Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2014-12-15

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

All times shown according to UTC.

Time Nick Message
00:06 jnials joined #salt
00:06 adilroot joined #salt
00:08 davesque i see my confusion now.  i didn’t realize that minion ids default to their fqdn or hostname (and that this value is cached).
00:08 kevinforbes joined #salt
00:10 kevinforbes Any salt-cloud users here?
00:11 kevinforbes The salt-cloud module requires you to hardcode your id and key (AWS credentials) in your cloud.providers file
00:11 kevinforbes I’m trying to figure out a way to not check my credentials into Github
00:11 kevinforbes Any idea on how I might accomplish that?
00:14 kevinforbes I tried using standard pillars, but got a syntax error on the id: {{ pillar['AWS_ACCESS_KEY_ID'] }} format
00:21 yomilk joined #salt
00:26 budman joined #salt
00:27 budman_ joined #salt
00:28 jnials joined #salt
00:31 foulou joined #salt
00:36 brendanashworth joined #salt
00:36 smcquay_ joined #salt
00:38 otter768 joined #salt
00:46 jnials joined #salt
00:52 smcquay_ joined #salt
00:56 smcquay joined #salt
01:03 aqua^mac joined #salt
01:18 genediazjr joined #salt
01:25 smcquay joined #salt
01:39 Daemonik joined #salt
01:43 TyrfingMjolnir joined #salt
01:47 aurynn the answer is no, you can't use mine data in a pillar
01:48 TheThing joined #salt
01:56 malinoff joined #salt
02:06 rocket joined #salt
02:08 bhosmer joined #salt
02:13 aparsons joined #salt
02:23 quickdry21 joined #salt
02:31 TyrfingMjolnir joined #salt
02:32 pdayton joined #salt
02:34 malinoff joined #salt
02:49 genediazjr joined #salt
02:50 yomilk_ joined #salt
02:50 bhosmer joined #salt
02:54 aw110f joined #salt
02:57 aw110f_ joined #salt
03:03 vbabiy joined #salt
03:04 CeBe1 joined #salt
03:06 genediazjr joined #salt
03:12 genediazjr joined #salt
03:23 NightMonkey joined #salt
03:36 otter768 joined #salt
03:50 genediazjr joined #salt
03:54 genediazjr joined #salt
04:06 ajolo joined #salt
04:06 favadi_ joined #salt
04:11 pdayton joined #salt
04:17 ndrei joined #salt
04:20 pdayton joined #salt
04:21 hunmonk_ joined #salt
04:24 hunmonk_ trying to get a jinja macro working as here: http://pastebin.com/qW9EWaum — getting Jinja variable 'iptables_managed' is undefined. putting the macros directly in the main SLS file works. is there something with the way include works that the macro is being called before the include?
04:26 yetAnotherZero joined #salt
04:30 jonbrefe joined #salt
04:31 genediaz_ joined #salt
04:41 otter768 joined #salt
04:42 emid_ joined #salt
04:43 badon joined #salt
04:55 kevinforbes joined #salt
05:02 atbell joined #salt
05:06 genediazjr joined #salt
05:22 yetAnotherZero joined #salt
05:22 pdayton joined #salt
05:38 peters-tx joined #salt
05:39 felskrone joined #salt
05:42 felskrone1 joined #salt
05:47 ramteid joined #salt
05:50 genediazjr joined #salt
05:54 seshan joined #salt
05:55 freelock joined #salt
06:00 jeddi joined #salt
06:01 tsotf joined #salt
06:08 jhauser joined #salt
06:15 tsotf joined #salt
06:17 tsotf joined #salt
06:21 jchen joined #salt
06:26 Emantor joined #salt
06:27 simonmcc_ joined #salt
06:27 tmmt joined #salt
06:27 flamin_scotsman joined #salt
06:27 cb joined #salt
06:27 APLU joined #salt
06:27 forze joined #salt
06:28 mschiff joined #salt
06:28 mschiff joined #salt
06:28 manytrees joined #salt
06:28 NightMonkey joined #salt
06:30 linjan joined #salt
06:40 NightMonkey joined #salt
06:48 gildegoma joined #salt
06:49 catpiggest joined #salt
06:52 otter768 joined #salt
06:55 Outlander_ joined #salt
07:01 colttt joined #salt
07:02 aw110f joined #salt
07:05 genediazjr joined #salt
07:06 oyvjel joined #salt
07:07 oyvjel1 joined #salt
07:08 oyvjel2 joined #salt
07:11 kermit joined #salt
07:13 JlRd joined #salt
07:14 oyvjel joined #salt
07:21 lempa joined #salt
07:22 dooshtuRabbit joined #salt
07:24 flyboy joined #salt
07:27 ajolo joined #salt
07:29 felskrone joined #salt
07:33 spiksius joined #salt
07:35 genediazjr joined #salt
07:43 genediazjr joined #salt
07:45 slafs joined #salt
07:45 jgelens joined #salt
07:45 slafs left #salt
07:52 yomilk joined #salt
07:54 trikke joined #salt
07:57 aw110f_ joined #salt
08:08 pf_moore joined #salt
08:11 lb1a joined #salt
08:11 che-arne joined #salt
08:11 che-arn4 joined #salt
08:18 genediazjr joined #salt
08:22 akafred joined #salt
08:24 vectra joined #salt
08:32 bill joined #salt
08:32 Guest63734 hello,I now want to use salt-api,but I can start it
08:35 Guest63734 hello,I now want to use salt-api,but I cannot start it
08:38 seshan joined #salt
08:41 kawa2014 joined #salt
08:45 iwishiwerearobot joined #salt
08:48 genediazjr joined #salt
08:51 karimb joined #salt
08:53 otter768 joined #salt
08:54 williambill joined #salt
08:54 williambill hello,can I use salt-api execute "salt cmd.run"?
08:55 lothiraldan joined #salt
08:55 tomspur joined #salt
08:56 monkey66 joined #salt
08:58 lb1a joined #salt
08:59 jeddi joined #salt
09:00 monkey66 joined #salt
09:00 monkey66 left #salt
09:03 intellix joined #salt
09:03 Auroch joined #salt
09:06 wnkz joined #salt
09:09 bbnix joined #salt
09:15 wvds-nl joined #salt
09:17 BigBear joined #salt
09:17 Dw_Sn joined #salt
09:19 thehaven joined #salt
09:20 elfixit joined #salt
09:29 ttrumm joined #salt
09:34 glyf joined #salt
09:35 favadi_ joined #salt
09:37 lothiraldan joined #salt
09:39 Flusher joined #salt
09:41 lothiraldan joined #salt
09:44 CeBe joined #salt
09:46 yomilk joined #salt
09:49 intellix joined #salt
10:11 iggywig123 joined #salt
10:12 iggywig123 Hey. Can anyone help me out with a salt-cloud rackspace/openstack question? I’m trying to use salt-cloud to provision into a private network on Rackspace..
10:12 iggywig123 Do I need to use the nova module?
