Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-02-10

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

All times shown according to UTC.

Time Nick Message
00:03 digitalhero joined #salt
00:05 onlyanegg joined #salt
00:06 shaggy_surfer joined #salt
00:10 baweaver joined #salt
00:11 liskl joined #salt
00:16 mephx joined #salt
00:21 mohae joined #salt
00:21 zmalone joined #salt
00:24 NightMonkey joined #salt
00:25 aqua^c joined #salt
00:30 ssplatt joined #salt
00:32 krymzon joined #salt
00:38 hasues joined #salt
00:39 hasues left #salt
00:44 abednarik joined #salt
00:45 baweaver joined #salt
00:48 hoonetorg joined #salt
00:49 brianfeister joined #salt
00:50 mavhq joined #salt
00:50 shoemonkey joined #salt
01:04 tracphil joined #salt
01:06 mavhq joined #salt
01:06 bhosmer_ joined #salt
01:15 mavhq joined #salt
01:16 shoemonkey joined #salt
01:16 abednarik joined #salt
01:17 jgm3 joined #salt
01:20 jack__ joined #salt
01:21 erjohnso joined #salt
01:25 baweaver joined #salt
01:32 mavhq joined #salt
01:36 Jimlad joined #salt
01:38 shoemonkey joined #salt
01:42 abednarik joined #salt
01:51 snc joined #salt
01:54 mavhq joined #salt
01:55 shaggy_surfer joined #salt
02:00 antpa joined #salt
02:01 antpa joined #salt
02:02 antpa joined #salt
02:05 totzky joined #salt
02:05 malinoff joined #salt
02:11 tweakism argh, this is driving me nuts
02:12 tweakism all I want is, to be able to map pillar data directly to minionids.
02:12 tracphil joined #salt
02:13 nyx_ joined #salt
02:16 jack__ joined #salt
02:16 edrocks joined #salt
02:18 tweakism I guess ext_pillar is the easiest way; I was sure I was missing something.
02:18 mavhq joined #salt
02:18 jmaguire joined #salt
02:22 catpigger joined #salt
02:24 subsignal joined #salt
02:28 jack__ joined #salt
02:29 edrocks joined #salt
02:32 donmichelangelo joined #salt
02:38 brianfeister joined #salt
02:46 digitalhero joined #salt
02:48 shoemonkey joined #salt
02:50 mavhq joined #salt
02:52 antpa joined #salt
02:53 hax404 joined #salt
02:54 jmaguire $toke
02:54 jmaguire err
02:54 jmaguire left #salt
02:55 PeterO joined #salt
02:59 ntropy now now :)
03:01 brianfeister joined #salt
03:01 irctc397 joined #salt
03:01 kusams joined #salt
03:02 digitalhero joined #salt
03:10 Pie_Mage joined #salt
03:20 neilf__ joined #salt
03:21 antpa joined #salt
03:22 mohae joined #salt
03:23 cilkay joined #salt
03:25 evle joined #salt
03:26 rem5 joined #salt
03:33 mavhq joined #salt
03:34 krymzon joined #salt
03:43 snicers-work joined #salt
03:44 justanot1eruser joined #salt
03:45 mavhq joined #salt
03:47 rem5 joined #salt
03:50 totzky joined #salt
03:50 antpa joined #salt
03:52 voidspacexyz joined #salt
03:56 ramteid joined #salt
03:57 antpa joined #salt
03:58 joe_n joined #salt
03:59 digitalhero joined #salt
04:00 mavhq joined #salt
04:02 antpa joined #salt
04:05 hacfi joined #salt
04:06 hightekvagabond joined #salt
04:10 treaki joined #salt
04:10 morissette joined #salt
04:19 FreeSpencer joined #salt
04:19 FreeSpencer joined #salt
04:23 antpa joined #salt
04:33 mohae joined #salt
04:35 mavhq joined #salt
04:35 krymzon joined #salt
04:38 mavhq joined #salt
04:47 terratoma joined #salt
04:54 ffredrikk joined #salt
04:58 Bryson joined #salt
04:58 justanotheruser joined #salt
05:03 ajw0100 joined #salt
05:04 nidr0x joined #salt
05:27 brycebudd joined #salt
05:28 ffredrikk joined #salt
05:31 ruxu joined #salt
05:32 brycebudd left #salt
05:32 nidr0x joined #salt
05:33 antpa joined #salt
05:45 cswang joined #salt
05:52 totzky joined #salt
05:52 impi joined #salt
05:55 mapu joined #salt
05:59 jhauser joined #salt
06:03 nidr0x joined #salt
06:03 kshlm joined #salt
06:09 rdas joined #salt
06:10 anmol joined #salt
06:11 mavhq joined #salt
06:15 cangiani joined #salt
06:25 mechleg left #salt
06:25 jxm_ joined #salt
06:32 bhosmer_ joined #salt
06:37 krymzon joined #salt
06:42 ffredrikk Can I somehow place a package (for the Windows package manager) within /srv/salt/ and have it automatically picked up by a pkg.refresh_db?
06:44 snicers-work joined #salt
06:44 ffredrikk I currently have it here: /srv/salt/win/repo-ng/salt-winrepo-ng/ - but I would rather put it in e.g. /srv/salt/_pkg
06:44 ffredrikk But then it isn't found when doing a pkg.refresh_db
06:46 symphorien left #salt
06:49 felskrone joined #salt
06:53 antpa joined #salt
06:59 antpa joined #salt
07:00 JDiPierro joined #salt
07:10 ruxu joined #salt
07:12 danielcb joined #salt
07:14 ffredrikk hm, I think I hit a bug with states module file.copy: http://pastebin.com/95pt6AQ9 - or do you think I did something wrong?
07:14 antpa joined #salt
07:16 ffredrikk ah, yes indeed... https://github.com/saltstack/salt/issues/26450
07:16 saltstackbot [#26450]title: file.copy: source file is not present. | According to http://docs.saltstack.com/en/latest/ref/states/all/salt.states.file.html#salt.states.file.copy, ...
07:19 antpa joined #salt
07:24 Bryson joined #salt
07:26 bhosmer_ joined #salt
07:26 ffredrikk joined #salt
07:28 ruxu joined #salt
07:29 ffredrikk joined #salt
07:32 cyborg-one joined #salt
07:34 cyborg-one joined #salt
07:35 jhauser joined #salt
07:36 penguinpowernz joined #salt
07:38 krymzon joined #salt
07:38 anthpa joined #salt
07:57 antpa joined #salt
07:58 catarrhine joined #salt
07:59 catarrhine Hi folks is the TCP tornado transport stable? I just looked in the salt master config example and did not see anything related to tornado, transport, or dare I say RAET
08:00 antpa joined #salt
08:01 joe_n joined #salt
08:04 krymzon joined #salt
08:07 opdude @ffredrikk I think the location of the win_pkg_repo is a configurable option in the minion file, but that is just from a very distant memory :)
08:08 ntropy catarrhine: 2015.8 release came with a "not for production" warning for tornado, as few issues were reported shortly after it was introduced
08:10 dariusjs joined #salt
08:13 ntropy catarrhine: https://docs.saltstack.com/en/develop/topics/releases/2015.8.0.html#tornado-tcp-transport - the 3 referenced issues are resolved & closed though
08:14 kawa2014 joined #salt
08:16 joe_n joined #salt
08:18 rotbeard joined #salt
08:20 impi joined #salt
08:20 catarrhine @ntropy thanks, maybe I'll try it out, 2015.8.5 is out already after all
08:21 bhosmer_ joined #salt
08:25 mavhq joined #salt
08:27 XenophonF joined #salt
08:32 keimlink joined #salt
08:32 slav0nic joined #salt
08:40 garphy joined #salt
08:43 krymzon joined #salt
08:50 krymzon_ joined #salt
08:54 netcho_ joined #salt
09:02 ronnix joined #salt
09:02 GreatSnoopy joined #salt
09:04 keimlink_ joined #salt
09:05 geomacy joined #salt
09:06 totzky joined #salt
09:09 pwalsh joined #salt
09:09 joe_n joined #salt
09:11 cangiani joined #salt
09:12 keimlink joined #salt
09:25 impi joined #salt
09:30 sfxandy joined #salt
09:30 akhter joined #salt
09:32 zerthimon joined #salt
09:39 ruxu joined #salt
09:39 LostSoul Hi
09:40 LostSoul Is is possible to use @reboot option with cron module?
09:40 LostSoul If so, how?
09:44 alxf LostSoul: check the set_special method of the cron module
09:52 antpa joined #salt
09:56 s_kunk joined #salt
09:57 joe_n joined #salt
09:59 snicers-work joined #salt
10:14 ffredrikk @opdude thanks, I'll check that out. I'd like to specify where on the master the package resides, preferably.
10:16 ffredrikk I can't find anything on the Windows command cmdkey: https://technet.microsoft.com/en-us/library/cc754243.aspx  - which looks like a fairly good module to try to write myself. Just want to make sure it does not already exist.
10:18 ronnix joined #salt
10:24 N-Mi joined #salt
10:24 N-Mi joined #salt
10:27 CeBe joined #salt
10:28 mavhq joined #salt
10:35 ronnix joined #salt
10:36 antpa joined #salt
10:36 icbt_nl joined #salt
10:39 ecdhe joined #salt
10:41 antpa joined #salt
10:44 sjorge joined #salt
10:44 sjorge joined #salt
10:45 mavhq joined #salt
10:51 amcorreia joined #salt
10:51 kshlm joined #salt
10:51 malinoff joined #salt
10:54 giantlock joined #salt
10:55 Muchoz joined #salt
10:56 rainbowtux joined #salt
10:57 rainbowtux Hi all, I am rather new to salt and have a question: is there any way to use salt-ssh and sudo with a password? I know NOPASSWD is advised, but unfortunately it is not that easy to change in my environment...
10:59 Modulus_ joined #salt
11:01 mavhq joined #salt
11:05 netcho_ joined #salt
11:11 ruxu joined #salt
11:17 RandyT joined #salt
11:17 krymzon joined #salt
11:21 mavhq joined #salt
11:38 pwalsh joined #salt
11:45 antpa joined #salt
11:46 evle joined #salt
11:48 fredvd joined #salt
11:48 opdude joined #salt
11:50 ggoZ joined #salt
11:50 giantlock joined #salt
11:55 permalac joined #salt
11:59 catpig joined #salt
12:00 nekrodesk joined #salt
12:04 amcorreia joined #salt
12:06 ingslovak joined #salt
12:07 antpa joined #salt
12:09 aqua^c joined #salt
12:21 petesmall joined #salt
12:21 mavhq joined #salt
12:21 malinoff joined #salt
12:21 ruxu joined #salt
12:24 teryx510 joined #salt
12:26 bluenemo joined #salt
12:28 bluenemo hi guys. My salt master logs "[ERROR   ][24335] Exception 'len([]) != len(['Permission denied, please try again.', '', '', '', '', 'and the repository exists.'])' caught while fetching gitfs remote git@git.example.org:salt/bareos.git". my /etc/salt/master: http://paste.debian.net/379511/ I can clone the formula as root user wo problems.
12:31 antpa joined #salt
12:33 ggoZ joined #salt
12:41 nahkiss_ How should I proceed if I want to create user foobar on all my minions and IF there's user account "www", then lock it. Can't really use salt.states.user.present because I don't want the "www" user created, I just want it locked if it's present
12:50 chesty you want foobar locked if www is present?
12:51 theanalyst joined #salt
12:54 AndreasLutro nahkiss_: just wrap it in an {% if salt['user.info']('www') %}
12:54 AndreasLutro I think user.info will return None if the user doesn't exist (you should double check this)
12:54 evle joined #salt
12:55 joe_n joined #salt
12:58 Muchoz joined #salt
13:01 snicers-work joined #salt
13:02 av_ joined #salt
13:02 DammitJim joined #salt
13:03 nahkiss_ ah, that sounds something that could work AndreasLutro
13:04 rem5 joined #salt
13:06 mavhq joined #salt
13:06 petesmall hi all, is it possible to use the output from a command in one state as the input to another?
13:09 anmol joined #salt
13:17 yuhlw joined #salt
13:18 teryx510 joined #salt
13:21 giantlock joined #salt
13:26 jeddi joined #salt
13:30 JDiPierro joined #salt
13:30 akhter joined #salt
13:34 rainbowtux Hi all, I am rather new to salt and have a question: is there any way to use salt-ssh and sudo with a password? I know NOPASSWD is advised (required?), but unfortunately it is not that easy to change in my environment...
13:35 kshlm joined #salt
13:36 AndreasLutro rainbowtux: from the docs: sudo works only if NOPASSWD is set for user in /etc/sudoers
13:36 rainbowtux AndreasLutro: I was hoping for a workaround because i have an army of hosts with read-only systems, so pretty hard to change it :)
13:37 AndreasLutro out of curiosity, what do you hope to achieve with salt-ssh on a "read-only" system?
