Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2013-09-23

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

All times shown according to UTC.

Time Nick Message
00:06 krissaxton joined #salt
00:10 gatoralli joined #salt
00:10 gatoralli what's the best file.copy alternative for older salt versions?
00:19 that_guy_ joined #salt
00:21 jtrucks_ joined #salt
00:23 mgw joined #salt
00:25 AlcariTheMad joined #salt
00:25 kenbolton_ joined #salt
00:26 jefferai joined #salt
00:27 NV joined #salt
00:27 sashka_ua joined #salt
00:27 Bosse_ joined #salt
00:27 spacedentist joined #salt
00:28 CheKoLyN joined #salt
00:28 denstark joined #salt
00:28 Lue_4911 joined #salt
00:34 Gwayne joined #salt
00:59 minaguib joined #salt
01:00 jinnko joined #salt
01:01 Furao joined #salt
01:07 krissaxton joined #salt
01:18 sibsibsib joined #salt
01:26 Ryan_Lane joined #salt
01:30 mmilano joined #salt
01:33 Nexpro joined #salt
01:34 mgw joined #salt
01:36 jkleckner joined #salt
01:43 nu7hatch joined #salt
01:49 deepakmd_oc joined #salt
02:01 xl1 joined #salt
02:05 malinoff joined #salt
02:07 krissaxton joined #salt
02:13 Ryan_Lane joined #salt
02:16 bhosmer joined #salt
02:17 g3cko joined #salt
02:17 xerxas joined #salt
02:17 goki joined #salt
02:17 mfournier joined #salt
02:17 Gareth joined #salt
02:17 f47h3r joined #salt
02:17 madduck joined #salt
02:17 arapaho joined #salt
02:17 Niichan joined #salt
02:17 EnTeQuAk joined #salt
02:17 hughsaunders joined #salt
02:17 jgelens joined #salt
02:31 vvoody joined #salt
02:31 StDiluted joined #salt
02:41 Brew joined #salt
02:49 piffio joined #salt
02:51 Furao joined #salt
02:53 jslatts joined #salt
03:00 redondos joined #salt
03:09 krissaxton joined #salt
03:45 Guest51 joined #salt
03:50 sibsibsib joined #salt
03:53 jfalco joined #salt
03:53 pdayton joined #salt
03:55 sgviking joined #salt
04:00 McMurphy joined #salt
04:00 McMurphy hi
04:01 forrest joined #salt
04:03 UtahDave hey, McMurphy!
04:04 offby1 R P McMurphy?
04:04 offby1 From Salem?
04:04 offby1 Thought they lobotomized you!
04:09 krissaxton joined #salt
04:15 stephenjy joined #salt
04:34 madduck joined #salt
04:43 krissaxton joined #salt
04:43 krissaxton1 joined #salt
04:49 Katafalkas joined #salt
04:55 jkleckner1 joined #salt
05:35 malinoff Hi all
05:35 malinoff What's the easiest way to copy a file from one minion to another?
05:36 robawt malinoff: is the file on the salt server?
05:36 malinoff robawt: Nope
05:37 malinoff robawt: I have a build server that builds a .war file; i want to transfer it to the other minion
05:37 malinoff both of them are minions
05:38 robawt malinoff: based off of this google discussion it appears it's in progress
05:38 robawt https://groups.google.com/forum/#!topic/salt-users/kWbKpi4xD5A
05:38 malinoff robawt: Thanks
05:38 robawt perhaps for now you can make a cmd.run that uses pillar to dynamically select the scp target?
05:39 forrest was just going to link that robawt
05:39 malinoff robawt: Yeah, something like that
05:39 robawt forrest: my salt fu is slowly progressing
05:39 forrest I wonder if you could run a state, that then executes this as a command
05:39 forrest http://docs.saltstack.com/ref/cli/salt-cp.html
05:41 forrest In reality though, it seems like a better idea to move your war file onto the salt master, or into a repo which you pull it from
05:42 robawt forrest: it does bring it up as an interesting deployment question
05:42 forrest which is?
05:42 robawt or maybe build the war in jenkins/buildbot/whatever and push it to master from there
05:42 forrest someone was talking about how they build java stuff with Jenkins, and part of the build process just moves it to the salt master
05:42 malinoff forrest: We will make rpm packages in future, right now I'm tied up with my client :(
05:42 robawt deploying a war
05:43 robawt forrest and i are sharing the same brain tonight apparently
05:43 malinoff the point is I don't need this war on the master
05:43 forrest but what is the reason it can't go on the master?
05:43 forrest that makes the deployment much easier
05:43 forrest robawt, we're both just thinking of the easiest solution.
05:43 forrest lazy sunday night brain is lazy
05:44 forrest or is the war file just massive
05:44 robawt heh
05:44 robawt truth forrest
05:45 forrest I'm not trying to say you can't or shouldn't do it that way malinoff
05:45 forrest it just sounds like a pain
05:45 robawt jenkins is really good at java, because it's java.  let it do the legwork
05:45 forrest do you remember who was talking about that procedure robawt?
05:45 forrest I don't feel like scrolling back through all the logs
05:46 Jahkeup_ joined #salt
05:46 robawt it gets brought up at every talk about java ever, not sure where to link exactly
05:46 forrest yea
05:46 forrest lol
05:46 robawt https://wiki.jenkins-ci.org/display/JENKINS/Building+Jenkins
05:46 forrest you COULD set up that location on the minion as a git repo, then use git_fs to pull from there, to deploy it onto the other minions.
05:46 malinoff forrest: I'm trying to not distribute files i don't need to distribute.
05:46 malinoff *on the master
05:47 malinoff This makes deployment much complex for a man who's not familiar with salt
05:47 forrest Well, it's because you want to copy from one minion to another
05:47 forrest which is a security issue
05:48 scristian_ joined #salt
05:48 malinoff more complicated* ofc
05:49 malinoff I'd rather specify something like ACL: what can i distribute and where instead of doing that through the master
05:49 robawt malinoff you could do that in a cmd and specify ACLs via pillar
05:49 robawt that'd be cool
05:50 malinoff yeah, thinking of it right now
05:50 forrest I mean if you REALLY wanted, you could write a state that simply executes scp file.war user@location:/path/to/place
05:50 forrest and set up ssh keys for the root user
05:50 forrest or just keys in general
05:50 scristian joined #salt
05:50 forrest I mean it would be very ghetto
05:50 forrest but it might do what you want
05:51 forrest you'd have to write some pillar logic/jinja2 logic for looping through all the boxes though
05:51 krissaxton joined #salt
05:51 malinoff forrest: Well, it's quite strange - we're leaving ssh and scp world to the brand new saltstack, but deep in it we HAVE to use our old scp friend
05:52 forrest Well, you're trying to use salt in an old school style way
05:52 deepakmd_oc joined #salt
05:52 forrest as opposed to either using a configured file server, git, or the salt master itself
05:52 malinoff forrest: Nope :) Just thinking about what's the best way to do something that salt is not supporting right now
05:52 ronc joined #salt
05:53 forrest Oh yea I understand
05:53 forrest it's just tough when none of the implemented solutions will resolve what you want to do
05:54 malinoff yeah
05:54 forrest Gotta think of less than stellar work arounds
05:54 forrest I've gotta go for the evening, let me know if you guys come up with a really cool solution to this.