10:21 felskrone joined #salt
10:22 ganes joined #salt
10:22 ganes hey
10:22 ganes Need Saltstack REST-API list
10:22 ganes ??
10:22 giantlock joined #salt
10:23 babilen No, thank you.
10:24 workingcats joined #salt
10:25 ganes any ref links or documents
10:27 glyf joined #salt
10:30 oyvjel joined #salt
10:36 TheThing joined #salt
10:38 brayn joined #salt
10:41 fredvd joined #salt
10:50 oyvjel1 joined #salt
10:50 oyvjel joined #salt
10:52 oyvjel joined #salt
10:53 bhosmer_ joined #salt
10:54 otter768 joined #salt
11:01 iggywig123 joined #salt
11:03 colttt joined #salt
11:04 claytron joined #salt
11:09 tsotf left #salt
11:18 bbnix joined #salt
11:20 seshan joined #salt
11:20 jonatas_oliveira joined #salt
11:26 seshan I am basically using the 2nd methd referenced here to push files from minion to master
11:26 seshan but everytime I try to push it is givning me this error :
11:26 seshan {'minion1': {'cp.push': 'Traceback (most recent call last):\n  File "/usr/lib/python2.7/dist-packages/salt/minion.py", line 1131, in _thread_multi_return\n    ret[\'return\'][data[\'fun\'][ind]] = func(*args, **kwargs)\nTypeError: push() takes exactly 1 argument (15 given)\n'}}
11:30 bhosmer joined #salt
11:31 babilen "here" ? Note that deictic expressions are tricky to resolve on IRC
11:32 forcer seshan: Can you post the command line you used to invoke cp.push?
11:32 seshan babilen, ya i got it working,
11:32 seshan some small syntax error
11:33 seshan just kinda overlooked it
11:33 seshan forcer, ya command line to worked :)
11:34 felskrone joined #salt
11:38 jeffrubic joined #salt
11:46 bram__ joined #salt
11:47 bram__ hi all - quick question… After an archive extract I would like to copy the extracted directory somewhere… How can I do this? I tried file.copy but that seems to fail for directories…
11:48 eXistenZNL file.recursive maybe?
11:48 eXistenZNL that works for other sources so maybe for local filesystem sources as well?
11:48 eXistenZNL *file.recurse
11:49 bram__ Recurse gives me: Comment: Invalid source ‘/bleep’ (must be a salt:// URI)
11:49 giantlock joined #salt
11:51 bram__ anyone else got any ideas?
11:52 linjan joined #salt
11:54 Sypher bram__, easiest is to use archive.extracted which will take care of moving it to a selected directory
11:54 Sypher http://docs.saltstack.com/en/latest/ref/states/all/salt.states.archive.html
11:54 bram__ Sypher: I am already using that
11:55 bram__ Sypher: after extracting I want to copy the files elsewhere
11:55 bram__ Sypher: as you can not define the name of the extracted directory
11:56 Sypher the fastest is a simple cmd.run that does it. but iirc it does allow you to set a name
11:57 bram__ Sypher: it doesn’t. It allows you to set the name of the containing directory
11:57 bram__ so I want to have something_1.5.4.tar.gz in /usr/share/something/
11:57 bram__ archive can create /usr/share/something_1.5.4/ though
11:57 babilen A cmd.wait with a suitable unless test and require sounds like a reasonable approach
12:00 bram__ maybe I should just use a symlink...
12:00 bram__ create /usr/share/something_1.5.4/ with archive and then symlink something to something_1.5.4
12:04 babilen /usr/share sounds like a weird place, but yeah. That would be an option.
12:07 mick3y babilen: sounds like a weird place, but would you ever guess where emc client's rpm installs it's binaries? ;)
12:07 diegows joined #salt
12:10 babilen Well, unpacking an archive is one thing and performing operations with the package manager another. I'd argue that only files under the control of the package manager should live in /usr/share (as opposed to, say, /usr/local/share)) which is why I said it is a "weird place"
12:10 mick3y babilen: oh i agree with that. emc powerpath binaries were installed under /etc :)
12:11 mick3y now that's a weird place ;-)
12:12 mick3y I agree on respecting the FHS :) http://www.linuxbase.org/betaspecs/fhs/fhs.html#usrshareArchitectureindependentData
12:12 iggywig123 joined #salt
12:23 ramteid joined #salt
12:26 yomilk joined #salt
12:27 Dw_Sn joined #salt
12:30 hobakill joined #salt
12:30 yomilk joined #salt
12:35 wm-bot4 joined #salt
12:36 diegows joined #salt
12:36 felskrone joined #salt
12:38 brayn joined #salt
12:47 tkharju joined #salt
12:50 iggywig123 Does anyone know how I can install salt.cloud.nova?
12:50 yomilk joined #salt
12:51 bram__ hey guys, what difference can there be between doing A: pkg.installed: […] service: [running] and separating A-pkg: pkg.installed + A: service: running
12:51 bram__ because for one of our services the first is faining and the second is working
12:51 dimeshake you may need a require: pkgname for the first to execute in the right order
12:52 dimeshake under service:
12:52 bram__ oh wow!
12:52 bram__ I didn;t know that
12:52 bram__ thanks, because now looking at the output I can see that it is trying the service first
12:52 kesyer joined #salt
12:55 otter768 joined #salt
12:57 malinoff joined #salt
13:02 dimeshake anyone else doing the enterprise training this week?
13:04 Fbraad joined #salt
13:10 genediazjr joined #salt
13:14 iggywig123 joined #salt
13:14 dimeshake iggywig123: did you read this? http://docs.saltstack.com/en/latest/ref/clouds/all/salt.cloud.clouds.nova.html
13:16 iggywig123 I did but I’m not really understanding it..
13:16 dimeshake i think the most relevant part for ya is: "This module currently requires the latest develop branch of Salt to be installed."
13:17 iggywig123 Ok. How do I do that?
13:18 dimeshake https://github.com/saltstack/salt-bootstrap
13:18 dimeshake see examples
13:21 intellix joined #salt
13:28 thawes joined #salt
13:32 lothiraldan joined #salt
13:32 elfixit joined #salt
13:33 jeddi joined #salt
13:35 jakubek joined #salt
13:36 jakubek anyone is using reactor and have more than one local.state.sls in one reactor file?
