Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2015-04-08

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

All times shown according to UTC.

Time Nick Message
00:00 KyleG I don't want to restart all my salt minions every single time I do a deploy….there must be some way to clear out the caches. I even tried removing the .pyc files
00:00 murrdoc joined #salt
00:01 nafg_ joined #salt
00:06 __ale__ joined #salt
00:08 baweaver joined #salt
00:09 nafg joined #salt
00:09 jerematic joined #salt
00:10 sunkist joined #salt
00:19 _ale1_ joined #salt
00:20 XenophonF KyleG: what does the state look like?
00:20 XenophonF if you use pip.installed or pkg.installed, it should automatically refresh the minion
00:20 XenophonF if not, there's a way to force a refresh in the state - give me a sec while i stfw
00:21 KyleG yeah the state is pip.installed upgrade=true
00:21 XenophonF http://docs.saltstack.com/en/latest/ref/states/index.html#reloading-modules
00:22 XenophonF it shouldn't have to do it with a pip.installed state, though
00:22 KyleG I'm on v.2014.1.13
00:22 KyleG if that makes a difference
00:23 KyleG - reload_modules: true
00:23 KyleG That could be worth a shot
00:23 XenophonF yeah, set that
00:23 KyleG Okay, let me test that out
00:23 XenophonF i just rtfs and the pip.installed function doesn't include a call to reload_modules at the end like the different pkg states do
00:23 XenophonF dunno why that is
00:24 KyleG :(
00:24 XenophonF you can use the module state to call reload_modules, too, if you'd rather do a bunch of pip.installed states and only reload once
00:24 KyleG nah I'll want to reload modules every time it's run
00:24 KyleG let's see here..
00:25 KyleG reload_modules: true
00:25 KyleG made no difference. :(
00:25 KyleG Returns an output that confirms it's not reading the latest python code for the module I installed via pip
00:28 KyleG Just to clarify, the module being cached is not the salt module itself, but a system level package I'm installing via PIP and locates itself at /usr/local/lib/python2.7/site-packages/ appropriately
00:29 XenophonF no i mean if you have two pip.installed states and both have reload_modules set, the minion will reload twice when you only need it to reload once
00:29 KyleG ah gotcha.
00:29 KyleG either way no dice :\
00:29 XenophonF FreeBSD?
00:29 KyleG yes
00:29 XenophonF yay!
00:29 XenophonF i'm not the only one!
00:29 KyleG haha :D
00:29 XenophonF why aren't you running 2014.7.2?
00:30 KyleG Just haven't updated yet, we stopped aggressively updating salt after a few heavy hitting bugs severely impacted our ability to work
00:30 XenophonF but regardless, i figured you weren't talking about salt itself
00:30 MTecknology I tried out 2014.7.2 today and had pretty wicked issues
00:30 KyleG There was a bug a year or two ago, where someone messed with the chmod calculations, and it chmod'd all my stuff to 420
00:30 KyleG lol
00:30 XenophonF ouch
00:31 XenophonF also hilarious
00:31 KyleG yeah, I don't even use high states anymore
00:31 KyleG state.sls all day
00:31 MTecknology 2014.7.0 comes with some insane changes, there are more insane changes coming
00:31 XenophonF i can't wait for 2015.2
00:31 KyleG I'm curious as to why Salt would cache this module outside of itself anyways
00:31 MTecknology That one will be crazy as well
00:31 XenophonF there's stuff in there i want to use so bad
00:31 XenophonF anyway
00:31 XenophonF back to your problem
00:31 KyleG :}
00:31 XenophonF all the reload_modules does is get the minion to reload the runtime environment
00:32 XenophonF so e.g., if you install boto via pip, you need to reload the minion so it knows about the boto library
00:32 KyleG like, service salt_minion reload type of reload?
00:33 KyleG I need to do this in an automated fashion, I somehow dobut that my minions would be able to reload themselves and return that they did so
00:33 XenophonF reload_modules doesn't restart the minion
00:33 KyleG rather, it returns False lol
00:33 KyleG Oh I know that.
00:33 XenophonF ah ok
00:33 KyleG But reload_modules does not seem to be solving the problem :\
00:35 ChesFTC Hi, I'm wondering if there's any way to set ec2 tags on instance creation rather than afterwards?
00:36 ChesFTC in my /etc/salt/cloud.profiles.d, can I have a stanza that does something like set_tags: { Billing: Foo, Roles: Bar,Baz}
00:36 XenophonF KyleG: take a look at http://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.saltutil.html and http://docs.saltstack.com/en/latest/ref/states/all/salt.states.module.html
00:36 KyleG yeah I've been all over them :\
00:36 murrdoc have u KyleG
00:36 murrdoc have u  /
00:36 murrdoc :D
00:36 KyleG yes
00:36 XenophonF heh
00:37 murrdoc so whats the issue
00:37 KyleG unfortunately I don't have clear_cache in my current version, so no idea if that would be the problem solver
00:37 murrdoc sync_all might help
00:37 murrdoc if u have that
00:37 KyleG murrdoc: salt is caching a pip package even if I update it. Installing update with pip.installed
00:37 KyleG yes, I tried all the sync_*'s
00:38 KyleG I even tried sys.reload_modules
00:38 mcm0245 joined #salt
00:38 murrdoc pip freeze shows the right version ?
00:38 murrdoc and u have restarted the minion
00:38 KyleG yup
00:38 KyleG nope. That's what I'm trying NOT to do.
00:38 murrdoc ah
00:38 MTecknology 19:31 < XenophonF> there's stuff in there i want to use so bad
00:38 MTecknology XenophonF: badly*
00:39 murrdoc i mean its a python daemon
00:39 KyleG I want it to release whatever cache it's decided to have of a package not even part of salt itself or the states
00:39 mcm0245 Hey all. I’m attempting to call a saltrunner from the API, but am having problems with syntax. Does anyone have a working sample json?
00:39 murrdoc if it needs new libs i would assume it needs to be restarted to pick up new files from pip
00:39 murrdoc in the python path
00:39 murrdoc i would assume
00:39 mcm0245 I’ve got local and local_batch going, but the arguments are a bit strange for runner
00:39 XenophonF MTecknology: ack
00:40 KyleG I guess I thought there would be some method to tell salt "Hey…release that cache and go reload yourself and let me know when you're done"
00:40 MTecknology ;)
00:40 KyleG and being able to return still
00:41 dude051 joined #salt
00:41 catpig joined #salt
00:42 mcm0245 Specifically I’m attempting to call runner through the api to run state.orchestrate orchestration.script pillar=‘{“target”:target.node}’
00:44 mcm0245 but according to the docs, runner wraps runnerclient.cmd_sync. Which would make fun=state.orchestrate. But I don’t know what to do with my other arguments
00:45 jerematic joined #salt
00:46 KyleG apparently this has happened for a long time now, and was only brought to my attention because our release engineer is quitting so we're going over everything lol
00:46 KyleG -_-
00:46 KyleG "I was just restarting the minions every time"
00:46 KyleG Developers...
00:47 dude051 joined #salt
00:47 XenophonF heh
00:53 ITChap joined #salt
00:53 jeremati_ joined #salt
00:54 opnsrc @XenophonF: yes, desktop-some.conf lives in /srv/salt/dir/to on the master
00:55 g3cko joined #salt
00:56 XenophonF opnsrc: are you doing /path/to/file:\n  file.managed:\n    - source: salt://dir/to/file
00:56 XenophonF (sorry for the formatting don't have editor/bpaste handy)
00:57 MatthewsFace joined #salt
00:57 renoirb_ joined #salt
00:57 opnsrc yes
00:58 XenophonF that should work
00:58 XenophonF can you pastebin the state?
00:58 XenophonF er, not pastebin
00:58 opnsrc it does, but randomly it will fail, it's odd
00:58 XenophonF use bpaste.net or something
00:59 XenophonF what kind of error do you get?
01:00 opnsrc -managed': {'comment': 'Source file salt://salt/dir/to/desktop-some.conf not found',
01:03 desposo joined #salt
01:04 fusionx86 joined #salt
01:04 Tyrm joined #salt
01:05 pdayton joined #salt
01:07 jerematic joined #salt
01:08 XenophonF so /srv/salt is the fileserver root directory, right?
01:08 XenophonF does it have a subdirectory named "salt"?
01:08 murrdoc are u in windows ?
01:08 murrdoc do u have POWERshell
01:08 murrdoc (tm)
01:08 XenophonF murrdoc: I do!
01:09 murrdoc is it fun
01:09 XenophonF opnsrc: salt://salt/dir/to/desktop.conf expands to /srv/salt/salt/dir/to/desktop.conf
01:09 murrdoc i havent used it
01:09 XenophonF murrdoc: i'm using it more and more
01:09 XenophonF that it ties so closely into .NET and the Windows API is a real boon
01:10 jeremati_ joined #salt
01:11 opnsrc yes, it has a sub directory 'salt', /srv/salt/salt/
01:11 XenophonF opnsrc: any errors on the master?
01:11 XenophonF i'm kind of shooting in the dark at this point, opnsrc. i'm not entirely sure what your problem could be.
01:12 XenophonF murrdoc: before I found the Lord, er, I mean before I started using Salt, I was getting ready to automate my Windows server builds using PowerShell
01:12 opnsrc 'dir' is a soft link to another location
01:12 CeBe1 joined #salt
01:12 murrdoc XenophonF:  oh you poor child
01:13 opnsrc no, none other than the 'not found' error
01:13 murrdoc :D
01:13 Furao joined #salt
01:13 murrdoc XenophonF:  u havent made any windows specific formulas yet
01:13 murrdoc slacking
01:13 XenophonF oh I have!