05:56 malinoff forrest: 'Night. I think, I'll prepare a big full usage experience for the New Year and put it it in salt-users
06:02 packeteer hey guys, trying to use an sls file to install a package. at what point do i specify which package it is?
06:03 scristian joined #salt
06:04 malinoff packeteer: http://docs.saltstack.com/ref/states/all/salt.states.pkg.html?highlight=pkg.installed#salt.states.pkg.installed
06:04 malinoff "usage" block
06:04 packeteer thx
06:04 malinoff np
06:05 packeteer yeah, figured it out. the first line needs to be the exact package name
06:07 packeteer now to build on that  :)
06:13 jason27409 joined #salt
06:16 jason27409 see me
06:17 jason27409 quit
06:20 jason27409 joined #salt
06:22 sebbe_ joined #salt
06:24 packeteer hmm.. No matching sls found for 'packages.sls' in env 'base'
06:26 davidone joined #salt
06:30 sebbe_ To install salt on OSX, it says that i need the Xcode command line tools. Do i also need the Xcode IDE? it's almost at 2GB and seem a bit to much to get salt up and running.
06:37 packeteer i believe that depends what you use to install ie. brew or macports
06:37 packeteer i could be wrong tho  :)
06:38 packeteer iirc brew can get away with just the command line tools, but macports requires the full enchilada
06:38 ml_1 joined #salt
06:39 sebbe_ Aha ok, hmm then i will install it using brew. Thank you for a quick answer :)
06:43 Ryan_Lane joined #salt
06:44 packeteer yeah, macports supports a lot more packages than brew. some of those packages require more than what is provided by Xcode command line tools. and as Apple have only provided 2 'packages' there is a simple choice
06:56 packeteer hmm, i seem to be finding ways to break salt  :/
07:00 jason27409 中国
07:01 packeteer prob nothing, but here's a partial paste: http://pastebin.com/Sp8gYQZe
07:05 pentabular joined #salt
07:05 __number5__ packeteer: what's in your sls file?
07:06 __number5__ and you probably want to use at least python2.7
07:06 packeteer centos6 defaults
07:06 packeteer packages:
07:06 packeteer pkg.installed
07:06 packeteer - pkgs:
07:06 packeteer - vim-enhanced
07:06 packeteer - tmux
07:07 packeteer - ntp
07:07 packeteer trying to create a state file to install packages post install
07:07 packeteer / minimum packages on all matching systems
07:08 __number5__ you missing a ':' after installed
07:09 packeteer ahah!
07:09 packeteer that did it, thx
07:09 __number5__ np
07:10 sgviking joined #salt
07:16 packeteer is it best to seperate installing of packages and their post install config, or squeeze it all into the 1 file?
07:18 ronc joined #salt
07:19 packeteer maybe i should just browse other users state files
07:25 mnemonikk joined #salt
07:26 aleszoulek joined #salt
07:44 nu7hatch joined #salt
08:01 gaoyang_ joined #salt
08:07 redondos joined #salt
08:09 ciss joined #salt
08:09 wanico_ joined #salt
08:10 pentabular joined #salt
08:27 krissaxton joined #salt
08:28 Ryan_Lane joined #salt
08:35 qba73 joined #salt
08:39 Moon joined #salt
08:43 balltongu joined #salt
08:45 nliadm joined #salt
08:46 carlos joined #salt
08:58 Ryan_Lane joined #salt
08:59 carlos_ joined #salt
09:02 adepasquale joined #salt
09:08 elsmorian joined #salt
09:11 pentabular joined #salt
09:12 middleman_ joined #salt
09:28 Katafalkas joined #salt
09:28 yota joined #salt
09:30 deepakmd_oc joined #salt
09:32 elsmorian joined #salt
09:37 jamiem joined #salt
09:38 lemao joined #salt
09:44 middleman_ joined #salt
09:44 elsmorian1 joined #salt
09:58 carlos joined #salt
10:04 balltongu joined #salt
10:07 EugeneKay joined #salt
10:08 ggoZ joined #salt
10:11 pentabular joined #salt
10:12 middleman_ joined #salt
10:20 ollins joined #salt
10:28 fredvd joined #salt
10:36 fredvd joined #salt
11:03 nliadm joined #salt
11:11 krissaxton joined #salt
11:11 backjlack joined #salt
11:12 pentabular joined #salt
11:12 LucasCozy joined #salt
11:15 felixhummel joined #salt
11:16 xl1 left #salt
11:18 gildegoma joined #salt
11:20 Katafalkas joined #salt
11:21 backjlack joined #salt
11:22 jslatts joined #salt
11:22 felixhummel joined #salt
11:30 kenbolton joined #salt
11:30 backjlack joined #salt
11:37 bud_ joined #salt
11:38 bud_ Hello to all :-).
11:38 bud_ How can I pass the user argument to cmd.run from a module.run state that executes cmd.run via publish.publish on another host.
11:39 bud_ With ? :-)
11:41 bud_ The following doesn't work - https://gist.github.com/anonymous/6669368.
11:41 oz_akan_ joined #salt
11:44 oz_akan_ joined #salt
11:44 AviMarcus joined #salt
11:44 nu7hatch joined #salt
11:45 AviMarcus Does file.managed and file.append mix? Does the ordering matter? I didn't test this yet...
11:47 bhosmer joined #salt
11:50 felixhummel joined #salt
11:53 sibsibsib joined #salt
12:05 felixhummel_ joined #salt
12:07 kadel joined #salt
12:09 MrTango joined #salt
12:10 piffio joined #salt
12:10 krissaxton joined #salt
12:12 pentabular joined #salt
12:17 santagada joined #salt
12:21 diegows joined #salt
12:22 jslatts joined #salt
12:37 mwillhite joined #salt
12:46 santagada left #salt
12:48 blee joined #salt
12:50 absolute joined #salt
12:52 deepakmd_ joined #salt
12:57 AviMarcus I tried mixing with an extend but it seemed to hang
12:57 oz_akan_ joined #salt
12:57 felixhummel joined #salt
12:58 AviMarcus I added an order=last so it takes it out and adds it back...