13:38 vbabiy joined #salt
13:40 seshan joined #salt
13:43 nitti joined #salt
13:44 FRANK_I joined #salt
13:47 thawes joined #salt
13:53 thawes joined #salt
13:55 glyf joined #salt
14:14 gngsk joined #salt
14:16 ajolo joined #salt
14:19 glyf joined #salt
14:24 P0bailey joined #salt
14:24 P0bailey joined #salt
14:28 jeremyr1 joined #salt
14:28 TrafficMan joined #salt
14:29 patrek joined #salt
14:31 mpanetta joined #salt
14:35 mpanetta joined #salt
14:35 thawes_ joined #salt
14:37 mpanetta joined #salt
14:40 mpanetta joined #salt
14:40 orion__ joined #salt
14:41 mpanetta_ joined #salt
14:42 ninkotech joined #salt
14:44 iggywig123 joined #salt
14:45 perfectsine joined #salt
14:47 kaptk2 joined #salt
14:47 pppd joined #salt
14:49 genediazjr joined #salt
14:50 _prime_ joined #salt
14:51 orion___ joined #salt
14:55 Auroch joined #salt
14:56 otter768 joined #salt
14:59 lothiraldan joined #salt
15:03 perfectsine joined #salt
15:04 iggywig123 joined #salt
15:04 quickdry21 joined #salt
15:05 hobakill good morning / afternoon! is anyone using a git fileserver backend that has adjusted the pull times? i know the default is 60 seconds but i'm troubleshooting an issue and i'd like to increase that.
15:06 hobakill if you're interested this is the error bogging down my logs: https://www.hobapolis.com/paste/?361cef352f4c6afb#KcCQB9KohIq7k+cFfUg/LUejMgBjDYP3vYPwYyE9LpM=
15:12 rojem joined #salt
15:13 ggoZ joined #salt
15:13 perfectsine joined #salt
15:15 bhosmer joined #salt
15:16 jnials joined #salt
15:19 _JZ_ joined #salt
15:21 thawes joined #salt
15:23 rickh563 joined #salt
15:25 nitti_ joined #salt
15:25 mapu joined #salt
15:27 perfectsine joined #salt
15:28 iwishiwerearobot joined #salt
15:29 mick3y question: does --no-color option works for cmd.run for anyone on 2014.7.0 ?
15:30 mick3y when i run for example: salt --no-color 'host' cmd.run  'cat /etc/issue' it still colors the output
15:30 phx joined #salt
15:30 tyler-baker joined #salt
15:32 hobakill mick3y, no color for me
15:32 Ozack joined #salt
15:36 DaveQB joined #salt
15:37 pdayton joined #salt
15:37 dynamicudpate joined #salt
15:38 dimeshake mick3y: does /etc/issue have color codes in it, itself?
15:39 fredvd joined #salt
15:44 mick3y dimeshake: no it doesn't
15:44 dimeshake works 4 me
15:44 mick3y hobakill: thank's for checking. did you use exact same syntax and did you run the command from your salt master?
15:44 dimeshake salt 'salt*' cmd.run 'cat /etc/issue' --no-color
15:44 dimeshake is the form that I used
15:44 hobakill mick3y, yup
15:44 * mick3y thinks it might be down to some master config settings then
15:46 mick3y cound you show me: grep -E 'state_(verbose|output)' /etc/salt/master ?
15:46 mick3y if you have them set up
15:47 dimeshake i dont' have it set in master config
15:47 dimeshake but i have salt itself aliased to salt -v
15:47 mick3y mkay. i have 'color: True' in my config. might be that
15:48 dimeshake I would imagine so - but I'd also expect --no-color on cli to override
15:48 DaveQB joined #salt
15:49 mick3y dimeshake: likewise - it doesn't work though
15:49 dimeshake mick3y: same behavior when i switch to color: True in master conf
15:49 dimeshake i get color
15:49 mick3y yeah. the cli option doesn't take priority over config setting
15:50 dimeshake to the issues queue with you!
15:50 mick3y i don't have an account @ github :-(
15:50 perfectsine joined #salt
15:50 dimeshake why do you need color: True though?
15:50 LittUp left #salt
15:50 dimeshake default is to color. I think they probably only anticipated that var to be used to turn of color all the time
15:50 sroegner joined #salt
15:51 mick3y dimeshake: i'm getting quickly bored with black and white :)
15:51 dimeshake it should color output by default
15:51 N-Mi joined #salt
15:51 mick3y makes sense. sort of
15:51 dimeshake i've never had to set it, is what i mean :)
15:51 dimeshake is your $TERM set to something without color support?
15:52 mick3y i don't think so
15:53 agend joined #salt
15:53 mick3y nvm. i'll unset it
15:55 che-arne joined #salt
15:56 FRANK_I I have a question is any way that I can installed the minion
15:57 FRANK_I sorry.
15:57 hunmonk_ joined #salt
15:57 iggywig123 joined #salt
15:57 FRANK_I I want to install the minion to 1500 Nodes is anyway that I can do it with salt-ssh or something like that?
15:57 conan_the_destro joined #salt
15:58 racooper joined #salt
15:58 FRANK_I I can do it from my kickstart i am just exploring my options
15:59 iggy FRANK_I: yes
15:59 jaimed joined #salt
15:59 iggy there are actually a few ways iirc
16:00 diegows hi
16:00 iggy salt-ssh is one, another would be the "saltify" provider in salt-cloud
16:00 aquinas joined #salt
16:00 iggy there are probably more
16:00 diegows any idea why I'm getting NameError: name '__salt__' is not defined in a custom module?
16:01 FRANK_I ok Let me check salt-ssh
16:01 nitti joined #salt
16:01 iggy diegows: is this coming when salt imports it?
16:01 jonbrefe joined #salt
16:01 SheetiS joined #salt
16:02 diegows yes, it's in the minion log: https://gist.github.com/diegows/53cf1da755ecbcf6099d
16:03 diegows iggy, ^^
16:03 N-Mi joined #salt
16:03 iggy odd, loader.py should be loading those dunder dicts before the import
16:03 diegows not sure what's missing, it was working in another node
16:04 diegows well, may be another bug to report, I have other non blocking about this :)
16:04 iggy is there a reason you're setting it like that?
16:04 diegows but this one is blocing
16:04 diegows iggy, setting what exactly?
16:05 iggy i.e. why mine_get = __salt__['mine.get'] vs data_I_really_want = __salt__['mine.get']('foo')
16:05 diegows I lke to use mine.get() :)
16:05 iggy I don't think that's causing your problem, but it certainly seems odd
16:05 diegows but I could try to change it
16:06 DaveQB joined #salt
16:07 diegows I moved mine_get = __salt__....
16:07 diegows to a function
16:07 diegows and works
16:07 diegows :P
16:07 diegows adding this to my todo list to check it better
16:09 iggy in that case, I suspect what is happening is that setting it that way (vs calling the function directly) is causing some sort of early lookup before the dunder dicts are loaded
16:10 iggy as to why it worked on another minion, and not this one... nfc
16:14 ndrei joined #salt
16:14 babilen The dunder dicts are monkey patched later on (which is super ugly) and are *not* available during evaluation time, but *only* during runtime. You will therefore need to enclose every call to them in a function and it is impossible to refer to them on a module level
16:15 iggy diegows: there you have it
16:15 babilen I typically have "__salt__ = None" in my modules so that I can still use static code checkers (e.g. syntastic) without them going bonkers because I constantly make calls to things that shouldn't be there.