01:13 XenophonF just did one for WINS recently
01:13 XenophonF had to shell out to run netsh for that one though
01:14 XenophonF and DCPROMO
01:14 XenophonF for a replica DC though
01:14 XenophonF next on my list is a formula for a new forest root PDC
01:14 XenophonF those are easy though - dump an INF file on the target and run DCPROMO
01:15 VSpike joined #salt
01:15 murrdoc O-O
01:15 murrdoc links
01:15 XenophonF and tonight I'm writing a formula for SharePoint 2013 SP1
01:15 XenophonF again, cheating a little by using https://autospinstaller.codeplex.com/
01:17 SheetiS joined #salt
01:17 XenophonF most of our infrastructure is Windows
01:17 murrdoc this is all good
01:17 XenophonF although now that we've got some funding for hpc and bioinformatics stuff, we're starting to deploy linux stuff
01:18 murrdoc damn
01:18 murrdoc u are too happy at your job
01:18 XenophonF we just installed one of these babies in west africa - http://bioteam.net/slipstream/
01:18 murrdoc :)
01:18 murrdoc nice
01:18 murrdoc do have any windows specific formulas in git
01:19 murrdoc i wants to read for nerdy reasons
01:19 jerematic joined #salt
01:19 murrdoc what company do you work for
01:19 XenophonF well our salt states repo is private, but let me see what i can copy over to my personal repo
01:19 XenophonF i'll git (hah keeping that typo) the link to you tomorrow
01:20 murrdoc :D
01:20 murrdoc later
01:20 murrdoc hook it up
01:20 XenophonF i work for www.nete.com, on a contract to niaid
01:20 murrdoc i am now pushing any formula we use internal back up
01:20 fusionx86 joined #salt
01:21 Xevian joined #salt
01:22 baweaver joined #salt
01:24 jeremati_ joined #salt
01:26 The-spiki joined #salt
01:31 CeBe joined #salt
01:33 zwi joined #salt
01:38 fusionx86 joined #salt
01:44 tmh1999 joined #salt
01:46 pass_by_value joined #salt
01:46 dendazen joined #salt
01:49 ITChap joined #salt
01:53 SeeDickCode joined #salt
01:54 otter768 joined #salt
02:01 XenophonF gotta get kids to bed - ttfn
02:03 munhitsu joined #salt
02:03 madduck_ joined #salt
02:04 manytrees joined #salt
02:06 bfoxwell joined #salt
02:06 cws joined #salt
02:06 Vye joined #salt
02:07 [vaelen] joined #salt
02:07 JPaul joined #salt
02:07 pipeep joined #salt
02:08 randomuser joined #salt
02:09 JoeHazzers joined #salt
02:14 analogbyte joined #salt
02:14 sysrich joined #salt
02:16 JordanTesting joined #salt
02:16 izibi joined #salt
02:17 ssm joined #salt
02:18 evle joined #salt
02:21 chandankumar joined #salt
02:24 jessie_ joined #salt
02:24 jessie_ morning all !
02:24 fusionx86 joined #salt
02:25 nafg joined #salt
02:26 fusionx86 joined #salt
02:28 jalaziz joined #salt
02:35 donmichelangelo joined #salt
02:36 MatthewsFace joined #salt
02:39 favadi joined #salt
02:44 _JZ_ joined #salt
02:45 fusionx8_ joined #salt
02:45 scbunn joined #salt
02:46 fusionx__ joined #salt
02:46 raygunsix joined #salt
02:47 _1_Ken joined #salt
02:47 pdayton joined #salt
02:48 otter768 joined #salt
02:58 loz-- joined #salt
02:58 echo joined #salt
02:59 dalexander joined #salt
03:12 yuhl_work_ joined #salt
03:13 zwi joined #salt
03:13 Tyrm joined #salt
03:14 joehughes joined #salt
03:16 joehughes I am having trouble defining mine_functions with args in my minion config. Specifically passing connection_args to mysql.get_master_status is not working
03:17 joehughes when I use mine.send the call works without any issues
03:17 joehughes it seems like maybe I am not defining the connection_user and connection_pass correctly
03:21 pdayton joined #salt
03:21 joehughes https://gist.github.com/anonymous/434a9bdc9e01169214db here is an example of what I am trying to do... based off what I saw in the docs at http://docs.saltstack.com/en/latest/topics/mine/
03:26 lnr joined #salt
03:28 viq joined #salt
03:33 subsignal joined #salt
03:37 stanchan joined #salt
03:46 lnr joined #salt
03:56 babilen joined #salt
04:05 lnr joined #salt
04:15 Tyrm joined #salt
04:18 rdas joined #salt
04:22 I3olle joined #salt
04:23 bhosmer joined #salt
04:26 nafg_ joined #salt
04:27 swazman joined #salt
04:31 SpX joined #salt
04:32 swazman joined #salt
04:33 _JZ_ joined #salt
04:37 rdas joined #salt
04:38 MatthewsFace joined #salt
04:39 chandankumar joined #salt
04:41 Matthews_ joined #salt
04:55 mpanetta joined #salt
04:57 catpig joined #salt
05:04 lnr joined #salt
05:09 hasues joined #salt
05:09 hasues left #salt
05:21 lnr joined #salt
05:26 randomuser joined #salt
05:28 lnr joined #salt
05:28 lnr joined #salt
05:29 lnr joined #salt
05:35 ja-s joined #salt
05:35 writtenoff joined #salt
05:35 stoogenmeyer joined #salt
05:37 yuhl_work_ joined #salt
05:40 otter768 joined #salt
05:43 lnr joined #salt
05:50 randomuser joined #salt
05:50 catpigger joined #salt
05:55 colttt joined #salt
05:59 AndreasLutro joined #salt
06:02 jerematic joined #salt
06:03 ashmckenzie joined #salt
06:05 ashmckenzie Hi all, is there known issue as far as anyone is aware with v2015.2.0rc2 as far as deleting a schedule returning success (True), but is still present?
06:08 hojgaard joined #salt
06:10 ITChap joined #salt
06:12 _ale1_ joined #salt
06:22 hasues joined #salt
06:22 hasues left #salt
06:24 __ale__ joined #salt
06:27 rsanchez joined #salt
06:32 flyboy joined #salt
06:33 KermitTheFragger joined #salt
06:35 Furao joined #salt
06:35 _ale1_ joined #salt
06:35 kawa2014 joined #salt
06:41 ITChap joined #salt
06:45 __ale__ joined #salt
06:51 _ale1_ joined #salt
06:54 __ale__ joined #salt
06:56 iwishiwerearobot joined #salt
06:56 mpanetta joined #salt
07:01 hebz0rl joined #salt
07:09 KermitTheFragger joined #salt
07:11 Romlok joined #salt
07:12 eseyman joined #salt
07:13 illern joined #salt
07:17 lietu joined #salt
07:21 ajw0100 joined #salt
07:30 jhauser joined #salt
07:38 slav0nic joined #salt
07:38 Andre-B joined #salt
07:40 otter768 joined #salt
07:44 Vynce joined #salt
07:45 I3olle joined #salt
07:45 Vynce let's say a minion named bob tries to connect to the master, but before master accepts his key, i decide he's a pod person and shoot him in the head.  then i hire a new bob, and then salt-key -A … did i get bob1's key instead of bob2's?
07:48 fredvd joined #salt
07:49 ashmckenzie @Vynce: bob2 would have a new key so I would expect it to replace the previous bob..  easy enough to test!
07:51 ndrei joined #salt
07:52 Vynce bob1's was never accepted until after bob2 came online; however, when i then did accept it, i seem to have gotten bob1's.  anybody have an opinion on whether that's a bug?
07:52 Tyrm joined #salt
07:53 greggynapalm joined #salt
07:55 fredvd joined #salt
07:56 ITChap joined #salt
07:56 lempa joined #salt
07:59 marnom joined #salt
08:01 SpX joined #salt
08:04 ITChap joined #salt
08:07 lb1a joined #salt
08:09 kevc what's preferred for managing an application deployed on a server running salt?
08:10 kevc we were using git-deploy to do heroku style deployment of a python based application, but it's quite limited
08:13 dkrae joined #salt
08:14 bluenemo joined #salt
08:15 Vynce kevc: i suspect there are about 626 different answers to that question in this channel, if you could get everybody to unidle long enough
08:15 AndreasLutro I set up a systemd service for it
08:16 GrueMaster joined #salt
08:17 antonw joined #salt
08:17 imanc_ joined #salt
08:17 abele joined #salt
08:18 Xevian joined #salt
08:20 garthk Vynce: +1
08:21 Vynce1 joined #salt
08:21 andrej joined #salt
08:22 garthk kevc: main trick is: the code in the repo or the binaries in the build artifact change on one axis, and the config data coming from Salt needs to change vs both server-specific info (e.g. pillar.which_db_server_to_use) AND changes to the config dictated by the changing software version
08:22 XenophonF Vynce: i kind of wish the master-minion key agreement went both directions
08:23 XenophonF from a minion's perspective, how do you know you're talking to the right master?
08:23 garthk kevc: so, you need to comprimise: do you make it easier to deploy config changes along with code, which then makes it harder to control server-specific info in the config? Or, do you make it easier to reconfigure, which then makes it harder to deploy format upgrades along with the code?
08:23 glyf joined #salt
08:24 ktosiek joined #salt
08:25 ckao joined #salt
08:25 garthk kevc: oh, hang on: maybe you can use a deployed-with-the-code template as the source… that'd help. Not sure how to pull it off. -source {{ some function call to run a salt command to pull a file from the minion? }}
08:27 babilen XenophonF: You have the master key on the minion
08:28 Vynce1 XenophonF: well, it does prevent a maste rin the middle, right?  you get a copy of the master's key — as long as that first master is correct, everything's golden years, right?
08:28 XenophonF i guess
08:28 babilen Vynce1: That holds true for every key based authentication
08:28 XenophonF i worry about that initial key agreement
08:28 XenophonF perhaps unreasonably
08:29 Romlok is there a way to manually set/override the master key on a minion?