13:38 geomacy joined #salt
13:39 tracphil joined #salt
13:39 shoemonkey joined #salt
13:41 wangofett joined #salt
13:41 dyasny joined #salt
13:42 subsignal joined #salt
13:43 damona joined #salt
13:45 subsignal joined #salt
13:45 lompik joined #salt
13:46 bhosmer joined #salt
13:46 numkem joined #salt
13:47 rsys joined #salt
13:47 subsignal joined #salt
13:49 rainbowtux AndreasLutro: running checks, changing values temperarily to test stuff... :)
13:50 rainbowtux AndreasLutro: I am starting to use salt for servers etc and it would be nice to be able to use the same tool as remote command execution platform as well.. already plenty of tools to master :)
13:51 subsignal joined #salt
13:51 malinoff joined #salt
13:51 mavhq joined #salt
13:54 AirOnSkin I'd like to write an if-then-else condition with jinja2 where the "then"-part does nothing. Something like: if role in my list do nothing, else apply the following state. Anyone know how to do that?
13:54 tedski let's say i have a pillar {'foo': {'bar': 'baz'}} ... why when i do pillar.get foo:bar does it return nothing, but if i do pillar.item foo, it returns the proper dict?
13:55 kaushal_ joined #salt
13:56 gothix joined #salt
13:57 tedski AirOnSkin: why not only test the negative like https://bpaste.net/show/03fa690a0364
13:57 gothix left #salt
13:58 M-MadsRC joined #salt
13:59 cangiani joined #salt
13:59 AndreasLutro or just replace "if" with "if not"
13:59 kusams joined #salt
14:00 AndreasLutro tedski: can you share the output of `salt my-minion pillar.get foo` ?
14:00 tedski AndreasLutro: yeah
14:00 _gothix_ joined #salt
14:00 AirOnSkin tedski, AndreasLutro: Ah, I didn't know you could do it like that. Sadly googling jinja2 examples is rather hard ... I mostly never find what I'm looking for. Thanks for the tips :)
14:01 tedski AirOnSkin: the docs are pretty good for examples
14:01 tedski AirOnSkin: http://jinja.pocoo.org/docs/dev/templates/
14:01 AndreasLutro AirOnSkin: just try python syntax if you ever wonder how something could be done
14:01 tedski and that, yeah
14:02 gothix joined #salt
14:03 edrocks joined #salt
14:03 AirOnSkin I see. I'll remember that :)
14:03 cangiani joined #salt
14:04 tedski AndreasLutro: https://bpaste.net/show/29b426f6351f
14:06 AndreasLutro tedski: does it make a difference if you run salt 'cas1' saltutil.refresh_pillar
14:06 dyasny joined #salt
14:06 JDiPierro joined #salt
14:06 JDiPierr_ joined #salt
14:06 tedski heh, yes... but not in a good way
14:07 tedski minion fell over it seems... let me dig into that
14:08 tedski isn't refresh_pillar a part of the highstate?
14:09 AndreasLutro not necessarily, in my experience
14:09 tedski okay
14:09 tedski poor assumption on my part
14:09 AndreasLutro I think it's cached for some time so if your pillar data changes often, you may have to refresh manually
14:10 tedski i had just highstated, so i was assuming refresh_pillar was taken care of
14:10 tedski and items gets pillar from the master... which explains the delta between .get() and .items() output
14:10 tedski moving along :)
14:10 tedski thanks AndreasLutro
14:10 penguinpowernz joined #salt
14:11 shoemonkey joined #salt
14:12 rsys joined #salt
14:13 KennethWilke joined #salt
14:15 M-liberdiko joined #salt
14:16 ffredrikk joined #salt
14:19 Muchoz Does Salt support Python 3?
14:22 Muchoz Because I think the problem I've been having with pyenv is because of Python 3
14:22 permalac joined #salt
14:22 Muchoz Gonna switch to 2.7 and see whether that solves it
14:24 ssplatt joined #salt
14:25 racooper joined #salt
14:28 toastedpenguin joined #salt
14:28 rm_jorge joined #salt
14:29 tkharju joined #salt
14:32 ronnix joined #salt
14:33 cyborg-one joined #salt
14:34 tedski Muchoz: https://github.com/saltstack/salt/issues/11995
14:34 saltstackbot [#11995]title: Python 3 compatibilty | Salt is currently not compatible with python 3. Ubuntu trusty and above default to using python 3, so it would be really nice to be able to run salt in the default python version.
14:37 zsoftich4 joined #salt
14:40 bhosmer joined #salt
14:41 philiaagape joined #salt
14:42 shoemonkey joined #salt
14:43 scoates joined #salt
14:44 Hazelesque Hi there
14:45 Hazelesque I noticed an inconsistency between the expected behaviour (indicated by a comment in the code) when using salt-run survey.diff, and the actual behaviour
14:45 Hazelesque namely that the kwargs are not passed on to LocalClient
14:45 Hazelesque I've created a pull request with a patch > https://github.com/saltstack/salt/pull/31084
14:45 saltstackbot [#31084]title: Survey runner: actually pass kwargs to LocalClient | Dear Salt developers,...
14:46 Hazelesque I hope that my patch is acceptable
14:46 anmol joined #salt
14:47 shiriru joined #salt
14:48 Tyrm joined #salt
14:51 rainbowtux salt-ssh '*' test.ping returns " ImportError: No module named pstats" although i do find the pstats module on my system. Any ideas?
14:52 hasues joined #salt
14:52 hasues left #salt
14:53 netcho_ joined #salt
14:53 Hazelesque And I found a problem with my patch already, oops.
14:55 hexhaxtron_ joined #salt
14:59 Brew joined #salt
15:01 sjorge joined #salt
15:01 sjorge joined #salt
15:01 Hazelesque Right, I've added a comment with a change that looks like it should fix the oversight I made.
15:02 hexhaxtron_ left #salt
15:02 danemacmillan joined #salt
15:05 impi joined #salt
15:06 rofl____ anyone seen salt appending \n to pillar data via pillar.get?
15:06 rofl____ i dont want it, breaks my nginx syntax:<
15:09 zmalone joined #salt
15:10 netcho joined #salt
15:10 rem5 joined #salt
15:10 teebes joined #salt
15:10 teebes left #salt
15:11 nethershaw joined #salt
15:12 jhauser joined #salt
15:13 andrew_v joined #salt
15:13 mpanetta joined #salt
15:15 _JZ_ joined #salt
15:15 subsigna_ joined #salt
15:16 BogdanR joined #salt
15:17 BogdanR Hello. I would like to detect when updates are available in Debian with salt-stack
15:17 BogdanR Any tips how I can do that?
15:17 brianfeister joined #salt
15:18 cangiani joined #salt
15:18 leev joined #salt
15:19 beardedeagle joined #salt
15:19 BogdanR Something like "apt-get --just-print upgrade" but maybe distro agnostic and with a pretty output from salt
15:22 kusams joined #salt
15:26 netcho_ joined #salt
15:33 AndreasLutro BogdanR: https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.aptpkg.html#salt.modules.aptpkg.list_upgrades
15:33 digitalhero joined #salt
15:34 gh34 joined #salt
15:35 bhosmer joined #salt
15:36 Muchoz joined #salt
15:38 _JZ__ joined #salt
15:41 Muchoz tedski, thank you.
15:41 insanegeek joined #salt
15:42 colegatron joined #salt
15:45 tedski Muchoz: no prob
15:47 mavhq joined #salt
15:47 petesmall joined #salt
15:48 slav0nic joined #salt
15:49 rem5 joined #salt
15:49 jnilsson joined #salt
15:50 WesleyTech joined #salt
15:51 ronnix joined #salt
15:52 snicers-work joined #salt
15:59 Taz_ joined #salt
16:02 jerryc joined #salt
16:02 genunix joined #salt
16:03 pppingme joined #salt
16:08 BogdanR AndreasLutro: Thanks. I suppose there isn't also something which shows from which version we are upgrading
16:08 cognos joined #salt
16:10 cognos Hi, first time around here. I have an issue with http.query while using it on an state. I'm getting a tornado exception: TypeError: Expected bytes, unicode, or None; got <type 'dict'> and it doesn't seem to be rendering my templates. Any advise on how to debug that?
16:11 abednarik joined #salt
16:11 pppingme joined #salt
16:11 nekrodesk joined #salt
16:12 numkem joined #salt
16:14 mew1033 joined #salt
16:15 mew1033 Anyone know of a way to disable logging/change logging level/squelch a specific log message while running a module?
16:15 abednarik_ joined #salt
16:16 Chuck_0326 joined #salt
16:20 cangiani joined #salt
16:20 Muchoz joined #salt
16:21 cognos Published the question on the google group. Thanks anyway!
16:22 genunix left #salt
16:22 PeterO_ joined #salt
16:23 RandyT wanted to share the following keystore challenge with the group and see if anyone has any input to make this any cleaner: https://github.com/saltstack/salt/issues/10237#issuecomment-182456317
16:23 saltstackbot [#10237]title: Feature Request - Manage Java keystore | We currently "manage" java keystores by simply pushing out a cacerts (or trustedcerts) we've created, but this is not really maintainable since we update once and sort of forget about it....
16:23 RandyT still learning here
16:26 tligda joined #salt
16:27 dendazen joined #salt
16:28 dendazen hey guys how can i troubleshoot connection from the minion client to the salt master
16:28 dendazen i get Minion did not return. [Not connected]
16:28 dendazen key on master is accepted
16:29 dendazen and salt-minion service is running
16:29 onlyanegg joined #salt
16:30 dendazen the log in /var/log/salt/minion is empty.
16:30 Taz_ RandyT: Looks neat! Where are you touching the . files so the cmds don't run overtime?
16:30 dendazen lsof shows me on the minion pid that TCP connection to the master is established
16:30 Taz_ everytime*
16:31 RandyT Taz_: touching in /var/cache/salt since I can count on that existing..
16:31 hightekvagabond joined #salt
16:31 Taz_ RandyT: yea I understand that, but where does your commands do that?
16:32 RandyT Taz_: and perhaps you meant to ask "Why", which is to prevent this from happening multiple times. It will fail if those same states are called again since the password has changed in the first state_id
16:32 RandyT Taz_: - creates: touches the file if it does not exist
16:33 Taz_ RandyT: oh wow TIL creates will actually create the file for you
16:33 Taz_ Ive always put ;touch whatever at the end of my cmds
16:33 RandyT Actually, I see it does not touch the file, and have apparently misread the doc...
16:34 Taz_ RandyT: oh
16:34 Taz_ Thats what I was asking :)
16:34 Taz_ generally I've used it for like tars, so you know what directories tars make after they are extracted
16:35 RandyT Taz_: I think that should be a feature request. :-) if the state runs successfully with a - creates: conditional, create the file.
16:35 Taz_ RandyT: sounds cool to me :)
16:36 RandyT Taz_: yes, debate if the touch file approach is correct, but it has the nice benefit that you can easily remove these files and rerun the state if needed.
16:36 Taz_ RandyT: ive also had use cases for adding/removing certs from truststores and keystores
16:36 Taz_ not sure if you need to do that too
16:36 RandyT I bit easier than storing grains and dealing with those
16:37 RandyT Taz_: I may need to deal with the truststores as well. I've avoided dealing with these certstores in java and am working my way through this
16:38 Taz_ ahh ok
16:38 Taz_ I ended up making a pki formula that handles certs and keystores since generally we want both
16:38 Taz_ but i also noticed there is a new x509 state
16:38 Taz_ so ill get around to using that one day
16:39 zmalone I'd be careful with the x509 modules, they were independently contributed, and I'm not sure how much faith I have in them
16:39 Taz_ yea hense my reluctance :P
16:40 Taz_ if it aint broke right?
16:40 Taz_ :P
16:40 rm_jorge joined #salt
16:40 zmalone it's a little broken right now, and you really need to inspect what it's doing and what the results are
16:40 ffredrikk joined #salt
16:41 giantlock joined #salt
16:41 Deevolution joined #salt
16:41 Taz_ i saw some guys trying it yesterday but left before i saw them finish or not
16:42 whytewolf Taz_: yeah turned out that ca_server required publish.publish working. and the ca_server: setting is a minion_id of the ca_server
16:42 whytewolf not ideal
16:42 Taz_ oh interesting
16:42 akhter joined #salt
16:42 Taz_ yea sounds confusing
16:43 cpowell joined #salt
16:43 whytewolf the example showed setting publish.publish working, however they used a sls file as a conf file.
16:43 whytewolf which doesn't work
16:44 Muchoz I found that something has been missed in the salt documentation, where do I edit that in the repo?
16:44 Taz_ whytewolf: thanks for the info :)
16:45 whytewolf Muchoz: it really depends. if it is a function part of the documentation it might be in the module. other wise it will be in the sphinx documentation
16:45 Muchoz It turns out the pyenv state requires git to be installed but it not mentioned in the docs.
16:45 whytewolf ahh that should be in the pyenv salt module.