16:15 babilen Rather ugly that
16:16 scalability-junk joined #salt
16:16 FRANK_I iggy, I created my roster ssh I was able to ping the computer but when i send the command salt-ssh '*' cmd.run 'ifconfig' I get the error /bin/bash: ifconfig: command not found
16:16 FRANK_I Any idea?
16:16 thawes joined #salt
16:16 Ozack joined #salt
16:17 iggy FRANK_I: ifconfig is either not installed or not in the user's path
16:17 zooz joined #salt
16:18 babilen ifconfig is in /sbin/ifconfig which is normally not in a user's PATH (nor necessarily in SUDO secure_path).
16:18 babilen I personally configure all my accounts to include /sbin and /usr/sbin in my path (regardless of the fact if I "can" call/use those commands)
16:18 FRANK_I babilen got it!
16:18 babilen But then: You shouldn't use ifconfig anyway, but ip!
16:19 debian112 joined #salt
16:20 iggy ^
16:20 iggy I think there are a few distros that don't include ifconfig any longer
16:20 mpanetta None should, iproute2 is much more modern and has much better features.
16:20 FRANK_I i never use  ipconfig
16:21 mpanetta IMO anyway
16:21 intellix joined #salt
16:21 FRANK_I used*
16:22 diegows babilen, iggy I don't undertand why this work in this way.. not only about my issue, in general. Why does not use  a simple from salt import whatever to access to the pillars, function,etc?
16:22 iggy because that would involve (re)reading all the config, etc.
16:23 babilen diegows: I have no idea. I consider monkey patching to be, to put it mildly, rather suboptimal
16:23 iggy it's already in memory, it just needs to be monkey-patched into the module before it runs code
16:24 diegows ok, it's supposed to be faster
16:24 StDiluted joined #salt
16:25 babilen iggy: There is no reason why they have to be monkey patched. You could easily call a function that references datastructures that are populated during runtime. The main problem is the wild "cross-pollination" between pillars, grains, modules, ... that makes it really hard to write one without the other. I would have much preferred a tiered system in which certain functions simply aren't available (with proper exceptions).
16:26 iggy I certainly would not want every module/state/pillar/grain/etc in salt re-reading the config (for reasons other than the crazy ordering issues you'd end up with)
16:26 mpanetta config servers FTW :P
16:27 iggy and when I say (re)reading the config, I mean doing the whole process of config reading, pillar setting, grain setting, etc
16:28 jalbretsen joined #salt
16:29 mortini joined #salt
16:30 perfectsine joined #salt
16:31 babilen iggy: I would be happy to "import salt.modules.pip as pip" and then run "pip.foo('bar', 'baz')" and be greeted with an exception if pip cannot be executed, rather than having to refer to __salt__['pip.foo'] and having __salt__ monkey patched into the module. (and getting a keyerror rather than an explicit error)
16:32 babilen Modules simply should be imported at a time when grains and pillars are "done". It would also have been nice to make the stages during the "bootstrapping" much more explicit...
16:33 babilen I mean the whole "pillars can't reference themselves" (which also makes it impossible to target by pillar in the pillar top.sls).
16:33 iggy I don't think it would work the way you think
16:33 DaveQB joined #salt
16:37 smcquay joined #salt
16:39 babilen Well, I don't say that I have it all figured out (far from it), but monkey patching is tricky and causes all sorts of problems. You might not necessarily want to conflate an interface (i.e. "call function foo") with the data being used to call it. The functions you call could also, easily, be a closure around the data or whatnot.
16:40 anotherZero aiight guys, what's the syntax for checking multiple apache modules at once?  http://docs.saltstack.com/en/latest/ref/states/all/salt.states.apache_module.html
16:40 cnelsonsic joined #salt
16:40 P0bailey joined #salt
16:41 housl joined #salt
16:41 babilen Anyway, it's not going to change so it is moot to discuss this. I would have simply preferred a more defined appraoch to this. (e.g. "you can *only* access grain1, grain2, ... and statically defined pillars at import time (stage "foo" during salt bootstrap)
16:41 P0bailey joined #salt
16:41 babilen )
16:42 cnelsonsic left #salt
16:44 babilen anotherZero: Use a for loop?
16:44 anotherZero bravo.
16:45 anotherZero it's the simple things in life...
16:45 babilen anotherZero: I typically read those from the pillar (module name/package name) and then iterate over the pillar to generate the pkg.install and apache_module.enable states.
16:46 anotherZero perfect.  I'm obviously sleep deprived today.  I'm trying to add things without going to the pillar... :/
16:48 babilen Another approach would be to define the modulename/packagename combo in a map.jinja that you ship with your state (index them by modulename) and then provide the modulename in the pillar. You would them get the exact values for the pkg.installed and apache_module.enable states from the lookup table and read "apache:modules:enabled" and "apache:modules:disabled" from the pillar (or something along those lines).
16:49 babilen I actually like that even more, but had to cut corners earlier ;)
16:54 davet joined #salt
16:55 jonatas_oliveira joined #salt
16:56 iggywig123 joined #salt
16:56 JlRd joined #salt
16:56 perfectsine joined #salt
16:57 otter768 joined #salt
16:57 eXistenZNL joined #salt
16:59 jngd joined #salt
17:00 DaveQB joined #salt
17:07 jonatas_oliveira joined #salt
17:09 shanemhansen left #salt
17:11 spookah joined #salt
17:12 nyx_ joined #salt
17:12 wendall911 joined #salt
17:13 bhosmer joined #salt
17:13 nyx_ joined #salt
17:15 pppd joined #salt
17:15 linjan joined #salt
17:17 zlhgo joined #salt
17:23 glyf joined #salt
17:24 housl_ joined #salt
17:25 glyf joined #salt
17:26 bhosmer joined #salt
17:27 KyleG joined #salt
17:27 Ryan_Lane joined #salt
17:28 Zachary_DuBois joined #salt
17:29 hebz0rl joined #salt
17:30 patarr joined #salt
17:30 patarr joined #salt
17:35 smcquay_ joined #salt
17:35 crane hey ho, anyone a tipp for me to set an env variable? :)
17:36 crane i'm running 2014.7 i think the new fancy environ.set is not build in yet
17:36 hal58th joined #salt
17:37 smcquay__ joined #salt
17:38 theologian joined #salt
17:40 manfred i see environ.setenv and enrivon.setval
17:40 manfred no just environ.set
17:41 manfred but both are in 2014.7 https://github.com/saltstack/salt/blob/2014.7/salt/modules/environ.py
17:42 crane manfred: then something is broken. i just put in the example from http://salt.readthedocs.org/en/latest/ref/states/all/salt.states.environ.html and get an error that State 'environ.set' found in SLS 'env' is unavailable
17:43 manfred oh, that is a state, not a module
17:43 crane *doh*
17:44 manfred and that does only have environ.setenv
17:44 manfred so that example should get fixed
17:45 crane https://www.refheap.com/94946
17:46 crane but i still see no output even after success...?
