Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2013-09-30

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

All times shown according to UTC.

Time Nick Message
00:05 cshuman joined #salt
00:16 Heartsbane joined #salt
00:16 Heartsbane joined #salt
00:40 jtruckz joined #salt
00:40 jdenning joined #salt
00:42 jtruckz joined #salt
00:45 Gwayne joined #salt
00:45 Gwayne joined #salt
00:53 L0j1k joined #salt
00:54 cshuman joined #salt
00:55 mgw joined #salt
00:57 kleinishere joined #salt
00:57 xl1 joined #salt
01:08 higgs001 joined #salt
01:11 mianos joined #salt
01:20 waldo323 joined #salt
01:20 L0j1k joined #salt
01:23 m_george|away joined #salt
01:24 m_george left #salt
01:26 L0j1k_ joined #salt
01:27 felixhummel joined #salt
01:41 malinoff joined #salt
01:45 mianos joined #salt
01:49 justlooks joined #salt
01:50 justlooks how can i use python to write sls file ,any example avariable?
01:51 malinoff Hi all
01:51 malinoff justlooks: Give me a second
01:52 justlooks malinoff:  a second is gone :D
01:57 malinoff justlooks: http://pastebin.com/TBWPtqEe
02:03 justlooks malinoff: thank you
02:04 sixninetynine joined #salt
02:04 malinoff justlooks: No problem :)
02:07 justlooks malinoff: some time i want to debug the python ,so i will print some info ,how can i do this so command ' salt '*' state.sls myfile ' can print my debug info ,not run really in minion?
02:08 cnelsonsic joined #salt
02:09 malinoff justlooks: You can use test=True: 'salt \* state.sls myfile test=True'. This command will say you something like "Okay, your state will be applied" if there no syntax errors in your state, or it will fail with a message about that fail.
02:10 malinoff justlooks: If you want to debug right on a minion, you can use logger (because salt will never print anything from the code)
02:12 justlooks malinoff: i just need get debug info from master ,because i want to make sure the python logic is correct just like i write python script
02:14 malinoff justlooks: I'm not sure about 'debug info from master', because a state is running on the specified minion... Could you clarify your question please?
02:15 cnelsonsic left #salt
02:15 Gwayne left #salt
02:16 jtrucks joined #salt
02:17 justlooks malinoff: i am not very familiar with python, for instance ,when i write a loop ,i need print loop n start to make sure it work fine
02:19 malinoff justlooks: You can't do that with 'print' statement. In Salt, you're saying 'apply this state on the specified minion' on the master, but the state as python's function is running on the minion. You should use logger to log some events (as you say, the start of a loop), and look into /var/log/salt/minion on the minion to see what happened
02:23 justlooks malinoff: i follow your example code ,write myself  sls file ,but i get error when i run it https://gist.github.com/justlooks/6758673
02:24 pabelanger justlooks, you code is not formatted properly
02:24 pabelanger def run():
02:24 pabelanger state= {}
02:24 pabelanger note the 4 spaces
02:25 malinoff justlooks: http://docs.python.org/2/tutorial/index.html you should totally start here
02:26 justlooks it's worked ,i fogot the python need format code
02:29 jtrucks joined #salt
02:36 justlooks can i write pthon and jinja_yaml in one file ? like this https://gist.github.com/justlooks/6758752
02:49 malinoff justlooks: http://docs.saltstack.com/ref/renderers/index.html#composing-renderers just try
02:49 malinoff justlooks: But i'm sure you can't mix py and yaml
02:53 ifnull_ joined #salt
03:30 snewbie joined #salt
03:32 snewbie reading the docs and it says root is the default user. how does this work on an os like ubuntu where the root user is locked by default?
03:35 dthom91 joined #salt
03:36 malinoff snewbie: Nothing stops you to start daemons as root user
03:36 snewbie so basically it's just sudo?
03:39 malinoff snewbie: I think, yes. I have no minions running on ubuntu, so i can't say properly :)
03:50 mianos joined #salt
03:51 cro joined #salt
04:11 cshuman joined #salt
04:21 Lue_4911 joined #salt
04:32 raghavp joined #salt
04:33 raghavp left #salt
04:38 robawt man alive
04:38 robawt 0 length binary files from salt://
04:39 redondos joined #salt
04:40 JesseCW joined #salt
04:43 t_chjohnswork joined #salt
04:43 f47h3r_ joined #salt
04:44 _ioni joined #salt
04:44 jrgifford__ joined #salt
04:44 johnsocp_ joined #salt
04:44 pabelanger_ joined #salt
04:44 Katafalkas joined #salt
04:44 tseNkiN1 joined #salt
04:47 jasiek_ joined #salt
04:48 cshuman joined #salt
04:52 mirko joined #salt
04:52 nocturn joined #salt
04:53 nocturn joined #salt
04:56 evax joined #salt
04:59 dthom91 joined #salt
05:01 mirko joined #salt
05:01 nocturn joined #salt
05:09 tim5464 joined #salt
05:11 ifnull joined #salt
05:11 redondos joined #salt
05:12 malinoff robawt: local: True
05:14 * goodwill eyes robawt
05:16 aleszoulek joined #salt
05:19 __number5__ snewbie: give salt a user with sudo permission it will be happy
05:25 gildegoma joined #salt
05:28 snewbie __number5__ just for the minions right? the master shouldn't need sudo?
05:29 snewbie __number5__: forgot the slash there
05:29 Ryan_Lane joined #salt
05:30 forrest snewbie, yea just the minions should be fine.
05:38 malinoff snewbie: Why do you need any access, but not just root?
05:41 snewbie malinoff: just trying to minimize the number of things that have root access
05:42 malinoff snewbie: I'd rather create new user (smth like saltminion) and give him any needed access.
05:45 snewbie malinoff: yea that's what I'm doing for the minions. but if the master doesn't need to be a sudoer then I'd rather not have him as one.
05:45 malinoff snewbie: Right.
05:59 davidone joined #salt
05:59 Ryan_Lane joined #salt
06:01 mianos joined #salt
06:04 robawt malinoff: not sure if you mean the minion config or the salt state. minion config is set to local
06:08 malinoff robawt: wat
06:10 robawt malinoff: the zero length binary file issue
06:10 robawt it's cool, headed to bed, i'll bug the channel tomorrow
06:11 malinoff robawt: Ah, that was a question.
06:13 rmt It sure would be nice if Salt and Puppet could easily share PKI infrastructure.. Every man and his dog reimplements this.
06:15 mianos use salt-ssh ?
06:17 rmt Explain yourself. ;-)
06:20 middleman_ joined #salt
06:21 mianos well, if you use salt-ssh you can use existing pki
06:22 rmt But that has nothing to do with Puppet's PKI.
06:23 mianos oh you mean share as in 'share between salt and puppet'
06:23 mianos not share between en existing PKI CA etc with salt
06:24 mianos because openssh has had CA support for a few years now
06:25 rmt Puppet runs its own 'CA'.. Salt runs its own 'CA'.. You can easily plug in client side SSL support to apache etc. to support auth'd access to resources, as well as lots of other technologies.. stunnel can reuse these certs..
06:27 rmt I want one way for a machine to create, request signing, and receive a signed cert.. I'd like these certs to be easily shared server-side so I could have Puppet use/trust a Salt cert, or vise-versa.
06:34 Ryan_Lane joined #salt
06:34 ml_1 joined #salt
06:41 malinoff rmt: Do you know it is the most holest security hole ever?
06:42 bpgoldsb joined #salt
06:44 rmt When it comes to services requiring root access, I'd rather have one well managed internal "CA" than several.
06:46 rmt It's also only less code that needs to be audited for silly crypto mistakes.
06:51 yes456 joined #salt
06:51 yes456 http://docs.saltstack.com/topics/installation/ubuntu.html ubuntu repor dp not have 0.17 pkg
06:52 rmt It would also be nice to be able to easily issue/sign certs with less privileges from the same "CA", which are linked to the node.. ie. it'd be nice to be able to have Salt or Puppet states deploy sub-certs for specific services.
06:52 rmt yes456, Look at the topic.
06:53 yes456 rmt: you mean this "0.16.4 is the latest |
06:53 rmt Yes.
06:54 rmt As far as I know, 0.17 hasn't been officially released yet.