16:46 dendazen what does error "Exception ValueError: 'I/O operation on closed file'" mean on salt minion?
16:46 Muchoz Ya, it's in states. Not modules
16:47 whytewolf Muchoz: does the state require it without the module? or is it the module that needs it?
16:47 Muchoz The code in the modules uses the clone command, the docs mention required packages but it doesn't contain git
16:49 whytewolf I'm asking cause of how it should be documented. if the state is using the clone and the exacution module isn't then the requirement should be tagged into the module with a note in the states pointing to the module
16:49 whytewolf wait, reverse that
16:49 whytewolf if the exacution module needs git, then the exacution module should have the requirement. and the state should have a note to check the requirments for the module
16:50 brianfeister joined #salt
16:50 whytewolf i think i need coffee
16:51 Muchoz Yes, in modules/pyenv.py it uses git clone. I'm pretty new to how Salt works in the background, but the state installs pyenv if it is not installed and the install stuff (git clone) is in the module.
16:52 whytewolf okay, then in the modules/pyenv.py it should have the git requirment. and states/pyenv.py should have a note with a link to the modules/pyenv.py documentation
16:53 cangiani joined #salt
16:53 cyborg-one joined #salt
16:54 whytewolf the documentation for both should be at the top of both
16:54 Muchoz How do I add that requirement?
16:55 Muchoz I'm looking at the module documentation for making one
16:55 mavhq joined #salt
16:56 ronnix joined #salt
16:56 impi joined #salt
16:56 whytewolf the mysql module has an example of how that should look. but there is this page for submitting documentation https://docs.saltstack.com/en/latest/topics/development/conventions/documentation.html#salt-docs
16:57 abednarik_ joined #salt
16:57 whytewolf Muchoz: and about the PR when you get there https://docs.saltstack.com/en/latest/topics/development/contributing.html
16:58 Muchoz I know how to edit the documentation. But do I just mention it in the documentation or is there something in code I can do?
16:59 whytewolf Muchoz: here is the note from mysql.py as an example. https://github.com/saltstack/salt/blob/develop/salt/modules/mysql.py#L7-L10
16:59 whytewolf the inline documentation is pulled into the online documentation
17:00 Muchoz I had something similar written now. I thought there was something that would make it auto install git then.
17:00 whytewolf no
17:00 mapu joined #salt
17:02 whytewolf i mean it would be possable. just a little __salt__.pkg.install('git') in the code. but i wouldn't recomend doing that.
17:03 mavhq joined #salt
17:04 PeterO_ joined #salt
17:05 nidr0x joined #salt
17:06 jrgochan Morning all! What's the best way to debug possible jinja related issues?
17:06 Eugene Alcohol.
17:06 radhac +1
17:07 Bryson joined #salt
17:07 whytewolf +5
17:07 Eugene Trial+error is really about it. Fix all the parse errors, then fix all the variable-are-wrong ones
17:07 zmalone https://github.com/mitsuhiko/jinja2/issues/508 is still my favorite jinja issue report
17:08 saltstackbot [#508]title: Nonstandard operator precedence | In most programming languages, operators are grouped into families having equal precedence.  In Jinja2's parser, each operator has its own precedence level.  This affects the associativity of some expressions....
17:09 rmnuvg joined #salt
17:09 whytewolf jrgochan: liberal use of show_highstate or this module https://github.com/whytewolf/salt-debug help
17:10 rijalati joined #salt
17:10 jrgochan mmmmm alcohol
17:10 jrgochan I'll give it a shot in a few. thanks for the recommendations folks!
17:11 rmnuvg joined #salt
17:16 jfindlay silly jinja
17:17 evilrob so we're having "discussions" (not quite arguments yet) in my team about using grains vs using pillar variables.  thoughts?
17:18 Fiber^ joined #salt
17:18 zmalone evilrob: I've had weird luck with grains on 2015.8.3, and wish we hadn't moved some information out of pillars and into grains
17:18 gdot joined #salt
17:19 whytewolf evilrob: depends. if you need to change it and/or it needs to be secure. pillar. if you are fine about it being minion. and being insecure, and don't mind that anyone with root can change it
17:19 zmalone I don't know if those problems still exist in 2015.8.5
17:19 Muchoz Eugene, well isn't he a bit right? Shouldn't it be compliant with Python's way?
17:19 zmalone (in particular, the different salt processes on a minion seeming to end up with different grain values after having their grain values updated from a master)
17:20 Eugene Yup, sure is.
17:20 numkem joined #salt
17:20 Muchoz That is why I always use brackets
17:23 bhosmer joined #salt
17:25 evilrob zmalone: when you made the choice to move things out of pillars what was the goal?  We're contemplating that now.
17:25 _JZ_ joined #salt
17:26 Taz_ evilrob: our main usecase was autoscaling
17:26 Taz_ evilrob: we didnt want to define minions on our masters pillar, so we needed to use grains as they went up and down for some systems
17:26 writtenoff joined #salt
17:27 evilrob that makes sense
17:27 Taz_ other than that i think pillar is nice to have everything in one place
17:27 zmalone getting ephemeral information (versions numbers of internal code) out of git / our salt repo
17:27 zmalone in an effort to streamline a release process
17:27 zmalone I think we'll likely go back to pillars though
17:30 rihannon joined #salt
17:33 aqua^c joined #salt
17:33 XenophonF gtmanfred, whytewolf: got my openstack dev environment back up
17:34 RandyT moved sensitive info to pillar on s3 recent as it allows me to separate that stuff from other folks doing work on the state development (when I have the luxury of hiring those people)
17:34 RandyT Looking forward to getting kms support in S3...
17:34 XenophonF ubuntu 14.04 + openzfs
17:34 XenophonF ubuntu 14.04 + openzfs + luks, actually
17:34 Taz_ XenophonF: sounds cool man!
17:34 whytewolf nice
17:34 Taz_ openstack is still a little crash for me
17:34 Taz_ crazy
17:34 zmalone XenophonF: I saw some complaints about file.managed on zfs on linux
17:35 djgerm is there a way to get documentation for for a module on a master?
17:35 XenophonF me too Taz_
17:35 zmalone https://github.com/saltstack/salt/issues/31047
17:35 saltstackbot [#31047]title: file.copy reports failed even when file was copied on ZFS | From the sls file:...
17:35 djgerm i want the docs for the glusterfs module
17:35 gtmanfred XenophonF: nice
17:35 XenophonF zmalone: i had no idea
17:36 XenophonF i'm gonna restart work on my openstack formula in my Copious Free Time
17:36 XenophonF am aiming for something that'd work across all supported operating systems (Ubuntu/RHEL/SUSE)
17:37 XenophonF and at various scales (from devstack-style on up)
17:38 djgerm for example, I running "salt '*' glusterfs.list_peers" and get "glusterfs.list_peers" is not available… and I don't get it.
17:38 Charly_ joined #salt
17:39 gtmanfred if it isn't loading, then you are missing what is required in __virtual__
17:39 gtmanfred which should just be the gluster binary
17:39 whytewolf djgerm: you can get documentation for modules that are loadable. however if it is saying it isn't available then it isn't going to give you docs
17:40 gtmanfred https://github.com/saltstack/salt/blob/develop/salt/modules/glusterfs.py#L29
17:40 djgerm huh… interesting. so I get result when I run "cmd.run 'which gluster'"
17:40 djgerm it's there… on my minions.
17:40 numkem joined #salt
17:41 gtmanfred python -c "from salt.util import which; print(which('gluster'))"
17:42 djgerm "ImportError: No module named util"
17:42 whytewolf should be noted that should be the python that actually runs salt incase multiple versions of salt are installed
17:42 whytewolf er multiple copies of python i mean
17:43 whytewolf and it is utils
17:43 Taz_ XenophonF: thats a lofty goal!
17:44 djgerm huh…
17:44 djgerm weird
17:44 djgerm NameError: name 'gluster' is not defined ….
17:44 cangiani joined #salt
17:46 Charly_ Hello. From what I read in the github issues the 2015.8 version of salt does not properly handle HTTP and HTTPS proxies in states like file.managed and archive.extracted. I also saw that a fix was merged to the codebase recently. I was wondering when the version with this fix would be released? I cannot seem to find a page announcing the next release date.
17:46 djgerm meh when I run from command line instead of cmd.run
17:46 djgerm # python -c "from salt.utils import which; print(which('gluster'))"
17:46 djgerm /usr/sbin/gluster
17:46 djgerm so, that's good.
17:46 ffredrikk joined #salt
17:47 whytewolf Charly_: 2015.8.5 was released like last week. what is the link to the bug you are refering to
17:47 murrdoc joined #salt
17:47 gtmanfred Charly_: do you have a link where they showed the fix was merged?
17:48 Charly_ give me a second to find them again. It was a github issue and a pull request for the fix
17:49 dariusjs joined #salt
17:52 Charly_ the issue: https://github.com/saltstack/salt/issues/23617
17:52 saltstackbot [#23617]title: file.managed with proxy broken in 2015.5 | Hey all, I have a state that looks something like this:...
17:52 Charly_ the fix: https://github.com/saltstack/salt/pull/29322
17:52 saltstackbot [#29322]title: add http proxy support for tornado | the tornado backend lacks support for http proxies from environment variables....
17:52 fredvd joined #salt
17:52 djgerm so… since my "python which" is coming back with the gluster binary… what else can I look at for why this module isn't loading?
17:53 linjan joined #salt
17:54 ssplatt if i put a date in pillar, like ‘ date: 2015-09-01 13:22:04 ‘  it looks like it is being converted somehow.  is there a way to change how that is converted…or tell it to not convert it?
17:54 snarfy joined #salt
17:54 ssplatt for example, ¸20151209T17:32:05.000000   is one output i got
17:54 jfindlay djgerm: does `salt minion sys.doc glusterfs` return the docs for that module?
17:55 jfindlay ssplatt: quote the date string: `date: '2015-09-01 13:22:04'`
17:55 ssplatt ok
17:56 jfindlay pyyaml opportunistically interpolates some info, like datetimes
17:56 djgerm no it doesn't  jfindlay….
17:56 ssplatt that could be useful tho, is there some docs on where and when it does that?
17:57 ssplatt i’m assuming its the salt renderer that does it
17:57 jfindlay djgerm: so the problem is then `salt.utils.which` is not finding `/usr/sbin/gluster`
17:58 djgerm but when I run this on my minions, it does
17:58 djgerm ~# python -c "from salt.utils import which; print(which('gluster'))"
17:58 djgerm /usr/sbin/gluster
17:58 abednarik_ joined #salt
17:58 djgerm sorry…. copt pasting
17:58 Charly_ Do not see the fix in the release notes it seems
17:58 jfindlay ssplatt: I'm not sure, you will probably find that in the PyYAML docs
17:59 jfindlay djgerm: did you run that python command on the master or on the minion?
17:59 djgerm minion
17:59 numkem joined #salt
17:59 Crazy67 joined #salt
17:59 djgerm the master does not have gluster installed
17:59 jfindlay and salt-call sys.doc glusterfs returns nothing?
18:00 ssplatt oh ok i guess pyyaml does return a real datetime.datetime
18:00 djgerm local: ———
18:00 Charly_ related: https://github.com/saltstack/salt/issues/8177
18:00 saltstackbot [#8177]title: Support http_proxy/https_proxy for sources | For example, a file.managed with source pointing to a file on a HTTP location will not work if you are behind a proxy....
18:02 shaggy_surfer joined #salt
18:02 gtmanfred djgerm: can you run salt-call -l debug glusterfs.list_peers, and see if something errors out?
18:02 djgerm ok one mo
18:03 Crazy67 joined #salt
18:03 antpa joined #salt
18:03 djgerm no errors. just "Function glusterfs.list_peers is not available" along with the debug output
18:03 djgerm i can pastebin that if you like
18:03 gtmanfred no stack traces?
18:03 djgerm nada
18:03 gtmanfred you are running it from the minion yeah?
18:03 djgerm yes
18:04 antpa joined #salt
18:04 gtmanfred hrm... unfortunately i have no idea
18:04 Bryson joined #salt
18:04 djgerm yeah it's a weird one....
18:04 mpanetta joined #salt
18:04 djgerm I had it working two days ago. I blew away the VMs and started from scratch with high confidence that my states would fix everything on new minions.
18:04 gtmanfred python -c 'from salt.modules import glusterfs' is working yeah?
18:05 djgerm no
18:05 djgerm ImportError: cannot import name glusterfs
18:05 gtmanfred lemme check that it is the same on mine
18:05 gtmanfred that is hte problem
18:05 gtmanfred [root@salt ~]# python -c 'from salt.modules import glusterfs'
18:05 gtmanfred [root@salt ~]#
18:06 djgerm bad… python?
18:06 snicers-work I am getting `Your local changes to the following files would be overwritten by merge:` with git.latest
18:06 gtmanfred might not have everything in the right path?
18:06 snicers-work I am trying to force the latest, any ideas?