17:50 giantlock joined #salt
17:50 nitti joined #salt
17:51 desposo joined #salt
17:55 aparsons joined #salt
17:58 monkey66 joined #salt
17:59 kevinforbes joined #salt
17:59 manfred exec $SHELL -l
17:59 manfred and reexec your shell
17:59 manfred it can't inject it into the currently running one, just into your configuratinos
18:00 manfred actually
18:00 manfred it only sets them for the environment of the salt processes
18:00 manfred not your bash environment
18:01 manfred https://github.com/saltstack/salt/blob/2014.7/salt/states/environ.py#L27
18:02 crane ah, okay. but how do i get them into my bash?
18:03 monkey661 joined #salt
18:04 forrest joined #salt
18:14 intellix joined #salt
18:17 murrdoc joined #salt
18:17 hal58th manfred: Probably have to edit some files using salt. Check out system-wide environment variables. https://help.ubuntu.com/community/EnvironmentVariables
18:20 kballou joined #salt
18:22 bivers joined #salt
18:23 garret1987 joined #salt
18:23 Ryan_Lane joined #salt
18:27 shaggy_surfer joined #salt
18:28 manfred hal58th: i know :P tell crane
18:28 hal58th ah, crane do that
18:28 manfred crane:  salt doesn't do that, you will want a file.managed state that drops a file into /etc/profile.d/something.sh
18:28 manfred that exports your stuff in there
18:32 glyf joined #salt
18:32 iggy how can the postgres module _not_ support tablespaces (well, creating/deleting them anyway)
18:32 iggy guess this project just got more complicated
18:37 hal58th1 joined #salt
18:40 bivers joined #salt
18:40 pppd joined #salt
18:41 N-Mi joined #salt
18:43 jab416171 joined #salt
18:44 murrdoc iggy:  what does your .gitignore look like
18:44 murrdoc i got *pyc and *slsc
18:45 iggy I... don't have one
18:46 iggy oh, our pillars have *~ and #*#
18:46 iggy that should probably be other places
18:47 iggy the way we(I) work, we push all our local code to the development salt server (and not the other way around), so compiled stuff never ends up in our tree
18:47 murrdoc :P
18:48 bivers anyone know if there's been any progress with this bug? https://github.com/saltstack/salt-bootstrap/issues/515
18:50 Diaoul joined #salt
18:51 iggy ping the bug, not the irc channel (generally speaking)
18:52 Ryan_Lane joined #salt
18:52 bhosmer joined #salt
18:56 iggywig123 joined #salt
18:57 MTecknology viq: I don't suppose you ever got around to poking at gitlab-ci, have you?
18:58 otter768 joined #salt
18:59 hal58th joined #salt
19:00 hal58th2 joined #salt
19:00 Mso150_a joined #salt
19:01 rap424 joined #salt
19:04 ajolo joined #salt
19:05 ggoZ joined #salt
19:11 nyx_ joined #salt
19:13 perfectsine joined #salt
19:21 CatPlusPlus joined #salt
19:22 CatPlusPlus joined #salt
19:22 pdayton joined #salt
19:24 P0bailey joined #salt
19:25 P0bailey joined #salt
19:27 glyf joined #salt
19:32 noway_ joined #salt
19:32 kawa2014 joined #salt
19:37 bhosmer_ joined #salt
19:39 housl joined #salt
19:45 schlueter joined #salt
19:51 boredatwork joined #salt
19:53 Steve73141 left #salt
20:05 ad4m joined #salt
20:08 rypeck joined #salt
20:09 chitown does anyone know where i put custom external pillars?
20:09 chitown i tried putting it in the "_modules" in the file roots
20:09 chitown but, the loader never saw it
20:10 chitown i tried modifying extension_modules, but then salt didnt load any pillars
20:11 Ryan_Lane joined #salt
20:13 viq joined #salt
20:13 ndrei joined #salt
20:15 schlueter joined #salt
20:16 iggy _modules is definitely not the correct place
20:17 jY chitown: if your extension_modules  is set i put all mine in a pillar dir in there and it works
20:18 giantlock joined #salt
20:21 schlueter1 joined #salt
20:25 hunmonk_1 joined #salt
20:26 ad4m left #salt
20:31 drybjed joined #salt
20:31 jaylark joined #salt
20:31 UForgotten joined #salt
20:31 tazle joined #salt
20:31 eagles0513875 joined #salt
20:31 mirko joined #salt
20:31 chrischris joined #salt
20:31 rockey joined #salt
20:31 wintamute joined #salt
20:31 gfa joined #salt
20:31 MikaT joined #salt
20:31 sirtaj joined #salt
20:31 Ymage joined #salt
20:31 CSa_ joined #salt
20:31 mortis joined #salt
20:31 masterkorp joined #salt
20:31 rnts joined #salt
20:31 dstokes joined #salt
20:31 nkuttler joined #salt
20:31 amatas joined #salt
20:31 tru_tru joined #salt
20:31 AnswerGuy joined #salt
20:31 retr0h joined #salt
20:31 davromaniak joined #salt
20:31 utahcon joined #salt
20:31 lazybear joined #salt
20:31 ShibMcNe joined #salt
20:31 stotch joined #salt
20:31 amontalban joined #salt
20:31 nlb joined #salt
20:31 skullone joined #salt
20:31 marcinkuzminski joined #salt
20:31 crazysim joined #salt
20:31 nyov joined #salt
20:34 oldmantaiter joined #salt
20:42 schlueter joined #salt
20:47 perfectsine joined #salt
20:47 wnkz joined #salt
20:48 chitown jY: yes, but it seems to crash the rest of the pillar data
20:48 chitown i can send teh stacktrace when i get home
20:54 perfectsine joined #salt
20:57 alexhayes I'm working on a formula to install freepbx (see https://github.com/alexhayes/freepbx-formula ) and I'm wondering what the best way to do requisites. At the moment I have freepbx.init, which essentially defines the ordering but obviously not the requirements. Would a good approach be to group the requirements (ie... formulas for git, mysql, apache etc.. ) into it's own state and then require this for all the other states?
20:58 bfoxwell joined #salt
20:59 otter768 joined #salt
20:59 iggy cross formula dependencies are very much a "whatever works best" kind of things right now
20:59 iggy there is some talk about eventually codifying that a little more
21:01 alexhayes Ok, are there any examples of what works best? Am I correct in thinking that 'include' doesn't support require?
