Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2013-10-09

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

All times shown according to UTC.

Time Nick Message
00:09 micah_chatt joined #salt
00:09 oz_akan_ joined #salt
00:09 danielmcbawse joined #salt
00:11 redondos joined #salt
00:13 m_george|away joined #salt
00:17 justlooks joined #salt
00:18 Marion salt/pki directory did not have correct permissions.
00:19 DanGarthwaite I'm quite happy with vbox.  You just have to set up host only mode: http://www.virtualbox.org/manual/ch06.html#network_hostonly
00:19 troyready joined #salt
00:20 Marion Also software-properties-common not installed.  I have to bring up the VM, ssh in and install it manually.  Is there anyway to change my config so this will install during provision?
00:22 xmltok joined #salt
00:22 mafrosis joined #salt
00:23 pentabular1 joined #salt
00:24 justlooks hi ,in python script how to use pillar dict?
00:27 cshuman joined #salt
00:29 Ryan_Lane joined #salt
00:30 cshuman joined #salt
00:31 m_george left #salt
00:31 larstr joined #salt
00:33 Ryan_Lane joined #salt
00:34 troyready joined #salt
00:41 jbunting joined #salt
00:42 Ryan_Lane joined #salt
00:43 druonysus can I use pillar['something'] inside pillar?
00:44 Ryan_Lane joined #salt
00:45 CheKoLyN joined #salt
00:57 forrest joined #salt
00:57 justlooks hi ,i have a lot of file need to be create they have same user and group ,can i write the user and group as alias then write the alias in file state ?
00:57 jesusaurus druonysus: i would hope not. at best you would have incomplete data because the data is still being compiled
00:59 druonysus jesusaurus: hmm... okay. I suppose that makes sense
01:05 jkleckner joined #salt
01:10 micah_chatt joined #salt
01:10 justlooks how to use grains dict in python code?
01:11 justlooks i try __grains__['key'] no use ,and salt.grains['key'] still no use
01:12 dthom91 joined #salt
01:14 JaredR left #salt
01:16 halfss joined #salt
01:16 justlooks anyone know how to use grains dict in sls file (use python code) ??
01:17 rgbkrk joined #salt
01:19 packeteer looks like i've landed a job :) now I get to use salt for realz
01:20 mwillhite joined #salt
01:22 akasto joined #salt
01:22 redondos joined #salt
01:22 redondos joined #salt
01:24 justlooks anyone know how to use grains dict in sls file with python template??
01:25 cachedout_home left #salt
01:27 srini_ joined #salt
01:27 packeteer i've been wresting with grains, but haven't made it that far
01:29 deepakmd_oc joined #salt
01:29 justlooks packeteer:  i think if i can use grain dict in jinjia template sls file ,i can also use it in python template sls file ,no one knows?
01:30 srini_ left #salt
01:31 Lue_4911 joined #salt
01:34 Ryan_Lane joined #salt
01:39 jslatts joined #salt
01:41 mapu joined #salt
01:42 Lue_4911 joined #salt
01:44 rgbkrk joined #salt
01:47 mgw joined #salt
01:48 _0x4b joined #salt
01:51 justlooks anyone know how to use grains dict in sls file with python template??
01:54 cetex joined #salt
01:58 jslatts joined #salt
02:00 jslatts joined #salt
02:04 dthom91 joined #salt
02:06 xmltok_ joined #salt
02:08 xl1 joined #salt
02:08 carmony joined #salt
02:08 fxhp joined #salt
02:09 jesusaurus justlooks: im not sure about a python template, but from jinja its grains['key'] or salt['grains.get']('key', 'default_value')
02:13 justlooks jesusaurus: thanks i tried ,just use grains ,it's work fine ,my sls file has another error , so it get runtime error
02:23 mua joined #salt
02:28 justlooks can i use multi-value for one key in pillar file?( key : [ val1,val2,val3 ]) if can ,how can i iterate the value list?
02:36 justlooks i want to list this in my pillar file ( hostname: [packagename1,pkgname2,pkgname3] ) ,and if my minion name is XX  let it install XX packages ,if name is YY ,let it install YY packages ,and i write "{% for i in pillar['grains['host']'] %} in sls file
02:36 justlooks is it can work?
02:37 tyler-baker joined #salt
02:37 tyler-baker joined #salt
02:41 hummus3 joined #salt
02:42 justlooks i tested it ,it worked  the code need modified as {% for i in pillar[grains['host']] %}  lol
02:44 ubercore joined #salt
02:57 jdenning joined #salt
03:07 jefimenko joined #salt
03:15 cedwards1 joined #salt
03:16 backjlack_ joined #salt
03:16 emilisto_ joined #salt
03:16 tempspace_ joined #salt
03:17 Ahlee joined #salt
03:17 puppet__ joined #salt
03:17 LarsN_ joined #salt
03:17 darrend_ joined #salt
03:17 lemao_ joined #salt
03:17 drogoh_ joined #salt
03:18 ze-_ joined #salt
03:20 zz_dane joined #salt
03:20 BrunoPolaco joined #salt
03:21 melinath_ joined #salt
03:21 marcinkuzminski_ joined #salt
03:24 ebodine joined #salt
03:24 Parabola joined #salt
03:24 dccc joined #salt
03:25 xl1 joined #salt
03:28 fxhp joined #salt
03:34 justlooks did salt has notify function? i want to format minion A service ,then start the service from A,then notify minion B to sync data from A service ,then start service in minion B
03:34 justlooks any good suggestion?
03:36 Corey justlooks: Peer Publish
03:36 justlooks watch function only useful in same host
03:46 cshuman joined #salt
03:47 cshuman joined #salt
03:49 justlooks Corey: so in peer communication ,minion has the ability to order master to send command to other minion?
03:50 Corey justlooks: Ah, if that's what you're trying to do, look at the event reactor subsystem.
03:51 jaequery joined #salt
03:51 justlooks event reactor subsystem and peer communication is the same thing?
03:51 Nexpro joined #salt
03:52 Corey No, it is not.
03:57 jgelens joined #salt
03:59 raghavp joined #salt
04:00 raghavp left #salt
04:01 dthom91 joined #salt
04:12 danielmcbawse joined #salt
04:16 redondos joined #salt
04:24 rgbkrk joined #salt
04:24 jefimenko joined #salt
04:31 pentabular joined #salt
04:52 jalbretsen joined #salt
04:56 mua joined #salt
05:04 Furao joined #salt
05:09 rawzone joined #salt
05:10 druonysus joined #salt
05:11 jaequery joined #salt
05:12 andrewclegg_ joined #salt
05:13 jaequery joined #salt
05:15 ggherdov joined #salt
05:17 pentabular joined #salt
05:19 oz_akan_ joined #salt
05:25 redondos joined #salt
05:32 Sypher joined #salt
05:34 Katafalkas joined #salt
05:38 redondos joined #salt
05:38 redondos joined #salt
05:41 Furao kiorky: ce que luminous demande ici https://github.com/saltstack/salt-cloud/issues/615#issuecomment-25768815 est quand meme justifié
05:56 Marion joined #salt
06:02 gamingrobot joined #salt
06:03 Furao 0.17.0 ppa are ready now
06:05 packeteer nice! thanks
06:05 toastedpenguin joined #salt
06:05 Corey Baaaaaaaaaaaah!