18:06 djgerm hmm I intalled from the ubuntu apt source
18:06 XenophonF of course as soon as I say "oh I have something cool working" the hard drive goes kaput
18:07 jfindlay gtmanfred: the glusterfs module doesn't import glusterfs, though?
18:07 gtmanfred if you can't import it from python, then salt can't import it either
18:07 djgerm XenophonF: I am in that same boat! As soon as I had this all working, I was gonna show it off and blamo everything is broken cause I can't install this module
18:08 gtmanfred jfindlay: [root@salt ~]# python -c 'from salt.modules import glusterfs; print(glusterfs.list_peers)'
18:08 gtmanfred <function list_peers at 0x385b6e0>
18:08 gtmanfred it is just a python module, you should be able to import it
18:08 gtmanfred you probably can't run it, cause it doesn't have __salt__ __opt__ or any other things in it
18:08 gtmanfred but you should be able to import it
18:09 shaggy_surfer joined #salt
18:09 hasues joined #salt
18:09 jfindlay gtmanfred: the glusterfs module does everything through a cmd.run to glusterfs, and doesn't import the glusterfs python module
18:09 gtmanfred [root@salt ~]# python -c 'from salt.modules import glusterfs; print(glusterfs.__virtual__())'
18:09 gtmanfred (False, 'glusterfs server is not installed')
18:09 jfindlay at least on 2015.8
18:09 gtmanfred yes
18:10 gtmanfred but he doesn't have a glusterfs salt module that python can import
18:10 murrdoc jfindlay:  when is 2015.5 EOL'ed
18:10 hasues left #salt
18:10 jfindlay gtmanfred: oh, sorry, I misread your comment
18:10 gtmanfred jfindlay: it looks like his install is missing the salt/modules/glusterfs.py file
18:10 gtmanfred djgerm: you should be able to run this
18:10 gtmanfred [root@salt ~]# python -c 'from salt.modules import glusterfs; print(glusterfs.__virtual__())'
18:10 gtmanfred (False, 'glusterfs server is not installed')
18:10 jfindlay gtmanfred: good point
18:10 gtmanfred and get True back
18:11 gtmanfred but sounds like the actualy python file is missing for some reason, maybe try apt-get install --reinstall salt-common ? or whatever holds the module file
18:11 gtmanfred dpkg -s /usr/lib/python2.7/dist-packages/salt/modules/glusterfs.py
18:11 gtmanfred bah
18:11 jfindlay murrdoc: https://saltstack.com/product-support-lifecycle/
18:11 gtmanfred dpkg -S /usr/lib/python2.7/dist-packages/salt/modules/glusterfs.py
18:11 gtmanfred i think it is -S for what owns a file in dpkg
18:12 murrdoc oh lord
18:14 mew1033 joined #salt
18:16 jim__ joined #salt
18:16 jim__ anyone using consul as an ext_pillar with some examples I could see?
18:17 debian112 joined #salt
18:17 bhosmer joined #salt
18:18 linjan joined #salt
18:20 murrdoc i PoC'ed using consul_template to write out a pillar
18:21 murrdoc so it worked as a regular pillar
18:21 jhauser joined #salt
18:21 jim__ @murrdoc what did you think after doing that?
18:22 jim__ something to pursue or nah?
18:22 murrdoc for sure
18:22 murrdoc consul_template made it trivial
18:22 djgerm murrdoc: that's awesome! any thing you can share?
18:23 murrdoc i would if i could
18:23 murrdoc but i nuked the vm
18:23 murrdoc (sorry)
18:23 NV joined #salt
18:23 murrdoc https://github.com/hashicorp/consul-template
18:24 jim__ @murrdoc consul-template updates very quickly, correct?
18:24 murrdoc its configurable
18:24 murrdoc basically u dont have to write a lick of code
18:24 murrdoc to talk to consul
18:24 murrdoc just a template
18:24 snarfy joined #salt
18:25 nebuchadnezzar joined #salt
18:25 hightekvagabond joined #salt
18:26 jim__ https://www.opencredo.com/2015/09/14/saltstack-using-consul-as-an-external-pillar-source/
18:26 jim__ that seems to indicate that consul as an external pillar works pretty well
18:27 djgerm hey jfindlay and gtmanfred The fact that the glusterfs module isnt on the minion probably has something to do with it huh
18:28 gtmanfred yeah
18:29 djgerm so…. huh i just did a reinstall of salt-common, no love
18:29 baweaver joined #salt
18:29 jfindlay djgerm: what version of salt?
18:30 jfindlay how did you install?
18:30 gtmanfred salt-call --versions-report | curl -F 'f:1=<-' ix.io
18:30 djgerm salt-minion 0.17.5
18:30 jfindlay that module has been around for a while
18:30 gtmanfred that is the problem
18:30 gtmanfred it was added in 2014 iirc
18:30 djgerm salt-minion 0.17.5?
18:30 gtmanfred when i wrot eit
18:31 djgerm it worked like a charm the other day!
18:31 gtmanfred what version of ubuntu do you have installed?
18:31 gtmanfred and what repo are you using?
18:31 akhter joined #salt
18:31 gtmanfred http://repo.saltstack.com/#ubuntu
18:33 djgerm i fixed it with "apt-add-repository ppa:saltstack/salt && apt-get update && apt-get upgrade salt-minion salt-common"
18:33 djgerm But I was using the deb http://repo.saltstack.com/apt/ubuntu/14.04/amd64/latest trusty main
18:33 gtmanfred 0.17.5 came out in 2013 iirc
18:33 gtmanfred can't know what was going on there, but
18:33 gtmanfred yeah, that was the problem
18:34 djgerm huh. somebody should fix the salt-minion coming off of that repo :)
18:34 djgerm cause i JUST installed those this morning.
18:35 pwalsh joined #salt
18:35 zmalone joined #salt
18:35 liskl joined #salt
18:35 cyborg-one joined #salt
18:35 gtmanfred there is no 0.17.5 minion in the repo here http://repo.saltstack.com/apt/ubuntu/14.04/amd64/latest/
18:36 gtmanfred so, i don't know what you are getting
18:36 gtmanfred but i do not believe the ppa is managed anymore?
18:36 zmalone the default ubuntu salt (if you don't have the ppa or repo.saltstack.com) is 0.17
18:37 zmalone the ppa one is 2015.5.3, last I checked
18:37 djgerm maybe I…. messed up my repo?
18:37 djgerm *sigh*
18:37 ruxu joined #salt
18:37 djgerm thanks guys for dealing with me being dumb
18:39 gtmanfred lol
18:39 gtmanfred no problem :)
18:40 gtmanfred djgerm: i bet you didn't do an apt-get update after adding the repo
18:40 gtmanfred and got the default salt install from ubuntu
18:40 djgerm ~$ grep salt /etc/apt/sources.list
18:40 djgerm ~$
18:40 djgerm no. I am JUST dumb
18:40 gtmanfred heh
18:41 djgerm *face palm*
18:41 djgerm you wrote the glusterfs module?
18:41 djgerm It's great!
18:41 gtmanfred :)
18:41 gtmanfred i wrote the initial one
18:41 gtmanfred someone went back and made it better recently
18:41 djgerm I hope, before I die, that I can do ONE thing as well as that module turned out
18:41 gtmanfred heh awesome
18:43 breakingmatter joined #salt
18:45 jimklo joined #salt
18:45 numkem joined #salt
18:46 murrdoc thats pretty fucking morose djgerm
18:47 djgerm i meant in the field of IT!
18:47 murrdoc but uh
18:47 murrdoc to each his own
18:47 murrdoc i guess
18:47 jfindlay lol
18:52 nZac joined #salt
19:02 zsoftich4 joined #salt
19:03 JDiPierro joined #salt
19:06 kevinquinnyo hey all, i changed the hostname of a minioin and ran from master: salt-run cache.clear_grains tgt=my_minion_id_here
19:06 kevinquinnyo and the fqdn grain remains unchanged
19:06 kevinquinnyo any ideas?
19:06 dariusjs joined #salt
19:07 slav0nic joined #salt
19:07 radhac Don't you need to change the minion_id file as well?
19:08 kevinquinnyo to change the fqdn grain?
19:08 kevinquinnyo that's listed as the id grain i believe
19:08 Taz_ i dont think thats the right command to run
19:08 kevinquinnyo Taz_: do you know what the correct command is?
19:08 kevinquinnyo i also tried cache.clear_all
19:08 Netwizard joined #salt
19:09 Taz_ kevinquinnyo: i think the hack is just to restart the minion, trying to look for a command tho
19:09 Taz_ could try that see if it works tho
19:09 Taz_ had issues with grain caching but its been a while
19:10 kevinquinnyo yeah that probably will work, but if i didnt do anything wrong then that's gotta be a bug
19:10 lemur joined #salt
19:10 kevinquinnyo [DEBUG   ] Clearing cached grains data for:  ['my_minion_id_here']
19:10 kevinquinnyo it even says it does it
19:11 Taz_ hm
19:11 bhosmer_ joined #salt
19:11 geomacy joined #salt
19:12 kevinquinnyo restarting the minion did work
19:12 Brew joined #salt
19:12 kevinquinnyo so i wonder if there's a minion cache and a master cache
19:12 djgerm it'd be really awesome if https://docs.saltstack.com/en/latest/ref/states/all/salt.states.pkg.html had an example to add the saltstack repo https://docs.saltstack.com/en/latest/topics/installation/ubuntu.html
19:12 Taz_ i think the salt-call isnt doing what we think
19:12 Taz_ yea i think so kevinquinnyo
19:12 kevinquinnyo salt-run runs on the master, so it wouldnt be able to go and clear the cache of the target
19:14 Taz_ i think its in the command, i dont run salt-run anymore unless im running something locally
19:14 Taz_ i dont know enough about them to be of use though sorry :/
19:15 lempa joined #salt
19:15 kevinquinnyo ohhh Taz_ i figured it out
19:15 kevinquinnyo https://github.com/saltstack/salt/issues/19853
19:15 saltstackbot [#19853]title: master needs a way to invalidate grains on the minion | I am using 'grains_cache: True' on the minion and there appears to be no clear the minion grains from the master. --refresh-grains-cache only works on the minion, while saltutil.sync_grains only "Sync the grains from the _grains directory on the salt master file server". I ended up running cmd.run from the master and deleting /var/cache/salt/minion/grains.cache.p. ...
19:15 kevinquinnyo you have to sync the grains from the master
19:16 akhter joined #salt
19:16 kevinquinnyo hmm maybe not
19:16 abednarik_ joined #salt
19:17 kevinquinnyo nope, that did actually work
19:17 kevinquinnyo ok that makes sense
19:17 dariusjs joined #salt
19:17 Taz_ kevinquinnyo: nice catch!
19:19 kevinquinnyo so i guess if you want to clear all cache on master and minion and resync all modified modules, etc, you can run salt-run cache.clear_grains tgt='*' && salt '*' saltutil.sync_all
19:19 jim__ [salt.pillar      ][CRITICAL][16149] Specified ext_pillar interface consul is unavailable
19:19 jim__ sadface.jpg
19:20 brianfeister joined #salt
19:20 jim__ oh man
19:20 jim__ it's always something
19:20 jim__ my config was just wrong
19:20 woodtablet joined #salt
19:20 jim__ I had 'port:' instead of 'consul.port:' in /etc/master
19:20 ajw0100 joined #salt
19:25 jfindlay kevinquinnyo: do you mean sync the grains to the master?
19:25 teryx510 joined #salt
19:26 rem5 joined #salt
19:27 kevinquinnyo jfindlay: no i was just trying to figure out how to clear the grains cache but realized salt-run only clears the cache on the master (becuase salt-run doesn't talk to minions)
19:27 kevinquinnyo then i realized that a secondary function of syncing is to invalidate minion cache
19:27 kevinquinnyo at least for grains
19:28 kevinquinnyo at least that's what i gathered
19:28 forrest joined #salt
19:29 hasues joined #salt
19:29 hasues left #salt
19:30 jfindlay I see how that could be confusing though
19:31 Glasswalker joined #salt
19:31 teryx510 joined #salt
19:32 Glasswalker Hey guys, so I'm using a state, which applies a user.present to create a user. It depends on about 10 groups to which the user belongs. The groups are all created successfully beforehand, but the user fails to create. But only on first attempt.
19:32 nidr0x joined #salt
19:32 Glasswalker If I then re-apply the state a second time, user is created just fine...
19:32 Glasswalker I'm fairly new to salt, so any suggestions how I should start troubleshooting this?
19:32 kevinquinnyo Glasswalker: are you using requisites to make sure the user is created after the groups are present
19:32 ffredrikk joined #salt
19:32 forrest Glasswalker, What did the output say when the user failed to create the first time?
19:33 hasues joined #salt
19:33 hasues left #salt
19:33 Glasswalker kevinquinnyo: yes that's how I'm doing it. On a fresh RHEL7 box, when I apply the state, it does the requisites just fine, runs them first in the order, but user fails. Second run of the exact same state, user passes.