08:30 Romlok I guess that'd be the only way to be sure
08:30 Vynce1 babilen: yes, point being that there is one.
08:31 babilen Vynce: What do you mean? There are two keys (minion and master)
08:31 Vynce XenophonF: paranoia is a survival skill for soldiers and sysadmins
08:32 Vynce babilen: yes. well… four keys, right?  private and public for each of those. not sure what your point is there.
08:32 babilen XenophonF: Feel free to check fingerprints
08:33 babilen Vynce: My point was "You are in trouble if somebody gets hold of your private keys" and argued that that holds true for every key based authentication. Salt is not special in that.
08:33 harkx joined #salt
08:34 marnom XenophonF, I found something about man in the middle attacks here: http://docs.saltstack.com/en/latest/topics/releases/0.10.3.html - the 'Master Finger Option' - perhaps it's of use to you
08:35 marnom XenophonF, We deploy using Vagrant/Salt and we pre generate the minion keys on the master and distribute it over SSH, including (if I recall correctly) the ID of the master so this should be pretty secure. The minions won't blindly connect to another master..
08:37 XenophonF hey thanks for that marnom!
08:37 XenophonF and babilen you are absolutely correct
08:38 subsignal joined #salt
08:42 Vynce babilen: yeah, no disagreement. it seemed like you were contesting a point of mine that was just "at least you do need a key to be a master" which it seemed like XenophonF was discounting; but it became clear that was intended to be concern over the initial key exchange, which the master gets a chance to reject but the minion doesn't (or at least, not as easily)
08:43 babilen yes, which is where fingerprint checking comes in (much like it does during initial connects via ssh for example)
08:43 babilen You simply need some pre-shared data to verify against.
08:44 XenophonF i definitely do fingerprint checking, but isn't that only for the minion key (salt-call key-finger --local on the minion and salt-key -f on the master)?
08:44 XenophonF i need to dig into the docs to see how to check the master key
08:44 XenophonF or i could compare it manually as babilen suggested
08:44 babilen marnom digged out the particular configuration for minions
08:44 babilen dug
08:45 marnom XenophonF, yeah I just added that option in our dev environment and it seems to verify the key correctly. Checked by changing fingerprint and minion then refuses to talk to the master
08:45 XenophonF i'm definitely going to add that into my minion configs
08:46 marnom yeah seems like a nice protection during the initial key exchange
08:46 XenophonF and then for the initial key exchange just manually check the master key
08:46 XenophonF i don't have salt-ssh or salt-cloud working quite yet, so i'm still doing the minion installs by hand
08:47 ekle joined #salt
08:47 XenophonF and we have several cases of physical servers located in remote locations that need to be built by hand
08:47 XenophonF speaking of salt-cloud, anyone here using 2014.7.2 with AWS EC2?  i'm having all kinds of problems deploying CentOS 7 and Windows Server 2012 R2 instances
08:48 ndrei joined #salt
08:48 XenophonF i think for the windows installs, i'm just going to have to wait until 2015.2 comes out, because that includes the userdata_file cloud profile option that will let me send a powershell script to the windows VM to open SMB in the windows firewall
08:49 XenophonF but for the linux installs, it's very strange - it will push keying material to the new CentOS instance, and then it immediately deletes it
08:49 XenophonF it doesn't install the minion
08:49 XenophonF i don't get it
08:50 marnom XenophonF, no experience with AWS2, though we do use salt-cloud with Digital Ocean and Proxmox which works great
08:50 marnom also on 2014.7.2
08:50 babilen Well, I am desperately waiting for *any* new release. https://github.com/saltstack/salt/issues/21480 broke my infrastructure and I can never be sure if something made it to a minion or not
08:52 babilen Quite surprised that not everybody ran into that, but maybe you need a relatively high number of minions to notice it
08:53 blast_hardcheese joined #salt
08:53 faust joined #salt
08:54 faust left #salt
08:55 XenophonF that's a nasty looking bug
08:56 bhosmer_ joined #salt
08:57 mpanetta joined #salt
08:58 babilen I hate it with a passion.
08:59 SteamWells joined #salt
09:00 ITChap joined #salt
09:01 wnkz joined #salt
09:04 packeteer joined #salt
09:04 __ale__ left #salt
09:05 swazman joined #salt
09:10 ITChap joined #salt
09:12 MatthewsFace joined #salt
09:18 kossy joined #salt
09:20 blast_hardcheese joined #salt
09:20 kevc garthk_away: it's a django application. The config for the application servers is in the main git repo, but you need to set ENV vars to use it.
09:21 kevc It's wrapped behind wsgi so actually that's all mostly taken care of. Big issue remaining is how to handle doing database migrations, etc
09:21 kevc seems that salt has 'djangomod' which looks helpful, but seems it's either lacking in features or documentation
09:22 fredvd joined #salt
09:22 Furao kevc: some inspiration for you https://github.com/bclermont/states/blob/master/states/sentry/init.sls
09:23 Furao but this is quite old
09:24 kevc looks helpful, thanks!
09:24 N-Mi_ joined #salt
09:24 N-Mi_ joined #salt
09:26 echo joined #salt
09:29 johnchalekson joined #salt
09:39 johnchalekson joined #salt
09:39 lolol joined #salt
09:40 johnchalekson joined #salt
09:41 johnchalekson joined #salt
09:41 otter768 joined #salt
09:41 johnchalekson joined #salt
09:43 johnchalekson joined #salt
09:44 johnchalekson joined #salt
09:45 johnchalekson joined #salt
09:45 johnchalekson joined #salt
09:46 johnchalekson joined #salt
09:46 johnchalekson joined #salt
09:52 kawa2014 joined #salt
09:53 ltsampros Hey guys
09:53 ltsampros I'm trying out some ext_pillar configuration using mysql
09:53 ltsampros but when I call salt '*' pillar.items I cannot see anything executed in the salt-master log files
09:54 ltsampros configuratoin looks like this: http://pastebin.com/bu9uLWCC
09:55 Sjigke joined #salt
09:55 kawa2014 joined #salt
09:57 denys joined #salt
09:57 gijs_ joined #salt
09:57 keltim joined #salt
10:03 CaptainMagnus joined #salt
10:03 XenophonF time to make the donuts - bbl
10:04 dvestal joined #salt
10:05 chutzpah joined #salt
10:08 supermike_ joined #salt
10:08 bhosmer__ joined #salt
10:22 ange salt-cloud question for a change : is there an option to pass to define the root volume type and size for an EC2 instance?
10:27 dRiN joined #salt
10:32 giantlock joined #salt
10:34 Grokzen joined #salt
10:39 lauri joined #salt
10:39 lauri Hi, could anyone elaborate about Salt communication channels? Is it using RabbiqMQ?
10:42 Tyrm joined #salt
10:43 babilen lauri: 0MQ or RAET
10:46 lauri babilen: Where is this documented?
10:47 lauri nvm, found something already
10:47 lauri so they're moving from 0mq to RAET?
10:47 aquassaut joined #salt
10:48 lauri what about Salt security model?
10:52 evle joined #salt
10:57 johnchalekson joined #salt
10:58 johnchalekson joined #salt
11:04 johnchalekson joined #salt
11:08 johnchalekson joined #salt
11:08 andreyst joined #salt
11:09 johnchalekson joined #salt
11:10 johnchalekson joined #salt
11:11 johnchalekson joined #salt
11:14 johnchalekson joined #salt
11:14 johnchalekson joined #salt
11:16 johnchalekson joined #salt
11:18 Romlok anyone know if it's possible to set the hwclock to UTC on a Windows minion?
11:18 Romlok I haven't been able to, but that might be because I'm using a VM
11:19 johnchalekson joined #salt
11:20 gijs_ joined #salt
11:21 johnchalekson joined #salt
11:22 johnchalekson joined #salt
11:27 johnchalekson joined #salt
11:28 simmel lauri: "no promises are yet made as to its reliability or security." ? ; )
11:31 johnchalekson joined #salt
11:35 johnchalekson joined #salt
11:36 johnchalekson joined #salt
11:37 johnchalekson joined #salt
11:38 johnchalekson joined #salt
11:39 ndrei joined #salt
11:40 subsignal joined #salt
11:42 johnchalekson joined #salt
11:42 otter768 joined #salt
11:43 johnchalekson joined #salt
11:46 johnchalekson joined #salt
11:48 malinoff joined #salt
11:49 dvestal left #salt
11:49 CeBe joined #salt
11:49 Grokzen joined #salt
11:54 amcorreia joined #salt
11:57 Romlok agh, why does salt have such a terrible interface for accessing pillars and grains in templates
11:58 Romlok I mean, `salt['pillar.get']('foo:bar:baz', 'qux')` is the new, better, way to do it? Really?
11:59 ltsampro` joined #salt
12:01 jerematic joined #salt
12:02 babilen Romlok: If you have something better we'll be happy to do it. But then you shouldn't have direct references to your pillar in most places anyway, but handle all of that in map.jinja (see salt, collect, influxdb, ... formulas)
12:02 jessie_ joined #salt
12:03 jessie_ anyone assist around gitfs
12:04 jessie_ i basically believe i haven't set things up right because i can't find the branch that its suppose to be pulling from and don't know the directory in which its suppose to be in
12:04 Romlok babilen: the docs I've read say nothing about any map.jinja :/
12:05 babilen Romlok: It would be discussed in the best practices and formula documentation, but I'm not sure if the style that some people agreed upon after this years saltconf is documented in detail. That's why I suggested taking a look at those formulas.
12:08 ekle hi, how do i share pillars which are in a git in all environments ?
12:09 ekle with this i cant access the pillars within an environment git: master ssh://git@abc/devops/saltstack.git root=pillar
12:13 andreyst joined #salt
12:16 pf_moore joined #salt
12:18 Romlok sigh. And I thought salt config looked so clear and simple
12:18 Tyrm joined #salt
12:19 Romlok but as soon as you delve into anything more than a homogeneous environment, it seems to become a sea of hacks, obscure syntax, and mystical rites :(
12:20 Romlok yet from things I've read about the 'nets, salt is the best in this regard ;_;
12:23 I3olle joined #salt
12:23 Romlok I thought the entire point of pillars was to do what map.jinja appears to also do?