06:05 Corey That SUCKED.
06:06 Furao Corey: you mean the sphinx error?
06:06 Corey Furao: All of it.
06:06 Corey The Sphinx issue is the most recent incarnation, but not worth holding up the rest over.
06:07 Corey It's not a packaging error as best I can tell, it's the same packaging instructions.
06:09 Furao https://groups.google.com/forum/#!topic/sphinx-users/tkByKOJR2Zw not very usefull
06:09 justlooks i use salt 0.16.3 ,why i can not find reactor option in /etc/salt/master configuration file?
06:09 Corey justlooks: Because it's not comprehensive. Go read: http://docs.saltstack.com/topics/reactor/index.html
06:11 justlooks Corey: so how can i react a event?
06:11 justlooks Corey: which from minion
06:11 Katafalkas joined #salt
06:12 esogas_ joined #salt
06:14 Furao Corey: python-sphinx in raring is 1.1.3+dfsg-7ubuntu2 and precise is 1.1.3+dfsg-2ubuntu2
06:14 Furao maybe you should try to install precise .deb in raring and try again
06:15 MKFG joined #salt
06:15 Corey Furao: yeah, that's the hell of it. The diff isn't significant.
06:15 marcinkuzminski joined #salt
06:16 imil_ joined #salt
06:16 minaguib_ joined #salt
06:17 andyshinn` joined #salt
06:18 \ask joined #salt
06:19 woebtz joined #salt
06:21 gmoro joined #salt
06:22 nkuttler_ joined #salt
06:23 Furao test suite launched against 0.17.0, let's wait 12 hours for results :)
06:24 packeteer how do I set global grains? i have some entries in /etc/salt/grains but they're not being read
06:24 Corey Furao: Let me know what I broke.
06:25 denstark_ joined #salt
06:25 blast_hardcheese joined #salt
06:26 Furao Corey: your packaging is never broken, it's always the "content" of it that is, such as 0.15.x -> 0.16.x that renamed few grains that broke all my states
06:26 Ryan_Lane joined #salt
06:26 matanya_ joined #salt
06:27 \ask joined #salt
06:28 Furao first thing it does is upgrade 0.15.x to 0.17, the upgrade seem to had been succesfull
06:28 Furao 06:27:47 2013-10-09 06:27:47,500 Minion integration-salt-0-17-0-1 is finally up after 4 seconds
06:31 Corey Furao: Yeah, this is a different style issue-- namely the same thing works just fine on other releases.
06:31 oz_akan_ hi
06:32 Furao 06:27:42   File "/usr/lib/pymodules/python2.7/salt/modules/cmdmod.py", line 384, in run
06:32 Furao 06:27:42 AttributeError: 'NoneType' object has no attribute 'STDOUT'
06:32 craig_ joined #salt
06:32 bashcode` joined #salt
06:32 Katafalkas joined #salt
06:32 kvbik_ joined #salt
06:32 oz_akan_ does anyone have issues with mine on salt 0.16.4
06:32 Furao weird
06:32 Corey Furao: Right?
06:32 Corey I changed *nothing*. This was programmatic, not a manual process, FYI.
06:32 Furao Corey: yes it is :)
06:32 Furao yes I know I did debian packaging for years
06:33 Furao I know what it is to be blamed for packaged content :P
06:34 Corey Furao: This may not be helpful, but I despise sphinx / ronn. "Dynamically generate the docs!" How about you statically compile that, and put it in a branch like a bloody adult?
06:35 Furao Corey: what is great with sphinx is the ability to generate doc from pydoc strings
06:35 Furao and also produce multiple output format
06:36 Furao in raring case it's manpages that fail
06:36 Corey Furao: Right, great, I don't care. :-) Do that RIGHT before a tag, stuff it somewhere.
06:36 Corey The packaging will pick THAT up and remove this problem.
06:36 joehh Furao: how do you tell that?
06:36 Katafalkas joined #salt
06:37 Furao joehh:   File "/usr/lib/python2.7/dist-packages/sphinx/writers/manpage.py", line 344, in unknown_visit
06:37 Furao raise NotImplementedError('Unknown node: ' + node.__class__.__name__)
06:37 Furao NotImplementedError: Unknown node: inline
06:38 Corey Yes.
06:38 Corey THAT is the issue.
06:38 joehh that is pretty obvious isn't it :)
06:39 Corey So... "what changed."
06:39 * Furao looking into the code...
06:39 Corey It's not the packaging, I promise. :-)
06:40 Furao seem to be in python-docutils and not python-sphinx
06:40 Furao precise/quantal == 0.8.1-4 raring == 0.10-1
06:41 Furao saucy 0.11-2
06:41 joehh sid is 0.11-2 too
06:41 Corey And sid works, in case that wasn't clear. :-)
06:41 joehh looks like debian missed the 0.10 release so didn't have the issue
06:41 Furao joehh: what is version python-sphinx?
06:41 joehh in sid?
06:41 tomspur joined #salt
06:41 tomspur joined #salt
06:42 Furao joehh: yes
06:42 joehh sid: 1.1.3+dfsg-8
06:42 joehh wheezy/stable: 1.1.3+dfsg-4
06:42 Furao probably 1.1.3
06:42 Furao yes
06:42 joehh ys
06:42 joehh with very minor diffs between the ubuntu one
06:42 Furao Corey: you can try to downgrade/upgrade 0.10?
06:43 joehh We can probably backport 0.11
06:43 Furao joehh: you mean move saucy package to salt ppa?
06:44 danielmcbawse joined #salt
06:44 joehh effectively
06:44 Furao that might break other non salt related stuff, but that should be first tested (I mean downgrade/upgrade)
06:45 joehh or now we know a bit more what to look for, modify the docs
06:45 jkleckner joined #salt
06:45 joehh I suspec that might be easier
06:45 Corey Furao: I can't. That's the hell of it. I can't change the build environment for the PPA.
06:46 Furao ah I see, it's their build setup
06:46 joehh I'll see if I can find the bit that is causing the problem tonight and patch the docs
06:46 Corey Exactly. I can do anything to the build scripts, but nothing to the build environment.
06:47 bud joined #salt
06:47 joehh joehh: gotta go - I'll catch up on the irc logs tonight
06:47 joehh thanks corey
06:47 Corey joehh: Talking to yourself is a sign of insanity.
06:47 Corey joehh: No, thank you.
06:48 joehh thanks Furao
06:48 joehh yeah true :)
06:49 matanya_ joined #salt
06:50 Furao joehh: no problem, sorry I can't help much as I don't have anything running raring
06:52 Katafalkas joined #salt
06:53 kiorky Furao: if i say 4 times it needs my fork, it needs my fork, point.
06:53 larstr joined #salt
06:55 Furao ahah
07:00 \ask joined #salt
07:01 balboah joined #salt
07:13 iMil joined #salt
07:14 ckao joined #salt
07:16 justlooks Corey:  which version support reactor?
07:16 jesusaurus joined #salt
07:18 alunduil joined #salt
07:19 [M7] joined #salt
07:20 [M7] joined #salt
07:20 freelock joined #salt
07:20 mortis joined #salt
07:21 Furao joined #salt
07:25 justlooks Corey: Salt version 0.11.0 introduced the reactor system ,why i can not use it in 0.16.3 version?