12:58 AviMarcus I could have just put it in, or used a req, but then they are dependent on each other
12:58 juicer2 joined #salt
12:59 oz_akan_ joined #salt
13:00 krissaxton joined #salt
13:03 sibsibsib joined #salt
13:03 Gifflen joined #salt
13:04 pdayton joined #salt
13:05 ipmb joined #salt
13:05 Kholloway joined #salt
13:07 shel3over joined #salt
13:10 Jahkeup_ joined #salt
13:13 pentabular joined #salt
13:14 piffio joined #salt
13:15 racooper joined #salt
13:16 blee joined #salt
13:20 kadel joined #salt
13:23 jfalco joined #salt
13:23 brianhicks joined #salt
13:27 nliadm joined #salt
13:29 nu7hatch joined #salt
13:32 toastedpenguin joined #salt
13:32 mgw joined #salt
13:32 krissaxton joined #salt
13:32 sebbe_ joined #salt
13:33 sebbe_ Can you write commentst inside your .sls files somehow? I have tried with # but that didn't work :)
13:38 toastedpenguin joined #salt
13:41 tmmt sebbe_: # works
13:42 felixhummel_ joined #salt
13:43 sebbe_ Okay i was lying you can make comment using #, i had a tabb before it but now it works :P
13:43 pdayton joined #salt
13:44 mgw joined #salt
13:45 juicer2 joined #salt
13:45 micah_chatt joined #salt
13:47 mohae joined #salt
13:48 piffio sebbe_: :)
13:50 sgviking joined #salt
13:50 deepakmd_ joined #salt
13:54 pdayton joined #salt
13:55 Jahkeup__ joined #salt
13:57 kyriakos joined #salt
13:58 balltongu joined #salt
13:59 kyriakos left #salt
14:00 kermit joined #salt
14:01 scristian trying to do a master/slave mysql state. on top sls based on role I load master.sls or slave.sls problem is that I have a confict name, both work with same my.cnf
14:04 az87c joined #salt
14:04 az87c_ joined #salt
14:04 juicer2 joined #salt
14:07 absolute joined #salt
14:12 m_george left #salt
14:13 pentabular joined #salt
14:14 Brew joined #salt
14:14 nu7hatch joined #salt
14:17 pentabular joined #salt
14:19 jbunting joined #salt
14:19 mannyt joined #salt
14:20 higgs001 joined #salt
14:23 y0j joined #salt
14:26 copelco joined #salt
14:27 pdayton joined #salt
14:29 cedwards scristian: update the ID declaration to master_my.cnf or slave_my.cnf and use - name: my.cnf
14:29 ccase joined #salt
14:29 balltongu joined #salt
14:32 alunduil joined #salt
14:36 scristian thanks cedwards, I figured out eventually that ID can be anything :)
14:36 Morton_ joined #salt
14:37 pnl joined #salt
14:37 chutzpah joined #salt
14:38 tyler-baker joined #salt
14:39 rgbkrk joined #salt
14:42 imaginarysteve joined #salt
14:48 cnelsonsic joined #salt
14:48 ml_1 joined #salt
14:51 jimallman joined #salt
14:52 copelco joined #salt
14:54 teskew joined #salt
14:55 jalbretsen joined #salt
14:57 ldlework When is 0.17 due?
14:58 gaoyang joined #salt
15:01 EugeneKay joined #salt
15:04 nu7hatch joined #salt
15:04 Lue_4911 joined #salt
15:04 devinus joined #salt
15:04 Jahkeup_ joined #salt
15:07 krissaxton joined #salt
15:12 ccase joined #salt
15:12 JoAkKiNeN joined #salt
15:14 faldridge joined #salt
15:21 multani hum, I have troubles running the Salt test suite, it can't find a "salttesting" module, which doesn't seem to be in the repository, or am I missing something?
15:24 rgbkrk joined #salt
15:24 KyleG joined #salt
15:24 KyleG joined #salt
15:27 lyddonb joined #salt
15:29 StDiluted joined #salt
15:33 arcleo joined #salt
15:38 kaptk2 joined #salt
15:41 arcleo Hi all, I am trying to use salt to manage a binary file. I've set the "show_diff" flag to "False", however, when we call the state it shows the diff between the source file and the file we are looking to replace
15:41 jcockhren multani: install this inside your virtual env -> https://github.com/saltstack/salt-testing
15:42 multani jcockhren: yeah, I just found it, thanks!
15:43 kyvn joined #salt
15:46 pjs joined #salt
15:46 imaginarysteve joined #salt
15:47 multani jcockhren: I tried to play with USE_SETUPTOOLS while installing Salt with "pip install -e ./", or using easy_install, but without luck
15:47 pentabular joined #salt
15:48 multani jcockhren: is there something I missed from the manual, or do you think I should fill a ticket for this?
15:48 pipps joined #salt
15:51 jcockhren multani: https://github.com/saltstack/salt/blob/develop/HACKING.rst#configure-your-virtual-environment
15:51 jcockhren however, this is slightly off
15:51 jcockhren you need to pip install -r 'dev_requirements_python2x.txt'
15:51 jcockhren multani: ^
15:52 jcockhren depending on which version of python you're using to run the tests
15:53 multani jcockhren: hum ok, I was reading this: http://docs.saltstack.com/topics/hacking.html#running-the-tests
15:53 multani well, which is the same actually -_-
15:54 multani jcockhren: oh no, it's at the end of your page: https://github.com/saltstack/salt/blob/develop/HACKING.rst#running-the-tests
15:54 miva joined #salt
15:58 piffio joined #salt
15:58 [diecast] joined #salt
15:58 KyleG joined #salt
15:58 KyleG joined #salt
16:00 ldlework does anyone know when .17 is due?
16:01 Ahlee psutil intalled, yet calls to ps.cached_physical_memory returns "ps.cached_physical_memory" is not available.
16:01 Ahlee thoughts?
16:02 arcleo joined #salt
16:02 Ahlee psutil certainly working as salt host cmd.run "python -c 'import psutil; print psutil.cpu_times()'" returns values
16:02 Boohbah joined #salt
16:03 jcockhren multani: you good now?
16:04 multani jcockhren: yep
16:04 multani jcockhren: thx
16:07 cro joined #salt
16:07 morton__ joined #salt
16:08 terminalmage Anyone here using salt on FreeBSD? I rewrote FreeBSD package management last week and would like more people to test it before I submit a pull request.
16:09 zloidemon terminalmage: What did you implement?
16:09 Boohbah joined #salt
16:11 cro joined #salt
16:11 redondos joined #salt
16:12 terminalmage freebsdpkg.py (before my rewrite) tried to use pkg_add -r to install packages, but it would also check for a pkgng database and if found, it would try to use pkgng
16:12 zloidemon terminalmage: Does it support pkgng?