06:55 yes456 rmt: sorry i just last salt air on youtube and jumped on irc :(
06:55 rmt It should be out shortly, though. :-)
06:55 rmt Someone mentioned yesterday some packaging issues that needed to be resolved first.
06:59 balboah joined #salt
06:59 ollins joined #salt
07:01 xl1 left #salt
07:02 xl1 joined #salt
07:02 yes456 good waiting  ....
07:04 Ryan_Lane joined #salt
07:05 LucasCozy joined #salt
07:06 raghavp80 joined #salt
07:06 [M7] joined #salt
07:07 linjan joined #salt
07:09 druonysuse joined #salt
07:09 druonysuse joined #salt
07:13 xl1 left #salt
07:18 raghavp joined #salt
07:20 zloidemon joined #salt
07:23 Valdo joined #salt
07:24 adepasquale joined #salt
07:35 Ryan_Lane joined #salt
07:42 linuxnewbie joined #salt
07:43 linuxnewbie hello, i can't make salt-minion start as a normal user ...shell i create the user with specific permissions (like sudo) ?
07:45 antsygeek joined #salt
07:47 Damoun joined #salt
07:55 ricicle joined #salt
07:56 bdenning joined #salt
08:01 paluh joined #salt
08:02 ricicle joined #salt
08:05 Ryan_Lane joined #salt
08:09 nocturn Hi, I'm trying to template a configfile with pillars
08:10 nocturn I have in the config: {% for domain in pillar['maildomains'] %}{{ domain,  }}{% endfor %}, but it prints () arround the values: ('orions2.org',)('vsbnet.be',)
08:10 nocturn How do I get rid of the braces and the quotes?
08:10 raghavp80 joined #salt
08:11 paluh @nocturn semicolon is a tuple constructor in python, so try to get rid of it in {{ domain }} tag
08:12 nocturn paluh, I set the pillar with YAML, how do I get rid of them in the tag?
08:13 raghavp joined #salt
08:15 paluh @nocturn sorry I've thought about "comma" not "semicolon"
08:15 paluh @nocturn so probably you should write something like: {% for... %} {{ domain }}, {% endfor %}
08:16 paluh @nocturn BUT I'm newbie salt user so I'm guessing a bit here ;-)
08:17 nocturn paluh, thanks.  It still prints ('domain') though
08:18 qba73 joined #salt
08:22 linuxnewbie i can't make salt-minion start as a normal user ...shell i create the user with specific permissions (like sudo) ?
08:23 raghavp80 joined #salt
08:27 raghavp joined #salt
08:29 t_chjohnswork joined #salt
08:31 TheCodeAssassin joined #salt
08:34 paluh @nocturn - so resulting config file contains: ('domain') instead of just: domain, yes?
08:35 zooz joined #salt
08:35 nocturn paluh, sorry, I don't know why but it took a second run
08:35 nocturn it works nog with {% for... %} {{ domain }}, {% endfor %}
08:35 paluh @nocturn - cool :-)
08:35 nocturn the problems seems to have been that I did {{ domain,  }}
08:35 nocturn Thanks very much paluh
08:36 ronc joined #salt
08:38 goodbytes joined #salt
08:38 raghavp joined #salt
08:38 ciss joined #salt
08:38 winux joined #salt
08:39 winux joined #salt
08:40 winux can salt state files be used to create databases and users for those databases?
08:40 Ryan_Lane joined #salt
08:40 winux i see a module for the mysql, but im not sure they can be used in state files
08:43 multani winux: there're states to create databases and users, http://docs.saltstack.com/ref/states/all/salt.states.mysql_user.html and http://docs.saltstack.com/ref/states/all/salt.states.mysql_database.html
08:44 bpgoldsb|too joined #salt
08:45 winux multani: excellent, thanks
08:45 nocturn joined #salt
08:45 mirko joined #salt
08:46 nocturn joined #salt
08:47 Katafalkas joined #salt
08:48 paluh I'm trying to run simple ping thorugh salt-ssh (and I'm comming from others ssh based tools). Do I have to install salt library or configure minion on destination hosts in order to use salt-ssh (because after some playing with "salt-ssh test.ping" I've got: Failed to create directory path "/etc/salt/pki/minion" - [Errno 13] Permission denied: '/etc/salt')?
08:50 justlooks i set one pillar variable to a key-value pair list ,how can i get the value in python ?
08:51 robthoms joined #salt
08:51 justlooks paluh: Permission denied check if auth is correct
08:51 martoss joined #salt
08:53 martoss hey folks, I am trying to fire a host_announce event to my salt master.
08:54 martoss I added the following in a state: http://pastebin.com/LnruAGGB
08:54 martoss any ideas what could be wring?
09:03 paluh @justlooks I'm not sure what you mean. From error message it seems that salt is trying to create /etc/salt/pki/minion file (or dir), probably on destination host. I'm interested if this is necessary and what is it pki/minion.
09:06 justlooks paluh: salt-ssh is belongs to salt package? i am not find this command in my salt installation
09:09 paluh @justlooks - yes, it was introduced in 0.17 realese. here is quick preview of this feature ("Salt Air 16"): http://www.youtube.com/watch?v=uWGDC1PdySQ
09:10 yota joined #salt
09:13 ricicle joined #salt
09:17 mollerstrand joined #salt
09:21 ollins joined #salt
09:24 ricicle joined #salt
09:34 p3rror joined #salt
09:34 paluh @justlooks it's ok, fresh version from repository fixed this problem, but... it introduced another one:
09:36 paluh I'm going to describe it on salt-devel, anyway @justooks - thanks for your affort!
09:41 Ryan_Lane joined #salt
09:41 Ryan_Lane joined #salt
09:51 giantlock joined #salt
09:56 faldridge joined #salt
09:57 giantlock joined #salt
10:11 Ryan_Lane joined #salt
10:11 Ryan_Lane joined #salt
10:25 scott_w joined #salt
10:26 krissaxton joined #salt
10:42 Ryan_Lane joined #salt
10:55 honestly whelp, something's fucky.
11:04 ingwaem joined #salt
11:09 aparashar joined #salt
11:10 gildegoma joined #salt
11:12 Ryan_Lane joined #salt
11:15 viq morfternoon
11:17 honestly morn
11:17 viq akoumjian: https://github.com/saltstack/salty-vagrant says to poke you ;) I'm trying to use the "complete" example, with arch, and it complains: python2-msgpack-0.3.0-2-x86_64.pkg.tar.xz is corrupted (invalid or corrupted package (PGP signature))
11:19 scristian joined #salt
11:20 viq akoumjian: and with the 'precise64' box it does not seem to set the hostname properly, thus does not recognize the minion keys
11:21 bhosmer joined #salt
11:25 unicoletti_ joined #salt
11:25 felixhummel joined #salt
11:26 ricicle joined #salt
11:33 jslatts joined #salt
11:35 HaxCore joined #salt
11:36 krissaxton joined #salt
11:37 p3rror joined #salt
11:39 ml_1 joined #salt
11:42 zooz joined #salt
11:43 Ryan_Lane joined #salt
11:46 paluh joined #salt
11:53 faldridge joined #salt
12:01 blee joined #salt
12:01 krissaxton joined #salt
12:01 honestly so.
12:02 honestly one or some of my salt minions are broken.
12:02 honestly let's investigate!
12:03 honestly EOFError: EOF read where object expected
12:04 rmt2 (1) how many minions do you have?
12:05 honestly 11
12:05 honestly not all of those are up though
12:06 honestly the minion on the master node is up
12:07 honestly 4 of my slave nodes have working minions
12:07 honestly 1 slave node is up, but the minion is broken
12:07 cshuman joined #salt
12:07 honestly throws a python stacktrace
12:08 honestly > Version: 0.16.3-1precise
12:09 rmt2 It sounds easy to figure out which one(s) are causing problems, then. ;-)
12:09 honestly oh, I have identified the one that is causing the problem
12:11 honestly well, this is nice
12:11 honestly Reading package lists... Error!
12:11 honestly E: Read error - read (5: Input/output error)
12:11 honestly E: The package lists or status file could not be parsed or opened.