07:27 adepasquale joined #salt
07:32 elfixit joined #salt
07:33 Katafalkas joined #salt
07:36 redondos joined #salt
07:36 Furao The conflicting ID is "virtualenv" and is found in SLS "base:virtualenv.absent" and SLS "base:virtualenv.absent"
07:39 Furao fun just started with 0.17
07:40 kallek joined #salt
07:44 matanya_ joined #salt
07:45 aleszoulek joined #salt
07:47 adepasquale joined #salt
07:50 felixhummel joined #salt
07:51 zooz joined #salt
07:55 c0bra joined #salt
08:07 krissaxton joined #salt
08:07 travisfischer joined #salt
08:11 TheCodeAssassin joined #salt
08:13 Furao kiorky: why you where is working version of your fork?
08:14 Furao bfdsafsad
08:14 Furao I mean
08:14 Furao kiorky: where is working version of your fork?
08:16 Furao in https://github.com/makinacorpus/salt/blob/docker/salt/modules/lxcdocker.py#L175 you called your module "docker" but here you refer to it with https://github.com/makinacorpus/salt/blob/docker/salt/states/lxcdocker.py#L82 as "lxcdocker"
08:19 kiorky Furao: because we added with techhat the virtual later
08:19 kiorky Furao: state module is not done and not usable right now.
08:19 Furao ok pas de prob
08:20 az87c joined #salt
08:21 az87c_ joined #salt
08:22 krissaxton left #salt
08:36 ml_1 joined #salt
08:38 whiskybar joined #salt
08:38 justlooks hi,anyone can help me https://gist.github.com/justlooks/6898169
08:40 s0undt3ch joined #salt
08:43 justlooks no one?
08:44 rmt2 justlooks, most of the developers (which are currently the experts) are based in the US, and tend to only show up for a short period during the US day and then for a while in the evening, it appears. I haven't been around long enough to see if this is a permanent trend or something temporary, though.
08:45 rmt2 justlooks, you might have more luck with the mailing list.
08:46 justlooks rmt2: thank you ,
08:46 jpihl joined #salt
08:46 rmt2 I've also been having trouble getting answers from them.
08:46 crashmag joined #salt
08:50 backjlack joined #salt
08:55 Furao joined #salt
08:57 Katafalkas joined #salt
09:00 raghavp joined #salt
09:00 raghavp left #salt
09:01 jkleckner joined #salt
09:09 jinnko joined #salt
09:13 shiin joined #salt
09:13 shiin how comes theres no PEAR module for salt stack? I didnt find anything on that topic.
09:20 unicoletti_ joined #salt
09:28 Furao shiin: module that does what?
09:29 backjlack joined #salt
09:32 shiin Furao: installs pear packages
09:32 shiin there is one for PECL, pip, npm, ruby gems..
09:32 shiin but no pear
09:35 rmt2 PECL..How did that get in there.. I'll file a bug to have it removed.
09:36 shiin right, it could be people dont like pear
09:36 shiin but then there should be a discussion somewhere?
09:36 rmt2 Or that noone has contributed a pear module, therefore one doesn't exist. It is an open source project.  You could of course pay SaltStack the company to write one for you.
09:37 shiin indeed I could. or I could write one. but then I wouldnt be the only person who wants that, and there still might be some discussion about it already.
09:39 Furao shiin: I wrote plenty of modules/states some are in official repos, it's not that hard
09:40 shiin so I see. Im a python beginner though.
09:41 bhosmer joined #salt
09:41 ninkotech shiin: dont be afraid, only 1-2 hours or days are needed to learn python.  then you are good to go :)
09:41 Furao yes, get away from PHP it's good for mental health :P
09:42 ninkotech ... well, but salt is somewhat hard for beginners, thats true
09:43 Furao ninkotech: true
09:44 ninkotech Furao: all other content management systems are even more crazy, if i can say -- except fabric, which is simple and fast to learn
09:44 Furao ninkotech: content management?
09:44 Furao config!
09:44 ninkotech but again, fabric is more like for deployment, right
09:44 Furao fabric is just python over bash
09:44 ninkotech Furao: oah, wrong term, wrong decade, lol
09:44 ninkotech config management, yes, :)
09:44 Furao http://en.wikipedia.org/wiki/Content_management_system
09:44 Furao is totally different
09:45 ninkotech Furao:  i was doing content management 10 years ago a lot, so its in my fingers :)
09:46 ninkotech now i write my large scale filesystem using python -- lots of fun :)
09:49 Furao if I was going to start something like that in 2013, I would do it in go instead
09:49 * Furao never wrote a line of go, but I wish to taste the koolaid
09:50 Furao I switched from perl to python 10 years ago, I can't blame you :)
09:53 Furao I wonder if it's hard to run go on android...
09:57 xiaoqiangnk joined #salt
10:02 shiin Ive opened a github issue now to discuss the matter.
10:02 jkleckner joined #salt
10:04 gildegoma joined #salt
10:05 jbunting joined #salt
10:10 Katafalkas joined #salt
10:32 terminalmage joined #salt
10:45 ml_1 joined #salt
10:52 jumperswitch joined #salt
10:56 giantlock joined #salt
10:56 mugsie joined #salt
11:02 ranl joined #salt
11:04 Marion joined #salt
11:10 ranl hi guys, did someone played a bit with the state.sls runner ?
11:12 aptiko joined #salt
11:22 uta joined #salt
11:25 matanya_ joined #salt
11:26 ml_1 joined #salt
11:28 ninkotech Furao: i was tempted to try javascript :)    but python is so lovely
11:28 ninkotech erlang would be fun to learn...
11:28 ninkotech life is short -> python
11:31 uta hey does anybody know when 0.17 is going to be the current release?
11:32 rmt2 uta, 0.17.0 is the current release. I think there's been some delay in getting the ubuntu & debian releases out the door, though.
11:35 uta rmt2: ah okay. Do you know if the redhat/centos version usually gets out about the same time as ubuntu/debian?
11:36 rmt2 I don't, sorry.
11:37 uta rmt2: no worries, thanks for the help
11:39 HastaJun joined #salt
11:41 ninkotech hmm topic says 0.16.4
11:41 shiin only the ubuntu raring one seems broken
11:42 shiin now that 0.17 hotfix changed the way nodenames are composed again. great.
11:47 jetblack joined #salt
11:55 diegows joined #salt
12:04 jkleckner joined #salt
12:06 uta joined #salt
12:12 copelco joined #salt
12:13 blee joined #salt
12:18 ml_1 joined #salt
12:19 jrdx joined #salt
12:23 jslatts joined #salt
12:30 bhosmer joined #salt
12:35 jumperswitch joined #salt
12:39 ram joined #salt
12:39 ram hi all
12:39 Guest98168 hi
12:40 HastaJun left #salt
12:40 ipmb joined #salt
12:40 Guest98168 i need a help guys
12:42 mwillhite joined #salt
12:45 oz_akan_ joined #salt
12:45 danielbachhuber joined #salt
12:47 oz_akan_ joined #salt
12:47 Guest98168 hi all
12:47 Guest98168 i like to learn saltstack like programming language
12:48 Guest98168 i need to write dictionary based exact coding
12:50 shiin so what kind of help do you need?