16:12 terminalmage then there was a pkgng module, which had more support
16:13 terminalmage but was not actually a true "pkg" provider in that it also lacked key functions
16:13 UtahDave joined #salt
16:13 terminalmage so I ripped all of the pkgng stuff out of the freebsdpkg module, and made it strictly pkg_add
16:13 zloidemon I able to check that, but I don't permanently use saltstack on FreeBSD
16:14 terminalmage and then I fleshed out the pkgng module and made it fully-supported as a pkg provider
16:14 abe_music joined #salt
16:14 Lue_4911 joined #salt
16:14 zloidemon hmm... Interesting.
16:14 zloidemon terminalmage: Let me see a patch
16:15 pipps1 joined #salt
16:15 terminalmage zloidemon: it's in my fork. it required changes elsewhere (in the pkg states, etc.) so it's best to just clone my fork (or add it as a remote) and checkout the bsd-pkg-rewrite branch.
16:16 terminalmage https://github.com/terminalmage/salt.git
16:17 shinylasers joined #salt
16:18 jbunting joined #salt
16:18 zloidemon terminalmage: Do you have a port or PR for use a github?
16:19 bitz joined #salt
16:19 terminalmage so, moving forward, the freebsdpkg module will be default for FreeBSD 9 and earlier, and the pkgng module will be default for FreeBSD 10 and newer. Those wishing to use pkgng to manage packages on FreeBSD 9 will need to override the provider in the minion config.
16:19 zloidemon I might change salt port in portstree for test that
16:19 terminalmage zloidemon: no. I usually test by installing a git checkout using pip install -e /path/to/git/checkout
16:19 zloidemon I not shure about that. I think need to change flag for use a pkg or pkgng
16:19 gatoralli joined #salt
16:20 zloidemon Because I only have FreeBSD 9.1
16:20 terminalmage is your 9.1 box converted to use pkgng yet?
16:21 zloidemon I only use pkgng on all my FreeBSD boxes and jails
16:21 terminalmage there's a flag in the make.conf that you need to set, WITH_PKGNG=yes, but that's just to tell ports to register installed ports in the pkgng database
16:22 zloidemon terminalmage: The best way to use a specific param for pkgng or pkg
16:22 zloidemon Yes. I know.
16:22 terminalmage zloidemon: ????
16:23 terminalmage I don't know what you mean by "The best way to use a specific param for pkgng or pkg"
16:23 zloidemon terminalmage: You said about use pkgng only for 10.x, this isn't correct need to do choise for users
16:23 terminalmage zloidemon: that's what the provider override is for
16:24 terminalmage zloidemon: http://docs.saltstack.com/ref/configuration/minion.html#providers
16:24 ggoZ joined #salt
16:24 terminalmage pkgng is not the default on FreeBSD 9, so it will not be the default package manager salt uses for FreeBSD 9
16:24 zloidemon terminalmage: Oh nice. I able to install your stuff
16:25 zloidemon And plese write me a list with tests to vg@FreeBSD.org
16:26 mgw joined #salt
16:30 devinus joined #salt
16:34 djn joined #salt
16:35 arcleo left #salt
16:38 travisfischer I'm using file.directory to create a base folder. I'm than having various other states put files inside this folder via file.managed and git.latest, however, one of my git.latest commands needs to be run with a user which doesn't create the permissions I want inside this folder. Is there a good sol'n to this. I would like to set recursive perms on my initial file.directory state but it gets run before the git.latest. My hac
16:38 travisfischer k is a cmd.run sudo chmod & chown but I'm not liking that as a final sol'n. suggestions?
16:38 terminalmage zloidemon: sent, thanks
16:39 abe_music anyone experiencing issues with salt bootstrap failing dueo to unsigned RPMs during the yum update?
16:39 abe_music in particular, "Package chkconfig-1.3.49.3-2.el6_4.1.x86_64.rpm is not signed"
16:39 zloidemon terminalmage: Ok. I'll check it in the morning.
16:40 terminalmage zloidemon: cool
16:41 felixhummel joined #salt
16:45 santagada joined #salt
16:46 lacrymology joined #salt
16:47 lacrymology from python custom modules, when calling other functions on the same module, is there any advantage in calling __salt__['module.function'] instead of function()?
16:48 krissaxton joined #salt
16:50 pipps1 joined #salt
16:51 santagada joined #salt
16:54 nu7hatch hi guys, what's the best/easiest way to get results of an async highstate job?
16:55 abe_music nu7hatch: might take a look at http://docs.saltstack.com/ref/returners/
16:56 nu7hatch abe_music: thanks!
16:56 abe_music np :)
16:58 Katafalkas joined #salt
17:00 ldlework Hello? Can anyone please give me a hint as to when 0.17 is intended to land? I cannot find any information on this.
17:00 eculver joined #salt
17:00 eculver joined #salt
17:01 jcockhren ldlework: RC is supposed to be cut "soon" as I was told
17:02 supplicant joined #salt
17:02 krissaxton joined #salt
17:02 carlos joined #salt
17:02 jcockhren ldlework: I imagine right now, the final PRs for the next released are being merged
17:02 ldlework thanks
17:03 UtahDave 0.17 RC 1 has been cut already
17:03 jcockhren ldlework: ^
17:04 [diecast] are there any 0.16 -> 0.17 upgrade gotchas?
17:04 mgw joined #salt
17:06 ldlework jcockhren: I need to wait for the PPA to be updated
17:06 jdenning joined #salt
17:06 jcockhren makes sense. we all are anxious
17:06 jcockhren ;)
17:07 mgw joined #salt
17:09 jpeach joined #salt
17:09 ldlework Is there anyway to dynamically set grain info on aribtrary minions from the master?
17:11 jcockhren ldlework: yep
17:11 jcockhren http://docs.saltstack.com/ref/modules/all/salt.modules.grains.html#salt.modules.grains.setval
17:11 jcockhren ldlework: ^
17:11 ldlework thx
17:11 devinus joined #salt
17:11 jcockhren at this point, I'm sure I've memorized this docs
17:12 jcockhren s/this/the
17:12 forrest joined #salt
17:12 ldlework jcockhren: hehe
17:15 jcockhren UtahDave: did mech442 ever ping/message you about an idea he had?
17:16 UtahDave jcockhren: yeah, he did. I haven't had a chance to respond, though.  Has he discussed it with you?
17:16 KyleG left #salt
17:16 Katafalkas joined #salt
17:17 jcockhren UtahDave: yeah at length. lol. it took a while to clean it up. but we got there
17:17 UtahDave :) cool. I should have a little time today to go over that again.