12:11 honestly time to remount / ro
12:12 honestly root@bb8:~# mount -r -o remount /
12:12 honestly mount: you must specify the filesystem type
12:12 honestly ...ouch
12:13 rmt2 ;-)
12:13 Ryan_Lane joined #salt
12:13 rmt2 honestly, reinstall the bastard. :)
12:14 honestly yeah, sdcard must be busted
12:14 honestly why the fuck is there an apache running here o.O
12:15 honestly there is an apache running on all my slaves
12:15 honestly wtf
12:15 honestly ...teaches me to just take the image my boss gives me
12:17 abele joined #salt
12:18 TheCodeAssassin joined #salt
12:21 Chocobo joined #salt
12:22 juicer2 joined #salt
12:24 pabelanger joined #salt
12:26 Chocobo joined #salt
12:27 Chocobo joined #salt
12:27 Chocobo joined #salt
12:27 mwillhite joined #salt
12:29 honestly fsck.ext4: Superblock invalid, trying backup blocks...
12:29 honestly yeah, that one's a loss
12:30 rmt2 He probably gave you an image with some subtle filesystem corruption too.  Checked for hidden cameras in your office?
12:30 honestly lol
12:31 honestly this thing has been running for a while
12:31 honestly it's a beagle-board
12:31 rmt2 Ah, then there's a more-than-zero chance it's a practical joke.
12:31 honestly haha
12:31 honestly it's not a practical joke
12:32 honestly the beagleboards just like eating sd-cards
12:33 dthom91 joined #salt
12:34 mgw joined #salt
12:37 copelco joined #salt
12:38 krissaxton joined #salt
12:43 Ryan_Lane joined #salt
12:47 timoguin joined #salt
12:50 markm joined #salt
12:50 ipmb joined #salt
12:54 linjan_ joined #salt
12:55 piffio joined #salt
12:56 graingert joined #salt
12:57 graingert how do I call salt modules from my salt .sls
12:57 graingert eg salt.modules.djangomod
13:00 geak joined #salt
13:02 ggoZ joined #salt
13:03 ggoZ Hi
13:03 ggoZ quick question - does state.sls refresh pillar before applying state?
13:08 robinsmidsrod joined #salt
13:09 brianhicks joined #salt
13:10 racooper joined #salt
13:12 micah_chatt joined #salt
13:13 Gifflen joined #salt
13:14 Ryan_Lane joined #salt
13:15 matanya joined #salt
13:17 jrgifford joined #salt
13:19 markm joined #salt
13:22 Ahlee anybody successfully used tags with a gitfs backend?  docs say tags should be converted to a branch, but I can't seem to get targeting to work with salt '*' env=tag
13:23 m_george|away joined #salt
13:26 unicoletti_ left #salt
13:29 uta joined #salt
13:33 faldridge joined #salt
13:36 redbeard2 joined #salt
13:44 Ryan_Lane joined #salt
13:45 jergerber joined #salt
13:45 Kholloway joined #salt
13:48 Ahlee er, converted to an environment
13:48 micah_chatt joined #salt
13:49 m_george left #salt
13:50 pdayton joined #salt
13:50 krissaxton left #salt
13:55 mgw joined #salt
13:56 rgbkrk joined #salt
13:57 markm joined #salt
13:58 imaginarysteve joined #salt
14:02 micah_chatt For anyone who is a docker user, please check out https://github.com/saltstack/salt/pull/7517
14:12 opapo joined #salt
14:13 cshuman joined #salt
14:14 redbeard2 pabelanger: ping
14:14 Ryan_Lane joined #salt
14:15 redbeard2 micah_chatt: did you see my note on #7517?
14:17 jbunting joined #salt
14:19 micah_chatt redbeard2: yes! just saw it
14:20 torandu joined #salt
14:21 micah_chatt thank  you much!
14:21 JaredR joined #salt
14:22 anteaya joined #salt
14:22 diegows joined #salt
14:23 adepasquale joined #salt
14:24 Boohbah joined #salt
14:26 redbeard2 micah_chatt: i've very interested in developments with docker/salt. let me know if there's anything i can do to help.
14:26 cnelsonsic joined #salt
14:26 micah_chatt if you could run the tests I wrote, that would be AWESOME
14:27 cnelsonsic left #salt
14:27 redbeard2 let me hop over to that box and take a look
14:28 ingwaem joined #salt
14:30 micah_chatt redbeard2: I just updated the test too
14:31 micah_chatt redbeard2: I think it may be best to just leave the requirements at lxc-docker'
14:31 micah_chatt as its rapidly developing
14:31 micah_chatt thoughts?
14:31 martoss hey folks, is there a way to list all cloud profiles?
14:31 redbeard2 micah_chatt: that's probably good for now, since the travis tests are run on ubuntu anyway.
14:32 matanya joined #salt
14:32 martoss what I currently get is 'Profile mongodb is not defined' and no error which / if at all profiles were read in.
14:32 micah_chatt redbeard2: Great. Docker only officially supports ubuntu 12.04 with a  kernel upgrade and 13.04
14:32 micah_chatt and dotcloud just announced RedHat support
14:33 krissaxton joined #salt
14:40 MTecknology I'm to the point now where running state.highstate can take >2hr. Setting the timeout waiting for a return is a bit crazy. Is there any way to have a minion push the return to a separate server if it times out?
14:40 linuxnewbie joined #salt
14:40 linuxnewbie joined #salt
14:41 tyler-baker joined #salt
14:41 tyler-baker joined #salt
14:42 timoguin MTecknology, only if it times out?
14:42 timoguin I know you can set minions to return to wherever you want, e.g., logstash, but I don't know if you can do it only if it times out
14:42 MTecknology logstash?
14:43 timoguin or wherever you want, a database etc.
14:43 MTecknology Could I have it always send somewhere else and then also to the master if it returns in time?
14:43 timoguin logstash is a log indexer
14:43 timoguin I'm not sure, but I'd think it'd be possible to always return to both
14:43 ingwaem Eeek, I have a big problem with my salt stack, I'm getting messages ImportError: cannot import name Queue. anyone seen this before?
14:43 timoguin and obviously if it times out it's not going back to the master
14:44 MTecknology I feel like something already exists for this... I need to run off for 30min and look after that.. back soon!
14:44 timoguin MTecknology, http://docs.saltstack.com/ref/returners/
14:45 Ryan_Lane joined #salt
14:45 redbeard2 micah_chatt: looks like your tests are failing on my end
14:46 micah_chatt hmmm
14:46 micah_chatt cause I had them passing on ubuntu 13.04
14:46 micah_chatt redbeard2: are they all failing?
14:46 redbeard2 micah_chatt: well, centos over here and all
14:47 kiorky micah_chatt: no offense but i would prefer you to collaborte with me on my fork
14:47 micah_chatt kiorky: I just saw that
14:47 kiorky micah_chatt: i began to work on this a bit longer, and i have focused on a well designed api
14:47 micah_chatt Yea, yours is much more complete
14:47 kiorky micah_chatt: so gaining your module in salt directly will just make crap in it :(
14:47 micah_chatt I just whipped mine up over the weekend
14:48 kiorky micah_chatt: well, the discussion began on salt-cloud github page i dont remember well where ^^
14:48 micah_chatt kiorky: I had not seen any discussion and asked on the IRC channel a week or so ago if anyone knew of anything
14:48 kiorky micah_chatt: and i started to hack on it after that, but was not hurry to push as i want it to have enought quality to the initial merge
14:48 redbeard2 kiorky and micah_chatt: https://github.com/saltstack/salt-cloud/issues/615
14:49 micah_chatt Ahh. I'd only searched in the salt repo
14:51 micah_chatt kiorky: I'll go ahead and close it. Mind if I make some additions to yours and PR to you?
14:51 kiorky micah_chatt: yep, but atm, its my work project so im heavily hacking on it, so we need to coordinate if you want to work on it
14:52 kiorky micah_chatt: help is welcome :pp
14:52 micah_chatt kiorky: Yea, I have a passion for full documentation
14:52 micah_chatt So I may just start there to get a feel for what all is done and what needs to be completed
14:52 kiorky micah_chatt: im pretty sparse on comments and docs, i usually document only the neccessary
14:53 sgviking joined #salt
14:53 kiorky micah_chatt: just read the header, problem is that its very sequential to code :(, you cant write b if a is not written.