12:51 amahon joined #salt
12:51 mosulica joined #salt
12:56 mgw joined #salt
13:00 wkf joined #salt
13:01 anteaya joined #salt
13:03 wkf joined #salt
13:06 Boohbah joined #salt
13:08 racooper joined #salt
13:08 jergerber joined #salt
13:09 mwillhite joined #salt
13:11 Guest98168 hi thanks for ur reply
13:12 Guest98168 i need to write state file
13:12 shiin state files are fun.
13:12 Guest98168 so i need to learn salt coding
13:12 Guest98168 yes your right
13:12 shiin I think its crucial to do so by getting practice.
13:14 Guest98168 s. but like learning c language or other programming language.. the same way i like to learn salt
13:14 Guest98168 syntax... coding style
13:14 Guest98168 writing for loop
13:14 Guest98168 if loop
13:14 Guest98168 while ... do while
13:14 Guest98168 functions
13:15 Chocobo joined #salt
13:17 Chocobo_ joined #salt
13:17 juicer2 joined #salt
13:18 jcockhren Guest98168: don't you mean learning python, not necessarily salt?
13:18 jcockhren ;)
13:18 jcockhren (like a programming language)
13:19 Guest98168 i dont have exact idea to learn what for salt state files
13:19 jcockhren Guest98168: yaml?
13:19 bhosmer joined #salt
13:19 jcockhren state files are in valid yaml
13:20 Guest98168 then what is jinja.....
13:20 Guest98168 then how do we use python?
13:20 jcockhren however, yaml is just a markup language
13:20 jcockhren jinja is a templating system
13:20 Guest98168 how to integrate python coding in state files
13:20 jcockhren as a templating system, jinja have programming like constructs
13:21 Kholloway joined #salt
13:21 Guest98168 thanks can you explain little brief and share some url to start
13:22 natim joined #salt
13:22 natim Hello
13:22 natim guys
13:22 natim Does the pillar overridding thing already work in last salt version ?
13:23 Guest98168 left #salt
13:23 pdayton joined #salt
13:24 jcockhren saltconf prices though. :(
13:25 wkf joined #salt
13:28 natim Do you know when the pkg.compare bug will be released ?
13:32 cedwards joined #salt
13:35 krissaxton joined #salt
13:37 agend joined #salt
13:40 micah_chatt joined #salt
13:48 agend joined #salt
13:51 mofomikes joined #salt
13:54 teebes joined #salt
13:57 Chocobo joined #salt
13:59 timoguin joined #salt
14:02 mannyt joined #salt
14:05 jrdx joined #salt
14:05 ldlework Saltcon prices jeeeeze
14:07 mapu joined #salt
14:08 Brew joined #salt
14:12 supplicant ldlework: ? how much?
14:12 ldlework supplicant: "too" much
14:13 ldlework unfortunately
14:13 supplicant ldlework: link?
14:13 ldlework its in the topic
14:13 deepakmd_ joined #salt
14:13 supplicant o_o
14:14 ldlework indeed
14:17 supplicant and it's in salt lake city
14:17 ldlework Djangocon had pretty reduced attendance because of its 800 dollar sticker price
14:19 ldlework Ah they only have 250 tickets
14:19 ldlework explains the price
14:20 shiin are there and Dash.app users here? would you mind voting for a salt stack docset?
14:20 kaptk2 joined #salt
14:21 agtilden joined #salt
14:23 Sypher joined #salt
14:24 cwright did something change with the way logging works in 0.17?
14:24 cwright I just upgraded from 0.16.4 to 0.17 and now I can't run salt commands as my user, even though client_acl is configured
14:24 cwright https://gist.github.com/corywright/6902132
14:25 jumperswitch_ joined #salt
14:26 cwright of course i can change ownership of the logs and log directory, but I'd like to know if something changed
14:26 Furao joined #salt
14:26 krissaxton joined #salt
14:30 mike_perdide joined #salt
14:30 cnelsonsic joined #salt
14:32 danielmcbawse joined #salt
14:34 mike_perdide Hi, I'm trying to build a setup where minion could potentially be compromised, and I have 2 questions: 1. is there a preferred way to limit the access of a minion to the master's files (salt://my_file), 2. when using pillar.items (or .data), I see some information I could base my sls files upon (grains being too compromisable to my knowledge), but for instance pillar["id"] or pillar["master"]["id"] won't work. Any suggestion on what I might be
14:34 StDiluted joined #salt
14:37 cshuman joined #salt
14:42 wincus joined #salt
14:42 jbunting joined #salt
14:43 timoguin joined #salt
14:44 tyler-baker joined #salt
14:45 cshuman joined #salt
14:47 pentabular joined #salt
14:47 pentabular joined #salt
14:56 fragamus joined #salt
14:59 andrewclegg weird custom module prob. I have a function that returns True (expected result) when I run it from the minion using salt-call, but when I run it from the master, it returns False
14:59 andrewclegg for that minion
14:59 andrewclegg any idea why?
15:02 brimpa joined #salt
15:04 [M7] joined #salt
15:04 whiskybar joined #salt
15:05 cshuman joined #salt
15:06 jkleckner joined #salt
15:06 cron0 joined #salt
15:08 jalbretsen joined #salt
15:14 jalbretsen joined #salt
15:15 jj_ joined #salt
15:15 shinylasers joined #salt
15:21 oz_akan_ hi
15:21 oz_akan_ I looks like mine doesn't work with multiple masters
15:21 oz_akan_ I created a bug report: https://github.com/saltstack/salt/issues/7697
15:21 JaredR joined #salt
15:22 iamaporkchop joined #salt
15:22 jacob_s joined #salt
15:25 jacob_s I'm having some issues that I believe started with Salt 17 and I'm not sure if the default behavior changed or I need to file a bug.
15:25 jacob_s Salt seems to be installing some programs in my home directory as the root user, so that it leaves files in my $HOME owned by root.
15:26 jacob_s for instance my .pip and .wine folders are both owned by root.
15:32 fragamus joined #salt
15:34 jrdx joined #salt
15:35 jaequery joined #salt
15:35 jumperswitch joined #salt
15:36 shiin left #salt
15:38 elfixit joined #salt
15:39 jacob_s Hm, it looks like it might have broken apt-get as well.
15:39 mmilano joined #salt
15:42 jdenning joined #salt
15:43 troyready joined #salt
15:44 fllr_ joined #salt
15:44 fllr_ Hey guys, how do I install virtualenv wrapper using salt?
15:44 g4rlic joined #salt
15:46 teebes joined #salt
15:51 g4rlic dear salt team: client_acl officially saved my security bacon today.  Thank you.
15:53 terminalmage g4rlic: great to hear! if you'd like to send a short description of how to info@saltstack, we'd love to hear more
15:54 terminalmage info@saltstack.com that is
15:55 higgs001 joined #salt
15:55 jkleckner joined #salt
15:59 BrendanGilmore joined #salt
16:01 unicoletti left #salt
16:01 mwillhite joined #salt
16:02 amahon joined #salt
16:02 mofomikes i've made a function addition to a salt module.  I copied the adjusted module to both the minion and master.  however when i try executing the new function from the master, I get '<function> is not available'.  When i do a salt-call on the minion to troubleshoot, it executes successfully.   Any idea how I can get the master to recognize the new function?