17:18 whit joined #salt
17:18 elsmorian joined #salt
17:18 forrest robawt, did you figure out a solution last night for copying that war file between minions?
17:18 mgw joined #salt
17:22 jcockhren UtahDave: word. Aside from his gist, that was confusing. I can sum it up pretty well. but take a look to get confused, then I'll make it make sense
17:24 mgw joined #salt
17:31 terminalmage I've just written up some details on my FreeBSD package management rewrite, in case anyone else is able to help test: https://gist.github.com/terminalmage/6673729/
17:34 hblanks joined #salt
17:36 pdayton joined #salt
17:36 pipps1 joined #salt
17:38 faldridge joined #salt
17:38 ciss joined #salt
17:40 a1j so the fact that schedule on all machines fire up simultaneously does not play nice on master.
17:41 a1j need to stagger by random time or something
17:42 ciss a1j: haven't used salt yet, but that sounds like a job for a queue with delayed execution
17:44 forrest You could also use batch size http://docs.saltstack.com/topics/targeting/batch.html
17:47 abe_music UtahDave: who can i talk to about salt-bootstrap issues?
17:47 a1j forrest: the schedule run is initiated from minion, so i cant control batch size.
17:47 abe_music it's not a salt issue really, but still something that probably should be addressed
17:47 forrest oh I see a1j
17:47 forrest abe_music, either s0undt3ch, or I can try to help
17:47 a1j forrest: there needs to be time slew for the first schedule run, + random slew for each consequtive run
17:47 abe_music cool forrest, thanks…
17:47 forrest Yea I understand what it is you want to get at
17:48 s0undt3ch abe_music: whats up?
17:48 abe_music forrest, s0undt3ch: the issue is during the install_centos_*_deps() function where we do a yum update
17:48 s0undt3ch ah! the update
17:48 abe_music one of the packages for me hasn't been gpg signed (weird i know)
17:48 s0undt3ch abe_music: ah, well, that weird
17:48 abe_music here's the particular package: http://mirror.centos.org/centos/6.4/updates/x86_64/Packages/chkconfig-1.3.49.3-2.el6_4.1.x86_64.rpm
17:48 forrest ok someone ran into this previously abe_music
17:49 forrest and it had to do with the mirror they were pulling from, which was very odd
17:49 abe_music i mean, i could disable gpg signing in the repo on the box, buttttt that's potentially unsafe
17:49 forrest yea I understand your concern
17:49 s0undt3ch abe_music: hmm, true
17:49 forrest do you know which mirror you're trying to pull chkconfig down from?
17:49 abe_music i could also blacklist the mirror and rebuild my AMI
17:50 s0undt3ch abe_music: and that issue is actually unrelated to the actual bootstrap script :|
17:50 s0undt3ch abe_music: which should also not disable gpg checking
17:50 abe_music s0undt3ch: right :(
17:50 forrest abe_music, can you blacklist that mirror first?
17:50 s0undt3ch abe_music: maybe that mirror blacklist option?
17:50 abe_music forrest, yeah i will try that…FYI it's mirrors.advancedhosters.com
17:50 forrest maybe there is something weird going on, I don't really see how chkconfig wouldn't be gpg signed, lol
17:50 UtahDave abe_music: s0undt3ch has done a lot of the work on bootstrap, terminalmage, too.   If you want to open an issue on the bootstrap git repo that might be a good way to open a discussion
17:51 a1j UtahDave: https://github.com/saltstack/salt/issues/7324  can this be assigned to 0.17 milestone? This is really 5 lines of code fix which will make world difference?
17:52 UtahDave a1j: I'm not sure it's going to make it. We've already cut the first RC.   I'll talk to Tom about it and see what he thinks.
17:52 abe_music forrest: i know, very strange…all the other 8-9 packages that need to be updated are signed fine. i pulled chkconfig manually and ran it through --checksig and it wasn't reporting the pgp flag
17:52 a1j oytherwise 2000 of minions firing schedule at the same time is not very proper thing
17:52 abe_music s0undt3ch: what's the terminalImage?
17:52 a1j UtahDave: ah ok then.
17:52 s0undt3ch abe_music: that's a nick ;)
17:52 forrest when you pulled it from the centos repo directly abe_music?
17:52 abe_music OHHH hahahah
17:52 abe_music my bad :)
17:53 abe_music thought it was some new fancy image lol
17:53 abe_music forrest, yup…pulled it using that link i pasted a few lines up
17:53 forrest that's very odd
17:54 abe_music forrest: black listed the mirror yum was using and the new mirror is also unsigned...geez
17:54 forrest That package was only added today
17:54 forrest well, they're just pulling from your link, so if that package is unsigned they all are
17:54 forrest I wonder if someone borked the build
17:54 forrest 23-Sep-2013 10:53 159K
17:54 forrest that's what I'm seeing as the date
17:54 forrest probably on GMT
17:55 abe_music forrest, yeah that has to be what happened. it's very surprising though. can't remember seeing this before
17:55 forrest If they don't upload a signed version today, it might be worth opening an issue with them
17:55 forrest yea I don't think I've ever seen an unsigned package from the centos repo
17:55 dthom91 joined #salt
17:56 abe_music forrest, will do…for now i'll rebuild my AMI after manually updating
17:56 robawt forrest: it wasn't my problem, i was just helping malinoff, but i feel like it's something that'll come up for me or my team
17:56 abe_music thanks forrest & s0undt3ch
17:56 forrest Ok, it might be worth installing on a VM first just for security purposes abe_music
17:56 forrest that would concern me
17:56 Ryan_Lane joined #salt
17:57 s0undt3ch abe_music: sorry I couldn't be of more help
17:57 abe_music no worries :)
17:57 forrest robawt, yea I know, I just know you and I were both working it, so I wasn't sure if you came up with a great solution that neither of us had already suggested.
17:57 bhosmer joined #salt
17:57 robawt i highly doubt it forrest, as you seem to be the one fixing all of my broken stuff
17:57 robawt :)
17:57 forrest lol
17:57 abe_music s0undt3ch & forrest, i'm wondering if maybe having an option to disable the repo updates would be alright? in our environment we'll have stock AMIs that will be updated manually by our IT staff. they might not like us messing with their workflow
17:58 abe_music just a thought
17:58 s0undt3ch abe_music: the current dev version of the script only does updates on request
17:58 forrest -U  If optional, don't fully upgrade the system prior to bootstrapping salt
17:58 s0undt3ch abe_music: still not stable
17:58 s0undt3ch forrest: that's still dev ;)
17:58 s0undt3ch I think
17:58 forrest shhhhhhhhhh
17:58 forrest dev works fine!