14:54 micah_chatt kiorky: sure
14:54 alunduil joined #salt
14:54 micah_chatt kiorky: I've found it the most helpful as a newer salt user to see docs with use examples
14:54 kiorky micah_chatt: what you can do is the "run" method, which use lxc-attach -n <ID> grabbed from inspect_container to run an arbitrary command on a container
14:54 kiorky micah_chatt: my idea was to mimic the cmd.run command but for containers
14:55 micah_chatt kiorky: that sounds like a good idea
14:55 kiorky micah_chatt: right now, on my own i m working to finish the create_container* functions, but i am fighting with docker API which doesnt want to do what i want :p
14:56 matanya joined #salt
14:56 micah_chatt kiorky: yea, I created 3 issues just last night to the docker-py repo. The API is still pretty young and changing a lot
14:56 L0j1k joined #salt
14:56 kiorky micah_chatt: why did you not use the python bindings ?
14:57 micah_chatt kiroky: In some cases, the python bindings didn't return anything, or were just broken.
14:57 micah_chatt *kiorky:
14:57 kiorky which methods ?
14:58 micah_chatt the quiet param is broken on client.containers
14:58 kiorky ha
14:58 kiorky did not tried it ^^
14:59 micah_chatt kiorky: yea, it was annoying as hell to try to figure out everything. I almost thought about doing an end run and just interacting directly with the docker API
15:01 kiorky micah_chatt: uhm, that surprises me, i had not that much problems with those bindings, they are just working for me
15:01 kiorky micah_chatt: with some buggix, thought
15:01 kiorky micah_chatt: see the openpullrequests :p
15:01 micah_chatt kiorky: I just noticed those
15:02 kiorky micah_chatt: what worries me more is that they are not merged yet
15:02 kiorky hope this is not in abandon status ...
15:02 micah_chatt kiorky: the docker-py seems a little half-assed
15:03 micah_chatt I just had a PR merged this morning though
15:03 kiorky micah_chatt: well, if it does not make it, fork is a solution ;)
15:03 micah_chatt or last night rather https://github.com/dotcloud/docker-py/pull/58
15:03 kiorky micah_chatt: uhm, so why mine did not get any activity.
15:04 markm joined #salt
15:04 kiorky micah_chatt: seems i will spam with /cc @shin- :p
15:04 micah_chatt kiorky: doooo it
15:05 kiorky micah_chatt: done.
15:05 kiorky micah_chatt: so do you want to start to do someting around my fork ?
15:05 micah_chatt kiorky redbeard2 I closed and commented on my PR
15:05 micah_chatt yes
15:06 kiorky micah_chatt: because the idea is that i always rebase to have only one commit the incoming pull request , and if you fork it i ll stop to do that.
15:06 redbeard2 micah_chatt and kiorky: i look forward to seeing another pull req... i'm excited to have docker support in salt
15:06 micah_chatt kiorky: yea, I like that rebase a lot
15:06 micah_chatt I'll fork it as soon as you do
15:06 pentabular joined #salt
15:06 kiorky micah_chatt: but rebase while collaborating isnt praticall at all ;)
15:06 micah_chatt right
15:07 micah_chatt kiorky: sure, it makes sense.
15:07 micah_chatt kiorky: it keeps the commit history from being bloated
15:07 kiorky micah_chatt: we will just do it before the PR...
15:08 micah_chatt kiorky: sounds great
15:08 kiorky micah_chatt: so what are you doing, the run method or elsething ?
15:08 micah_chatt I'll start with run
15:08 kiorky ok !
15:09 kiorky micah_chatt: so sorry for your work
15:09 kiorky micah_chatt: i know what it is to discover already done work ...
15:09 kiorky so sorry about that :(
15:09 micah_chatt kiorky: and add a few doc notes. Docker seems particularly hard for people to get their heads around when learning it
15:10 kiorky micah_chatt: well, i think we do not use it at makinacorpus like everyones does :p
15:10 kaptk2 joined #salt
15:10 kiorky micah_chatt: we use it more to manage our wms based on LXC... just to have a more friendly layer around it
15:11 Ryan_Lane joined #salt
15:11 kiorky micah_chatt: so i am not the best person to document it in the mainstream use of a monoprocess thing :)
15:11 lineman60 joined #salt
15:11 micah_chatt kiorky: thanks. Its a little bummer, but I'll be glad to see a docker support in salt. I'll also make an issue in Salt for it so nobody else does the same thign
15:12 zooz joined #salt
15:12 kiorky micah_chatt: good idea !
15:12 mannyt joined #salt
15:13 pentabular joined #salt
15:14 kiorky micah_chatt: any chance you know shin- IRC nick ?
15:14 micah_chatt kiorky: nope. Sorry
15:14 micah_chatt kiorky: he works for dotCloud, ping the #docker channel
15:16 micah_chatt or tweet @docker, that'll get their attention
15:17 forrest joined #salt
15:18 kiorky micah_chatt: i tweeted him directly, i hate those social medias
15:20 danielbachhuber joined #salt
15:21 micah_chatt kiorky: do you want PR's to your fork or makinacorpus?
15:21 pabelanger redbeard2, pong
15:21 kiorky micah_chatt: If you dont mind yes, it would be great to do the PR here !
15:21 hazzadous joined #salt
15:21 kiorky micah_chatt: (it is a yes)
15:22 jalbretsen joined #salt
15:24 MTecknology and back
15:24 MTecknology timoguin: thanks!
15:26 micah_chatt kiorky: haha, to your fork or makinacorpus' fork?
15:26 cro joined #salt
15:27 kiorky micah_chatt: yep
15:27 kiorky micah_chatt: docker branch
15:27 micah_chatt kiorky: ha its an OR question
15:27 Mef1 joined #salt
15:29 kiorky micah_chatt: uhm, i just want you to know that the code resides in the docker branch of makinacorpus fork and that you need to PR in this repo
15:29 Mef1 Quick question:  The documentation claims that in pillar you can include another sly file.  I've tried to include another pillar file (basically to put in setting of parameters in one file) but it does not seem to be working properly.  I've looked and there appears to be some bugs on this .  Has anybody made this work?
15:29 micah_chatt kiorky: ah ok. sorry for the confusion
15:29 morty_ joined #salt
15:30 APLU joined #salt
15:31 krissaxton joined #salt
15:31 forrest Mef1, did you take a look at https://groups.google.com/forum/#!topic/salt-users/gH7DHC0Ck88/discussion
15:31 torandu joined #salt
15:31 torandu joined #salt
15:32 lahwran joined #salt
15:33 Mef1 yes that is from 4/18 so I was checking in.  So this is still work in progress, correct?  You can't say include: <sls file> and then use the values in that pillar file that does the include?
15:36 forrest Mef1, I don't know off hand.
15:37 Mef1 ok, well I haven't been able to make it work so I may just move on and be less "clean" for now.  If anybody has made it work that would be great to know, otherwise I'll just move on
15:37 forrest It states that it works with version 0.16.0, but if it isn't working for you, then it might still be a problem .
15:37 forrest the file lives in your /srv/pillar dir right?
15:37 Mef1 Yes forrest I'll send you the pastebin
15:39 markm joined #salt
15:42 MTecknology snmpd is a pain in the ass... https://gist.github.com/MTecknology/6cb2d6fcd708cfbecf93 <-- This is what I had to do to finally get it working correctly
15:42 MTecknology I actually have to have the service off in order to add users... :S
15:43 Mef1 forrest here it is http://pastebin.com/CmcF4Teg
15:43 jdenning joined #salt
15:43 TheCodeAssassin joined #salt
15:44 forrest MTecknology, You should consider splitting that out, addning a jinja mapping for users and such, and then ask to have it added to saltstack-formulas
15:45 troyready joined #salt
15:45 MTecknology I really feel like there has to be something better than this. It requires libsnmp15 but the latest snmpd requires libsnmp30.
15:46 MTecknology If I had a test environment I'd be playing to see if maybe it's much easier in the latest snmpd.
15:46 canci joined #salt
15:46 forrest Mef1, can you try to move the agn_keystone_jinja_params to /srv/pillar, then change your include to reference it there?
15:50 Mef1 move it into /srv/pillar/agn..sls  changed the include state to get rid of the namespaces and just included agn...  BTW In all cases there was no error indicating that it couldn't find the include file, I don't know if this is relevant but I would have hoped that if it couldn't include it it would have given an error
15:51 Mef1 Oh it did not work after moving
15:51 forrest ok
15:51 Mef1 From the syntax I believe what I'm doing is correct so it makes me believe it just doesn't work.
