
21:04 dave_den hunter: I can target via pillar on 0.17.5 without issue
21:05 dave_den hunter: are you sure you're calling salt with -C?  salt -C 'I@webserver_role:dev' test.ping
21:06 MindDrive_ joined #salt
21:06 forrest sroegner, for shame, merging your own commit
21:08 faldridge joined #salt
21:08 hunter dave_den: I wasn't trying that command line at all - I was using salt -I/--pillar from the states tutorial.
21:09 hunter i just filed github 10454
21:09 hunter dave_den: let me try that
21:09 dave_den hunter: using -I also works for me.
21:10 ipmb hunter: your problem is interesting because the targeting is working as expected
21:10 ipmb the return value is not what you expect though
21:10 dwiden In my state file, how do I require that a certain version of my software to be installed?
21:10 rojem joined #salt
21:10 ipmb dwiden: depends on how you are installing it
21:10 hunter http://pastebin.com/dha92byt
21:11 hunter dave_den, ipmb: http://pastebin.com/dha92byt
21:11 ipmb right. it *does* find the minions
21:11 ipmb or match them
21:11 ipmb they don't return though
21:11 hunter Oh shit
21:11 hunter I just noticed that.
21:12 dwiden the minion is a windows machine.  I have ..win/repo/nasm/ directory with my init.sls and the installer
21:12 dwiden (those files are on my master, of course)
21:12 hunter Ummm. Something is still broke though - why would a manual test.ping work but a test.ping from a pillar target not work?
21:12 ipmb you lost me at "windows machine" :)
21:12 hunter ipmb: let me try targetting with a different state call
21:13 dave_den hunter: try to update the pillars on those minions
21:13 dwiden unfortunately this is the "easy" stuff, I'm gonna need to get licensed software installed on my windows minions at some point
21:13 hunter dave_den: how do I do that?
21:13 JasonSwindle MTecknology:  Did you ever get your odd issue fixed?
21:13 ipmb dwiden: http://docs.saltstack.com/ref/windows-package-manager.html
21:13 JasonSwindle The one that made your Highstate last forever?
21:13 dave_den hunter:  salt -L 'france.igsb.anl.gov,leipzig.mcs.anl.gov' saltutil.refresh_pillar
21:14 yidclare joined #salt
21:14 hunter dave_den: from that call both minions report "None"
21:14 dwiden ipmb: I'm trying to figure out how to specify version in my state file for this, but that page doesn't have an example of it
21:14 hunter dave_den: is that the expected output?
21:14 jesusaurus is there a way to trigger a gitfs refresh?
21:14 dave_den hunter: refresh_pillar always returns 'None'.
21:15 hunter dave_den: that did it -- now a test.ping comes back.
21:15 hunter dave_den: WHAT the hell did I do wrong there?
21:15 jesusaurus im finally taking the plunge and using gitfs for my states, and ive updated the state on the git remote, but highstate is still using the old (~1 hour) state
21:15 yidclare I'm wondering if there are going to be videos available from the SaltConf last month?  Does anyone know?
21:15 dave_den your minions did not have a current set of pillar data.
21:16 hunter dave_den: but a) why not and b) why would it matter if the minions didn't for a test.ping ?
21:16 dave_den hunter: minions only update pillar data during certain things. the most common being highstates and saltutil.refresh_pillar
21:16 dave_den hunter: the minions themselves determine if the job applies to them
21:17 jesusaurus basepi: when does gitfs update its state cache?
21:17 hunter dave_den: even if I've done "salt-call pillar.items" on the minions  - and that salt-call was successful?
21:17 dave_den the master just published the job saying "hey minions, if you meet these parameters then run this job'
21:17 Gifflen joined #salt
21:17 hunter dave_den: *HEAD*DESK*
21:18 dave_den heh
21:18 hunter dave_den: so a common working pattern would be to dork around with my pillar data, call refresh_pillar, and THEN try if things work?
21:19 hunter dave_den: I claim thats a bug in the 4th part of the state tutorial then - it NEVER mentions updating the minion pillar data
21:19 dave_den yes, if you are targeting on pillar data and you change the pillar on the master then the salt-minion doesn't know that until a function that refreshes the pillar data is executed
21:19 dave_den hunter: i dunno, i havent read the tutorial but it should definitely be mentioned
21:20 smcquay joined #salt
21:21 hunter https://github.com/saltstack/salt/issues/10454
21:21 hunter You get the credit -thanks dave_den.
21:21 dave_den glad to help!
21:22 StDiluted joined #salt
21:22 hunter I feel an ass for typing all that up and filing a bug on a known ... thing. But the docs led me astray (waves fist in air)
21:23 jeremyBass joined #salt
21:25 basepi jesusaurus: pretty sure it's updated every loop_interval
21:25 schimmy1 joined #salt
21:25 basepi terminalmage may know better, he's spent some time there recently
21:26 jesusaurus thanks
21:26 jesusaurus i'll keep digging to find out why its not updating
21:27 jimallman joined #salt
21:30 terminalmage jesusaurus: yes it's loop interval
21:30 terminalmage which defaults to 60 sec
21:31 terminalmage there is a dedicated thread for the master that does maintenance stuff, including updating fileserver cache, and then sleeps for loop_interval second
21:31 terminalmage *seconds
21:34 jesusaurus weird so it is in fact updated on the master
21:34 * jesusaurus kicks salt with a refresh_modules
21:34 hunter joined #salt
21:34 jesusaurus and its happy...