12:23 Romlok at least, that's the impression I got from the pillar introduction
12:24 AndreasLutro pillars are for storing configuration values or other data that differs between each host
12:24 AndreasLutro but really, pillars can be used however you see fit
12:24 AndreasLutro you could easily move the logic typically found in map.jinja into your pillars, if you want
12:24 Romlok so why have map.jinja at all?
12:25 AndreasLutro because the stuff that's in map.jinja "belongs" to the state and shouldn't clutter up your pillar data
12:27 Romlok "belongs to the state" - darn socialists
12:27 Romlok ;)
12:27 AndreasLutro haha
12:27 AndreasLutro salt ☭
12:27 Romlok need to think about that one
12:28 Romlok atm it just seems bad to do effectively the same thing in two different places (which two different syntaxes)
12:28 Romlok give me something to ponder while people drone on about sales in this meeting, at least
12:28 ekristen joined #salt
12:29 malinoff Romlok, ansible has 15 possible ways to define a variable. Two different places is actually ok
12:31 cmcmacken joined #salt
12:32 numkem joined #salt
12:37 numkem joined #salt
12:41 elfixit joined #salt
12:43 PI-Lloyd joined #salt
12:45 stamper joined #salt
12:46 mapu joined #salt
12:47 dendazen joined #salt
12:48 ndrei joined #salt
12:52 dendazen joined #salt
12:53 PI-Lloyd joined #salt
12:55 XenophonF Romlok: am glancing back over the irc log
12:55 XenophonF Romlok: i want to emphasize a major difference between states and pillars
12:56 XenophonF Romlok: state SLS files get cached by the minion
12:56 XenophonF Romlok: and any file under fileserver_roots is readable by any minion
12:56 johnchalekson joined #salt
12:57 XenophonF Romlok: so if you have data that is security-sensitive, whether that means only certain minions should be able to access it or it means that the data shouldn't be cached, you would put that data into pillar
12:58 subsignal joined #salt
12:58 XenophonF Romlok: the rendered state files (state SLS after processing by Jinja, which could include Pillar data) do not get cached by the minion
12:59 XenophonF Romlok: and the pillar data doesn't get cached by the minion either
12:59 subsigna_ joined #salt
12:59 XenophonF Romlok: i don't know how that compares with other CM tools, as i only have experience with salt
12:59 MAbeeTT_ joined #salt
13:01 johnchalekson joined #salt
13:01 ekristen basepi: I’m still seeing manage.status and test.ping returning inconsistent results, any thoughts on what to do, I’m upgrading to 2014.7.2 right now
13:01 dyasny joined #salt
13:01 JDiPierro joined #salt
13:01 ekristen but even with 7.2 each time I run manage.status I see minions that I now are online showing as down
13:02 johnchalekson joined #salt
13:02 JDiPierro joined #salt
13:02 TinuvaMac joined #salt
13:02 johnchalekson joined #salt
13:03 ndrei joined #salt
13:04 johnchalekson joined #salt
13:04 clinta joined #salt
13:05 johnchalekson joined #salt
13:08 johnchalekson joined #salt
13:09 johnchalekson joined #salt
13:10 johnchalekson joined #salt
13:11 onmeac joined #salt
13:11 D-Spair joined #salt
13:11 johnchalekson joined #salt
13:12 vstoniest joined #salt
13:12 johnchalekson joined #salt
13:13 johnchalekson joined #salt
13:13 onmeac Greetings; I am trying to enforce one or more key=value in an unknown number of files, for doing this i chose augeas. Upon running the highstate i get this error message "Error: Command setm is not supported (yet)"; Do i create a feature request on salts github or is there another way to do what i want to do :)
13:15 glyf joined #salt
13:16 fusionx86 joined #salt
13:18 PI-Lloyd joined #salt
13:19 saltilicious90 joined #salt
13:20 mats_ joined #salt
13:20 saltilicious90 Hello all, I have installed Salt 2014.7.1 on a node with python 26 and 27 present.
13:21 saltilicious90 How could I make Salt run against python27?
13:21 saltilicious90 works fine with 26
13:21 penguinpowernz joined #salt
13:22 saltilicious90 With 27 I get: ImportError: No module named salt.scripts
13:22 andreyst joined #salt
13:22 racooper joined #salt
13:23 renoirb_ joined #salt
13:24 saltilicious90 Neverdmind, covered here: https://github.com/saltstack/salt/issues/22003
13:26 niseak joined #salt
13:27 echo joined #salt
13:27 jdesilet joined #salt
13:29 mpanetta joined #salt
13:29 saltilicious90 left #salt
13:30 CeBe left #salt
13:31 renoirb_ joined #salt
13:32 Micromus joined #salt
13:33 Rory joined #salt
13:33 Rory #saltstack is full of people
13:33 Rory is that unofficial or something?
13:34 glyf joined #salt
13:34 clinta left #salt
13:37 babilen Not really, but I wonder why it doesn't forward here
13:41 PI-Lloyd joined #salt
13:41 nafg joined #salt
13:42 Tyrm joined #salt
13:42 pdayton joined #salt
13:43 iwishiwerearobot joined #salt
13:43 otter768 joined #salt
13:44 johnchalekson joined #salt
13:45 johnchalekson joined #salt
13:45 PI-Lloyd joined #salt
13:45 Brew joined #salt
13:46 glyf joined #salt
13:49 JDiPierro joined #salt
13:49 timoguin joined #salt
13:50 yawniek anyone uses salt-cloud with aws eu-central-1 or china, seems that my version can not cope with the v4 signature stuff. anyone got it running, which version works?
13:52 ndrei joined #salt
13:52 _ether_ joined #salt
13:53 kaptk2 joined #salt
13:56 Romlok XenophonF: hmm, the state/pillar caching difference is good to note, thanks
13:57 XenophonF yawniek: someone mentioned a workaround for eu-central-1 earlier today - check the chat logs
13:57 Romlok I decided in the meeting (and boy what a meeting) to just use pillars for the grains-dependent settings for now
13:57 yawniek XenophonF ty
13:57 XenophonF Romlok: babilen convinced me not to use grains for anything security-sensitive
13:58 XenophonF yawniek: np
13:58 Romlok no sense making things too complex, when I'm still in the evaluation phase
13:58 Romlok XenophonF: I'm not considering sensitive information right now
14:00 Romlok specifically, the only grain-dependent thing I have right now is to not set_hwclock on Windows minions
14:00 chandankumar joined #salt
14:00 Romlok which I've done by defining a time:set_hwclock in the pillars
14:02 yawniek seems that i need 2015.2, can it be installed via pip?
14:03 I3olle_ joined #salt
14:05 yawniek ok pip2 install salt==2015.2.0rc2 does the trick for aws v4 signature support
14:07 andrew_v joined #salt
14:12 favadi joined #salt
14:13 Brew joined #salt
14:13 pdayton joined #salt
14:14 debian112 joined #salt
14:14 zwi joined #salt
14:16 faust joined #salt
14:17 giantlock joined #salt
14:21 hasues joined #salt
14:21 hasues left #salt
14:21 faust left #salt
14:23 glyf joined #salt
14:27 StDiluted joined #salt
14:29 mpanetta joined #salt
14:30 rhodgin joined #salt
14:35 bhosmer_ joined #salt
14:35 zmalone joined #salt
14:36 bhosmer_ joined #salt
14:37 jalbretsen joined #salt
14:39 JDiPierro joined #salt
14:40 sandah joined #salt
14:41 froztbyte hmm
14:41 froztbyte if I wanted to tag hosts with some kind of data, without using grains, how would I go about doing that?
14:41 glyf joined #salt
14:41 froztbyte the one idea that I had now is I could have some pillars with that data in it, and target based on hostname
14:42 MatthewsFace joined #salt
14:42 murrdoc joined #salt
14:43 froztbyte the specific problem I'm poking at is integrating my sensu setup with salt; the sensu-formula I've seen doesn't seem to do much about subscriptions
14:43 froztbyte but in my case I have a couple of subscriptions for different host/role types
14:43 froztbyte (subscriptions are, for all intents and purposes, some tagging metadata)
14:44 froztbyte (which then gets used to fetch some content the sensu server)
14:45 murrdoc morning
14:49 SeeDickCode joined #salt
14:52 ToxaKniotee joined #salt
14:55 gijs_ left #salt
14:56 igorwidl joined #salt
14:59 rojem joined #salt
15:01 timoguin joined #salt
15:02 dopesong joined #salt
15:04 ALLmightySPIFF joined #salt
15:05 ALLmightySPIFF joined #salt
15:05 thehaven joined #salt
15:06 chandankumar joined #salt
15:06 chandankumar joined #salt
15:09 XenophonF hey murrdoc
15:13 my_name joined #salt
15:15 my_name hello.. I mostly have a standard image set via Salt-Cloud. The one missing peice is the boot volume size. I'm able to make the block device, /dev/sda1 100G, but the filesystem always comes up 8G. I'm using ec2 as my provider. Any thoughts on how to make the fs match the device?
15:15 XenophonF my_name: what does your profile look like?
15:15 my_name this is what I have in my profile:     block_device_mappings:       - DeviceName: /dev/sda1         Ebs.VolumeSize: 100     volumes:     - { size: 100, device: /dev/xvda1 }
15:17 dude051 joined #salt
15:18 scbunn joined #salt
15:18 dude051 joined #salt
15:22 andreyst1 joined #salt
15:22 my_name XenophonF: I get the same result if I only have block_device_mappings:       - DeviceName: /dev/sda1         Ebs.VolumeSize: 100
15:23 Vynce joined #salt
15:23 iggy your ami is probably 8G (10)... so it just drops that on the block device... it's up to you to resize that (resize2fs, btrfs grow, etc.)