21:01 alexhayes Because, that would probably make it very easy
21:01 robot_hands joined #salt
21:01 iggywig123_ joined #salt
21:01 jonatas_oliveira joined #salt
21:01 iggy you can do include at the top of the sls file and then require:  - sls: foo
21:02 iggy in a state
21:02 perfectsine joined #salt
21:02 bigl0af joined #salt
21:02 Mso150 joined #salt
21:02 pmcg_ joined #salt
21:02 bivers joined #salt
21:02 rocket joined #salt
21:03 schlueter joined #salt
21:03 iggy so far, I think most people just note requirements in the README and expect the user to set the top file up properly
21:03 pr_wilson joined #salt
21:03 alexhayes Ah ok
21:04 Auroch joined #salt
21:04 iggy i.e. something that uses mysql shouldn't really rely on the mysql formula specifically
21:04 eliasp joined #salt
21:04 dimeshake I've a state that's failing on one minion for no reason I can seem to figure out... the 'repoquery ...' command it runs to determine if the package is available doesn't return anything, and having trouble debugging this
21:04 dimeshake here's a gist https://gist.github.com/mshade/27de2cca41a421fe0ac0
21:04 sgate1 joined #salt
21:04 dimeshake it runs fine on other centos 6.6 minions, which is strange
21:04 sinh joined #salt
21:05 schlueter1 joined #salt
21:05 bhosmer_ joined #salt
21:05 manfred dimeshake: gist your yum repolist on all those devices?
21:05 dimeshake any ideas on how to debug this? the package name it searches for in the repoquery does indeed show as available with a simple 'yum search collectd'
21:05 kevinforbes joined #salt
21:06 dimeshake same on all the centos 6.6 minions, basic base repos
21:06 Dw_Sn joined #salt
21:06 dimeshake added repolist to that gist
21:07 alexhayes iggy: Yes, good point.
21:07 hal58th2 dimeshake: Have you tried running some of the commands that salt was running manually? "repoquery --plugins --queryformat="%{NAME}_|-%{VERSION}_|-%{RELEASE}_|-%{ARCH}_|-%{REPOID}"   --all --whatprovides collectd"
21:07 manfred dimeshake:  i don't see the repolist?
21:08 eightyeight joined #salt
21:08 dimeshake hal58th2 / manfred: it's in the state-output (ran from a salt-call on the minion). executing it alone does indeed return nothing for some reason
21:08 sudarkoff joined #salt
21:09 manfred at that point, i would just trouble shoot it like normal when you can't find the package you ar elooking for :P
21:09 hal58th2 dimeshake: Well if the command itself returns nothing…Then not sure since it seems to be a yum program.
21:09 dimeshake manfred: yeah but your normal yum search collectd does indeed find it!
21:09 manfred but repoquery doesn't
21:09 dimeshake aye
21:09 jonatas_oliveira joined #salt
21:10 manfred dimeshake: the only times i have seen it are if the repo is using includepkg: or something extra on the configurations
21:10 manfred and the time I saw something similar, was when *mysql* was in /etc/yum.conf excludes
21:10 alexhayes iggy: Perhaps a good idea would be to have a freepbx.requirements state that includes those formulas but modify the way I've structured the pillar data so it doesn't rely on the pillar structure for those formulas (ie.. mysql user/pass)
21:10 manfred dimeshake: can you run a yum clean all, and then try again?
21:10 conan_the_destro joined #salt
21:10 dimeshake did try that
21:11 hebz0rl joined #salt
21:12 hebz0rl hi is there an easy way to sign the hostkey for an sshd server? im using certificate authorities and wanted to automate the signingprocess of new machines
21:12 alexhayes Can you do a pillar.get inside a pillar file?
21:12 iggy alexhayes: what happens if someone wants to use freepbx with postgres? (or whatever, I don't know anything about freepbx)
21:12 manfred alexhayes:  you can' but you really shouldn't
21:13 iggy _sometimes_ you can
21:13 manfred alexhayes:  the pillars are rendered on the master, so it isn't predictable for using it all the time
21:13 manfred they will be rendered with master data
21:15 dimeshake my fault. epel data was being returned from yum search collectd, but not enabled - after the yum clean all, i didn't do another yum search
21:15 alexhayes iggy: Yes, I'm on my way to writing support for that, so you can either use postgres or mysql
21:15 sarlalian joined #salt
21:15 diegows joined #salt
21:16 philipsd6 joined #salt
21:16 shaggy_surfer joined #salt
21:16 repl1can1 joined #salt
21:16 schlueter joined #salt
21:17 CeBe joined #salt
21:17 wm-bot4 joined #salt
21:20 N-Mi joined #salt
21:20 N-Mi joined #salt
21:22 glyf joined #salt
21:24 jswanson_ joined #salt
21:26 jhauser joined #salt
21:33 viq MTecknology: no, maybe I'll have some time in the next couple of days or over christmas
21:35 pdayton joined #salt
21:36 alexhayes manfred/iggy: How would you suggest I do something like this http://pastebin.com/1a1emniy
21:37 mpanetta joined #salt
21:41 iggy alexhayes: why would you want to do that?
21:42 MTecknology viq: I'll send you some homemade rum! :D
21:42 ckao joined #salt
21:42 viq Too bad I don't drink ;)
21:42 MTecknology oh :(
21:43 alexhayes iggy: I need to reference that mysql user/pass in another state and I'd also like to take advantage of the mysql formula
21:43 alexhayes I'd prefer not to define the user/pass twice in pillar
21:44 iggy you can reference the mysql pillar data from your openpbx formula
21:45 alexhayes And just enforce the use of mysql.user.foo?
21:45 iggy so instead of pulling foo:mysql:user in your formula, just pull mysql:users:openpbx
21:46 iggy you could probably even make that variable if you wanted
21:46 alexhayes Ok, cool, that is what I'm already doing so I'll leave it like that
21:46 alexhayes Make what a variable?
21:46 iggy the user
21:46 alexhayes mysql:users:openbox?
21:46 alexhayes ah... thats what I'm trying to do
21:47 viq MTecknology: also depending which contintent your on this could be prohibitively expensive ;)
21:47 iggy yeah, so: set myuser = pillar.get('foo:myuser', 'openpbx')  pillar.get('mysql:user' + myuser)
21:48 iggy something similar anyway
21:49 iggy I would personally just hardcode the user and make sure the user already has that user setup in some way
21:49 MTecknology viq: Do you not live in the states?
21:49 iggy I would personally just hardcode the user and make sure the admin already has that user setup in some way
21:49 iggy less unclear
21:50 viq MTecknology: nope
21:50 alexhayes with regards to 'set myuser = pillar...' is that something that goes in pillar or in my state? I'm assuming state correct?
21:50 MTecknology oh
21:50 viq MTecknology: entirely wrong side of the pond ;)
21:50 iggy alexhayes: state... because the flying spaghetti monster kills kittens when you put pillar lookups in pillars
21:54 rojem joined #salt
21:58 yomilk joined #salt
21:59 alexhayes iggy: that guy, he's everywhere
22:00 bivers joined #salt
22:02 murrdoc joined #salt
22:02 diegows joined #salt
22:03 schlueter joined #salt
22:04 schlueter1 joined #salt
22:05 karimb joined #salt
22:09 loggyer joined #salt
22:10 loggyer Guys i'm trying to create tls certificates, but it fails giving, 'tls.create_self_signed_cert' is not available
22:10 loggyer Couldn't find anything to resolve this..Any suggestions ?