17:58 s0undt3ch I know it does
17:59 abe_music ahh, cool…i like living on the edge so i'll probably give it a shot :)
17:59 s0undt3ch it just hasn't reached the stable branch
17:59 s0undt3ch abe_music: the url is a diferent one
17:59 abe_music s0undt3ch: is there a way to get salt-cloud to use the dev branch for bootstrap?
18:00 forrest s0undt3ch, what testing still needs to be done before it's ready to go to stable?
18:00 s0undt3ch forrest: I need to clean up an issue with ubuntu 10.04
18:00 forrest ahh ok
18:00 whit joined #salt
18:01 s0undt3ch abe_music: sudo mkdir -p /etc/salt/cloud.deploy.d/ && sudo wget -O /etc/salt/cloud.deploy.d/bootstrap-salt.sh https://raw.github.com/s0undt3ch/salt-bootstrap/develop/bootstrap-salt.sh
18:01 s0undt3ch abe_music: that will make salt-cloud use whichever version is on /etc/salt/cloud.deploy.d/
18:02 kenbolton joined #salt
18:02 abe_music s0undt3ch: good to know, thanks…i'll get that going
18:02 morganfainberg joined #salt
18:02 s0undt3ch abe_music: anytime!
18:04 melinath joined #salt
18:05 hblanks left #salt
18:06 KyleG joined #salt
18:07 morganfainberg joined #salt
18:08 druonysus joined #salt
18:08 morganfainberg joined #salt
18:08 berto- joined #salt
18:09 pdayton joined #salt
18:10 mgw joined #salt
18:20 miva joined #salt
18:20 pdayton joined #salt
18:21 diegows joined #salt
18:21 tuxIO joined #salt
18:27 mmilano joined #salt
18:31 g4rlic joined #salt
18:35 fragamus joined #salt
18:36 pipps joined #salt
18:40 |Kellan| joined #salt
18:40 farra joined #salt
18:41 farra if I install the dev version of salt on my master, will my minions need to be updated?
18:41 Ahlee anybody successfully using psutil?  trying to call salt -G 'virtual:physical' ps.physical_memory_usage isn't returning anything
18:42 mgw joined #salt
18:42 Ahlee if i call python -c 'import psutil; ps.physical_memory_usage()' through cmd.run it returns
18:43 Ahlee and I know salt is running from the same python environment i'm calling with cmd.run
18:44 oz_akan_ joined #salt
18:50 farra joined #salt
18:53 jbunting joined #salt
18:55 oz_akan_ joined #salt
18:55 EugeneKay joined #salt
18:55 g4rlic question for any salt gurus in the house..
18:55 g4rlic What's the canonical way to handle iptables rules?
18:55 g4rlic I see that a state exists, which is good.
18:56 g4rlic But what would be the best way to use it?
18:56 g4rlic In my case: one particular node type will need say, 10 states applied to it.
18:56 g4rlic 4 of those states involve services that will need iptables configuration.
18:57 g4rlic Better to do a static firewall rule and link it in a single state? Or spread out the rules in each individual service state?
18:57 whit left #salt
18:57 morganfainberg joined #salt
18:57 bhosmer joined #salt
19:07 pipps joined #salt
19:08 racooper doing firewall rules (iptables, at least) in states was a bit tricky last time I looked at it...I know there was work on a state module for that, but I don't know the status right now
19:09 ldlework I don't understand /etc/salt/grains. Why would you want the same custom grains applied to all your servers?
19:10 rgbkrk joined #salt
19:13 faldridge joined #salt
19:14 faldridge left #salt
19:16 eaviles joined #salt
19:18 nu7hatch joined #salt
19:19 qba73 joined #salt
19:24 jbunting joined #salt
19:32 shinylasers joined #salt
19:33 nu7hatch joined #salt
19:36 cjh joined #salt
19:40 nu7hatch joined #salt
19:41 xinkeT joined #salt
19:42 matanya joined #salt
19:45 jcockhren ldlework: you mean the grains file on the master?
19:45 ldlework jcockhren: yep
19:46 jcockhren ldlework: you sure it's on the master?
19:47 jcockhren ldlework: did you happen to install the minion deamon as well
19:47 shinylasers joined #salt
19:48 ldlework jcockhren: I'm not sure what you think I'm asking
19:48 jcockhren noted
19:48 shinylasers joined #salt
19:48 ldlework I'm saying, what's the point of having a grains file on the master that is applied to all servers.
19:49 ldlework Isn't the grain info supposed to differentiate servers? If so, why would you apply some grain to all your servers (by defining it on the master and having it pushed to your servers)
19:49 ldlework (I admit I am probably confused, but asking to clear that up :)
19:49 jcockhren /etc/salt/grains applies to minions
19:50 jcockhren that section of the docs speaks of precedence of grains for minions
19:50 ldlework :|
19:50 jcockhren minion config > /etc/salt/grains
19:50 ldlework jcockhren: okay assume you don't put grains in a minion config
19:50 ldlework what's the point of a master-located grain source?
19:51 micah_chatt joined #salt
19:52 jcockhren why do you think grains are defined on the master?
19:52 pipps1 joined #salt
19:52 jcockhren there's http://docs.saltstack.com/topics/targeting/grains.html#writing-grains
19:53 ldlework Custom grains should be placed in a _grains directory located under the file_roots specified by the master config file. They will be distributed to the minions when state.highstate is run, or by executing the saltutil.sync_grains or saltutil.sync_all functions.
19:53 jcockhren now.... I could be missing you. and everyone here could be snickering
19:53 ldlework *Why* would you put grains on the server?
19:53 ldlework the master server, I mean
19:53 ldlework Wouldn't this apply the same grain info to all the minions, thereby making that grain information useless for differentiating between different minions?
19:54 jcockhren those grains are called to deteremine grain specific key:value parts dynamically
19:54 jcockhren notice they are functions
19:54 Ahlee ldlework: You can write python to return your values
19:54 Boohbah joined #salt
19:54 Ahlee i.e., i can write a grain that can query a database to determine something
19:55 Ahlee look at https://github.com/saltstack/salt/blob/develop/salt/grains/core.py
19:55 mesmer joined #salt
19:55 ldlework Ahlee: that makes wayyyy more sense
19:55 ldlework Ahlee: I thought grains were just static yaml data
19:56 Ahlee ldlework: so did I at first, and had the exact same thought process you did.
19:56 jcockhren Ahlee: 1, jcockhren: 0
19:56 ldlework jcockhren: thanks for listening though
19:56 * ldlework huggles jcockhren
19:58 Ahlee That said, I do'nt actually use _grains or know how to write them ;)
19:58 Ahlee hrm. I thought there were helper functions to return info, like help grains
19:59 jbunting joined #salt
20:01 nu7hatch joined #salt
20:02 pipps2 joined #salt
20:02 Ahlee https://github.com/saltstack/salt/issues/634 - yet closed without providing additional info
20:03 Ahlee the linked doc https://github.com/saltstack/salt/issues/634 is how you distro them (which I can attest with my custom _modules works), but not how you add write _grains or append to 'em
20:04 g4rlic racooper: OK, that's what I was worried about.