15:27 murrdoc joined #salt
15:29 niseak_ joined #salt
15:30 viq_ joined #salt
15:33 bluenemo_ joined #salt
15:34 thedodd joined #salt
15:36 niseak joined #salt
15:37 Gareth Has anyone used the bootstrap script to install Salt from the official Salt repositories?  Not the ones out of, for example, the official Debian repositories.
15:37 ciandro joined #salt
15:37 Seldaek left #salt
15:37 rm_jorge joined #salt
15:37 ciandro hey guys i am having some issues connecting to the master
15:38 ciandro it keeps saying i cannot authenticate
15:38 ciandro im running both master and minion in debug mode
15:38 ciandro the master shows nothing, like it's not even been touched
15:39 ciandro and the agent generates the key, connects to the IP, but fails to authenticate
15:39 ciandro im running CentOS 6.5
15:39 aparsons joined #salt
15:39 ciandro 2014.7.1 on both master and minion
15:40 iggy Gareth: salt has official repositories?
15:42 iggy ciandro: make sure you don't have multiple copies of the master or minion running
15:42 ciandro how can i do that?
15:43 MatthewsFace joined #salt
15:43 Aikar ps ?
15:43 murrdoc i love you
15:43 murrdoc wait what
15:44 otter768 joined #salt
15:45 northway joined #salt
15:46 noway_ joined #salt
15:46 ciandro i did ps -e | grep salt on my master
15:46 ciandro there were like 10 PID
15:46 ciandro :D
15:46 ciandro rebooting master server
15:47 ajw0100 joined #salt
15:47 ciandro i forgot the number one rule
15:47 iggy that's probably normal
15:47 Fiber^ joined #salt
15:47 ciandro oh it is?
15:47 iggy so long as they all belong to the same process tree
15:48 ciandro i mean, there seem to be a bug tracked for something like this: https://github.com/saltstack/salt/issues/19861
15:48 ciandro but it seems pretty unlikely to me that i cannot make a connection
15:49 ciandro im stopping the service, removing minion_id, cleaning up all keys
15:49 GabLeRoux joined #salt
15:50 iggy it is actually normal behavior for the minion to stop if it doesn't get ahold of the master
15:53 ciandro not sure if this could throw things off, but my process is I start an instance in ec2 with packer and use salt as provisioner.. this process installs base packages in masterless fashion
15:53 ciandro then when that's done, it packages the instance in an ami.. then i generate an instance from that ami and point it at the master, but it won't connect
15:54 Guest70 joined #salt
15:55 glyf joined #salt
15:55 ciandro [ERROR   ] Attempt to authenticate with the salt master failed
15:56 mpanetta joined #salt
15:56 mpanetta joined #salt
15:57 dyasny joined #salt
15:57 Guest70_ joined #salt
16:00 iggy do you see any unaccepted keys on the master?
16:00 pravka joined #salt
16:00 iggy is there a minion_id and/or key file in the actual ami (i.e. not generated when the minion spins up)?
16:01 GabLeRoux joined #salt
16:02 ciandro yeah i thought that could be the issue
16:02 ciandro so i wiped minion_id and all keys
16:02 ciandro and tried to force connection back, but still nothing
16:02 racooper joined #salt
16:02 _JZ_ joined #salt
16:02 ciandro and no, no sign of unaccepted keys.. there really is no trace on the master of any issues
16:03 ciandro even the debug logs i have on the console running, they don't show any event or rejection/denial
16:05 jcockhren ciandro: two things: 1. is the master set in your minion config properly? and 2. can the two ec2 instances see and talk to each other ok?
16:05 niseak joined #salt
16:05 stoogenmeyer joined #salt
16:05 jcockhren only asking b/c I didn't see any mention in the scrollback
16:05 ciandro 1) master is set as master in /etc/hosts for the minion; 2) telnet to port 4505 works but no reponse from 4506
16:07 iggy the default master name is "salt" not "master"
16:08 chandankumar joined #salt
16:08 jeremyalons clever way of determining how long highgstate took to run?
16:08 raygunsix joined #salt
16:08 iggy time?
16:09 Tecnico1931 joined #salt
16:09 ciandro iggy: yes, i meant i have in /etc/hosts salt salt-master salt-prd-01
16:09 iggy in newer versions times are tracked in the job log
16:09 iggy ciandro: okay, well, I got nothing then
16:09 writtenoff joined #salt
16:10 jcockhren ciandro: mind trying to set the 'master' key to the private IP of the master to be sure?
16:10 jcockhren this is assuming security groups and firewalls on both instances have been checked
16:11 ciandro i just tested connection from a salt-minion in the same VLAN as the master and it works..
16:11 jcockhren I had a similar issues when I had minion in a privtae VPC subnet
16:11 ciandro but sg allows 4505 and 4506
16:11 ciandro so i am not sure why this is happening, but i am assuming my master is fine
16:11 jcockhren can the master even see the minion?
16:11 spookah joined #salt
16:13 ciandro how can i try that?
16:13 Vynce left #salt
16:14 ciandro (I apologize for the immense noobness)
16:14 jcockhren nah. you're good.
16:14 jcockhren ping, if you allow for icmp traffic between the two
16:15 Guest70 joined #salt
16:16 desposo joined #salt
16:16 ciandro no we dont allow ping
16:16 ciandro but i can telnet port 4505 from minion to master successfully
16:16 ciandro but not port 4505
16:16 ciandro mean 4506
16:18 theologian joined #salt
16:21 jcockhren hmm
16:21 murrdoc joined #salt
16:21 jcockhren aah
16:22 ciandro im trying to change auth port on master and minion both on 4505
16:22 jcockhren ciandro: are you sure you're salt-minion service is running? try running it from the cli like: `sudo salt-minion` to see if it immediately exits
16:22 ciandro restarting services
16:24 ciandro ah
16:24 ciandro changing ports wasn't good
16:24 ciandro master never came up
16:24 jcockhren yeah. do the same check for the master as well
16:26 ciandro ok port changing does nothgin
16:26 ciandro same error on minon
16:26 ciandro master is up, working with other minons non-AWS
16:26 ciandro it's gotta be a fw issue
16:32 KyleG joined #salt
16:32 KyleG joined #salt
16:32 conan_the_destro joined #salt
16:32 basepi ekristen: we have some keepalive fixes which we've heard good things about. They aren't in 2014.7.4, but they should be in 2015.2.0 (I think, though they're not in the current release candidate). These are the fixes if you are interested: https://github.com/saltstack/salt/pull/21044
16:34 murrdoc basepi:  whats holding us back from putting them in 2014
16:34 murrdoc or are they raet dependant
16:34 iggy they are not raet dependent
16:34 basepi murrdoc: they're in 2014.7, just not 2014.7.4 =)
16:34 iggy (I'm not using raet yet)
16:34 iggy oh nice
16:34 basepi They backported yesterday
16:34 murrdoc is 2014.7.4 out ?
16:34 murrdoc i am still on .2
16:34 iggy tagged
16:35 basepi It is not officially announced, but it's tagged and being packaged
16:35 iggy but you know how salt does their releases
16:35 ekristen I guess I should have just upgraded to 7.4
16:35 ekristen but it wasn’t marked as the latest release in github so I missed it
16:35 murrdoc are u packaging before announcing !
16:35 iggy (the opposite way of every upstream I've ever seen)
16:35 shanemhansen joined #salt
16:35 murrdoc no i like it
16:35 murrdoc i want the deb when they have it
16:35 basepi murrdoc: ya, that's how we've been doing it for 1.5 years. Because people would complain about lack of packages. We may switch back to just announcing and letting packages do their thing....
16:36 losh joined #salt
16:36 basepi But then again, we hope to just take over packaging in the next 6 months and we'll be able to time it much better.
16:36 murrdoc no no
16:36 iggy ekristen: the fixes are _not_ in 2014.7.4
16:36 murrdoc i prefer when u say new build is available that u have hthe build available for it
16:36 ekristen oh
16:36 basepi ekristen: Ya, they're in 2014.7 branch, but not in 2014.7.4 unfortunately.
16:36 ekristen gotcha
16:36 shanemhansen So let's say I wanted to have salt-master generate a certificate and sign it for my minion and then push it out to the minion's fs. Assuming the signing key is on the master. How might I do that?
16:36 iggy that's silly... then everybody on deb/ubuntu has to wait for an ungodly long time while epel gets their shit together
16:36 ekristen are they in 2015.2.0rc2?
16:37 iggy ekristen: no
16:37 murrdoc iggy:  no they can maintain their own repos
16:37 basepi ekristen: nope, they were only backported yesterday
16:37 murrdoc shit i ll host the repos for them if they want
16:37 jcockhren murrdoc: do it
16:37 iggy but they are in the 2015.2 branch, so they should be in the next rc (or hopefully release)
16:37 ekristen basepi: ah
16:37 calve joined #salt
16:37 murrdoc jcockhren:  not the building part
16:37 murrdoc the hosting part :D
16:37 iggy murrdoc: yeah, in a perfect world... but that's not what we have now
16:38 ekristen hrm
16:38 Guest70_ joined #salt
16:38 basepi hehe, it's the building, not the hosting, that's the problem. ;)
16:39 murrdoc anyhow basepi you have my support if u build deb before announcing
16:39 murrdoc and totes if u want a mirror for hosting your stuff
16:39 murrdoc lmk
16:39 basepi After this is merged the fixes will be in the 2015.2 branch: https://github.com/saltstack/salt/pull/22468
16:39 basepi Ready to be in the next rc
16:40 murrdoc is matching fixed :)
16:40 ekristen basepi: or iggy can I just update my master to 2014.7 branch to get this fix? It seems like it is just on the master return port from the minions
16:40 iggy basepi: wait, so the keepalive stuff isn't in 2015.2 yet?