15:51 forrest Yes I believe what you are doing is correct as well
15:52 forrest I don't really see why it can't import though, it's just another sls like any other.
15:52 forrest I don't see what makes pillar unable to do that.
15:52 Mef1 Do you know anybody that has been able to make it work?
15:53 faldridge joined #salt
15:56 forrest Mef1, I don't know of any examples where it does, no
15:56 forrest I just see people using work arounds to import the variables they need
15:56 Mef1 Ok time to move on.  Thanks for your help, I'll just partition my pillar into a parameters section and then when it is fixed I can move them out of the file
15:57 forrest Sounds good, I'll try to remember to ask UtahDave when I see him around today.
15:57 forrest or one of the other devs if they pop in
15:58 geak_ joined #salt
16:02 canci joined #salt
16:05 pentabular1 joined #salt
16:05 JaredR joined #salt
16:06 SoR joined #salt
16:13 ifnull joined #salt
16:14 canci joined #salt
16:15 kaptk2 joined #salt
16:15 pipps joined #salt
16:16 SoR joined #salt
16:16 KyleG joined #salt
16:16 KyleG joined #salt
16:17 az87c joined #salt
16:20 Lue_4911 joined #salt
16:21 xuru joined #salt
16:27 geak joined #salt
16:36 robbyt joined #salt
16:43 c0bra joined #salt
16:44 jaequery joined #salt
16:47 c0bra hey guys
16:48 c0bra would there be any issues without setting up salt minion on a bunch of machines when the master isn't set up yet? Like if we had a bunch of laptops we wanted to eventually provision down the road, but weren't ready to do right away. Could we set up a master at the specified hostname later on and then the minions would eventually connect?
16:49 forrest c0bra, yes
16:49 forrest Just turn the minion service off
16:49 c0bra It couldn't be left on?
16:49 forrest and when you're ready turn it back on, and then accept all the keys on the master
16:49 forrest You could probably leave it on
16:49 forrest I just don't know if it will log that it can't connect to the master
16:50 forrest or if that's even a concern to you that it will be making requests to an IP/hostname that isn't assigned.
16:50 ggoZ joined #salt
16:50 forrest or available at least.
16:51 c0bra I don't think it would matter
16:51 jbunting joined #salt
16:51 c0bra as long as it's not blocking the machine or using up excessive cycles
16:52 forrest In that case you should be good to go
16:53 c0bra sweet
16:54 c0bra I still haven't figured out a way to pull from local state files on a Windows machine.  But maybe we can live without that
16:56 CheKoLyN joined #salt
17:02 pipps joined #salt
17:05 SunSparc joined #salt
17:06 KennethWilke joined #salt
17:09 morty_ joined #salt
17:13 xuru Could someone help me out with getting gitfs working correctly?
17:13 jcockhren xuru: I can help
17:14 xuru I have all the files up on a gist:  https://gist.github.com/xuru/6766985
17:14 danielbachhuber joined #salt
17:14 xuru I'm using the ntp-formula as an example to get started with
17:14 ml_1 joined #salt
17:15 xuru I get the error: no states found for this minion
17:15 xuru and [ERROR   ] No matching sls found for 'ntp.client' in env 'base'
17:17 jcockhren xuru: dot-notation is for finding state files
17:17 jcockhren xuru: ntp.client translates to: ntp/client.sls
17:17 xuru So how do I just install ntp client?
17:22 morganfainberg joined #salt
17:22 jcockhren xuru: your top file need to call a state file that exists
17:22 jcockhren needs*
17:23 jcockhren state top file
17:24 xuru so create another sls file under the salt directory, and include ntp.client (adding that sls file to top.sls)?
17:25 xuru If I do that, I get:  Unknown include: Specified SLS base: ntp.client is not available on the salt master in environment(s): base
17:25 CheKoLyN joined #salt
17:26 jcockhren xuru: I've added a comment
17:26 jcockhren to your just
17:26 jcockhren gist*
17:26 xuru ok, cool… looking
17:26 jcockhren what you're calling from your states top file is a states file
17:27 jcockhren doing "- ntp.client" means it's going to look for "ntp/client.sls"
17:27 jcockhren doing "- ntp" means it's going to look for "ntp/init.sls"
17:28 jcockhren what it does NOT do is pull from your pillar unless you explicitly tell it to
17:28 xuru but, there is an ntp/client.sls in the formua:  https://github.com/saltstack-formulas/ntp-formula
17:29 blee_ joined #salt
17:31 troyready joined #salt
17:31 Katafalkas joined #salt
17:31 xuru I guess I'm confused about how to call the ntp-formula to do the work.  putting in the pkg.installed doesn't call that ntp-formula to install the basic config file with my pillar data
17:33 mdrnstm joined #salt
17:33 jcockhren xuru: ok
17:33 jcockhren sorry. I forgot .17 dropped
17:33 lacrymology joined #salt
17:34 lacrymology does anyone know how can I run a state through a debugger (preferrably pycharm's debugger, but pdb is good too)?
17:35 forrest lacrymology, as far as I'm aware at this time there is no debugger for state files. You can run them with the 'test' option so the change don't get applied, but there is no syntax validater at this time.
17:36 forrest *changes
17:37 lacrymology forrest: no, not state files. I have a custom state module, and I want to debug it
17:37 jcockhren xuru: on the master, could you stop the daemon and run it like:
17:37 jcockhren salt-master -l debug
17:37 jcockhren gist that output
17:37 forrest lacrymology, oh I see. I'm not sure then.
17:37 xuru Sure, I'm running master on the same machine (i.e. through vagrant)
17:38 jbunting joined #salt
17:39 jcockhren and you've confirmed that it can reach the outside world right?
17:39 jcockhren also, git is installed?
17:39 xuru git is installed, and it can get out (i.e. apt-get update)
17:40 xuru but I see this:  Git fileserver backend is enabled in configuration but could not be loaded, is GitPython installed?
17:40 forrest what version of gitpython are you running?
17:41 xuru I haven't explicitly installed it.  I had assumed that the bootstrap script for salted-vagrant would have installed it :/
17:41 xuru Installing it now...
17:42 xuru GitPython exception caught while fetching: len(["8e1c659d9135c94a13fc31cf2a60466c9e6ce83d\tnot-for-merge\tbranch 'master' of https://github.com/saltstack-formulas/ntp-formula\n"]) != len(['POST git-upload-pack (148 bytes)', ' * [new branch]      master     -> origin/master'])
17:43 xuru GitPython 0.3.2.RC1
17:43 forrest what does your URL look like?
17:44 kiorky micah_chatt: where are you ? i have finished import and create, so now if you are not done i would like to go on run* methods
17:44 micah_chatt kiorky: go for it
17:44 micah_chatt kiorky: I've had back to back meetings
17:44 kiorky micah_chatt: i'm going for a footing right now :p so just ping me if you have done them else i should do it in 2 hours
17:45 kiorky micah_chatt: which timezone are you in?
17:45 micah_chatt kiorky: ok. I did a bit of doc clean-up and added the __virtual__ method
17:45 xuru forrest: https://github.com/saltstack-formulas/ntp-formula.git
17:45 micah_chatt kiorky: EST,
17:45 kiorky rah which GMT ?
17:45 kiorky im GMT+2 and i always have an hell to convert time
17:45 kiorky ^^
17:46 micah_chatt haha
17:46 micah_chatt GMT -4
17:46 forrest xuru, can you modify your url to be git://github.com/saltstack-formulas/ntp-formula.git .
17:46 kiorky micah_chatt: so 15h ?
17:46 kiorky micah_chatt: (its 20h here)
17:47 micah_chatt its almost 14:00 here
17:47 xuru forrest: that seemed to work.  Going to run the minion now
17:47 forrest Cool.
17:47 kiorky micah_chatt: when i say you that im really bad at time conversion :)
17:47 forrest You guys know the internet has time converters right? :P
17:48 micah_chatt kiorky: no worries, forrest: I am all over it
17:48 forrest lol ok
17:48 micah_chatt ;)
17:48 kiorky micah_chatt: ok, so ill try to see what you have done so far when i come back, push on your repo, and i ll pull from it  (can you give me the url)
17:48 micah_chatt kiorky, I can PR to your fork
17:48 xuru forrest: that didthe trick!