19:33 jfindlay also any minion debug log from that event
19:34 Glasswalker forrest: error is: "Failed to create user X" with no further explanation.
19:34 woodtablet :w
19:34 forrest Glasswalker, Alright, If you can I would say repro the box, then run the state for the first pass again with -l debug
19:34 woodtablet hi, and sorry
19:35 Glasswalker totally didn't think to look at minion debug logs lol... Pretty clearly says in there it failed to create the user home dir
19:35 forrest Glasswalker, Without a doubt it's an execution order problem, but getting some debug output is probably the fastest method to see some more details on why it fails.
19:35 Glasswalker wierd that it works second time around lol
19:35 notnotpeter joined #salt
19:35 hasues joined #salt
19:35 hasues left #salt
19:35 forrest Glasswalker, Subdirectory issue perhaps?
19:36 Glasswalker yeah user is the oracle user, home is in /u01/oracle, and /u01 doesn't necessarily exist at that point.
19:36 woodtablet anyone here have rhel and aws salt-cloud experience here ?
19:36 Glasswalker strange that /u01 is created on the second pass lol
19:36 nZac joined #salt
19:36 jfindlay woodtablet: I have RHEL+EC2 experience, so not sure if that's enough :)
19:37 woodtablet hi jfindlay!
19:37 Glasswalker forrest: that kind of leads to my second question I was going to ask... I have a state to build the /u01 folder, but I want to do the following:
19:38 Glasswalker Detect if a physical disk is presented (will always be /dev/xvdbc) and if it is, partition it, and mount the fs as /u01, but if disk doesn't exist, just make /u01 as a folder on filesystem.
19:38 nidr0x joined #salt
19:39 woodtablet jfindlay - so i am trying to launch a machine salt-cloud, works great, even with my custom ami, but what i want to do is have salt-cloud run a script after the ami starts up.  salt-cloud has a script: section in the salt-cloud profile, but i think the host run salt-cloud needs to ssh in and run it, is that correct?
19:39 forrest Glasswalker, I assume /dev/xvdbc isn't being created by salt?
19:40 jfindlay woodtablet: without looking at the source code, that seems like a correct assumption
19:40 Glasswalker So what I've done is this for an sls file: http://pastebin.com/0PUE3UTa
19:40 woodtablet jfindlay - crap
19:40 devopsprosiva joined #salt
19:40 nZac joined #salt
19:40 Glasswalker The problem I'm having is what is the requisite I should use on the partition.mkpart, to check if the disk is presented to the vm?
19:40 jfindlay woodtablet: salt-cloud already uses ssh to run the salt bootstrap script
19:41 hasues joined #salt
19:41 hasues left #salt
19:42 Glasswalker Also secondly, how do I actually install the partition and xfs modules lol, they don't appear to be installed by default ;)
19:42 forrest Glasswalker, So there are a couple ways to do it. The way I would do it is for systems that have that disk present (you would know this when you create the system), you set a grain, then do a matcher in your top.sls against that grain for this specific state.
19:42 mavhq joined #salt
19:42 jfindlay Glasswalker: what version of salt?
19:42 whytewolf Glasswalker: what linux distro also
19:42 woodtablet jfindlay - ok, i ll need to change my image to have some sort of deploy user with a deploy ssh key present.  next question, and looking for more like a suggestion.. i need the script in salt-cloud to some how take the input of the hostname that is passed to when the salt-cloud cmd line is run. suggestions ?
19:42 forrest xfs might not be included on the distro
19:43 Glasswalker jfindlay: salt 2015.8.4 (Beryllium)
19:43 whytewolf [I noticed a long time ago that the XFS module doesn't work on ubuntu]
19:43 Glasswalker whytewolf: RHEL 7
19:43 Glasswalker whytewolf: to be more accurate Oracle Linux 7
19:43 Glasswalker but for all intensive purposes it's RHEL7 lol
19:43 forrest You just pay 2-10X as much ;)
19:43 jfindlay woodtablet: a cloud map might work better for your needs
19:44 digitalhero joined #salt
19:44 woodtablet jfindlay - whats that ?
19:44 Glasswalker And yes XFS is included as a kernel module on the OS, it's actually the default filesystem
19:44 Glasswalker forrest: lol in most Oracle products, yes that's the case... In the case of linux it's just as "free" as RHEL is lol
19:44 jfindlay woodtablet: https://docs.saltstack.com/en/latest/topics/cloud/map.html
19:44 digitalhero joined #salt
19:45 Glasswalker (and if you happen to run it on Oracle hardware you get the OS support subscription for free, which actually puts it ahead of RHEL in that one edge case lol)
19:45 forrest Glasswalker, Do you have the xfs packages installed?
19:45 Eugene FWIW, Oracle and Amazon linux definitely are not the same as CentOS or upstream RHEL
19:45 forrest xfsprogs is what it's called for me on centos 7, but lol rhel/oracle so who knows
19:45 Eugene All three claim to have binary compatibility, but that's "mostly"
19:45 onlyanegg joined #salt
19:45 Glasswalker Eugene: yes that's true, it's more like 95% binary compatibility
19:46 Glasswalker there is a very small percentage of package differences
19:46 whytewolf does blkid exist on the system [it is used by the xfs module
19:46 Eugene All bets are out the window with Oracle's custom kernel or the AMI cloud-init stuff
19:46 Glasswalker the compatibility goes down if you're using the Oracle kernel
19:46 Glasswalker right
19:46 whytewolf [the xfs module seems to have been writen for SuSE]
19:46 ffredrikk joined #salt
19:46 Glasswalker yes xfsprogs is installed on the os
19:47 digitalhero joined #salt
19:47 Glasswalker whytewolf: can you confirm what you mean by blkid, as in a grain on the minion?
19:47 jfindlay woodtablet: you can also use the salt reactor to trigger a (high)state when the minion is created
19:47 keimlink joined #salt
19:47 forrest Glasswalker, can you run the module straight from the command line with something like salt-call --local xfs.devices?
19:47 jfindlay woodtablet: https://github.com/saltstack/salt/issues/23763#issuecomment-103180661
19:47 saltstackbot [#23763]title: Cloud: start_action prevented by highstate | Iḿ currently testing out the VMware cloud driver. For this I have created a VM template which gets cloned. There are however some system-specific tasks to deal with prior to it being provisioned (e.g. growing partitions, changing NIC naming, installing "base" updates)....
19:48 Glasswalker forrest: yes that call works on the minion directly
19:48 nZac joined #salt
19:48 Glasswalker Maybe that one is working
19:48 Glasswalker one sec
19:48 Glasswalker Hmm
19:48 jfindlay woodtablet: rather than running a custom script, you could be salty about it and state your script's content into sls data
19:48 Glasswalker odd partition is working too
19:49 forrest okay, the xfs module is what is used by the state call, so if that works then xfs should be good to go.
19:49 Glasswalker ok consider that issue resolved lol...
19:49 forrest okay
19:49 Muchoz joined #salt
19:49 whytewolf lol, love when things just fix themselves
19:49 spiette joined #salt
19:49 Glasswalker lol I don't, because then I'm usually left on the hook to explain HOW it fixed itself ;)
19:49 Glasswalker lol
19:49 Glasswalker ok
19:50 whytewolf I was being sartastic
19:50 whytewolf :P
19:50 Glasswalker so back to the issue of targeting it... your suggestion of using a grain is one option, and certainly not out of the question... But is it not possible to filter somehow based on the disk being presented itself?
19:50 forrest Glasswalker, Yeah that's possible as well, just use unless.
19:50 forrest Glasswalker, https://docs.saltstack.com/en/latest/ref/states/requisites.html#unless
19:51 Glasswalker Oh awesome! How did I miss that when reading the requisite docs? ;)
19:51 Glasswalker (as I said, new to salt... as in this is my first stab at implementing it, only introduced to it last week lol)
19:52 mgresser joined #salt
19:52 forrest Glasswalker, It happens, onlyif is also a good option
19:52 woodtablet jfindlay - hmm.. so the script or this portion of the script is like 2 line, but i could have salt generate the script, by pulling out the dynamic data out of a pillar, generate the script, then run it, is that right ?
19:53 forrest Glasswalker, onlyif might be better in this situation, just don't forget to create the requires chain for those other items in the partition state, otherwise it will see that device doesn't exist, not do the xfs work, then try to do the other states and fail, so they either need a require, or an unless/onlyif check as well.
19:53 jim__ question: does the consul ext_pillar only access the key-value store?
19:53 jim__ I don't have anything in /kv/ on my consul cluster, but I register services instead
19:53 jfindlay woodtablet: that sounds reasonable, although I'm not fully grasping all of the details of your use case
19:53 djgerm what would the keyserver be for the ubuntu package from https://docs.saltstack.com/en/latest/topics/installation/ubuntu.html  for the 14.04?
19:54 woodtablet jfindlay - naturally, i gave you little context =D. thanks, i think i can do this in a salty way, thanks!
19:54 jfindlay woodtablet: nice, I hope it works out well
19:54 Glasswalker forrest: yeah if you look in the pastebin I posted, they all have requisites chained (I think correctly lol)
19:55 danderson joined #salt
19:55 forrest Glasswalker, For sure, just wanted to note it again.
19:55 danderson Hi folks. Is security@saltstack.com still monitored? If not, what's the preferred way to report a security issue?
19:55 Glasswalker but the top one (the mkpart itself) didn't have a requisite... That solved that problem... Now I'm getting: State 'partition.mkpart' was not found in SLS 'system.storage.datadrive'
19:55 forrest danderson, You can double check with jfindlay but I believe that email is still monitored.
19:56 Glasswalker Reason: 'partition.mkpart' is not available.
19:56 Glasswalker but if I do for example: salt-call --local partition.list /dev/xvdb
19:56 Glasswalker on the minion directly, it lists partitions on that device... Implying the partition module is working
19:56 jfindlay danderson: I just saw your message
19:57 aqua^c joined #salt
19:57 danderson jfindlay: ah, cool. I was worried that I'd somehow misdirected into limbo.
19:57 forrest djgerm, Should just be that gpg-key.pub path on the install page
19:57 Glasswalker And if I call partition.mkpart directly at the minion, it shows the help (explaining mkpart needs min 2 params) which implies the module works
19:57 Glasswalker I think this is what lead to me assuming the module wasn't installed lol
19:58 Glasswalker (the earlier self-solving problem which wasn't really a problem) :)
19:58 forrest Glasswalker, You're targeting that specific minion right? salt 'minion' blah blah blah
19:58 djgerm forrest: i'll try that. I am asking in the context of pkgrepo.managed
19:58 Glasswalker yes I'm targeting that one minion
19:59 forrest djgerm, Yeah I figured.
19:59 whytewolf pst. there is no partition state module
19:59 Glasswalker OH! I was confusing a run module with a state module
20:00 Glasswalker ok... how do I call a run module from within a state?
20:00 whytewolf module.run
20:00 Glasswalker Excellent :) lol
20:00 Glasswalker ok trying that now
20:00 rem5 joined #salt
20:00 whytewolf be very careful cause it isn't stateful
20:00 whytewolf just like cmd.run
20:01 Glasswalker hmm... ok
20:01 Glasswalker can I combine requisites?
20:01 Glasswalker such as onlyif and unless
20:01 whytewolf yeap, you can have both
20:01 djgerm forrest:       Error: key retrieval failed: Executing: gpg --ignore-time-conflict --no-options --no-default-keyring --homedir /tmp/tmp.BwhdPXqVMe --no-auto-check-trustdb --trust-model always --keyring /etc/apt/trusted.gpg --primary-keyring /etc/apt/trusted.gpg --keyserver repo.saltstack.com/apt/ubuntu/14.04/amd64/latest/SALTSTACK-GPG-KEY.pub --logger-fd 1 --recv-keys DE57BFBE
20:01 whytewolf I often do
20:01 Glasswalker to say only if the /dev/xvdc exists but /dev/xvdc1 does not?
20:02 Glasswalker that should provide the (somewhat messy) statefulness that I need
20:03 forrest djgerm, Maybe pull the actual key value and just use the dir? I don't think the keyserver value usually includes that.
20:04 djgerm forrest: can you give me an example of that? like…. grab the pub file and?
20:04 djgerm forrest: thanks for helping me!
20:04 forrest djgerm, I mean just modify it to be  repo.saltstack.com/apt/ubuntu/14.04/amd64/latest/
20:04 djgerm oic.
20:04 forrest djgerm, it might even be repo.saltstack.com
20:04 djgerm just the servername failed
20:05 djgerm trying URI minus filename
20:05 forrest Cool
20:05 forrest I don't have any ubuntu boxes right now so I'm not sure what it SHOULD be.
20:05 forrest jfindlay, Are you familiar at all with what the keyserver should be for the saltstack ubuntu repo?