20:04 g4rlic It looked like the documentation for the state itself was kind of threadbare, and mentioned it might be deprecated later.
20:04 g4rlic I'm tempted to just apply a static ruleset based on a grain instead.
20:05 Ahlee ah, sys.doc is what i was forgetting
20:10 mesmer joined #salt
20:11 Heartmender joined #salt
20:13 kenbolton joined #salt
20:14 ronc joined #salt
20:15 diegows joined #salt
20:16 mesmer joined #salt
20:20 Tekni joined #salt
20:21 Heartmender joined #salt
20:21 Heartmender joined #salt
20:34 carlos__ joined #salt
20:35 Boohbah joined #salt
20:36 tuxIO_ joined #salt
20:37 dthom91 joined #salt
20:37 morganfainberg joined #salt
20:38 mgw joined #salt
20:39 dthom91 joined #salt
20:40 mesmer joined #salt
20:42 devinus joined #salt
20:46 mesmer joined #salt
20:49 nu7hatch joined #salt
20:52 mesmer joined #salt
20:54 jbunting joined #salt
21:01 pipps1 joined #salt
21:01 lacrymology how do I pass variables to the __opts__ dict when calling states with salt-call state.high?
21:02 drdran joined #salt
21:02 UtahDave lacrymology: the __opts__ dict is created by Salt's loader.  you can pass in pillar data from the cli, if you'd like
21:03 drdran left #salt
21:04 lacrymology UtahDave: yeah, that, how do I do it? sorry, I'm a python dev, not so used to salt itself yet, my clients know their needs pretty well, but I'm still getting used to all this
21:04 pipps1 joined #salt
21:06 UtahDave lacrymology: well, have tried using pillar data yet?
21:08 lacrymology UtahDave: no. I think I understand how it works, but I don't know how to set it up from the command line..
21:09 lacrymology and I'm not even sure what it is yet. I've seen it used to pass module parameters, but not to set up options ('test': True is what I need to test)
21:09 forrest lacrymology, have you looked at the pillar docs yet?
21:09 pipps1 joined #salt
21:10 lacrymology forrest: I know I started reading, but I'm not even sure if I finished. I can deduce from my very words that in any case I didn't absorve much of it
21:10 forrest ahh ok, well think about the pillar data like global variables almost
21:10 forrest https://github.com/terminalmage/djangocon2013-sls/blob/master/pillar.example
21:11 forrest if you look at that, it's an example of pillar being used in a real project. So you can see there that values are set (via jinja2), and then defined lower, so when you call them, you can do a look for something like foo:venv
21:11 lacrymology forrest: yes, I understood that much. And I understood you can specify which pillars go to which minions
21:11 forrest ok cool
21:12 oz_akan_ joined #salt
21:14 cwright i've got a pkgrepo.managed apt repo with an order: 1, and it gets added, but it seems apt-get update isn't run before my packages are attempted to be installed
21:14 ede joined #salt
21:14 cwright i've tried adding a cmd.wait watcher on it, but that isn't doing the trick either.  what is the recommended way to ensure apt-get update is run after a new pkgrepo is added?
21:14 cdunklau left #salt
21:15 cwright I'd rather not add a requires: pkgrepo for all my individual packages unless I have to
21:15 forrest cwright, are you on 0.17?
21:15 lacrymology forrest, UtahDave: I'm trying $ salt-call state.high '{"foo": {"uwsgi": ["enabled"]}}' pillar='{"test": true}' but it complains that state.high takes only one argument
21:16 forrest if not, how do you currently handle ensuring the pkg repo is created prior to installing the package without a require?
21:16 cwright not yet.  i didn't realize there were ubuntu packages available for 0.17 yet
21:16 cwright order: 1
21:16 forrest I don't know if there are honestly, I've been gone all weekend
21:16 forrest oh duh you said that earlier
21:16 cwright it installs the apt sources.list.d file correctly, it just doesn't apt-get update
21:16 forrest right
21:17 forrest how about
21:17 forrest require_in
21:17 forrest Set this to a list of pkg.installed or pkg.latest to trigger the running of apt-get update prior to attempting to install these packages. Setting a require in the pkg will not work for this.
21:17 pipps1 joined #salt
21:18 cwright well, that's not really any different than adding a require to each package.  i'd prefer to allow it to be more flexible so that when I add more packages I don't have to go back through my state files and update it elsewhere
21:18 cwright i just want it to apt-get update after adding the pkgrepo :)
21:19 forrest can you try to use a cmd.run?
21:20 forrest then it will run every time the state is applied, ensuring the update is there.
21:20 liquidautumn joined #salt
21:20 forrest or do you not want to do that either?
21:21 cwright well, the apt-get update adds quite a lot of time to the deploy.  i'd like to only run update when the aptrepo changes
21:21 forrest yea fair enough
21:21 forrest so cmd.wait doesn't work, even if you modify the file value?
21:21 cwright i could do that, but i am looking for a specific solution here, and i know it must exist
21:21 forrest yep, just going through possible solutions.
21:21 cwright i just started with a fresh virtual machine and it still didn't do it
21:22 druonysus joined #salt
21:22 cwright i guess i need to learn more about when salt will try to do an apt-get update
21:22 forrest what about trying to use cmd.mod_watch
21:23 forrest then you could set onlyif conditions on there.
21:23 forrest and could you show us your sls where you have the cmd.wait?
21:24 cwright the documentation for mod_watch seems kinda scarce.  is there an example anywhere that you know of that demonstrates how to use it?
21:24 forrest I honestly don't
21:25 forrest the docs for cmd.run are a bit better, where you can use onlif
21:25 forrest *onlyif
21:25 forrest but I don't see any examples there either
21:25 cwright forrest: this is what i've got https://gist.github.com/corywright/6677184
21:25 cwright i changed out the hostname but otherwise thats it
21:26 faldridge joined #salt
21:30 forrest Did you run that in debug yet cwright?
21:31 cwright no
21:31 cwright well
21:31 forrest can you try it? Maybe that will give us a hint
21:31 cwright i've got logging turned up to garbage and i use salt -v
21:31 forrest ahh
21:32 cwright its ok, i'll keep digging.  thanks for your help.  just thought I'd check to see if anyone had a quick answer
21:32 forrest yea sorry man, I don't see anything wrong with how you've got it going, let me know if you figure out what it is
21:33 cwright will do
21:33 forrest thanks!