16:02 bitz joined #salt
16:02 druonysus joined #salt
16:03 higgs001_ joined #salt
16:05 bhosmer joined #salt
16:06 g4rlic terminalmage: Done.
16:07 g4rlic mofomikes: you've restarted the master, right?
16:07 mofomikes yeah, still not recognized
16:08 terminalmage g4rlic: thanks!
16:08 mofomikes g4rlic: i deleted the .pyo and regenerated it
16:08 mofomikes deleted the .pyc file too
16:10 redondos joined #salt
16:10 * g4rlic is not a salt expert.  ;)
16:10 jbunting joined #salt
16:14 terminalmage mofomikes: where did you put this updated module?
16:15 terminalmage you shouldn't have to manually copy modules anywhere.
16:15 terminalmage http://docs.saltstack.com/ref/modules/index.html#modules-are-easy-to-write
16:15 terminalmage mofomikes: putting the updated module in the _modules on the master and running salt '*' saltutil.sync_modules should do it
16:17 fragamus joined #salt
16:18 troyready joined #salt
16:20 pipps joined #salt
16:21 Lue_4911 joined #salt
16:23 dthom91 joined #salt
16:25 mofomikes terminalmage: ok, i didnt do that at all
16:25 mofomikes terminalmage: i located the svn.py in the /usr/lib/python2.6/ directory, and made my changes.  I'll take a look at that URL.  thanks!
16:26 avienu joined #salt
16:27 xet7 joined #salt
16:27 mbbs joined #salt
16:28 mofomikes terminalmage: will modules in the _modules directory of the file_roots overload the standard modules that come with salt?
16:28 mbbs updated salt stack packages stopped working
16:28 mbbs 2013-10-09 16:27:17,186 [salt.crypt       ][CRITICAL] The Salt Master has rejected this minion's public key!
16:28 mbbs To repair this issue, delete the public key for this minion on the Salt Master and restart this minion.
16:28 mbbs Or restart the Salt Master in open mode to clean out the keys. The Salt Minion will now exit.
16:29 mbbs salt-minion 0.17.0
16:29 mbbs any issues with this version ?
16:30 mofomikes nm, rtfm
16:30 mofomikes thanks again terminalmage
16:39 terminalmage mofomikes: sorry, had to step away for a moment
16:40 terminalmage mbbs there was a change in minion id determination that may have caused that issue. Can you try manually setting the minion ID in /etc/salt/minion and restarting the minion?
16:40 terminalmage 0.17.1 will be out soon and will fix this
16:41 terminalmage by caching the minion ID
16:41 terminalmage changes in the algorithm that discovers the minion ID should not cause problems like this
16:41 terminalmage so we're sorry for the inconvenience if this is indeed the case
16:43 Sypher joined #salt
16:43 fllr_ joined #salt
16:43 jrdx I don't suppose anyone has any good resources/info on using salt to update Nagios along with server builds? The only interesting thing I see is YouVersion saying they're doing it... and are promising releasing their method "soon"... but no details.
16:43 higgs001 joined #salt
16:43 gmoro joined #salt
16:46 drags joined #salt
16:46 mbbs terminalmage: confirmed with setting id , restarting minion , I had to delete master key as well to make it work
16:47 terminalmage mbbs: ok, deleting master key should not have been necessary, unless the master's public key changed on the master
16:47 KyleG joined #salt
16:47 KyleG joined #salt
16:47 terminalmage which is not likely
16:47 mbbs that might be due to i am playing with keys
16:47 terminalmage ah
16:47 mbbs thanks i dont see i need to file a bug report since you are aware of it
16:48 mbbs or should i ?
16:48 terminalmage yeah, starting with 0.17.1 (which should be out very soon), when a minion auths the first time it will cache the minion ID. up to now, it would determine the minion ID every time the minion starts
16:48 dyim joined #salt
16:48 terminalmage mbbs: no, I believe this is already fixed in git, or will be fixed later today
16:49 mbbs thanks terminalmage
16:49 terminalmage mbbs: no prob, and sorry for the inconvenience
16:50 mofomikes awww hells yeah...
16:50 mbbs you are too kind thanks for your work
16:50 terminalmage :)
16:50 jkleckner joined #salt
16:50 JesseC joined #salt
16:51 mofomikes terminalmage: having the modules/grains/states/renderers getting pushed out from the file_root on the master is brillant
16:51 mofomikes things working now
16:51 mofomikes thanks again
16:52 terminalmage no prob. this is how you can push custom modules/grains/etc., but it is handy to override behavior in the standard modules
16:53 terminalmage mofomikes: just remember to remove that module from _modules and re-sync when there's a new release that fixes the issue, or your version in _modules will still be used over the default one
16:53 terminalmage mofomikes: is this a bug we've yet to fix, or some additional functionality?
16:53 terminalmage if it's not yet reported, we'd love a pull request
16:53 Katafalkas joined #salt
16:54 Ryan_Lane joined #salt
16:54 mofomikes I just adding a svn switch function to the built in svn module
16:54 mmilano joined #salt
16:55 mofomikes which is odd that its not included, since the syntax is pretty much identical to the checkout function
16:55 jimallman joined #salt
16:55 terminalmage ahh
16:55 terminalmage well, if it adds new functionality, we'd love to see a pull request
16:56 mofomikes pull request?
16:56 mofomikes (sorry im not a dev)
16:57 terminalmage ahh... well you'd need to fork the project on github and update your fork with the new code, then click the "Pull Request" button on github
16:57 terminalmage if you're not familiar with that, and would like to pastebin the new function that you added, I can get it in
16:57 bemehow joined #salt
17:00 amahon joined #salt
17:01 teebes joined #salt
17:02 mbbs so i found a bug in pip state
17:02 mbbs how should that be reported ?
17:02 jacob_s is it making .pip owned by root?
17:02 mbbs on google group ?
17:02 Jahkeup joined #salt
17:02 terminalmage mbbs: https://github.com/saltstack/salt/blob/develop/Contributing.rst
17:03 Kholloway joined #salt
17:03 mbbs jacob if requirement file has another requirement file imported saltstack breaks
17:04 jacob_s ah
17:04 jacob_s since salt 17, I'm having issues with salt leaving things owned by root in $HOME
17:05 mbbs it copies requirement files to tmp and proceeds to execute, break down occurs since imported file is not in /tmp
17:05 taylorgumgum joined #salt
17:05 mbbs thanks terminalmage
17:05 lineman60 joined #salt
17:05 terminalmage jacob_s: there have been a lot of changes to the pip module between 0.16.0 and 0.17.0, please report any issues you've found
17:07 terminalmage mbbs: no prob
17:07 mapu joined #salt
17:07 jacob_s_ joined #salt
17:08 winux joined #salt
17:09 logix812 joined #salt
17:09 jacob_s_ The only thing i'm seeing new that I don't quite understand is when I install wine with salt
17:09 jacob_s_ the .wine folder in home is owned by root
17:10 winux hello.. I can run individual state files fine, but when I run state.highstate I get: "AttributeError: 'list' object has no attribute 'get'" .. i can't seem to figure out what is causing this issue.. I would think it is an indentation issue but like i said state files work fine.. any ideas on what could be causing this?