16:40 Gareth basepi removed compound matching again?!
16:40 bash124512 joined #salt
16:40 ekristen or do the minions need to be upgraded to?
16:41 murrdoc ./topic basepi removed compound matching again?!
16:41 murrdoc HAHAHAHA
16:41 murrdoc its all <3 basepi
16:41 basepi iggy: no, it was actually merged back to 2014.7 yesterday, after I merged forward yesterday
16:41 basepi So today's merge forward picked it up
16:42 iggy basepi: hmmm, well, 2015.2 has been way better at keeping connections up even before that then
16:42 basepi I think that's true. There are so many core improvements in 2015.2
16:42 Aikar {% if pillar['foo'] is defined %} -- this what I want to check if a pillar is defined?
16:42 basepi jacksontj was showing off some graphs on salt-users yesterday, they're pretty impressive.
16:42 iggy I thought Mike said it was in 2015.2 already, that's what I've been telling everybody
16:42 basepi Maybe it was. I could be wrong
16:43 basepi The original pull request was to develop, but he might have made a second one to 2015.2
16:43 * basepi looks
16:43 iggy Aikar: no, that would give you a key error if it wasn't there
16:43 calve hi all
16:43 murrdoc what iggy said
16:43 calve any
16:43 basepi iggy: oh, they totally are, just kidding
16:43 Aikar so what do i want then?
16:43 basepi 2015.2 already has the fixes.
16:43 iggy Aikar: {% if salt['pillar.get']('foo') %}
16:44 calve is there any way to populate an entire file using a pillar variable ?
16:44 basepi Gareth: you're a terrible person
16:44 basepi Gareth: you coming to pycon?
16:44 iggy basepi: good, and that was merged like a week after saltconf (or so)?
16:44 Gareth basepi: Very true.
16:44 basepi ekristen: yes, upgrading your master to 2014.7 branch actually would incorporate the fixes, might be worth trying.
16:44 Gareth basepi: unfortunately not.
16:44 mpanetta_ joined #salt
16:44 basepi iggy: no idea when, I just checked the code. But that sounds right. =P
16:44 iggy calve: contents_pillar
16:44 basepi Gareth: bummer.
16:44 ekristen basepi: 2015.2 is backwards compatible with 2014.7.x minions right? cause I can always try that too ;)
16:45 Gareth basepi: yeah.  You're speaking again?
16:45 basepi ekristen: Yes, it should definitely be. =)
16:45 mpanetta joined #salt
16:45 basepi Gareth: indeed, salt this time
16:45 iggy ekristen: I didn't do thorough testing, but it worked well enough to upgrade the minions from the new master
16:45 Gareth basepi: Nice!
16:45 basepi I'm underprepared. o.O
16:45 basepi I mean, I know the stuff, but it needs to be *tight* to fit in 25 minutes.
16:45 ekristen k
16:45 ekristen thanks, I’ll got to 2014.7 first I think
16:45 basepi ekristen: sounds great.
16:46 ekristen then I’ll look into 2015.2
16:46 Gareth basepi: Lots of hand gestures and big words, if all else fails break into an interpretive dance.
16:46 basepi Gareth: the attendees would exit with PTSD.
16:46 iggy ugh... the mongodb-formula gives me the shivers
16:49 Grokzen joined #salt
16:49 murrdoc i would pay to watch basepi interpret dance raet
16:49 murrdoc bi directional data with one pipe ?
16:49 * murrdoc brings popcorns
16:49 basepi hahaha
16:49 basepi scarred for life, man.
16:50 noway__ joined #salt
16:50 calve joined #salt
16:52 murrdoc damnit iggy
16:52 murrdoc "My comment wasn't about why you moved the file. It was why you chose sources/ instead of the prevailing standard of other formulas and the SaltStack docs of using files/
16:52 murrdoc "
16:52 murrdoc because it interferes with tab completion
16:52 iggy comment on the issue
16:52 calve iggy: thank you very much
16:53 iggy I want to see if anyone else thinks your reason is as asinine as I do
16:53 iggy calve: yw... for what?
16:53 noway_ joined #salt
16:53 murrdoc "I have moved the one sources file that was sitting in the root dir to the sources/ directory. I did this to bucket all source files, templates/otherwise in a separate directory from the state files that the formula provides. Sources as a bucket is better to represent that it the directory has templates and other files. Also sources doesn't interfere with tab completion when working in the formula dir.
16:53 murrdoc "
16:54 mikeottinger joined #salt
16:54 murrdoc https://github.com/saltstack-formulas/sysstat-formula/pull/9#issuecomment-90586434
16:55 iggy it did not say that before
16:56 calve iggy: for reveling me the existence of contents_pillar, which will save me a few hours
16:56 robawt is Salt at Pycon again this year?
16:56 murrdoc iggy:  it didnt , ninja edits
16:57 iggy calve: ahh, yeah, handy feature (there's also an issue open to add pillar:// uri support which should help too... if anybody ever does the work)
16:59 Andre-B joined #salt
17:00 spookah joined #salt
17:11 ekristen is there a simple and easy way to perform a service record lookup in a state to get some info to be used?
17:11 writtenoff_ joined #salt
17:11 GabLeRoux joined #salt
17:12 arthabaska joined #salt
17:12 writtenoff_ joined #salt
17:13 arthabaska Hi all, quick question: does the salt-master service need to be restarted after a deploy? and why?
17:13 pravka joined #salt
17:13 arthabaska assuming the deploy changes states only, not the master config
17:14 babilen no
17:16 arthabaska what if the deploy contains updated custom runners or reactors?
17:16 murrdoc then u need to run a saltutil.sync_all
17:17 babilen That is being done with every highstate btw
17:17 subsignal joined #salt
17:17 babilen arthabaska: What's the actual problem you are trying to solve?
17:17 ekristen basepi: any reason this is in the Blocked milestone? https://github.com/saltstack/salt/issues/9805
17:17 arthabaska @babilen trying to avoid restarting the master after each (very frequent) deploy
17:17 arthabaska I'd just assumed it was necessary
17:19 babilen arthabaska: Well, you don't have to. You can run "salt-run fileserver.update" if you can't wait for the master to pick up changes automagically, run saltutil.sync_all explicitly or many other things. it really depends on what you want to do :)
17:20 arthabaska @babilen recently moved away from gitfs due to reliability issues (with gitpython and pygit2)..just using good old fabric instead
17:21 basepi ekristen: no idea, I apparently lost track of that one. Triaged properly now!
17:22 ekristen basepi: thanks! if you need any info, please let me know, I’d love to stop having to add saltenv to my commands, it is not the end of the world, but would save a bit of typing, lol
17:22 basepi ekristen: yep, i hear ya.
17:23 ekristen I’ve screwed up and typed the wrong env before, that ends up being a mess too
17:23 sijis joined #salt
17:23 sijis whiteinge: are you headed to pycon?
17:23 enarciso joined #salt
17:25 stanchan joined #salt
17:25 druonysus joined #salt
17:28 fannet joined #salt
17:29 fannet hey everyone. I'm on debian wheezy and upgraded my salt master and now high state fails 100% of the time returning '    TypeError encountered executing state.highstate: string indices must be integers, not str. See debug log for more info.'
17:30 babilen fannet: Oh, I've had that for days. Complete infrastructure is broken ever since .2
17:30 mapu joined #salt
17:30 pmcg joined #salt
17:30 babilen Let me get you the bug report
17:30 ekristen I’ve seen it too, but hasn’t been consistent in my environment
17:30 fannet fantastic
17:31 babilen https://github.com/saltstack/salt/issues/21480
17:31 babilen Problem is that nobody seems to be clear on what causes this or what it is in .4 that is supposed to fix this
17:31 fannet how can i upgrade to .4
17:31 babilen And given the time it takes for packages to arrive it'll be a week before salt becomes usable again
17:32 ekristen babilen: I
17:32 ekristen babilen: I’d suggest to stop using packages
17:32 ekristen they have too many problems in my experience
17:32 ekristen I use the bootstrap script and install from git tags/branches
17:32 babilen Really wonder how that made it through testing, but its a nightmare and I have around 300 minions in an inconsistent state
17:32 ekristen its way easier to upgrade
17:32 fannet ya I have about 300 as well waiting on a critical upgrade
17:32 babilen ekristen: Not an option
17:33 ekristen that is unfortunate
17:33 babilen yeah
17:33 ekristen roll your own package then?
17:33 fannet ekristen can I swap my package to a source pull w/out breaking my salt configs
17:33 perfectsine joined #salt
17:33 ekristen fannet: you should be able to
17:33 ekristen as long as you don’t purge the package
17:34 ekristen apt-get remove salt-minion && bootstrap.sh git 2014.7
17:34 ekristen I’d backup everything first
17:34 ekristen but I’ve done it
17:34 fannet f# me
17:34 babilen I am not doing that on all my minions
17:34 babilen I mean it is bad enough as it is :)
17:34 fannet can I revert my pacakge
17:34 ekristen babilen: it gives you better control in the future
17:34 fannet it was working fine until this
17:35 babilen fannet: We found that running jobs in *very* small batches allows most minion to continue
17:35 swa_work joined #salt
17:35 babilen fannet: Downgrades aren't really supported on Debian, but you can do it. The question is: Do you really want to run the risk of making this worse on even more minions?
17:36 fannet what a nightmare
17:36 babilen If only somebody would mention *what* exactly causes it.