20:05 viq joined #salt
20:05 djgerm URI failed :(
20:05 bhosmer joined #salt
20:06 djgerm how does Salt recommend we manage the repo with the pkgrepo.managed state ? hehe that should be like THE first example on the doc
20:07 hightekvagabond joined #salt
20:07 jfindlay djgerm: good idea
20:07 jfindlay do you think you could update the doc? :-)
20:07 jfindlay forrest: not sure.  I'll ask
20:07 murrdoc joined #salt
20:08 djgerm I am busy trying to convince my company we havent made a huge mistake adopting salt :)
20:08 jfindlay ok, well, no pressure
20:08 djgerm (they know we havent! I am just joshing)
20:08 djgerm (we just plopped down the big bucks for training and support and all that)
20:08 jfindlay djgerm: you may also be interested in https://github.com/saltstack-formulas/salt-formula
20:08 forrest djgerm, I'm sure it's something obvious, I just don't know what it is for the salt repos.
20:10 djgerm boosh! great thinking jfindlay
20:10 jfindlay djgerm: https://github.com/saltstack/salt/issues/31104
20:10 saltstackbot [#31104]title: Add repo.saltstack.com as an example for pkgrepo.managed |
20:11 Taz_ I think the biggest issues for companys to overcome is the initial training and ramp up for learning a configuration management tool
20:11 djgerm Oh man, you're so great!
20:11 forrest jfindlay, Should probably be in the ubuntu install instructions as well.
20:11 saffe joined #salt
20:11 forrest djgerm, https://github.com/saltstack-formulas/salt-formula/pull/187/files#diff-b84166e1678672d1d928c4371fb2ef74R9
20:11 saltstackbot [#187]title: Update salt.pkgrepo/Vagrant to use new APT repo on Ubuntu | Needed this for personal use. I'm very new to the SaltStack ecosystem - if there are more idiomatic ways to approach this, or a means of testing that I overlooked, I'm happy to update this PR as desired.
20:11 Taz_ Love the idea for the ubuntu install pkgrepo example!
20:12 forrest djgerm, Looks like it's okay if you use key_url instead of keyserver
20:12 nidr0x joined #salt
20:12 jfindlay forrest: repo.saltstack.com/#ubuntu should have a basic working example already, but we could add it to the formal docs as well
20:12 djgerm oh now you tell me! I used the https://github.com/saltstack-formulas/salt-formula/blob/master/salt/pkgrepo/ubuntu/init.sls
20:13 djgerm formulas are TOO HARD! :)
20:13 forrest jfindlay, For sure.
20:13 Taz_ haha
20:13 forrest djgerm, Lol, some people want to make them harder. That's specifically from some guy's PR we didn't merge though
20:14 djgerm i have like… a mental hangup on any instruction that includes gitfs or something. I don't know. They're not actualy hard
20:14 djgerm update
20:14 djgerm oops wrong window
20:16 sfz- joined #salt
20:17 abednarik joined #salt
20:20 digitalhero joined #salt
20:22 djgerm Ha! I wrote this state to make sure my minions are up to date.
20:22 djgerm but I think my minions are too old
20:22 djgerm State pkg.uptodate found in sls minion_update is unavailable
20:23 djgerm New in version 2014.7.0.
20:25 neogenix joined #salt
20:25 neogenix left #salt
20:26 baweaver joined #salt
20:28 digitalhero joined #salt
20:29 Trauma joined #salt
20:33 TyrfingMjolnir joined #salt
20:33 mavhq joined #salt
20:34 Trauma joined #salt
20:35 captain_magnus joined #salt
20:37 rmnuvg joined #salt
20:37 tehsu can a minion have multiple schedules in multiple pillars?
20:38 baweaver joined #salt
20:39 netcho_ joined #salt
20:42 djgerm gah! I fix one thing, and 4 more break…. how does this keep happening and what the heck does data failed to compile mean?
20:42 murrdoc joined #salt
20:43 ffredrikk joined #salt
20:44 ruxu joined #salt
20:44 forrest djgerm, Usually a typo or some other issue, rerun with -l debug and see what it is.
20:45 philiaagape joined #salt
20:45 ssplatt joined #salt
20:45 adis joined #salt
20:46 adis imal ko iz Bosne za razgovor?
20:47 adis bilo ko?
20:47 djgerm also no extra data from debug… but… my test.ping comes back with two results from each minion..
20:48 shaggy_surfer joined #salt
20:48 forrest adis,
20:48 forrest Ne znam nikoga ovdje da govori hrvatski, ispričavam se. (crappy google translate)
20:49 kevinquinnyo so i created my own salt module, and an execution module to go along with it.  i can run the module itself like this: salt dns1.whatever.com geodns.add_record zone_name=example.com record=foo ips="[ ["192.168.0.99", 0], ["192.168.10.100", 0] ]"
20:49 adis ok, my friend.
20:49 kevinquinnyo how might i do this when a new minion comes online with a reactor using the data from the new minion
20:51 forrest adis,
20:51 forrest Pokušajte objavljivanja na https://groups.google.com/forum/#!forum/salt-users, možda netko neće biti u mogućnosti pomoći
20:51 forrest adis,
20:51 forrest ili barem bolje šanse da bude viđen
20:51 adis aham, ja sam ovdje prvi put
20:52 adis i ne razumijem kako ovo sve funkcionise
20:52 shaggy_surfer joined #salt
20:53 forrest adis,
20:53 forrest Bez brige, možda ćete dobiti bolju vidljivost od nekoga tamo tko govori jezik, postavite pitanje, možda bismo mogli dobiti kroz s google prijevod.
20:53 forrest adis, Mislim ovdje na IRC
20:54 adis ok, hvala puno!
20:55 sinonick joined #salt
20:56 mavhq joined #salt
20:56 jfindlay forrest++
20:56 pppingme joined #salt
20:57 kevinquinnyo i keep hitting the same brick wall with the same pattern.  Can anyone help me with this?  It seems like I must be missing something fundamental here
20:58 shoemonkey joined #salt
20:58 kevinquinnyo i asked above, but i'll ask again:  i created my own salt module, and an execution module to go along with it.  i can run the module itself like this: salt dns1.whatever.com geodns.add_record zone_name=example.com record=foo ips="[ ["192.168.0.99", 0], ["192.168.10.100", 0] ]"
20:58 kevinquinnyo i want to be able to *target* specific DNS minions and add a record like that based on an event fired due to a new minion coming into existence
20:59 kevinquinnyo and i do not know how to design this.  Do i use a salt-runner, do i use local.state.sls directly in a reactor file?  something else?
21:00 bhosmer joined #salt
21:00 kevinquinnyo so there are two minions, the new minion that came into existence, event fired and i have it's data in the reactor.  then i need to pass some of it's data (hostname and ips mostly) to another minion, one of the dns servers.  it's that 'middle' part that i dont understand
21:04 insanegeek joined #salt
21:05 Muchoz joined #salt
21:05 forrest kevinquinnyo, You're going to have to use the output data from the reactor in the commands that you run, did you look at: https://docs.saltstack.com/en/latest/topics/reactor/index.html#a-complete-example already?
21:05 forrest That combined with the output provided here: https://docs.saltstack.com/en/latest/topics/reactor/index.html#knowing-what-event-is-being-fired should provide some help
21:06 adis left #salt
21:08 djgerm morrose
21:08 lemur joined #salt
21:08 XqueZme joined #salt
21:10 kevinquinnyo forrest: looks like that example uses local.state.sls
21:10 kevinquinnyo so i guess i'll try that
21:10 dariusjs joined #salt
21:11 kevinquinnyo i hope kwarg isn't a typo in that documentation
21:12 kevinquinnyo forrest: here's the thing though.  With state.sls is it looking for a "state module" named haproxy in that example?
21:12 kevinquinnyo or a salt module
21:12 kevinquinnyo local.state.sls
21:13 XqueZme Hi all, I'm trying to manage ssh keys using ssh_auth.present. When I issue salt state.highstate test=True it says SSH key needs to be updated, but when I remove test=True it says it failed to find source file. I have verified on minion (/var/cache/salt) as well as master and both files seem to be OK.
21:13 whytewolf kevinquinnyo: local.state.sls is just state.sls run locally on the minion.
21:13 whytewolf kevinquinnyo: local.module.function
21:13 djgerm gtmanfred, or anybody else. not that I have my salt-minions up to date, I am getting python errors with simple commands like ' salt-call glusterfs.list_peers'
21:15 djgerm http://pastebin.com/WUHFCErQ
21:16 djgerm if it's not one thing, it's another today. 48 hours ago I had this all working.
21:16 djgerm new minions *sigh*
21:16 whytewolf djgerm: did you get the multiple ping returns fixed?
21:16 kevinquinnyo whytewolf: ok so in that case i could call local.state.sls, send tgt: dns* with arg: record_present and with kwarg: <define the record_present function's values>
21:16 djgerm yeah. I rebooted the servers :)
21:16 kevinquinnyo thanks forrest whytewolf let me try that
21:17 djgerm whytewolf: now they're all on Beryllium
21:17 djgerm so now the module is loading!
21:18 djgerm ~# python -c 'from salt.modules import glusterfs; print(glusterfs.__virtual__())'
21:18 djgerm True
21:18 whytewolf djgerm: okay just making sure ;) looks like the xml returned by _gluster_xml isn't right
21:18 whytewolf ElementTree is choking on it
21:20 djgerm gee whiz…. but I am just trying to do a simple show peers… no config on my part or anything...
21:21 whytewolf try running the gluster command on a  minion gluster --xml --mode=script from the /root directory
21:22 whytewolf se what is being returned
21:22 XqueZme ssh_auth.present (minion running in debug mode) says Could not find file from saltenv 'base', u'salt://ssh-keys/files/key.pub', but I'm using env=dev
21:22 gtmanfred djgerm: what does gluster --xml --mode=script show on the minion?
21:22 gtmanfred what whytewolf said
21:23 djgerm ~# gluster --xml --mode=script
21:23 djgerm gluster>
21:23 gtmanfred oh
21:23 gtmanfred i think i remember seeing this bug
21:23 djgerm > peer status
21:23 djgerm <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
21:23 djgerm <cliOutput>
21:23 djgerm <opRet>0</opRet>
21:23 djgerm <opErrno>0</opErrno>
21:23 djgerm <opErrstr>peer status: No peers present</opErrstr>
21:23 djgerm <peerStatus/>
21:23 djgerm </cliOutput>
21:23 djgerm oh yeah?
21:24 djgerm maybe a regression in the module with beryllium or even tho I did reinstall salt-common I still have an old version?
21:24 djgerm do modules get cached?
21:24 gtmanfred yeah
21:24 Bloodshot joined #salt
21:24 gtmanfred it is a problem with how the cmd.run runs it
21:24 gtmanfred what version of gluster do you have?
21:24 djgerm looks likr 3.4.2
21:25 djgerm glusterfs 3.4.2 built on Jan 14 2014 18:05:37
21:25 Bloodshot Hey all. I'm trying to learn SaltStack, and would like to know if it's possible to target only hosts with a certain version of a package already installed.
21:25 gtmanfred djgerm: https://github.com/saltstack/salt/issues/30923
21:25 saltstackbot [#30923]title: Starting a glusterfs volume after creation fails with an exception in 2015.8 and after. | Looks like when the module got changed to use the _gluster_xml function, the start command was missed; using create now with start: True will raise an exception....
21:25 gtmanfred that is the problem
21:25 breakingmatter joined #salt
21:25 gtmanfred or something like that
21:25 Bloodshot To be specific, some hosts have an older version of a package which does not support a certain configuration option, and I would like to update the configuration of all hosts with new enough packages.
21:25 zmalone Bloodshot: I don't think you end up with a full package list in grains by default, but you could make it work.
21:26 zmalone okay, you can do that, it just wouldn't be with overall targeting
21:26 mavhq joined #salt
21:26 gtmanfred weird
21:26 gtmanfred https://github.com/saltstack/salt/blob/develop/salt/modules/glusterfs.py#L119
21:27 djgerm gtmanfred: even though I am just doing  salt-call glusterfs.list_peers?
21:27 brianfeister joined #salt
21:27 gtmanfred that should be fine
21:27 gtmanfred https://github.com/saltstack/salt/blob/v2015.8.5/salt/modules/glusterfs.py#L119
21:27 Bloodshot zmalone: Should I write my own grain?
21:27 djgerm gtmanfred: http://pastebin.com/WUHFCErQ
21:27 gtmanfred djgerm: hrm, yeah i don't know why that is failing, it looks like it should pass
21:27 zmalone I'd probably do it by making your formulas check for version of package prior to upgrading
21:27 zmalone or rather, writing a config
21:29 Bloodshot zmalone: I tried looking into having that in my states, but I got the impression that everything would *ensure* that the new version was installed (possibly breaking hosts that need the older version), rather than just checking.