17:12 jacob_s_ what does your top.sls look like?
17:13 winux jacob_s_: havent touched top in a while, the issue just started popping up today
17:13 winux but it looks good
17:13 jacob_s_ ah, so it might be a 17 issue.
17:13 JaredR joined #salt
17:13 g4rlic winux: my first thought would be to do something like this: find /srv/salt -type f -name "*.sls" -exec grep -Hn 'get' '{}' \; and see what pops up.
17:13 g4rlic and then check the revision history of any matching files and try to time corellate.
17:14 g4rlic But if you just updated, and things started breaking, start there.. ;)
17:15 winux g4rlic: ill try that line .. ill try to backtrack.. but the odd thing is afaik the files in git are the same as when it was working yesterday
17:27 g4rlic winux: do you highstate often?  If not, I'm not sure how easily you can time the breakage.
17:28 mofomikes terminalmage: http://pastebin.com/T8Hz8Dm3  lines 147-175 is the new function (which is just a copy of the checkout function with the actual svn command replaced)
17:29 danielmcbawse joined #salt
17:30 UtahDave joined #salt
17:30 bemehow joined #salt
17:31 cro joined #salt
17:32 UtahDave micah_chatt: So I'm having a bit of trouble configuring the django_orm external pillar
17:33 micah_chatt UtahDave: ok, are you getting an error message?
17:33 UtahDave I'm quite the django newbie. Someone else set up a very very basic django app to manage some config data from a web page.
17:33 UtahDave no, no errors at all
17:33 micah_chatt ok
17:33 UtahDave here's a pastebin of my config in my master config file: http://pastebin.com/Ea61Lgd3
17:34 danielmcbawse joined #salt
17:34 UtahDave I couldn't find any file that seemed to match what was requested for the env_file config item.
17:34 micah_chatt UtahDave: ok I think i might see what your problem is
17:35 micah_chatt First: the assumption is that you're running django in a virtualenv
17:35 terminalmage mofomikes: thanks!
17:35 jaequery joined #salt
17:35 micah_chatt UtahDave: so the `env` param is the path to that virtualenv
17:36 micah_chatt the 'env_file' is optional
17:36 UtahDave ok.  I'm not sure if that's a virtualenv, actually
17:37 matanya_ joined #salt
17:37 micah_chatt I'm guessing it is not. seeing that your project path is a subdir of the env
17:37 Furao joined #salt
17:38 micah_chatt its quite typical for a django project to use ENV variables to set certain settings, so the env_file is an optional bash file to set any necessary variables
17:38 UtahDave Hm. I may be totally off on the definitions of those.  :0
17:38 bemehow joined #salt
17:39 micah_chatt UtahDave: Often, django projects will set up a virtualenv in the home of the project, either 'venv' or 'env'
17:40 UtahDave micah_chatt: unfortunately I'm pretty ignorant of how django applications are set up.  So would the virtualenv actually be in the html_root? Or would that likely sit somewhere else?
17:40 micah_chatt it depends on how you have it set up. see ^^
17:40 micah_chatt the virtualenv can be wherever you want it
17:41 micah_chatt a typical place is in the root of your django project under /venv or /env
17:41 micah_chatt typically the .gitignore on django projects will filter out /venv and /env
17:42 micah_chatt and you'll build it with a pip install -r requirements.txt
17:42 UtahDave ok, I don't see those.  I guess I'm going to have to wait until I can get ahold of whiteinge, then.
17:43 UtahDave we might not be in a virtualenv
17:43 micah_chatt ah, you're using the system python then
17:44 UtahDave yeah, I'm thinking that's the case
17:44 UtahDave micah_chatt: can this work without a virtualenv?
17:44 micah_chatt ok, in that case, the orm pillar doesn't supper that *yet*
17:45 bemehow joined #salt
17:46 bemehow joined #salt
17:46 micah_chatt wellll actually it might, let me check real quick
17:46 UtahDave ok, thanks
17:49 cshuman_ joined #salt
17:50 micah_chatt UtahDave: Yea it wont' work with the system python. I can put in a patch to support that
17:50 UtahDave ok, that would be cool.
17:50 micah_chatt UtahDave: I won't be hard.
17:51 UtahDave thanks, micah_chatt .  how much work would that be?
17:51 Katafalkas joined #salt
17:51 UtahDave hey, Furao
17:51 micah_chatt It'll cost you precisely 1 SaltStack tee-shirt
17:51 UtahDave lol
17:51 micah_chatt UtahDave: but really, a few minuets
17:52 UtahDave I think we're down to little girly sizes, unless the next shipment has come in since I've been out of the office.  :)
17:53 terminalmage mofomikes: https://github.com/saltstack/salt/pull/7702
17:53 terminalmage thanks for the contribution!
17:56 cro joined #salt
17:56 srage joined #salt
17:56 bdf When I do a test.ping on a master of masters in a syndic setup, should I get a return from the minions behind the syndica
17:56 bdf syndics?
17:58 wincus joined #salt
17:58 bdf I keep seeing "2013-10-09 09:40:33,708 [salt.master                                 ][ERROR   ] An extra return was detected from minion zk1a.ec2.test.zpath.net, please verify the minion, this could be a replay attack"
17:59 mbbs joined #salt
17:59 Katafalkas joined #salt
18:01 DredTiger joined #salt
18:01 micah_chatt UtahDave: patch is hopefully complete, going to lunch, then I'll test/PR it
18:02 pipps joined #salt
18:02 feth joined #salt
18:03 feth hello,
18:03 feth I'm a bit lost, I found docs on modules writing, but I can't get my module to load
18:03 feth where should I put it ?
18:03 feth (on the minion)
18:04 feth (or in the /srv/salt dir, I don't know)
18:05 jacob_s_ should be something like /srv/salt/_modules
18:07 feth That's the place I store my file in.
18:07 feth it's a file named 'pelican.py' with only a stub 'def update(value=0): return value' in it (with docstrings everywhere)
18:08 Jahkeup joined #salt
18:09 feth oh I got it : salt \* saltutil.sync_modules
18:09 diegows joined #salt
18:10 feth it's so high in the doc page... ;)
18:11 taylorgumgum joined #salt
18:13 fllr joined #salt
18:13 fllr hey guys, How do I add an user to the list of sudoers using salt?
18:13 g4rlic fllr: I'm personally using file.managed
18:14 g4rlic and dropping something into /etc/sudoers.d/
18:14 g4rlic so I don't have to muck with modifying the default /etc/sudoers.
18:14 derelm joined #salt
18:16 jefimenko joined #salt
18:18 wincus joined #salt
18:18 cshuman joined #salt
18:18 mapu joined #salt
18:21 cbloss can someone help me with what I'm doing wrong? I know it is a really stupid error on my part… http://pastebin.com/hTRyWn5B
18:21 cbloss also, is it possible to combine both of those "pkgrepo.managed" fields into one?
18:25 taylorgumgum joined #salt
18:25 fllr g4rlic: that sounds like a good solution... is it easy to maintain like that?