22:11 Ryan_Lane loggyer: you need to have pyOpenSSL installed
22:11 Ryan_Lane https://pypi.python.org/pypi/pyOpenSSL
22:11 Ryan_Lane it's pyopenssl, if you're using Ubuntu
22:12 loggyer Ryan_Lane: I install that as a requires condition in the tls
22:12 murrdoc joined #salt
22:12 loggyer but the pyopenssl fails
22:12 Ryan_Lane installing the package fails?
22:12 loggyer yes, doing a pip install pyopenssl fails on centos6
22:13 Ryan_Lane install the PyOpenSSL package?
22:13 Ryan_Lane you don't have to use pip
22:13 Ryan_Lane err pyOpenSSL package
22:14 Ryan_Lane it may be in epel
22:15 murrdoc joined #salt
22:16 perfectsine joined #salt
22:17 N-Mi joined #salt
22:17 loggyer Ryan_Lane: Yes got that installed. but will check the tls error now
22:19 loggyer Ryan_Lane: That worked! Thanks
22:19 Ryan_Lane yw
22:21 Jahkeup joined #salt
22:22 hal58th joined #salt
22:23 pdayton left #salt
22:26 murrdoc joined #salt
22:27 aquinas joined #salt
22:29 StDiluted joined #salt
22:30 eykd joined #salt
22:32 eykd Hello. I’ve copied the dev hipchat returner from https://github.com/saltstack/salt/blob/develop/salt/returners/hipchat_return.py into `/srv/salt/_returners/hipchat.py` on my master, and confirmed that it’s been synced to my minions. However, when I run `salt myminion test.ping —return hipchat I get this error on the minion: https://dpaste.de/29wL
22:33 theo joined #salt
22:34 eykd Seems like the minion isn’t picking up the custom returner. Any idea what I might be doing wrong?
22:35 nitti joined #salt
22:35 hal58th1 joined #salt
22:35 schristensen joined #salt
22:36 dimeshake eykd: did you do a highstate on the minion after adding the returner?
22:36 dimeshake or saltutil.sync_returners
22:36 eykd Yes, and I confirmed it’s on the minion in /var/cache/salt/minion/files/base/_returners
22:37 murrdoc joined #salt
22:37 eykd I should probably mention I’m on 2014.7.0
22:37 theo__ joined #salt
22:39 perfectsine joined #salt
22:39 dimeshake not sure then, eykd - no expert :)
22:39 jhauser joined #salt
22:39 dimeshake did you just grab the hipchat_return.py from git?
22:39 mosen joined #salt
22:39 iggy thoughts on matching a distro "major" version in a map.jinja/grains.filter_by?
22:40 eykd dimeshake: Thanks anyway. :) Yes, I did. I’ve also got the configuration for it in the pillar, but this looks like it’s not even recognizing the pillar.
22:41 dimeshake is it supposed to look at the pillar? the comments seem to indicate it gets 'em from the minion conf
22:41 DaveQB joined #salt
22:42 iggy eykd: do you have python-requests installed on all the minions? (probably yes as I think it's a requirement for salt-minion on most distros
22:42 Ryan_Lane dimeshake: it should get them from the minion config or the pillars
22:42 Ryan_Lane via the same key
22:42 eykd iggy: Yes, requests 0.14.2
22:43 eykd dimeshake, Ryan_Lane Yes, my mistake, I’ve got a state that pushes the hipchat configuration to /etc/salt/minion.d/ and uses the pillar to get the API key for the conf template.
22:46 schlueter joined #salt
22:46 kermit joined #salt
22:50 UtahDave joined #salt
22:51 schlueter1 joined #salt
22:51 dimeshake i do get the same keyerror
22:51 dimeshake trying it out myself
22:52 eykd dimeshake: Weird. I wonder if maybe there’s an error on import that’s getting suppressed?
22:52 eykd dimeshake: since I’m borrowing the returner from dev.
22:52 alexhayes Is it possible to do a require: - sls: foo.bar?
22:53 alexhayes or just require: - sls: foo
22:53 perfectsine joined #salt
22:53 dimeshake yes
22:54 dimeshake http://docs.saltstack.com/en/latest/ref/states/ordering.html - look at sls: network
22:55 dimeshake eykd: get the same error if  i use a bogus returner name too
22:56 alexhayes dimeshake: yes that's what I was basing my assumptions on, just doesn't seem to be finding my foo.bar, even though I am referencing it in the include (which is working)
22:56 eykd dimeshake: Thanks for looking. I’m starting to suspect a suppressed error of some sort.
22:56 ecdhe Can the windows minion run masterless?
23:01 alexhayes dimeshake: So for instance, say my state depended on the php formula, would i simply do require: - sls: php ?
23:01 Ryan_Lane ecdhe: I'd imagine it should be able to
23:01 Ryan_Lane you'd use salt-call
23:01 Ryan_Lane you don't have to run a daemon if you're doing masterless
23:01 dimeshake alexhayes: that's how i would do it
23:02 dimeshake man i really want to get this hipchat returner working now, too, eykd :DS
23:02 alexhayes dimeshake: and does the include for the php formula need to be in the same sls file as the require?
23:02 eykd dimeshake: I think the problem is in this line: https://github.com/saltstack/salt/blob/develop/salt/returners/hipchat_return.py#L72
23:02 alexhayes dimeshake: I assume not, but it's not working...
23:03 eykd dimeshake: Yeah, that was it. Replacing it w/ `from urlparse import urljoin as _urljoin` gets me to a *new* error. ;)
23:04 alexhayes dimeshake: I'm getting a 'The following requisites were not found require: sls: php', but in a state in a separate .sls I have successfully included php
23:05 murrdoc joined #salt
23:05 dimeshake your pythonfu is stronger than mine :D
23:05 dimeshake the pylint stuff there is a clue for me, but otherwise i'd have no idea how to track down which line it was dumping on
23:05 jhauser joined #salt
23:06 dimeshake i'm not sure why i'd be bailing like that alexhayes  - can you pastebin the state that's not working with the require
23:06 eykd dimeshake: The next problem is with https://github.com/saltstack/salt/blob/develop/salt/returners/hipchat_return.py#L107-L114 which appears to be a newer API that doesn’t exist in 2014.7.0. I’m going to look at this further tomorrow. Thanks for the help. :)
23:06 ecdhe Ryan_Lane, I see salt-call here but not sure where to put some states to test.
23:07 Ryan_Lane the config file should be setup somewhat like a master
23:07 Ryan_Lane you define the location of the file root, the pillar root, etc
23:07 alexbst joined #salt
23:07 dimeshake follow the masterless quickstart and get your file_roots set up
23:09 schlueter joined #salt
23:09 alexhayes dimeshake: So in https://github.com/alexhayes/freepbx-formula/blob/feature/requisites/freepbx/pear.sls I reference https://github.com/alexhayes/freepbx-formula/blob/feature/requisites/freepbx/requirements.sls
23:09 ecdhe Ryan_Lane, dimeshake, thanks for the pointers.  The biggest problem I have with windows is all the manual installation of stuff.  Salt may redeem it!