17:36 babilen (in the bug report I mean)
17:36 fannet how is this not a 'code red' for the salt team
17:36 babilen ekristen: I might consider that, but I won't roll that out like that
17:36 babilen fannet: I have no idea
17:36 shanemhansen left #salt
17:37 babilen Feel free to comment on that bug
17:37 ekristen babilen: sure np, I made the switch to using bootstrap and git long ago, and have been way happier
17:37 ekristen I ran into so many problems with bad cached files in packages, and other debian-esk stuff with packages and I steer clear of them now
17:37 murrdoc thats crazy
17:37 murrdoc curl -o -  | bash is crazy
17:37 murrdoc for prod
17:37 murrdoc :
17:37 murrdoc :D
17:37 babilen I know
17:38 ekristen no more crazy than apt-get install salt-minion
17:38 ekristen when the package is coming from a public server
17:38 babilen Or even better: "curl -o - ...|sh" and expecting bash and not using https
17:38 ekristen either way I download the script to a local server first ;)
17:38 babilen ekristen: There is a *massive* difference: http://wiki.debian.org/SecureApt
17:39 ekristen babilen: you are still trusting the person who uploaded the debian package
17:39 ekristen that it isn’t evil in anyway
17:40 ekristen it is no different than pulling a script down off an https website and then running it with bash — you still have to trust the source
17:40 ekristen I personally have read through the bootstrap script many times
17:40 ekristen and I use a version i have locally in my datacenter
17:41 babilen absolutely right
17:41 ekristen do you download the deb package, extract it and expect every detail of it? or do you just trust the person who uploaded it to the public repo … I’d be willing to bet money on the latter not the former
17:41 babilen I've inspected the packaging, yeah
17:41 ekristen every version?
17:42 ekristen every line of code? ;)
17:42 babilen sure
17:42 babilen I do nothing else
17:42 ekristen haha
17:42 ekristen yeah me too — all day long ;)
17:42 babilen But seriously: I did take a good look at the packaging (being a DM myself)
17:42 babilen Not for every version though
17:43 ekristen sure np — either way — using git and the bootstrap script has made life a lot easier switching between versions and ensuring the right code is in place.
17:45 otter768 joined #salt
17:45 theo joined #salt
17:46 cozby joined #salt
17:46 cozby hey guys, currently experimenting with salt
17:47 cozby we're an ansible shop atm but we need something more
17:47 cozby I would like nodes in my cluster to be auto approved
17:47 babilen cozby: Weclcome! And you are full of hope and questions? How may we help?
17:47 cozby ie. sharing a key
17:47 Furao joined #salt
17:47 cozby I have this sorta workinf
17:47 cozby working
17:47 cozby but not really
17:48 cozby I have the same key on two minions, and the minion ids are the same
17:48 cozby so they auth with the master fine
17:48 babilen http://docs.saltstack.com/en/latest/topics/tutorials/preseed_key.html ?
17:48 cozby but it only recognizes one node
17:48 babilen Did you just copy the same key from one minion to another? (minion key that is?)
17:48 cozby so when I send it a command it works - it gets run on both machines but the master thinks somethings wrong because of the double response
17:48 cozby babilen: yep
17:49 babilen And are you aware of http://docs.saltstack.com/en/latest/ref/configuration/master.html#auto-accept ?
17:49 babilen Do they have the same ID ?
17:50 cozby yep
17:50 cozby same ID
17:50 cozby IDs on both minions are NodeApp and the keys are called the same thing
17:51 cozby but the master only recognizes them as one machine
17:51 babilen They need a different minion id
17:51 cozby but then I would have to rename the key
17:51 cozby as well
17:51 cozby for this to work, no?
17:51 cozby key must match minionid?
17:52 babilen You can preseed any key you want (or autoaccept whatever the master is being sent), but each minion needs a different id
17:52 murrdoc or use the example from the reactor documentation
17:52 cozby babilen: hmm OK
17:52 cozby reactor?
17:52 murrdoc if u have a list of moinions available u can improve that
17:53 babilen cozby: Reactor is a system in salt that allows you to react to events (e.g. "minion comes online"). It would allow you to accept specific minions.
17:53 ltsampro` yes reactors are nice
17:54 babilen cozby: But then you could just set "auto_accept: True" in your master .. It really depends on what you want and what kind of checks you need to keep it secure.
17:54 cozby reactors eh
17:54 cozby hmm
17:54 cozby ok
17:54 cozby I'll read up
17:54 murrdoc a whole new world
17:54 murrdoc a new excitement u never knew
17:54 iggy or autosign
17:54 babilen cozby: What prevents you from accepting any new incoming key/minion ?
17:54 spookah iggy: do you run your salt master as root or another user?
17:55 iggy root
17:55 murrdoc groot
17:55 murrdoc (sorry)
17:55 cozby babilen: autoscaling
17:55 ltsampros babilen: is there a way to fire a mysql query from within an SLS file ?
17:56 babilen cozby: Ah, definitely look into reactors then. https://github.com/saltstack-formulas/ec2-autoscale-reactor for inspiration
17:56 cozby babilen: its a dynamic env, I can't sit around and approve keys. I want them to bootstrap configs
17:56 murrdoc uh hmm
17:56 murrdoc what babilen said
17:56 babilen ltsampros: http://salt.readthedocs.org/en/latest/ref/states/all/salt.states.mysql_query.html
17:56 cozby cool thanks
17:57 iggy ugh... rtd?
17:57 babilen oooh
17:57 babilen sorry
17:57 babilen http://docs.saltstack.com/en/latest/ref/states/all/salt.states.mysql_query.html
17:58 ltsampros babilen: nice although what i want to achieve maybe is cleaner through a custom _modules/wheel/me.py file
17:58 babilen ltsampros: That might very well be
17:58 iggy you didn't say what you were trying to achieve, so...
17:58 ltsampros accepting minions based on an external database
17:59 Diaoul joined #salt
17:59 baweaver joined #salt
18:00 babilen That is a little bit more involved than "fire a mysql query from within an SLS file" :)
18:00 ltsampros babilen: :D :P
18:00 babilen ltsampros: You'd probably want to use that as external pillar and then use that in a reactor
18:00 iggy something that generates /etc/salt/autosign.conf ?
18:00 babilen Or that
18:00 babilen yes, autosign!
18:01 dendazen Guys today checked on other locations that resolver formula still doesn’t work
18:01 ltsampros my approach right now is that I override the wheel.key.accept function
18:01 dendazen on another server instead of 'chi1.veliosystems.com' it parses into this record somehow
18:01 dendazen domain veliosystems.com
18:01 dendazen search chi1
18:02 ltsampros I do my queries and own logic, and then I either accept() or deny() they key as done in the wheel.key module
18:02 ltsampros much cleaner and keeps lots of logic out of sls files
18:02 iggy dendazen: what does  'salt-call grains.item domain' return on said system?
18:03 dendazen even though on that server i took upstream functions which grains.get(domain) uses and add them to the script and result is [root@uat-rob tmp]# python my_test.py
18:03 dendazen ('uat-rob', 'chi1.veliosystems.com')
18:03 iggy the problem with custom modules/etc is they don't always work in the master context
18:03 dendazen the second item is domain
18:03 murrdoc ext_modules will control stuff on master
18:03 murrdoc but not on minions
18:04 ltsampros yes that's what I understood and my prototype showed.
18:04 murrdoc so if u want some modules to run outside of file_roots on a minion, u will have to add the second location to file_roots
18:05 iggy dendazen: well... it doesn't print everything broken up... so that's all I fixed... beyond that open an issue (or fix it yourself... that doesn't look like a highly maintained formula)
18:05 dude051 joined #salt
18:05 clintberry joined #salt
18:05 dendazen yeah, no problems
18:10 perfectsine joined #salt
18:12 dude051 joined #salt
18:12 mats_ joined #salt
18:12 mats_ hi everyone
18:12 mats_ I am having the following issue, I've set up pillar information env:eng and role:web
18:13 mats_ for a host, when I run salt -I "env:eng" test.ping and salt -I "role:web" test.ping
18:13 mats_ I get on each test the right host
18:13 mats_ however, salt -I "env:eng and roles:web" test.ping
18:13 mats_ No minions matched the target. No command was sent, no jid was assigned.
18:13 mats_ any idea why that might be?
18:13 babilen That's a compound matcher, you need -C
18:14 babilen And ... well, I get you the docs
18:14 babilen http://docs.saltstack.com/en/latest/topics/targeting/compound.html
18:14 ajw0100 joined #salt
18:14 mats_ yeah, I read that a million times
18:14 mats_ thought I could use I as it was all pillars
18:15 babilen the "and" is not
18:15 mats_ well silly me ..
18:15 mats_ thanks babilen :)
18:15 babilen You are welcome. Don't forget about the I@ :)
18:15 perfectsine joined #salt
18:15 mats_ yeah :D
18:18 chandankumar joined #salt
18:19 smcquay joined #salt
18:22 writtenoff_ left #salt
18:25 andreyst1 joined #salt
18:26 rojem joined #salt
18:29 ipmb joined #salt
18:30 nafg_ joined #salt
18:30 CeBe joined #salt
18:34 dopesong_ joined #salt
18:36 denys joined #salt
18:37 CeBe1 joined #salt
18:45 sxar joined #salt
18:47 stoogenmeyer joined #salt
18:48 stoogenmeyer joined #salt
18:48 ajw0100 joined #salt
18:49 CeBe joined #salt
18:49 aparsons_ joined #salt
18:51 jalaziz joined #salt
18:55 timoguin joined #salt
19:01 ndrei joined #salt
19:02 perfectsine joined #salt
19:03 dopesong joined #salt
19:05 ajw0100 joined #salt
19:11 MatthewsFace joined #salt
19:12 fusionx86 joined #salt
19:14 FRANK_T joined #salt
19:14 FRANK_T I am trying to install Firefox on my windows machine and I am getting this error https://www.refheap.com/2e029f145fe8dd48a0feb0a8c
19:14 FRANK_T I need help
19:17 dopesong_ joined #salt
19:17 FRANK_T This is my sls https://www.refheap.com/8d8f6420d546459be05ff85c7
19:19 djn joined #salt
19:19 pdayton joined #salt
19:21 FRANK_T help :(
19:21 murrdoc :|
19:21 jespada joined #salt
19:21 FRANK_T murrdoc any idea
19:21 FRANK_T TypeError encountered executing pkg.install: list indices must be integers,                                                                                 not str. See debug log for more info.