17:48 xuru thanks
17:48 forrest xuru, Excellent, and np.
17:49 xuru jcockhren: Thanks!
17:49 kiorky micah_chatt: just try to push often to your repo
17:49 micah_chatt sure
17:50 jcockhren xuru: np
17:56 xuru forrest: salty-vagrant pulls down https://github.com/saltstack/salt-bootstrap for it's bootstrap script.  I don't see anywhere in the script that it installs GitPython.  Would this be a bug?  Or is it by intent?
17:57 forrest xuru, let's ping akoumjian about that one, he does most of the vagrant stuff, and I'm not sure if that is by design or not.
17:57 forrest oh the bootstrap itself, duh
17:57 forrest let me look
17:57 forrest sorry akoumjian when you see this ignore it :P
17:58 xuru yeah, it seems to me that it should at least be an switch to the script...
17:58 forrest Hey s0undt3ch, is the lack of installing gitpython in the bootstrap by design?
18:01 forrest xuru, yea that might have just been overlooked, what distro are you on by the way?
18:01 xuru precise64
18:01 xuru ubuntu
18:02 xuru And I'm using type "stable" with the bootstrap script
18:02 forrest Yea it's not in the develop branch, I already double checked.
18:03 forrest If we don't hear from s0undt3ch in a few minutes, I'll file an issue about it.
18:03 xuru I noticed that I also had to install git
18:04 forrest on ubuntu it installs git-core for the develop, let's look at stable
18:05 xuru yeah, git-core isn't installed on it, but git is (that I did manually)
18:05 forrest git-core should be installed when it runs through the ubuntu deps
18:05 druonysuse joined #salt
18:05 druonysuse joined #salt
18:06 xuru maybe I'm blind, but I don't see install_ubuntu_git_deps being called at all
18:07 xuru ah, he constructs the list of functions to call...
18:08 alunduil joined #salt
18:09 xuru Ok, so it looks like install_ubuntu_git_deps are only called if using the install type of 'git'
18:09 StDiluted joined #salt
18:10 forrest salty-vagrant is just pulling down the stable right?
18:10 forrest when it does the call to the bootstrap?
18:11 xuru you can specify which you want
18:11 xuru I specified "stable"
18:11 forrest gotcha
18:12 xuru So really, it's just a matter of "Do we want git-core and GitPython to always be installed or not?"
18:12 forrest Yea I'm filing an issue on it now.
18:13 xuru Yeah, I'd favor always installing it, so when someone switches to using gitfs, they aren't surprised it doesn't work.  However, are there security issues with having it on a production machine?
18:13 forrest I don't know.
18:15 pmrowla joined #salt
18:16 forrest https://github.com/saltstack/salt-bootstrap/issues/245
18:16 forrest We'll see what the result of that is xuru, and if it's no biggie I'll modify the code when I get home.
18:16 xuru no problem.  It's just an example anyway.  Thanks!
18:16 forrest granted that will be the develop branch, there is still some code to finish up with some Ubuntu issues I believe.
18:16 forrest yea np.
18:17 forrest Oh actually it looks like the stable branch was updated a few days ago, awesome.
18:19 xuru cool.  I'll add some comments to the ticket.  If there is agreement, I can do a pull request
18:20 forrest sounds good!
18:23 nliadm is there any more concrete details on 0.17.0 release?
18:23 diegows joined #salt
18:24 cedwards nliadm: i believe it's supposed to be announced soon
18:24 forrest nliadm, there's http://docs.saltstack.com/topics/releases/0.17.0.html as well as http://www.youtube.com/watch?v=Z2G8jsp_14A&amp;feature=c4-overview&amp;list=UUpveTIucFx9ljGelW63-BWg
18:24 nliadm :D
18:24 bpgoldsb I've been a long time puppet user.  I'm interested in trying out salt, though mostly for orchestration.  Has anyone seen any good information on a hybrid environment, or using salt for ONLY orchestration?
18:24 forrest nliadm, 0.17 is out now.
18:24 cedwards nliadm: i heard today
18:25 cedwards forrest: it's been released to packagers but not formally announced yet
18:25 mmilano joined #salt
18:25 forrest pssh, it's available in git :P
18:25 s0undt3ch_ joined #salt
18:26 unicolet joined #salt
18:28 forrest bpgoldsb, are you talking about specifically for replacing MCollective?
18:28 KyleG joined #salt
18:28 KyleG joined #salt
18:28 forrest I have a meeting I have to go to bpgoldsb, if you don't get your questions answered I'll be back later
18:39 cachedout joined #salt
18:39 robertkeizer joined #salt
18:40 aleszoulek joined #salt
18:42 Katafalkas joined #salt
18:48 ml_11 joined #salt
18:49 pipps joined #salt
18:58 juicer2 joined #salt
18:58 vipul joined #salt
19:02 ifnull_ joined #salt
19:04 KyleG joined #salt
19:04 KyleG joined #salt
19:05 asachs joined #salt
19:06 asachs Hi all, does anyone know how to call a module like say rabbitmq from a sls file ?
19:07 lineman60__ joined #salt
19:10 pdayton joined #salt
19:14 timoguin_ joined #salt
19:15 felixhummel joined #salt
19:17 qba73_ joined #salt
19:19 RoboTeddy joined #salt
19:20 RoboTeddy salt 0.17.0 isn't the stable release right now, right?
19:20 Koma joined #salt
19:22 copelco joined #salt
19:22 RoboTeddy (salt-bootstrap seems to have installed it for me, instead of 0.16.4)
19:26 linuxnewbie joined #salt
19:28 linuxnewbie hello, should salt-minion create file_roots if i have defined them in the /etc/salt/minion ?
19:30 forrest RoboTeddy, it installed when you ran the stable right?
19:30 RoboTeddy forrest: `curl -L http://bootstrap.saltstack.org | sudo sh -s -- -N -M` on ubuntu 12.04.3 installed salt 0.17.0
19:30 forrest asachs, you mean like this: http://docs.saltstack.com/ref/modules/all/salt.modules.rabbitmq.html ?
19:31 forrest RoboTeddy, I thought that 0.17.0 was now the official release, but other people are saying otherwise. Did that install an RPM/DEB?
19:31 forrest linuxnewbie, you would need to create the file_roots directories.
19:32 RoboTeddy forrest: yep it used apt
19:32 linuxnewbie forrest, thank you
19:32 forrest np.
19:32 kenbolton joined #salt
19:32 asachs forrest: like that but i want to call the add_user from inside an sls, or have I missed the boat ?
19:32 forrest cedwards, see I told you 0.17 is the official :P
19:32 jinnko joined #salt
19:32 linuxnewbie can u show me how is your structure ?
19:32 lacrymology what am I doing wrong? `$ salt-call -c conf/ state.high '{"foo": { "uwsgi": ["enabled", {"source": "salt://path/to/file.jinja"}, {"template": "jinja"}, {"watch": [{"file": "/etc/config.conf"}]}]}'`  >>>> AttributeError: 'str' object has no attribute 'items'
19:32 linuxnewbie forrest, can u show me how is your structure ?
19:32 RoboTeddy forrest: http://docs.saltstack.com/ says 0.16.4
19:32 linuxnewbie just to have an example
19:33 forrest RoboTeddy, I know, there is question about that.
19:33 RoboTeddy but also, salt-cloud installed 0.16.4 salt-minions, and the bootstrap script installed 0.17.0 salt-master, and the two don't seem so friendly
19:33 havan joined #salt
19:33 * RoboTeddy wants a way to tell salt-cloud which version of salt-minion to install
19:33 forrest linuxnewbie I don't have an example on me, the docs: http://docs.saltstack.com/ref/file_server/file_roots.html have an example where they have the file_roots defined, and then those just exist
19:34 linuxnewbie thank you for everithing  forrest
19:34 forrest linuxnewbie, np, sorry I don't have more for you
19:34 linuxnewbie it's super OK ...i'll read documentation and i hope it will be ok :P
19:35 KyleG1 joined #salt
19:35 lacrymology ah
19:35 lacrymology I was missing a }
19:36 rgbkrk joined #salt
19:36 forrest asachs, http://docs.saltstack.com/ref/states/all/salt.states.rabbitmq_user.html
19:37 forrest RoboTeddy, from what I understood, 0.17.0 was released last week in a stable form for the first release (it's no longer 0.17.0rc1 in git).