23:09 Ryan_Lane :D
23:09 Ryan_Lane that's the biggest problem? :)
23:09 ecdhe Ryan_Lane, it's problem enough.
23:09 Ryan_Lane heh
23:10 ecdhe I want to be able to wipe a virus with an unattended install.
23:10 dimeshake alexhayes: ah. so you're require is actually a reference to another sls that does an include on the php state
23:10 dimeshake your*
23:10 alexhayes dimeshake: Even if I change that require from 'freepbx.requirements' to 'php' I still get the same error that it can't find the requirement (note I'm getting this issue when running: state.sls freepbx.pear test=True)
23:10 alexhayes dimeshake: Yes, that's ideally what I'd like to do
23:11 dimeshake that looks to be the way it *should* work to me, alexhayes. does it do the same with a state.highstate test=True?
23:15 bivers joined #salt
23:15 alexhayes dimeshake: yeh, same thing
23:16 dimeshake not sure where to go from there. i've not had that issue but i'm not using it quite like that
23:16 genediazjr joined #salt
23:17 nitti joined #salt
23:20 bivers can you execute other runners via the Orchestrate runner?
23:21 alexhayes dimeshake: Yes, even if I move the 'install-pear' state into the requirements.sls (that contains the include for php) and change the require to 'sls: php' I still get the same issue
23:22 alexhayes dimeshake: ie.. http://pastebin.com/rZpeQNeR
23:23 alexhayes dimeshake: seems to directly contradict what the docs say, which I can't imagine to be the case...
23:27 alexhayes dimeshake: Interestingly require: - pkg: php works as expected
23:28 iggy bivers: probably not
23:29 alexhayes Does anyone else know why http://pastebin.com/rZpeQNeR would be saying: The following requisites were not found: require: sls: php
23:30 loggyer Ryan_Lane: I'm trying to copy files (basically certs) from the master to the minion..but does really get copied.
23:31 loggyer This is what i'm trying, salt rsm1\* cp.get_file salt://etc/pki/tls/certs/test.crt /home/cloud-user
23:31 murrdoc joined #salt
23:32 Ryan_Lane I don't use a master :)
23:32 loggyer Aah :(
23:33 thawes joined #salt
23:33 iggy alexhayes: without more context, no try posting the command you're running, all the states involved, the full error message, etc.
23:33 alexhayes I'm just doing: salt foo state.highstate test=True
23:34 pppd_ joined #salt
23:34 alexhayes It returns this: http://pastebin.com/2v6tHe57
23:34 glyf joined #salt
23:34 iggy if the require'd state would produce changes, using test=True will fail that way
23:35 alexhayes It's just the require isn't behaving the same as if I change the require from sls to pkg.
23:35 iggy you might also consider using something other than pastebin.com... I know there are some people that won't even open the url if it's pastebin.com
23:36 alexhayes oh ok... apologies, any suggestions?
23:36 iggy refheap, gist, most distros have some paste site, etc.
23:37 mitsuhiko_ joined #salt
23:37 iggy try pasting the sls files as well
23:38 iggy and like I said, if using test=True and the require'd state says it would change things, the require'ing state will fail
23:42 foulou joined #salt
23:43 dynamicudpate joined #salt
23:43 alexbst_ joined #salt
23:43 chayak joined #salt
23:43 loggyer anybody could help me with the cp.get_file to get file from master, log output says, https://gist.github.com/Akshaykapoor/f88b1193e2563309e833
23:45 chayak joined #salt
23:45 UtahDave joined #salt
23:45 chayak left #salt
23:45 __number5__ loggyer: run `salt-call -l debug cp.get_file ...` on your minion and see what's in the log
23:45 cb joined #salt
23:46 mschiff joined #salt
23:46 mschiff joined #salt
23:48 iggy loggyer: try cp.list_master to make sure the master and your expectations match
23:48 Jarus joined #salt
23:49 iggy any grains.filter_by experts about? https://github.com/saltstack-formulas/postgres-formula/blob/master/postgres/map.jinja if you take the service out of the lower (14.04) dict wouldn't you expect that to be inherited from the Debian entry in the top dict?
23:50 loggyer iggy: cp.list_master list all the files on the master under /srv/salt...m trying to copy the certificate from /etc/pki/tls/certs/
23:50 iggy loggyer: oh, you can't
23:50 loggyer __number5__: This is what i get, https://gist.github.com/Akshaykapoor/50ba8dbadb8094b34f49
23:50 loggyer iggy: is there a way to achieve this or i just cant :/
23:51 iggy you can't copy random files from your master's root filesystem
23:51 iggy that would be a _huge_ security risk
23:51 alexhayes Ok, can anyone help me out with this - https://gist.github.com/alexhayes/bbffb9f435984fba11a0
23:51 __number5__ loggyer: ^^
23:51 DenkBrettl joined #salt
23:52 iggy loggyer: your best bet would be to have the files managed on the master (running a minion) and the other minions to keep things in sync
23:52 hunmonk_1 install_centos_git_post() function is broken at http://bootstrap.saltstack.com — for older versions of RHEL/CentOS, it incorrectly tries to install systemd stuff, which borks the rest of the install
23:52 __number5__ alexhayes: where is your php.sls?
23:52 iggy alexhayes: you still didn't paste your other state files (mostly the php one)
23:52 iggy hunmonk_1: there's a bug open about it
23:52 alexhayes The php one is just the php formula
23:53 alexhayes As in, php.init
23:53 loggyer iggy: I have these files generated with salt-state and i want minions to get those files from the master
23:54 iggy so generate them under your salt file_roots and deploy from there?
23:54 loggyer iggy: if you could point me to something so that i can implement it the way you told me, keep things in sync ?
23:54 speed145a joined #salt
23:55 __number5__ loggyer: a common practice to distribute tls/ssl certs is to store them in pillar
23:55 iggy ^
23:56 StDiluted joined #salt
23:56 loggyer __number5__: iggy: aahan. ok let me try that approach
23:56 paha joined #salt
23:56 iggy that kind of kills the idea of generating them from a state, but is way more secure
23:56 hunmonk_1 iggy: is it opened as a github issue? having some trouble finding it
23:56 iggy hunmonk_1: yes
23:58 alexhayes I've included a link to the php-formula/php/init.sls for clarity
23:58 iggy hunmonk_1: are you looking at bugs in teh salt-bootstrap repo?
23:58 iggy https://github.com/saltstack/salt-bootstrap/issues/515
23:58 jpaetzel joined #salt
23:58 hunmonk_1 iggy: well, then, that would explain it :) thx
23:59 jab416171 joined #salt
23:59 flebel joined #salt
23:59 iggy hunmonk_1: for now, I'd just pull an older version of the script
23:59 iggy i.e. https://github.com/saltstack/salt-bootstrap/raw/stable/bootstrap-salt.sh
23:59 hunmonk_1 iggy: the vagrant salt plugin kind of does that for me, not wanting to hack that

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