21:29 kevinquinnyo # salt-call event.fire "{'foo': 'bar'}" 'start'
21:29 kevinquinnyo [ERROR   ] The Salt Master has cached the public key for this node, this salt minion will wait for 10 seconds before attempting to re-authenticate
21:29 kevinquinnyo i dont understand the error
21:29 beardedeagle joined #salt
21:29 kevinquinnyo why would it tell me that based on what i did
21:30 gtmanfred check salt-key on the master, and make sure that minion is accepted
21:30 kevinquinnyo what minion
21:30 gtmanfred the one that you are doing salt-call from
21:30 kevinquinnyo oh
21:30 kevinquinnyo ohh ok i get it
21:30 kevinquinnyo thatnks
21:30 gtmanfred np
21:31 gtmanfred djgerm: i don't know why that is failing, i also unfortunately do not have a gluster setup right now, but i can probably take a look this weekend if you don't figure it out by then
21:31 djgerm something broken with beryllium
21:31 djgerm i downgrade to 2015.5.3
21:31 djgerm :~# salt-call glusterfs.list_peers
21:31 djgerm [INFO    ] Executing command "gluster peer status | awk '/Hostname/ {print $2}'" in directory '/root'
21:31 djgerm local:
21:31 djgerm None
21:31 snicers-work I need to downgrade nginx (ubuntu 14.04) but pkg.removed and .purged do not work.
21:32 buhman djgerm: does gluster --xml --mode=script actually return valid xml?
21:32 gtmanfred djgerm: it was the change to xml i think, i think that all happened in 2015.8.4/5, should have waited for a new major release
21:32 gtmanfred buhman: it does not, it returns the interactive shell to then put 'peer status' into
21:33 gtmanfred djgerm: looks like some of the gluster stuff got reverted in the 2015.8 branch
21:33 gtmanfred https://github.com/saltstack/salt/blob/v2015.8.5/salt/modules/glusterfs.py#L63
21:33 gtmanfred there is the 2015.8.5 tag
21:34 gtmanfred oh, nevermind, was looking at the wrong thing
21:34 hightekvagabond joined #salt
21:34 gtmanfred buhman: it sends it to the gluster shell https://github.com/saltstack/salt/blob/2015.8/salt/modules/glusterfs.py#L63
21:34 gtmanfred as stdin
21:35 djgerm wow and I was just saying what an amazing, intuitive, easy to use module it is
21:35 gtmanfred and his peer status from the shell, looked like it returned valid xml
21:35 djgerm Sorry :(
21:35 djgerm it did
21:35 gtmanfred djgerm: it used to be :P
21:35 gtmanfred then they added this xml stuff to make it be able to do more things
21:35 sasajuric joined #salt
21:36 gtmanfred because the xml is more consistent, and the plain output changes between releases
21:36 gtmanfred yeah, i need to sit down and figure out how all this works again, been too long
21:36 sasajuric left #salt
21:36 djgerm well I have a compeltely acceptable workaround for now
21:37 djgerm just downgrade to lithium
21:37 djgerm mmm lithium
21:37 gtmanfred heh
21:38 djgerm but now I have to rewrite the state I wrote to upgrade to beryllium! haha
21:40 VSpike_ joined #salt
21:40 mohae_ joined #salt
21:40 rawzone^ joined #salt
21:40 rihannon1 joined #salt
21:40 bmcorser_ joined #salt
21:41 dkrae1 joined #salt
21:41 renoirb_ joined #salt
21:41 yjmbo joined #salt
21:42 maartenpi joined #salt
21:42 sarasfox joined #salt
21:42 jnilsson joined #salt
21:42 nebuchad` joined #salt
21:43 kevinquinnyo any way to get grains of the minion that fired an event in a reactor
21:43 pmumby_ joined #salt
21:43 jfred_ joined #salt
21:44 whytewolf kevinquinnyo: you only have the event data in a reactor. you might work with mines to get grain data in the state that is run
21:44 honestly_ joined #salt
21:44 kevinquinnyo omg
21:44 mirko joined #salt
21:44 bonzibud1y joined #salt
21:44 muep_ joined #salt
21:45 CaptTofu__ joined #salt
21:45 whytewolf actually you might be able to use mines [as a runner] in the reactor. I'm not 100% sure on that though
21:45 whytewolf either way mines
21:45 kevinquinnyo i wonder what use is the reactor system then
21:45 ruxu joined #salt
21:45 eliasp_ joined #salt
21:46 kevinquinnyo if it has no knowledge of anything useful
21:46 TTimo_ joined #salt
21:46 whytewolf it does. it is used to trigger states based on events
21:46 jfindlay forrest, djgerm: https://github.com/saltstack/salt-pack/issues/33.  From what I remember, our current key should already be supported/ratified/whatever by an external keyserver; cro would know those details for sure
21:46 saltstackbot [#33]title: package key process and keyserver support | 1.  When we create a new major release we should consider whether to use a new encryption key to be used with that release based on the guaranteed support lifetime of that release and whether it terminates before the key expires.  This will not need to happen for every major release, but must happen for major releases whose lifetimes extend beyond the expiration of the current key.  In th
21:47 kevinquinnyo but i need to trigger a dns record add on a dns server for the new minion and without grains data, how do i know it's ip addresses to add records?
21:47 sarasfox is their a way to pass args to network.netstat
21:47 whytewolf kevinquinnyo: you have data['id'] to use as the target for the mine
21:47 GreatSnoopy joined #salt
21:47 KingJ_ joined #salt
21:47 alexlist` joined #salt
21:47 kevinquinnyo ok i guess i need to look into the mine system, i never have
21:48 Vaelater1 joined #salt
21:48 [BNC]aboe joined #salt
21:48 Vye_ joined #salt
21:48 snarfy^ joined #salt
21:48 quarcu_ joined #salt
21:48 garphy joined #salt
21:48 nZac joined #salt
21:49 xMopxShell joined #salt
21:49 Karunamon joined #salt
21:49 AndreasLutro joined #salt
21:49 etw joined #salt
21:49 yjmbo Reasonably new to Salt - where do I need to look to understand how to take an action on two systems at the same time; i.e. when I install a web server, I also want to add that service to my nagios checks ... ?
21:49 TomJepp joined #salt
21:49 ThomasJ joined #salt
21:50 whytewolf yjmbo: orchestrate
21:50 whytewolf yjmbo https://docs.saltstack.com/en/latest/topics/tutorials/states_pt5.html
21:50 winsalt joined #salt
21:50 CustosLimen joined #salt
21:50 giantlock joined #salt
21:50 yjmbo whytewolf: thanks, will read and hopefully understand :-)
21:51 whytewolf it might take some fidaling to understand it isn't the most intuative setup
21:51 babilen joined #salt
21:51 baweaver joined #salt
21:52 tyler-baker joined #salt
21:52 whytewolf but we are here to answer questions you might end up with
21:53 Gareth joined #salt
21:53 yjmbo whytewolf: thanks. Hopefully once I've read up on the topic and tried a few things, I'll understand enough to ask a good question (or, it'll all work just fine!)
21:53 paolo joined #salt
21:54 aRkadeFR joined #salt
21:54 dijit joined #salt
21:54 cberndt joined #salt
21:54 bhosmer joined #salt
21:54 izibi joined #salt
21:55 workthrick joined #salt
21:55 dimeshake joined #salt
21:55 tinyhippo joined #salt
21:55 monokrome joined #salt
21:55 ssplatt mypillar = __salt__['pillar.get’](‘mything’,merge=True)        should merge lists that are like    mything: another: - listitem     and      mything: another: - listitem2      to give me   mything: another: -listitem - listitem2   right?
21:55 Pixionus joined #salt
21:55 Pixionus joined #salt
21:55 nahkiss_ joined #salt
21:55 ramblinpeck joined #salt
21:55 chutzpah joined #salt
21:55 shoemonkey joined #salt
21:56 gtmanfred yjmbo: they key about state.orchestrate, is so that you can do stuff in the order that they need to be done for the deploy, like for my openstack deploy, my databases have to be up, then i need keystone up, then all the other bits can come up... so  https://github.com/gtmanfred/openstack-salt-states/blob/master/deploy/openstack.sls
21:56 gtmanfred does the highstates on the right pieces in the right order
21:56 jesusaur joined #salt
21:56 ecdhe joined #salt
21:56 czchen joined #salt
21:56 OliverMT joined #salt
21:56 ramishra joined #salt
21:57 trave joined #salt
21:57 tcolvin joined #salt
21:57 NachoDuck joined #salt
21:57 mattl joined #salt
21:57 wendall911 joined #salt
21:57 lompik joined #salt
21:58 Ryan_Lane_ joined #salt
21:58 upb joined #salt
21:59 geomacy joined #salt
21:59 bVector joined #salt
21:59 JonGretar joined #salt
22:00 mihait joined #salt
22:00 edrocks joined #salt
22:02 yjmbo gtmanfred: thanks, another entry for my research notes :-)
22:02 gtmanfred :)
22:02 kevinquinnyo whytewolf: reading through the "mine" documentation here.  Will the mine data be immediately available when the new minion comes into existence?  It looks like there is a polling delay?
22:03 kevinquinnyo since my event is based on a new minion coming online
22:03 whytewolf kevinquinnyo: have a reactor that causes a mine.update on the minion once it comes u
22:03 whytewolf p
22:03 kevinquinnyo ah ok
22:04 evilrob I got an invite to the saltstack meetup in mountainview tomorrow.  shame I'm in TX  :)
22:05 hal58th joined #salt
22:05 djgerm i am on the other side of the hill from mountain view, and I FEEL like I may as well be in TX. it's so far
22:05 whytewolf kinda wished someone [other then me] would start a saltstack meetup in Las Vegas
22:06 teryx510 joined #salt
22:06 ssplatt hmm oh i see i need a default= for merge=true.  how could i do that from the commandline tho, or in a python module?.  i see how i’ve been doing it in my maps.jinja file for my regular formulas.  any hints?
22:06 beneggett joined #salt
22:06 djgerm unrelated to my other work so far today, how would one prevent any commands like salt '*' cmd.run 'shutdown -h now' from being run?
22:07 jfindlay djgerm: there is blacklist support for that kind of thing
22:08 djgerm oh kewl kewl. I'll check that out :) not that I dont trust the people I work with
22:08 WesleyTech joined #salt
22:08 djgerm But I dont trust the people I work with
22:08 gtmanfred as it should be
22:08 jfindlay djgerm: some orgs have a process where only salt is allowed to touch the system directly and all changes to highstate have to be peer reviewed
22:08 whytewolf hell some days i don't trust myself
22:08 jfindlay lol
22:09 gtmanfred djgerm:if yall have ldap, we use ldap to control who can run commands against salt, and don't let anyone get root acess
22:09 gtmanfred access to the salt master
22:09 gtmanfred so you run it with the ldap auth
22:09 murrdoc 'djgerm: some orgs have a process where only salt is allowed to touch the system directly and all changes to highstate have to be peer reviewed'
22:09 * murrdoc is working towards it
22:10 djgerm brilliant stuff. I am saving these words of wisdom in my "To investigate" journal
22:10 whytewolf murrdoc: I also work on something simalar. only box i tyically login to is my salt box. everything else is off limits. on my cloud boxes i actualy had proccesses that would nuke and rebuild the box if the login count went over 1
22:10 Guest15678 joined #salt
22:10 buhman gtmanfred: hmm, so how do you deal with ugly hardware-related nonsense?
22:11 gtmanfred we don't?
22:11 buhman that sounds fun
22:11 gtmanfred delete the server and recreate it
22:11 buhman oh
22:11 whytewolf nuke and rebuild. it is the wave of the future
22:11 murrdoc i dont log into any box
22:11 buhman so this process is for not-actual-hardware
22:12 buhman nobody having root access on hardware sounded rather strange ;p
22:12 gtmanfred oh yeah
22:12 gtmanfred for drac and raid stuff, we don't need those things
22:12 gtmanfred my team lives entirely in the cloud
22:12 gtmanfred doing postbuild automation for all of the rackspace cloud customer stuff
22:12 buhman I was hoping you'd say "we have salt states for drac too"
22:12 gtmanfred heh
22:12 gtmanfred pass
22:12 whytewolf buhman: actuallly... possable ...
22:12 gtmanfred there was a module out there somewhere for racadm
22:13 gtmanfred buhman: https://docs.saltstack.com/en/latest/ref/states/all/salt.states.dellchassis.html
22:13 gtmanfred bam
22:13 gtmanfred just using proxy minions
22:13 yjmbo left #salt
22:13 whytewolf I have looked at the ipmi salt stuff as well, but the boxes i have that use ipmi are braindead.
22:13 buhman that looks like something you'd definitely want for playing with blades
22:13 Muchoz Is there any documentation on how to use the gpg renderer here? https://github.com/saltstack/salt/pull/12325
22:14 saltstackbot [#12325]title: Add a first pass at a gpg renderer | See issue #11147
22:14 gtmanfred https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.drac.html
22:14 gtmanfred Muchoz: https://docs.saltstack.com/en/latest/ref/renderers/all/salt.renderers.gpg.html
22:16 Muchoz gtmanfred, thank you. That last yaml code, is that the top.sls from the pillars or one of the pillars?