19:22 baweaver joined #salt
19:22 murrdoc i dont see the module name
19:23 murrdoc sorry state function in the sls
19:24 FRANK_T statu function?
19:24 FRANK_T state function*
19:24 murrdoc firefox:
19:24 murrdoc pkg.installed:
19:24 murrdoc -
19:24 murrdoc -
19:24 murrdoc so on
19:24 FRANK_T Sorry this is for a Windows client
19:25 FRANK_T I am following this http://docs.saltstack.com/en/latest/topics/windows/windows-package-manager.html
19:25 babilen https://github.com/saltstack/salt/issues/21480 ← FRANK_T
19:25 babilen Current salt release is broken and unusable due to that
19:25 teskew joined #salt
19:27 FRANK_T thank you
19:29 murrdoc joined #salt
19:35 ipmb joined #salt
19:35 baloo_the_bear joined #salt
19:36 baloo_the_bear guys, what ports do i need open for master/minion to work?
19:36 baloo_the_bear 4505 and 4506 only?
19:36 perfectsine joined #salt
19:37 giantlock joined #salt
19:37 babilen http://docs.saltstack.com/en/latest/topics/tutorials/firewall.html
19:38 dopesong joined #salt
19:39 rojem joined #salt
19:40 baloo_the_bear so it's request 4505 and auth on 4506?
19:43 glyf joined #salt
19:45 |Fiber^| joined #salt
19:46 otter768 joined #salt
19:51 fusionx86 Hey guys, question about pillars and environments. I set my minions environment using the environment parameter in the minion config file. I can target my minions based on that (master:environment:dev) from the CLI, but in a pillar file using 'master:environment:dev' and -match: pillar doesn't seem to work. I can match by hostname though. Any idea what I'm missing?
19:55 ajw0100 joined #salt
19:57 niseak_ joined #salt
19:59 thedodd joined #salt
19:59 iggy there's a setting to pass that master config as pillar data... dunno if it's on by default
19:59 baweaver_ joined #salt
19:59 Guest70 joined #salt
20:00 smcquay joined #salt
20:01 fusionx86 I'll hunt around some more, but change my googling to include passing master config. Thanks iggy.
20:01 glyf joined #salt
20:04 faust joined #salt
20:04 dwfreed_ joined #salt
20:05 unni joined #salt
20:06 unni Hi, Is there any way to simply rename/nickname the minions? I dont want the hostname as minion name for some servers.
20:06 jalbretsen joined #salt
20:06 iggy the minion
20:06 unni ?
20:06 iggy the minion_id doesn't need to be the hostname at all... it just is by default
20:07 unni I tried append_domain. The name is getting long.
20:07 unni Cant we use custom names?
20:07 iggy it can be whatever you want
20:07 unni where can I rename those?
20:08 iggy change the minion_id
20:08 iggy you'll have to reaccept the keys (iirc)
20:10 unni ok iggy. That worked. But I need to remove old keys along with accepting new keys.
20:10 iggy manage.down has an option to remove keys for non-existing hosts
20:11 thelonereaper joined #salt
20:11 thelonereaper How do I run salt as CoreOS containers?
20:11 unni Which is that option?
20:12 thelonereaper I tried to install salt on CoreOS , but it said unsupported OS
20:12 thelonereaper So I'm thinking of running salt as a container
20:13 iggy unni: see the docs
20:13 unni ok Brian. Thanks.
20:13 iggy thelonereaper: you could use salt-ssh (instead of running a minion)
20:14 timoguin joined #salt
20:15 druonysus joined #salt
20:15 druonysus joined #salt
20:15 evilrob joined #salt
20:16 Tecnico1931 joined #salt
20:17 ipmb joined #salt
20:18 glyf joined #salt
20:19 jeremyalons 2014.7.4, calt-call --locall --pillar-root doesn't appear to be looking there for pillars - known limitation?
20:21 baweaver joined #salt
20:22 hernan joined #salt
20:22 pdayton1 joined #salt
20:23 lnr joined #salt
20:23 synestine joined #salt
20:24 synestine Hello Salt peeps. I've got a newb question. What's the best way to target items from a grains list? e.g. a list called "role" with values "mysql" and "php". How is the best way to target that from an sls file other than the top.sls?
20:29 babilen You only target in top.sls, I'd also not necessarily keep roles in grains, but in pillars and you probably want to use jinja tests in your states: http://jinja.pocoo.org/docs/dev/templates/
20:29 clintberry joined #salt
20:29 nikogonzo iggy: doesn't salt-ssh still require a writable /usr/lib/python ?
20:29 nikogonzo thelonereaper: maybe you want something like https://github.com/thisissoon/Docker-Salt-Master ?
20:30 iggy nikogonzo: no clue never used it, was just a suggestion
20:30 iggy realistically, there are better tools to manage a bunch of coreos machines
20:31 nkuttler synestine: you can do stuff like {% if 'foo' in grains.get('roles', ()) %} etc., but pillars are probably better
20:32 synestine nkuttler: Thanks. I'll look into pillars and that if syntax looks like what I need.
20:32 iggy grains are fine for roles so long as you understand a rogue minion can set it's own grains to whatever it wants and possibly access pillar data that way (if you target pillars based off those roles)
20:32 nkuttler the syntax is just hardened against incomplete configs
20:33 nkuttler yeah, i also use grains in some configs, but grains didn't exist yet back then
20:33 nkuttler er, pillars didn't exist
20:35 faust joined #salt
20:40 paha joined #salt
20:41 conan_the_destro joined #salt
20:41 dopesong_ joined #salt
20:45 jespada joined #salt
20:48 JDiPierro joined #salt
20:49 rojem joined #salt
20:49 JayFK joined #salt
20:54 jeffin joined #salt
20:54 jeffin Hi, I was trying to installing minions on my ubuntu VM. But I am wondering how can I chkconfig salt-minion in an ubuntu. Anybody did that?
20:54 agend joined #salt
20:55 manfred update-rc.d salt-minion defaults
20:56 manfred if it is started with upstart, it is always chkconfiged on, unless you make a /etc/init/salt-minion.override file that does nothing
20:57 CheKoLyN joined #salt
20:57 zmalone update-rc.d, init.d scripts by hand, upstart, or (god forbid) install chkconfig on ubuntu.
20:57 zmalone systemd on the next ubuntu, I think.
20:57 manfred should be
20:57 jeffin managed to work with update-rc.d
20:59 XenophonF left #salt
20:59 jeffin left #salt
21:01 lothiraldan joined #salt
21:02 stoogenmeyer hi, can someone explain what's happening here. I have a state file (http://pastebin.com/MRsq7Ut8) which I invoked using sudo salt '*' state.highstate. The output said that the first step succeeded and that the next 2 steps failed. I then ran the highstate again and everything was green. I checked and the state was not really performed, but it keeps being green. what gives?
21:02 scooby2 left #salt
21:03 desposo joined #salt
21:06 druonysus joined #salt
21:06 druonysus joined #salt
21:06 niseak joined #salt
21:07 stoogenmeyer oh just got it.. cmd.wait doesn't run if the watch state didn't change
21:07 relidy stoogenmeyer: I was so close to saying just that. Glad you got it figured out.
21:11 beneggett joined #salt
21:13 nikogonzo stoogenmeyer: is the url for ohmyzsh right? i get 404: https://gist.github.com/anonymous/413ad337fa164486f7e2
21:13 Guest70 joined #salt
21:14 stoogenmeyer nikogonzo: no I removed one word cause the link I'm using is from a private repo, sorry about that
21:15 nikogonzo werd
21:18 subsigna_ joined #salt
21:20 Guest70 joined #salt
21:21 ona_matt left #salt
21:28 mats_ joined #salt
21:30 perfectsine joined #salt
21:30 baweaver joined #salt
21:32 losh joined #salt
21:38 glyf joined #salt
21:45 subsignal joined #salt
21:47 otter768 joined #salt
21:50 ndrei joined #salt
21:51 StDiluted joined #salt
21:53 Guest31320 joined #salt
22:00 renoirb_ joined #salt
22:01 jerematic joined #salt
22:02 druonysus joined #salt
22:06 GabLeRoux_ joined #salt
22:08 desposo joined #salt
22:09 rome_390 joined #salt
22:11 irctc166 joined #salt
22:12 enarciso joined #salt
22:13 aCodinMan joined #salt
22:14 spicyWith joined #salt
22:16 rome_390 joined #salt
22:18 jerematic joined #salt
22:18 SeeDickCode joined #salt
22:20 amcorreia joined #salt
22:20 crane_ joined #salt
22:22 bfoxwell joined #salt
22:22 Laogeodritt joined #salt
22:22 Georgyo joined #salt
22:30 druonysuse joined #salt
22:34 sunkist joined #salt
22:35 AlexStraunoff joined #salt
22:36 echo joined #salt
22:40 rhodgin_ joined #salt
22:43 spookah joined #salt
22:46 jespada joined #salt
22:48 aquassaut joined #salt
22:48 baweaver joined #salt
22:54 _JZ_ joined #salt
23:02 dopesong joined #salt
23:03 StDiluted joined #salt
23:05 baweaver Is there any way to limit the arguments to cmd.run?
23:06 baweaver and if not what would need to be done (if anything) for that to be an option?
23:11 zz_ashmckenzie joined #salt
23:11 mosen joined #salt
23:13 ipmb joined #salt
23:13 GabLeRoux joined #salt
23:17 murrdoc joined #salt
23:17 scoates joined #salt
23:22 StDiluted_ joined #salt
23:30 loz-- joined #salt
23:30 Tahm joined #salt
23:32 Singularo joined #salt
23:32 StDiluted joined #salt
23:32 sunkist1 joined #salt
23:48 otter768 joined #salt
23:59 coval3nce joined #salt

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