18:26 g4rlic fllr: has been so far.  ;)
18:26 g4rlic there's no daemon to restart with sudo, so file.managed seems to work nicely.
18:26 g4rlic you update the file as needed, and highstate, and your changes are made. easy peasy.
18:27 ggoZ joined #salt
18:31 cbloss mmmm, it looks like it should actually work… bug in 0.17 maybe?
18:32 aberant joined #salt
18:34 Brew joined #salt
18:34 giantlock joined #salt
18:38 Jahkeup joined #salt
18:44 bdf am I understanding correctly that syndic is more or less only for propagating commands, and things such as the fileserver/state actually connect to the local master?
18:45 pdayton joined #salt
18:48 felixhummel joined #salt
18:48 socks__ joined #salt
19:01 Kholloway joined #salt
19:01 copelco is it not possible to install 0.16 anymore from the ppa?
19:03 adepasquale joined #salt
19:03 copelco 0.16.4 that is
19:05 bhosmer_ joined #salt
19:06 cbloss copelco: doesn't look like it
19:06 copelco i'm seeing some weird issues with 0.17
19:06 copelco wanted to revert... doesn't look like i can :P
19:06 cbloss yeah I am too
19:07 rgbkrk joined #salt
19:09 copelco is there an alternative way to install it?
19:09 jbunting joined #salt
19:10 copelco is anyone else seeing conflicting ID error messages with 0.17? The conflicting ID is "ssh" and is found in SLS "base:sshd" and SLS "base:sshd"'
19:10 copelco I'm not sure how to diagnose it, since it appears to be the same thing
19:10 KyleG Are they named the same...
19:11 cshuman joined #salt
19:11 mgw1 joined #salt
19:11 rgbkrk I'm seeing old discussion of message "No Top file or external nodes data matches found" when using gitfs remotes with local git
19:12 rgbkrk But no solutions
19:12 rgbkrk Anybody have other tips?
19:12 copelco KyleG: well, they are, but i have multiple file_roots and one is overriding the other.
19:12 copelco it worked in 0.16
19:12 rgbkrk gitfs_remotes:
19:12 rgbkrk - file:///code/salt-nbviewer
19:13 rgbkrk Worked when I was pointing to the github repo
19:13 rgbkrk Haven't gotten local to work
19:14 amahon joined #salt
19:14 micah_chatt joined #salt
19:15 jdenning Does anyone know if 'service.running' uses the init.d status method to determine if a service is actually running?  I am getting 'Service freeradius failed to start' on an ubuntu 12.04 box, when the freeradius daemon *is* actually running.
19:16 jdenning If I stop the service manually and let salt start it via the state, there is no issue.  But if the service is already running, I get the "failed to start" error when running state.highstate.
19:16 jdenning I do not experience this error with e.g. sshd
19:17 jdenning '$ service freeradius status' does not work on ubuntu, so I'm wondering if this is the source of the issue?
19:17 StDiluted seems likely
19:17 StDiluted but I don't know the answer
19:17 cshuman joined #salt
19:18 StDiluted Is there a changelog/features added list for 17.0?
19:18 jdenning Is there a "warn only" flag or something similar I can apply to the state?
19:18 ramteid joined #salt
19:19 StDiluted neve rmind, found the release notes
19:19 Gifflen joined #salt
19:20 Kholloway joined #salt
19:20 ramteid Hi, is there anything known that file.managed + template: jinja is not working on Windows XP with Saltstack 0.16.3? if I don't use template it works, but with template: jinja I only get "Error: No changes made for …" And I run out of ideas and search results...
19:21 adepasquale joined #salt
19:24 adepasquale joined #salt
19:24 pipps joined #salt
19:25 Gifflen joined #salt
19:27 mgw1 rgbkrk, I had the same problem with gitfs a couple days ago. Unfortunately, it just started working about the time I tried to debug it.
19:28 rgbkrk Trying to clear various caches now to see if I can fix it.
19:29 StDiluted is ther ehalite docu
19:30 bemehow joined #salt
19:30 ramteid StDiluted,  https://github.com/saltstack/halite ?
19:30 StDiluted hm, not exactly what i was looking for but I suppose it's a start
19:31 ramteid StDiluted, I see, don't know of any other doc. :(
19:31 jbunting joined #salt
19:35 bemehow joined #salt
19:38 Jahkeup joined #salt
19:38 ranl_ joined #salt
19:40 oz_akan_ joined #salt
19:41 krissaxton joined #salt
19:47 taylorgumgum joined #salt
19:49 pipps joined #salt
19:49 mwillhite joined #salt
19:51 bhosmer joined #salt
19:53 taylorgumgum joined #salt
19:59 Marion joined #salt
20:00 ramteid FYI: -v helps, since -l debug doesn't return the errors in template… it seems that filter strftime is not available under windows...
20:01 jlund joined #salt
20:02 krissaxton joined #salt
20:03 carmony any recommendations on self-updating your salt minions? best way to do it?
20:03 cloud_noob joined #salt
20:03 cloud_neofight joined #salt
20:07 ramteid Ok, I assume the reason is that timelib is missing in windows installer.
20:08 saurabhs joined #salt
20:08 jdenning In case anyone cares - my issue above was indeed related to the freeradius init script not having a 'status' option.  I created a status init script option, and all is well now (patching the file as a requirement of the service.running state)
20:10 StDiluted awesome
20:10 StDiluted glad you got it working
20:12 wincus joined #salt
20:14 Brew joined #salt
20:14 krissaxton joined #salt
20:18 Jahkeup_ joined #salt
20:19 derelm joined #salt
20:20 toastedpenguin joined #salt
20:20 xmltok joined #salt
20:24 wincus joined #salt
20:33 bemehow joined #salt
20:34 druonysuse joined #salt
20:34 druonysuse joined #salt
20:43 bemehow joined #salt
20:43 teebes joined #salt
20:44 bhosmer joined #salt
20:50 amahon joined #salt
20:52 robbyt joined #salt
20:52 amahon joined #salt
20:54 amahon joined #salt
20:58 alrs joined #salt
20:58 pipps joined #salt
20:59 xmltok_ joined #salt
21:00 zwevans joined #salt
21:02 Jahkeup joined #salt
21:04 schannel joined #salt
21:05 schannel left #salt
21:05 schannel joined #salt
21:06 schannel left #salt
21:06 LarsN_ does anyone have any example state files that trigger pip installs?
21:09 jesusaurus LarsN_: you mean using the pip.installed function?
21:10 LarsN_ jesusaurus: didn't realize there was a pip.installed
21:10 LarsN_ that makes life easy.  I suspect I can figure that out ;)
21:11 jesusaurus LarsN_: glad i could help :)
21:11 LarsN joined #salt
21:11 mike_perdide left #salt
21:12 LarsN one more post installation script to be wiped off the list by a state file I suspect.
21:12 LarsN in the not distant future.
21:13 xmltok joined #salt
21:13 jbunting joined #salt
21:14 LarsN next up however is figuring out how to get SaltyVagrant working with KVM
21:14 LarsN ;)
21:20 ckao is there a way to manage grains centrally?  i'd like to have the master look at a flat file and then push a custom /etc/salt/grains file vs manually editing that file on each minion.  is this possible?