22:17 gtmanfred that is the encrypted data for the pillar
22:17 mihait_ joined #salt
22:17 gtmanfred so it would be in the /srv/pillar/something.sls
22:17 Muchoz Alright, thank you very much!
22:17 aboe joined #salt
22:17 gtmanfred np
22:18 amcorreia joined #salt
22:18 mirko_ joined #salt
22:18 asoc_ joined #salt
22:19 nebuchad` joined #salt
22:20 dyasny_ joined #salt
22:20 Guest76581 joined #salt
22:20 viq_ joined #salt
22:21 babilen_ joined #salt
22:21 dkrae joined #salt
22:22 kevinquinnyo https://github.com/saltstack/salt/issues/6655  someone posts an example here on using salt "mine" but i can't seem to figure out what network.interfaces and network.ip_addrs actually represent?
22:22 saltstackbot [#6655]title: Additional documentation regarding the salt mine. | It would be great to see some more documentation for [The Salt Mine](http://docs.saltstack.com/topics/mine/index.html), preferably showing an actual example of how it is set up on the master (as in a real world example, maybe built off the walkthrough data), the information that you're pulling from the minion, and then how the data gets returned, where it is stored, etc. . The c
22:22 gtmanfred kevinquinnyo: the modules
22:22 kevinquinnyo are those methods in some grains class?  they aren't actual keys in the grains as far as i can tell
22:22 gtmanfred salt \* network.interfaces
22:22 Vye joined #salt
22:22 kevinquinnyo ohhh
22:22 kevinquinnyo ok
22:22 kevinquinnyo thanks
22:22 gtmanfred it is the module the minion runs, and then reports the data back, and is saved in the mine
22:22 gtmanfred it is the same data that you would get back on the master if you ran the command
22:22 gtmanfred it is just cached
22:22 alexlist joined #salt
22:22 paolo joined #salt
22:22 paolo joined #salt
22:22 gtmanfred and saved
22:23 kevinquinnyo ok cool
22:23 kevinquinnyo thanks
22:23 gtmanfred you have to tell the minion what to report back in the mine though
22:23 dijit joined #salt
22:23 gtmanfred that is what the mine functions are
22:23 gtmanfred https://docs.saltstack.com/en/latest/topics/mine/
22:24 gtmanfred you tell it to run test.ping, and network.ip_addrs eth0 cidr='10.0.0.0/8'
22:24 gtmanfred well
22:24 onlyanegg joined #salt
22:24 gtmanfred network.ip_addrs interface=eth0 cidr='10.0.0.0/8'
22:24 gtmanfred since it is doing key value pairs
22:24 gtmanfred but it is just the same thing you would run on the commandline
22:25 gtmanfred and then make that data available to other minions
22:25 kevinquinnyo i read those docs, i must have missed where it explains that those are <salt module>.<method>
22:25 gtmanfred i don't think it explicitly says it, but that is what they are :)
22:26 kevinquinnyo well thanks for clarifying
22:26 gtmanfred and you can include aliases like in the example
22:26 gtmanfred ip_list:
22:26 gtmanfred - mine_function: grains.get
22:26 gtmanfred - ip_interfaces
22:26 whytewolf i recommend using alias. makes this a lot simpler.
22:27 kevinquinnyo cool
22:27 subsignal joined #salt
22:28 snarfy joined #salt
22:28 nidr0x joined #salt
22:29 gtmanfred +1 to that
22:29 gtmanfred if i specify the arguments in the alias, then i can just ask for the alias name in my mine.get?
22:29 gtmanfred and not have to pass the full command?
22:30 djgerm oh hey… gtmanfred you might THE one who can help me on this one!
22:30 joe_n joined #salt
22:30 gtmanfred maybe :/
22:31 djgerm so here's the deal. I have this gluster pool thing where I add in my state all four of my nodes
22:31 djgerm BUT
22:31 gtmanfred i haven't looked at gluster in over a year :/
22:31 djgerm since glusterfs-server just started, nodes in the same subnet add each other to their pools automagically.
22:31 pppingme joined #salt
22:31 gtmanfred oh
22:31 gtmanfred for peering or something?
22:31 djgerm so i get "minionX is already part of another cluster"
22:32 djgerm yeah
22:32 gtmanfred i used to use an orchestrate runner and choose one minion at random to perform all the peering tasks initially
22:32 jfindlay djgerm: JoeJulian has been working on gluster modules recently
22:32 gtmanfred djgerm: https://github.com/gtmanfred/salt-states/blob/oldmaster/overstate.sls
22:32 djgerm Ha! he's in here too?! That dudes amazing!
22:32 gtmanfred actually, an overstate
22:33 gtmanfred from back when that was a thing... i think that got removed
22:33 joe_n joined #salt
22:33 totzky joined #salt
22:34 jfindlay yes, the orchestrate runner replaced overstate
22:35 gtmanfred :)
22:36 djgerm hmm, well i think, even if I did do that, since of with my 4 nodes, there are two subnets, and thus when glusterfs-server starts for the first time it has  two disparate, unrelated pools, I'd still need to orchestrate somehow to remove all the nodes from their pools so I can create one pool
22:36 JoeJulian :)
22:37 djgerm Hi JoeJulian!
22:37 JoeJulian Howdy
22:37 pppingme joined #salt
22:37 djgerm geo segregated, no geo replicated volume has worked for my use case so far by the way :)
22:41 kevinquinnyo anyone know which salt module creates the grain for fqdn and domain, etc
22:41 JoeJulian djgerm, gtmanfred: I'll be presenting at the Incontro Devops Italia how I install clustered software (gluster and ceph) using reactor.
22:42 gtmanfred nice
22:43 djgerm awesome!
22:44 zenlot1 joined #salt
22:45 perfectsine joined #salt
22:45 rem5 joined #salt
22:46 jfindlay kevinquinnyo: that is probably a core grain: salt/grains/core.py
22:47 kevinquinnyo well wait i'm confused again
22:47 kevinquinnyo how would i put that into a mine
22:48 jim__ joined #salt
22:48 gtmanfred kevinquinnyo: put grains.get: [fqdn] and domain in the mine
22:48 kevinquinnyo the example used network.interfaces which is salt/modules/network
22:48 bhosmer_ joined #salt
22:49 jhauser joined #salt
22:49 kevinquinnyo ohh salt/grains is also a module
22:49 kevinquinnyo is it unusual that i'm confused by this
22:49 MindDrive joined #salt
22:51 jfindlay kevinquinnyo: grains are set upon minion startup by python files salt/grains.  You can also generate your own custom grains in a variety of ways, the simplest being the grains.set execution module function
22:52 jfindlay standard grains are set upon minion startup, that is
22:52 kevinquinnyo yeah that part i get for the most part
22:53 djgerm so I am thinking
22:53 me joined #salt
22:53 djgerm for the glusterfs, why not pull the peer and volume creation out and give that state to a single node?
22:53 gtmanfred djgerm: that is what i did
22:54 djgerm with your fancy overstate
22:54 beardedeagle joined #salt
22:54 gtmanfred yeah, but don't use that anymore
22:54 gtmanfred use orchestrate
22:54 gtmanfred it is better
22:54 beardedeagle has anyone actually gotten USE_KEYRING to work with salt-cloud? I remember trying it a while back but found out that while the flag was there it wasn't actually implemented yet.
22:54 gtmanfred it looks a little different too
22:54 geomacy joined #salt
22:55 gtmanfred beardedeagle: i think it was pulled out, cause no one actually talked about it
22:55 gtmanfred beardedeagle: but i think sdb was added for the same purpose
22:55 djgerm i dont know what that is :( can I accomplish that same thing with like… top.sls: minionglustmaster* gitmaster ….. glustminion*: other glust stuff….
22:55 gtmanfred beardedeagle: https://docs.saltstack.com/en/develop/topics/sdb/index.html
22:55 djgerm that was… a very bad communication
22:55 beardedeagle I thought USE_KEYRING was going to use sdb as the backend
22:55 gtmanfred oh, maybe?
22:56 djgerm know what I mean? like apply all the gluster states that arent pool and volume creation...
22:56 djgerm then
22:56 gtmanfred i never got around to testing it
22:56 beardedeagle Well time to start digging again
22:56 gtmanfred djgerm: this is what an overstate file looked like, https://github.com/gtmanfred/salt-states/blob/oldmaster/overstate.sls
22:56 gtmanfred and this is what an orchestrate looks like
22:56 gtmanfred https://github.com/gtmanfred/openstack-salt-states/blob/master/deploy/openstack.sls
22:57 gtmanfred and orchestrate can be run with salt-run state.orch deploy.openstack
22:57 gtmanfred and from your salt fileserver, i remember not being able to do that back when i was using overstate
22:57 JDiPierro joined #salt
22:58 baweaver joined #salt
22:58 djgerm can i put that in my file_root?
22:58 gtmanfred yeah
22:58 gtmanfred if it can be referenced from salt:// it can go there
22:59 djgerm so….tgt is the target i.e. the minion
22:59 djgerm to apply to
22:59 gtmanfred andi can call
22:59 gtmanfred salt-call cp.get_file_str salt://deploy/openstack.sls
22:59 gtmanfred tgt: is what you are matching to
23:00 djgerm but… what state is it applying to those matches?
23:00 gtmanfred and you can do tgt_type to specify compound, grain, cidr, pillar, pcre, etc
23:00 djgerm just highstate in that example?
23:00 gtmanfred it is doing a highstate
23:00 gtmanfred yeah
23:00 snarfy joined #salt
23:00 gtmanfred you can tell it a list of sls file
23:00 djgerm so i would my mystate: True and mysecondstate: True
23:00 gtmanfred or just functions
23:00 gtmanfred djgerm: https://docs.saltstack.com/en/latest/topics/tutorials/states_pt5.html#examples
23:01 djgerm oh yeah…
23:01 gtmanfred sls:
23:01 gtmanfred - nginx
23:01 gtmanfred - gluster.peer
23:01 gtmanfred or whatever your states are
23:01 gtmanfred same way you would specify them in the top.sls
23:01 gtmanfred but in a yaml list under sls:
23:02 kevinquinnyo gtmanfred: is this right? https://gist.github.com/kevinquinnyo/b655fac6d2882555cd9e
23:02 kevinquinnyo the aliases aren't working for me
23:03 gtmanfred looking
23:05 kevinquinnyo by the way can i write piullar data as python data structures?  YAML makes my head hurt
23:07 gtmanfred kevinquinnyo: it is working for me
23:07 kevinquinnyo hm
23:07 gtmanfred kevinquinnyo: you can write it as json? or using the pyrenderer i believe
23:07 gtmanfred i dropped your file in /etc/salt/minion.d/mine.conf on the minion i wanted data reported from
23:07 gtmanfred then restarted the minion
23:07 kevinquinnyo with the py renderer is it the same, you just have a function called run() required to return a dictionary?
23:07 gtmanfred and it worked
23:08 gtmanfred kevinquinnyo: i believe so
23:08 kevinquinnyo you're running salt \* mine.get \* fqdn
23:08 kevinquinnyo and it's returning an fqdn for the targeted minions?
23:08 gtmanfred [root@novacompute03 ~]# salt-call mine.get \* fqdn
23:08 gtmanfred [INFO    ] Determining pillar cache
23:08 gtmanfred local:
23:08 gtmanfred ----------
23:08 gtmanfred salt.manfred.io:
23:08 gtmanfred salt.manfred.io
23:09 gtmanfred it is retruning it for the minions that have the mine function setup
23:09 brianfeister joined #salt
23:10 kevinquinnyo # salt-call mine.get \* network.interfaces
23:10 kevinquinnyo that works
23:10 kevinquinnyo but my aliases dont
23:10 kevinquinnyo weird
23:10 gtmanfred https://gist.github.com/gtmanfred/76ac7c67c81d1773ed5c
23:11 kevinquinnyo i put this in a pillar file
23:11 kevinquinnyo and made sure '*' loads it in the pillar top.sls
23:12 gtmanfred ahh, never tried it in the pillars
23:12 gtmanfred did you do a salt \* saltutil.pillar_refresh ?
23:12 kevinquinnyo yep
23:13 kevinquinnyo and mine.update
23:13 kevinquinnyo on all of them
23:14 N-Mi joined #salt
23:14 kevinquinnyo oh well
23:15 kevinquinnyo thanks for your help gtmanfred, i gotta go sit in traffic now
23:15 gtmanfred good luck :)
23:15 kevinquinnyo thanks
23:23 gtmanfred yup, both those examples work for me :/
23:27 perfectsine joined #salt
23:28 beardedeagle joined #salt
23:31 freelock joined #salt
23:33 jim_ joined #salt
23:33 Antiarc joined #salt
23:33 aqua^c joined #salt
23:36 Antiarc :q
23:43 GrueMaster joined #salt
23:45 radhac joined #salt
23:49 wendall911 left #salt
23:54 garphy joined #salt

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