19:37 asachs forrest: wow, my google foo is super weak today, thanks a stack :)
19:37 forrest However several people in the IRC are disputing that, so without data from the saltstack team I can't confirm
19:37 forrest asachs, yea np, it actually wasn't coming up very easily, had to go somewhere else then find that.
19:38 RoboTeddy forrest: ok cool, thanks. but then, question is... why did salt-cloud 0.16.4? it should be using the same bootstrap script
19:38 forrest I thought it would just be salt.states.rabbitmq, but it wasn't
19:38 RoboTeddy *salt-cloud installed salt-minion 0.16.4
19:39 forrest RoboTeddy, so when you just ran the bootscript with the stable option, it installed 0.17.0, but when you installed via salt-cloud it installed 0.16.4?
19:39 RoboTeddy forrest: I ran the bootscript with no option about stability (which should default to `stable`) and 0.17.0 was installed; but salt-cloud installed 0.16.4 on the same ubunti image (correct)
19:40 RoboTeddy and 0.17.0 doesn't seem to be entirely backwards compatible, with no mention of it in the release notes
19:40 * RoboTeddy wonders if he should be using salt :(
19:42 LucasCozy joined #salt
19:42 forrest RoboTeddy, 1.5.8 is the default version of the bootstrap that will be run it looks like in salt-cloud. Can you confirm with another run of salt-cloud (or if you have those old runs) that it pulled from the same repo location as the bootstrap did, I don't see why it wouldn't...
19:42 forrest And what backwards compatibility is broken?
19:43 RoboTeddy forrest: perhaps it's because I ran apt-get update prior to installing salt-master, but perhaps salt-cloud doesn't do the same before it installs salt-minion
19:45 forrest RoboTeddy, the bootstrap has this option: -U  If set, fully upgrade the system prior to bootstrapping salt, but https://github.com/saltstack/salt-bootstrap/blob/develop/bootstrap-salt.sh#L1253 seems to say apt-get update should run
19:45 forrest I've gotta run to lunch real quick, can you try to provision another machine using salt-cloud after running an apt-get update to see what happens?
19:46 RoboTeddy forrest: OK, will do, thanks
19:51 lacrymology does anyone know any way of calling salt.script.salt_call from python?
19:53 pipps1 joined #salt
19:54 cachedout Could somebody clarify some documentation for me, please? http://docs.saltstack.com/ref/states/highstate.html references something called 'highdata' in the sentence before the last code block on the page. Is that a typo or something that I'm not yet familiar with?
19:55 RoboTeddy https://github.com/saltstack/salt-cloud/blob/develop/saltcloud/deploy/bootstrap-salt.sh was only updated to salt-boostrap 1.5.8 3 days ago; I'm guessing that the salt-cloud ubuntu package is on an older version
19:55 RoboTeddy hence different versions of salt-bootstrap were likely being run (curl grabs the current version; salt-cloud was using the old one)
19:59 RoboTeddy oh, no: that's not the issue at all: it looks like apt-get install salt-cloud installed salt-master 0.17.0 on top of my 0.16.4 install
19:59 RoboTeddy salt-master --version (shows 0.16.4); sudo USE_SETUPTOOLS=1 easy_install salt-cloud; salt-master --version (shows 0.17.0)
20:00 rlarkin I'm just getting started here with my first install/setup.  I want the salt master to do work, run commands, etc triggered by an event on a minion, like a new key accepted.  What salt specific vocabulary describes this?
20:03 RoboTeddy (nevermind me: I should've been installing salt-cloud using apt rather than easy_install, since the easy_install version also installed salt 0.17.0)
20:03 timoguin rlarkin, you want to look into the event and reactor systems
20:03 rlarkin thank you timoguin
20:03 hunter joined #salt
20:04 timoguin Is the salt windows installer still open source? I notice this hasn't had any commits in a year: https://github.com/saltstack/salt-windows-install
20:04 timoguin but new installers have been built since then
20:05 timoguin oh it's in the pkg directory of the salt repo.
20:05 timoguin carry on!
20:06 Damoun joined #salt
20:09 Thiggy joined #salt
20:19 EntropyWorks so just pushed some changes to my https://github.com/EntropyWorks/salt-openstack working out some issues like adding a script to create the mysql users since I did that originally in a different salt state but I should just add it to this instead.
20:21 EntropyWorks also need to redo my example pillar stuff since thats dated now. but if your using salt to deploy openstack on in a multi node environment using nova network instead of newtron you may find my stuff handy
20:21 ranl joined #salt
20:21 havan joined #salt
20:31 zooz joined #salt
20:35 giantlock joined #salt
20:36 xuru Has anyone ever used "roots" and "git" fileserver_backend together?
20:36 xuru It seems all I can do now is gitfs
20:38 woebtz joined #salt
20:40 backjlack joined #salt
20:43 ricicle joined #salt
20:44 APLU joined #salt
20:45 c0bra hey guys
20:46 LGSilva joined #salt
20:48 LGSilva hey guys, I was trying to run one of my states today but salt['pkg.perform_cmp']('2.99.22.4',  '2.99.22.6-1') function is mia. Isn't that supported anymore?
20:51 ronc joined #salt
20:52 APLU joined #salt
20:54 kiorky terminalmage: any chance you there ?
20:54 terminalmage kiorky: yes I'm here
20:54 kiorky terminalmage: i think regression just masked something weird :p
20:54 terminalmage LGSilva: that function has been removed
20:54 terminalmage that was never intended to be used on the CLI
20:54 kiorky terminalmage: https://github.com/saltstack/salt/pull/7522
20:56 LGSilva terminalmage: thanks for the info. I was using it on jinja to uninstall a package if a specific version was found.
20:56 bhosmer joined #salt
20:57 mordred joined #salt
20:57 terminalmage LGSilva: so, is there a reason you can't use salt['pkg.version'
20:57 LGSilva I just updated my code to use that
20:57 terminalmage LGSilva: salt['pkg.version']('pkgname')
20:57 terminalmage ok
20:58 terminalmage kiorky: I don't follow
20:58 pipps1 joined #salt
20:58 LGSilva is something broken with salt 0.17? I'm getting lots of conflictiong IDs
20:58 LGSilva that worked fine friday... :/
20:58 terminalmage LGSilva: more information required
20:59 LGSilva Data failed to compile:
20:59 LGSilva ----------
20:59 LGSilva Detected conflicting IDs, SLS IDs need to be globally unique.
20:59 LGSilva The conflicting ID is "add_server_key_to_github" and is found in SLS "base:github" and SLS "base:github"
20:59 kiorky terminalmage: what are you not following ?
21:00 terminalmage LGSilva: Please pastebin that SLS file
21:00 terminalmage kiorky: I don't understand your traceback
21:00 terminalmage sorry
21:00 tyler-baker joined #salt
21:00 LGSilva oh, I guess I know what's wrong. If i use "include" in a state then call that include by itself later on on the top.sls I will probably get those errors right?
21:00 kiorky terminalmage: with your changeset, saltcall is even more broken
21:00 kiorky terminalmage: uhm no
21:01 kiorky terminalmage: with output=highstate, before and with applying your changeset i get a traceback.
21:01 kiorky terminalmage: rah, there is another thing. (i did not have such traceback before)
21:01 terminalmage LGSilva: what do you mean by "call that include by itself"
21:02 kiorky terminalmage: with default output or yaml one, no traceback, so it must be really the hightstate output
21:02 ede joined #salt
21:02 terminalmage kiorky: possibly. it was modified between 0.16.0 and 0.17.0 to add a counter for failed and successful states
21:03 f47h3r joined #salt
21:04 kiorky terminalmage: will try to fix ...
21:05 LGSilva lets say on the top file use run the state github and inside the github state we include openssh state. Then after the github entry on the top file we have the openssh state entry.
21:05 LGSilva @terminalmage ^
21:06 robertkeizer joined #salt
21:06 terminalmage LGSilva: the state renderer should recognize those kind of duplicates and ignore them
21:07 terminalmage please open an issue on the tracker
21:07 terminalmage and please include examples of your top.sls, github.sls, and openssh.sls
21:07 LGSilva ok. thanks
21:07 hazzadous joined #salt
21:08 terminalmage np
21:08 kiorky terminalmage: uhm, there may be a sense for double nesting afterall, for function calls
21:13 lacrymology does anyone know a good spot to initialize python execution modules' "globals"?