21:26 jdenning joined #salt
21:27 JaredR joined #salt
21:31 shinylasers joined #salt
21:35 aberant joined #salt
21:41 woebtz joined #salt
21:43 DredTiger joined #salt
21:44 lmickh joined #salt
21:46 jcockhren ckao: you can set grains from the master
21:47 jcockhren http://docs.saltstack.com/ref/states/all/salt.states.grains.html
21:47 jcockhren ckao: ^
21:48 yidhra /help
21:48 stevetodd joined #salt
21:48 jcockhren ckao: as well -> http://docs.saltstack.com/ref/modules/all/salt.modules.grains.html
21:49 druonysus joined #salt
21:49 druonysus joined #salt
21:51 adepasquale joined #salt
21:53 stevetodd Salt T-Shirt Idea (someone has probably already thought of this): Salt logo plus the text "I'm always in highstate."
21:53 ipmb joined #salt
21:54 joshuagoodson joined #salt
21:56 joshuagoodson left #salt
21:57 jcockhren stevetodd: shit-it!
21:57 jcockhren s/shit/ship
21:57 jcockhren LOL
21:58 andrej heh
21:59 jcockhren totally meant ship the first time
22:00 jaequery joined #salt
22:00 andrej shi-it would have worked nicely! :)
22:01 DredTiger joined #salt
22:01 jcockhren digitalocean has the "I don't give a ssh" shirts
22:02 Thiggy joined #salt
22:03 Thiggy Is there anything like sys.doc but for runners on the master?
22:03 pentabular joined #salt
22:03 Thiggy nm, salt-run doc.runner
22:05 bemehow joined #salt
22:06 derelm i want to give salt-ssh a try - i installed it using pip as a normal user -- it seems it doesn't respect command line option -c and instead always complains about /etc/salt not existing and not being able to create it
22:09 __number5__ derelm: you need sudo
22:09 derelm why do i need root rights on the "master" server? doesn't make sense to me
22:10 __number5__ because salt-ssh try to create folder /etc/salt/pki/minion
22:11 derelm that's a hardcoded directory that cannot be changed?
22:11 jesusaurus derelm: the pki directory can be set in the config file /etc/salt/master
22:11 __number5__ also I guess you want -C instead of -c
22:12 jesusaurus you can change it to be e.g. inside the homedir of a 'salt' user
22:12 derelm __number5__: my salt-ssh reads:   -c CONFIG_DIR, --config-dir=CONFIG_DIR
22:12 derelm Pass in an alternative configuration directory.
22:12 derelm Default: /etc/salt
22:13 __number5__ derelm: sorry, my bad, need a pair of glasses
22:13 derelm __number5__: no problem, but as i said, that value seems to be ignored
22:14 jaequery joined #salt
22:15 mianos joined #salt
22:20 troyready joined #salt
22:23 __number5__ derelm: I've tried salt-ssh on my virtual env with -c, it works fine
22:23 derelm __number5__: are you running it as a user without a /etc/salt ?
22:23 __number5__ derelm: yes. do you have a roster file in your salt config dir?
22:23 derelm __number5__: my current salt dir is empty
22:24 __number5__ derelm: oh, then you need to at least create a master config and a roster
22:25 derelm __number5__: oh ok i see, can i create a default template somehow?
22:25 __number5__ derelm: yes, default one should work, just change the root_dir to proper path
22:27 bemehow joined #salt
22:27 copelco joined #salt
22:27 derelm __number5__: can i use relative paths as root_dir?
22:28 derelm __number5__: seems to work fine
22:28 alrs joined #salt
22:28 derelm __number5__: that get's me past the errors, thanks for your help
22:29 joehh joined #salt
22:30 kermit joined #salt
22:30 __number5__ derelm: no problem
22:30 derelm __number5__: hm one more question - i want to connect to a server where my users public ssh key is already deployed, do i have to pass some extra option?
22:31 ktenney joined #salt
22:31 __number5__ derelm: two way to do it, ssh-add the key to ssh-agent, or you salt-ssh --priv option
22:33 derelm __number5__: hm, salt-ssh --priv doesn't exist
22:34 __number5__ derelm: hmm, maybe it's only in develop, have you try adding the key to your ssh-agent?
22:35 dzderic joined #salt
22:36 McLovin joined #salt
22:38 derelm there seems to be another problem with ssh-keygen on the mac
22:39 derelm in combination with salt-ssh (and a minimal configuration)
22:40 derelm it prints "Too many arguments." and shows ssh-keygen options - later it asks for a password and after that it fails to copy the public key part to the server, because it was never generated
22:43 juicer2 joined #salt
22:44 Guest19625 Hello. A question if I may. I'd like to implement Salt state management on new servers only, leaving existing servers unmanaged because of the risk and complexity of bring them into the fold. But at the same time I want all servers - new and legacy - to run the minion for orchestration purposes. There is therefore the constant danger of accidentally applying state management to legacy systems. Is there a simple method of disabling t
22:44 Guest19625 Is there a simple method of disabling the state management capability on a minion, leaving it as an orchestration agent only?
22:44 honestly Guest19625: target your states?
22:44 honestly I don't think so
22:45 Brew joined #salt
22:45 felixhummel joined #salt
22:47 pipps joined #salt
22:49 felixhummel joined #salt
22:53 Guest19625 Thank you. I don't know how realistic it is, but it would be a nice minion feature. I'd guess that running such a mix is not uncommon(?)
22:53 dthom91 joined #salt
22:57 pentabular joined #salt
22:58 honestly you could always use two separate masters
23:00 joehh or use client_acl and use one uesr for state functions and another for orchestration
23:04 s0undt3ch joined #salt
23:08 wkf joined #salt
23:17 jeffrubic joined #salt
23:20 helderco joined #salt
23:20 cshuman joined #salt
23:21 Guest19625 Thank you, I will check out those ideas. Also, I've been playing with the pydsl renderer, but am not a Python programmer and have got stuck on what I suppose to be a trivial syntax issue, but for the life of me can't get past it. I iterate through hierarchical Pillar data which contains arbitrary network settings on a per-server basis ('name: bond0', 'mode: 802.3ad', etc.)
23:21 Guest19625 I successfully read these as KEY, VALUE pairs, and then wish to apply them to my state object, e.g. nic_state.network.managed(KEY=VALUE). However, from what I can make out, 'KEY' is being passed a a literal (if that's the right word), so instead of '(mode=802.3ad)', for example, the result is '(KEY=803.3ad)'.
23:21 Guest19625 I need the *value* of KEY to be passed, rather than the "KEY" string. Can anyone suggest what the magic syntax is?
23:28 kiorky Guest19625: .managed(**{KEY:Value})
23:28 kiorky Guest19625: .managed(**{KEY:Value,OTHERKEY:VAL23})
23:30 bemehow joined #salt
23:30 Guest19625 Woah. Thank you very much. I was not going to stumble across anything that looked like *that* any time soon!
23:35 mwillhite joined #salt
23:36 drags1 joined #salt
23:38 bemehow joined #salt
23:44 jrhorn424 joined #salt
23:55 jdenning joined #salt
23:57 darrend joined #salt
23:58 sibsibsib_ joined #salt

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