21:35 ccase joined #salt
21:35 jesusaurus terminalmage: for future reference, where in the codebase is that maintenance thread?
21:36 vbabiy_ joined #salt
21:37 hunter joined #salt
21:39 terminalmage jesusaurus: salt/master.py
21:39 terminalmage in class Master, function _clear_old_jobs
21:40 jesusaurus terminalmage: awesome, thanks
21:40 hunter joined #salt
21:40 terminalmage jesusaurus: why is it weird that the fileserver cache is updated on the master/
21:40 terminalmage ?
21:40 terminalmage that's where the fileserver is maintained and the place from which the files are served
21:41 jesusaurus no, i meant that it was weird that the master was not out of date but the minion was
21:41 terminalmage ah
21:42 terminalmage the minion doesn't constantly update the fileserver cache, it only does so when it needs new files
21:42 jalbretsen joined #salt
21:43 chrisjones joined #salt
21:44 jtang1 joined #salt
21:44 MindDrive joined #salt
21:45 MTecknology JasonSwindle: nope... and it's causing more and more of an issue :(
21:46 sroegner forrest: hu?
21:47 kballou joined #salt
21:47 Gifflen joined #salt
21:49 Gifflen_ joined #salt
21:53 TheRealBill joined #salt
21:57 juniorzoid Is there a way to get a salt minion to reload /etc/salt/grains without restarting?
21:59 joehoyle joined #salt
22:00 mgw1 joined #salt
22:01 kermit joined #salt
22:03 Gifflen joined #salt
22:04 Ryan_Lane1 joined #salt
22:09 juniorzoid I just noticed that salt-call -g on the minion shows the updated contents of /etc/salt/grains right away. I guess my problem is getting the salt master to see the new grains
22:15 joehoyle joined #salt
22:16 jeremyBass1 joined #salt
22:18 psyl0n joined #salt
22:18 vbabiy_ joined #salt
22:19 faldridge joined #salt
22:20 elfixit joined #salt
22:20 hunter Is there any reason NOT to change the minion id's to be the hostname without the domain bit?
22:26 btorch does anyone know how I can avoid salt from throwing an exception if for some reason a minion still doesn't have any roles defined ?
22:26 forrest sroegner, you merged your own pull request for hadoop
22:27 forrest hunter, I don't think so, it all seems to be preference.
22:27 hunter I just get tired of typing all the extra verbiage.
22:27 hunter And for now at least, our hostnames are unique even across domains.
22:29 btorch I was assuming  that it would just see the roles grain as being null and move on
22:31 bhosmer joined #salt
22:32 forrest btorch, I'm not sure
22:33 hunter Is there any relationship between the "environment" thats set in pillar data and the environment thats used in state data?
22:33 hunter Or are they seperateed completely?
22:34 nineteeneightd joined #salt
22:34 bemehow joined #salt
22:34 MindDrive joined #salt
22:35 MindDrive left #salt
22:35 MindDrive joined #salt
22:37 hunter I guess I'm asking if pillar data can override (or replace) the states top.sls file.
22:37 faldridge joined #salt
22:38 higgs001 joined #salt
22:40 JasonSwindle left #salt
22:41 btorch forrest: yeah seems like would have to check that roles is indeed defined everywhere :) or just make sure there is at least one role (even if it's a bogus one) during tests for things not to fail
22:56 vejdmn joined #salt
22:57 sroegner forrest: I did?
22:58 alexthegraham Is it possible to set backup_mode on the Salt master so that it applies for all minions?
22:59 sroegner forrest: i think i just push all the time
22:59 alexthegraham It doesn't give an error when I put backup_mode: minion in the master config, but I'm also not positive it's doing anything.
23:02 xmltok_ joined #salt
23:02 yomilk joined #salt
23:02 frasergr_ joined #salt
23:05 felixhummel joined #salt
23:10 swissknife joined #salt
23:15 ajw0100 joined #salt
23:15 cachedout joined #salt
23:16 alunduil joined #salt
23:20 kballou joined #salt
23:21 frasergraham joined #salt
23:25 hunter my ... salt config seems to work.
23:26 hunter The '*' in base spits out the motd formula and the one host mentioned in dev got that PLUS all the dev stuff.
23:26 hunter I feel like I should walk away from the keyboard right now.
23:29 rojem joined #salt
23:32 gnugnu4 joined #salt
23:33 alice|wl hunter: nooo! install all the reciepes :)
23:34 fragamus joined #salt
23:37 Sypher joined #salt
23:41 hunter Well, the salt mine and fixing whatever is wrong with the zfs module/state are on the list.
23:41 hunter But not tonight.
23:46 forrest joined #salt
23:46 ajw0100 joined #salt
23:52 brianhicks joined #salt
23:55 psyl0n joined #salt

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