21:14 lacrymology calling __salt__ or __pillar__ from the module level fails. I thought calling salt.module.pillar directly would avoid this, but no good
21:17 pipps joined #salt
21:20 ede joined #salt
21:26 kiorky terminalmage: was using with non highstate struct ... what an idiot
21:26 kiorky terminalmage: nice commit !
21:26 kiorky terminalmage: i know more about salt ouputters now :p
21:27 kiorky lacrymology: uhm there is something in the doc for that
21:27 kiorky lacrymology: http://docs.saltstack.com/ref/states/writing.html#mod-init-interface
21:27 terminalmage kiorky: oh, so it was just the struct you were returning?
21:28 kiorky terminalmage: yep
21:28 terminalmage got it
21:28 terminalmage we should probably handle that more gracefully though
21:28 kiorky terminalmage: like every error in the world :)
21:28 terminalmage :)
21:29 MTecknology bleck... I need to rework how I'm storing stuff for servers and telling salt what goes to which system...
21:29 MTecknology :(
21:30 terminalmage kiorky: can you add a comment to my pull request then?
21:30 terminalmage if you haven't already
21:30 terminalmage so that others don't think there's a blocker for that req to be merged
21:31 Thiggy joined #salt
21:31 pentabular1 left #salt
21:31 pentabular joined #salt
21:31 kiorky terminalmage: i have done it prior bugging you on irc :)
21:32 terminalmage kiorky: cool, thanks!
21:32 zooz joined #salt
21:33 cshuman joined #salt
21:33 ifnull__ joined #salt
21:35 Mef1 joined #salt
21:35 Mef1 Has anybody been able to get states file.copy to work?
21:35 Mef1 Here is a pastebin http://pastebin.com/EVz0WqVv that is a simple test that does not work.
21:40 boite joined #salt
21:40 forrest Mef1, you aren't mixing file.copy: with things above like file: - copy are you?
21:42 Mef1 Forrest, not sure I'm trying to use this salt.states.file.copy(name, source, force=False, makedirs=False)¶
21:42 Mef1 the pastebin has the approach I'm taking.
21:42 forrest I know you are, I should have asked more clearly. does file.copy exist in a state file with other calls?
21:43 lacrymology kiorky: that sounds like it's only for state modules, not execution modules
21:44 Mef1 If i understand your question no.  Its in its own "stanza" (not sure of the proper name for the label".  There are other "stanza" in my terminology in the file but not no other function/state in that label
21:45 forrest so what I'm asking is in your state file where that lives, you have other sections. Is there any mixing of formatting
21:46 tyler-baker joined #salt
21:48 hunter joined #salt
21:48 Mef1 forrest the answer is yes but I just commented out every line except for that file.copy (i.e.copy_test:) and it gave me the same error.
21:49 Mef1 Does file.copy work the way I'm writing it or am I missing something?
21:50 forrest I don't see why that wouldn't work, can you add the debug flag to see if we can get more details when you make the call Mef1?
21:51 Heartmender joined #salt
21:51 Heartmender joined #salt
21:52 Mef1 sure btw I'm running 0.16.3 if that would matter
21:54 forrest Well, it's not there in 0.16.0: https://github.com/saltstack/salt/blob/0.16/salt/states/file.py unless I'm missing it
21:54 ricicle joined #salt
21:54 forrest there's no tag for 0.16.3...
21:54 forrest oh nevermind there it is
21:54 forrest fail
21:55 forrest :\ Mef1, https://github.com/saltstack/salt/blob/v0.17/salt/states/file.py
21:55 forrest def copy is in there
21:55 forrest but it's not on 0.16.4, or 0.16.3. So either it was moved in some fashion, or it's new for 0.17
21:57 ifnull__ I'm trying to call salt.fileserver.gitfs.envs() from python but I keep getting an error. "global name '__opts__' is not defined". http://pastebin.com/hVGYFyaX
21:59 forrest ifnull__, https://github.com/saltstack/salt/blob/develop/salt/fileserver/gitfs.py#L179
22:00 Mef1 forrest I'm not clear on your comment, yes its in 0.17 but you didn't find it in 0.16.4 or 0.16.3, correct?  Then that is my issue.
22:00 Mef1 I did create a new pastebin but if its as simple as not being the right release then I won't bother you
22:00 ifnull__ @forrest, is there something i should be importing first that defined __opts__['gitfs_base']? .. or is that something i need to define manually
22:00 forrest ifnull__, I don't know
22:01 forrest I just figured maybe that would help put you on the right track.
22:01 ifnull__ thanks. i was just looking at that file.
22:01 forrest Mef1, Yea either it isn't in the earlier releases, or it was somehow moved. I would go with it only being in 0.17.0 based on https://github.com/saltstack/salt/issues/6440 and https://github.com/saltstack/salt/pull/6446 being 2 months old
22:01 MeanderingCode joined #salt
22:02 forrest but that was also completed 2 months ago, so I'm not sure why it isn't on 0.16.4
22:02 ifnull__ anyone else familiar with the gitfs module?
22:02 forrest ifnull__, no, sorry I can't be of more assistance
22:03 forrest *np
22:03 marcinkuzminski joined #salt
22:05 marcinkuzminski joined #salt
22:10 lynxman joined #salt
22:10 lynxman joined #salt
22:10 emilisto joined #salt
22:10 luminous joined #salt
22:11 scooby2 joined #salt
22:11 andyshinn` joined #salt
22:11 jeffrubic joined #salt
22:11 ewong- joined #salt
22:16 backjlack_ joined #salt
22:17 ifnull__ Does anyone know if there is a CLI equivalent for gitfs.envs
22:38 djn theres an error in the docs: http://docs.saltstack.com/ref/configuration/logging/handlers/salt.log.handlers.logstash_mod.html previous and next topic are the same link
22:40 forrest djn, that's because as far as I'm aware those are the only 2 logging handlers.
22:40 terminalmage this is not a bug
22:40 ifnull__ joined #salt
22:40 terminalmage forrest: correct
22:41 jcockhren djn: its a feature
22:41 terminalmage the sphinx autosummary is what generates these next/previous links
22:41 terminalmage and there are only two
22:41 terminalmage maybe there is a workaround, I dunno
22:42 terminalmage but it's not really a priority to investigate since as soon as a third log handler is added, this will be moot
22:42 terminalmage if someone wants to look into it and submit a pull req, go for it
22:45 mianos joined #salt
22:46 matanya joined #salt
22:54 cshuman joined #salt
23:05 alekibango joined #salt
23:09 micah_chatt joined #salt
23:11 kula so, this is an entirely new one to me: https://gist.github.com/kula/6771634
23:16 cachedout kula, is 'import os' present in yumpkg.py (probably around line 19)?
23:16 bhosmer joined #salt
23:17 kula yep
23:17 cachedout How odd then
23:18 kula if you re-load that gist and look at the first comment, you can see me loading yumpkg.py with 'python2.6 -i' and manually running list_repos() and it working just fine
23:18 kula i have absolutely no idea how this is failing
23:18 cachedout Ah, I was just taking that very approach.
23:19 cachedout Could be related: https://github.com/saltstack/salt/issues/7203
23:21 alekibango joined #salt
23:22 alekibango joined #salt
23:22 kula that was it
23:22 cachedout Cool. :]
23:24 faldridge joined #salt
23:28 kula and for the next sap that runs into this and finds my gist, i was nice and put a comment about which issue fixed the problem....
23:34 Code_Bleu left #salt
23:35 xinkeT It appears that the gitfs backend doesn't like relative repo paths, e.g (git+ssh://git@git.myhost.com:someproject/salt.git).  I haven't seen a bug reported against this? has anyone else has run into it?
23:35 kula whatever genius invented the _modules method of shoving out modules gets a tasty beverage of their choice the next time i see them.
23:37 UtahDave joined #salt
23:46 L0j1k joined #salt
23:47 Heartmender joined #salt
23:50 tyler-baker joined #salt
23:53 scristian_ joined #salt
23:56 lineman60 joined #salt
23:56 cachedout left #salt
23:57 jslatts joined #salt
23:59 cshuman joined #salt

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