21:43 critbit joined #salt
21:43 kermit joined #salt
21:45 anuvrat joined #salt
21:47 aparashar_ joined #salt
21:57 mesmer joined #salt
22:05 mesmer joined #salt
22:09 mesmer joined #salt
22:14 |Kellan| joined #salt
22:15 cdsrv joined #salt
22:15 piffio I'm playing with the grains.setval function
22:15 piffio but I can't figure out a way to set a list as a value
22:15 piffio what I want to achieve is basically run a setval that can set a grain such as
22:15 piffio roles:
22:15 piffio - webserver
22:15 piffio - memcache
22:16 nu7hatch joined #salt
22:16 piffio not sure this is possible to achieve with the grains module
22:16 cdsrv make a compound grain?
22:16 piffio yep
22:17 piffio I guess I need to use salt.modules.grains.append(key, val) instead of setval
22:17 piffio but it's going to be introduced with 0.17 accordingly to the documentation
22:18 cdsrv nice
22:18 piffio as well as grains.delval :(
22:19 cdsrv anyone using multiple file_roots ?
22:19 cdsrv the documentation says 'subdirectories in the multiple file roots cannot match' ..
22:19 cdsrv this seems problematic
22:20 cdsrv however the example seems to go against the recommendation
22:20 piffio cdsrv: I haven't tried it
22:20 piffio I'm trying to go with pure gitfs
22:20 cdsrv the notes seem to indicate some kind of ambiguity in pattern matching.. quite unusual
22:21 cdsrv will have to find out !
22:21 cdsrv example :  debian/etc/default ; redhat/etc/default
22:22 cdsrv seems reasonable, not sure what the comments in /etc/salt/master mean when referring to file_roots
22:22 oz_akan_ joined #salt
22:26 mesmer joined #salt
22:27 ronc joined #salt
22:27 lacrymology I've got to create a state that's a wrapper for file.managed and module.wait, examples I've seen for this do __salt__['state.high'] and pass it a state (dict). What's the real difference between this and doing `from salt.states.file import managed`....?
22:31 offby1 left #salt
22:33 sibsibsib_ joined #salt
22:34 alunduil joined #salt
22:43 luketheduke joined #salt
22:48 jfalco joined #salt
22:50 raghavp joined #salt
22:50 g4rlic I found something regarding salt and iptables, but it doesn't appear to be using the default iptables state. http://www.nineproductions.com/saltstack-iptables-state/
22:50 g4rlic hmm.
22:51 raghavp left #salt
22:53 mwillhite joined #salt
22:55 rgbkrk joined #salt
22:57 mesmer joined #salt
22:57 forrest g4rlic, that just looks like a managed file that's a template
22:58 g4rlic Right.
22:58 g4rlic I'm nto sure which way is "right", per se.
22:59 g4rlic Using templates, grains, etc., or using the iptables.append state
22:59 g4rlic eg: make my firewall rules modular and pinned to a configured service, or make them monolithic and per host?
22:59 forrest Yea, for me it would probably come down to, how many different OSs I had in my environment, and how many rules I had
23:00 g4rlic 2, ruleset depends on role.
23:00 forrest I mean if your whole env only has one set of rules, then the managed file is probably the easiest
23:00 forrest ahh
23:00 g4rlic The reason I was looking to control it via individual services, is that one node currently has a *ton* of states applied to it.
23:01 g4rlic Eventually I'd like to split out those states to > 1 machine.
23:01 forrest yea that's very fair reasoning
23:01 forrest it would make it a lot easier to expand on if you do it per service
23:02 g4rlic Right, but there's a question or two I have about that..  like, for example, is the iptables rules generated by iptables.append default allow, or default deny?  I suppose I should just set it up and see what happens.  I have console into these systems out of band.
23:02 forrest yea a good question
23:05 forrest In your situation I'd go for each one being service based
23:05 forrest yea it's a pain now, but what about 6 months from now, 2 years, etc.
23:05 forrest eventually you WILL have to break it out
23:05 forrest so you might as well do it now
23:05 forrest and avoid that extra work down the line
23:06 forrest unless you want to write some monolithic template based set of iptables :P
23:07 forrest I've done that before in puppet, and it's not nice
23:09 kenbolton joined #salt
23:15 bhosmer joined #salt
23:15 mesmer joined #salt
23:18 Brew joined #salt
23:20 jbunting joined #salt
23:23 jbunting joined #salt
23:25 mesmer joined #salt
23:25 GoKage joined #salt
23:29 LordOfLA|Weechat joined #salt
23:30 pjs_ joined #salt
23:31 mesmer joined #salt
23:32 pipps1 joined #salt
23:35 f47h3r hello folks... I've got a quick question regarding salt-ssh and the options it parses
23:35 f47h3r it would seem that the --roster option fails to change the roster file upon run...
23:35 piffio joined #salt
23:36 f47h3r they parsers.py file is kinda convoulted... where is the parser that adds the roster file to the config / targets structure?
23:36 f47h3r the*
23:37 f47h3r anyone alive in here?
23:37 g4rlic nope.
23:37 f47h3r should I log a bug against github?
23:38 f47h3r or is there another way to specify the roster file, such as in the master config?
23:38 joehh f47h3r: I suspect a bug is worthwhile - worst case it can be closed
23:39 f47h3r sweet, I'd just fix it and submit a pull request, but without digging into the code for awhile to figure out if it's missing in the optparse parse_args
23:40 f47h3r I'll log a bug... thanks!
23:42 lacrymology https://salt.readthedocs.org/en/latest/ref/states/writing.html#watcher-function << HOW is the watcher function declared, exactly?
23:43 joehh I don't know that bit of code at all, but in general pull requests are received gratefully
23:45 joehh lacrymology: I reckon that is worth a bug against the docs, but looking at salt/states/cmd.py, it appears to be def mod_watch(name)
23:45 joehh though I'm not sure what it returns
23:47 f47h3r joehh: isssue logged... is there someone that is more familar with that area of code?
23:48 joehh That would be a question for UtahDave (ie who is more familar, not necessarily the details of the code)
23:48 lacrymology joehh: yeah, I was looking at service, and I saw the watcher was called mod_watch, but I didn't realize it was a "magic name" until I saw the ame in module
23:49 lacrymology I read about states using module.run and module.wait, but I can't find anything about how to actually.. well, MAKE a state that uses them.. any doc pointers?
23:51 mesmer joined #salt
23:53 rasp joined #salt
23:55 pipps1 joined #salt
23:58 devinus joined #salt
23:58 UtahDave f47h3r: thanks for filing that bug. salt-ssh is very very  new.  You happened to catch us at a really busy time. Several of us are conducting interviews and others are traveling.
23:59 Ryan_Lane1 joined #salt
23:59 Ryan_Lane joined #salt

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