Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-02-22

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

All times shown according to UTC.

Time Nick Message
00:00 joe_n joined #salt
00:00 jeffspeff joined #salt
00:00 snc joined #salt
00:04 keimlink_ joined #salt
00:15 nyx_ joined #salt
00:15 keimlink joined #salt
00:26 ALLmightySPIFF joined #salt
00:36 hightekvagabond joined #salt
00:36 bucky joined #salt
00:41 bucky joined #salt
00:45 bucky_ joined #salt
00:50 drawsmcgraw1 joined #salt
00:50 andrej left #salt
01:00 iceyao joined #salt
01:07 bucky joined #salt
01:10 iceyao_ joined #salt
01:11 iceyao__ joined #salt
01:25 eodgooch joined #salt
01:36 bucky joined #salt
01:37 iceyao joined #salt
01:37 bucky_ joined #salt
01:38 bucky_ joined #salt
01:39 bucky_ joined #salt
01:42 iceyao_ joined #salt
01:44 bucky joined #salt
01:46 joe_n joined #salt
01:52 kevinquinnyo1 i doubt many people are watching this channel right now, but I'm wondering how i can ensure that a directory is a git repo, and that it's remote 'origin' is set up to the correct remote.  I don't want it to pull any code, i just want it to be setup properly.  I'm not seeing a git state function for that at first glance
01:56 bucky joined #salt
01:58 hightekvagabond joined #salt
02:01 bucky joined #salt
02:01 bucky joined #salt
02:02 srinivas joined #salt
02:03 iceyao joined #salt
02:04 tweakism if it doesn't have that, you could make a cmd state for it easily enough
02:13 bucky joined #salt
02:15 dyasny joined #salt
02:16 bucky_ joined #salt
02:16 ageorgop joined #salt
02:19 donmichelangelo joined #salt
02:21 darix joined #salt
02:26 bucky joined #salt
02:29 favadi joined #salt
02:35 darix joined #salt
02:46 ajw0100 joined #salt
02:48 ilbot3 joined #salt
02:48 Topic for #salt is now Welcome to #salt! | Latest Version: 2015.8.7 | Register for SaltConf16: http://saltconf.com/register/ | Paid support available for open source Salt! https://www.saltstack.com/support/ | Logs: http://irclog.perlgeek.de/salt/ | Paste: https://gist.github.com/ (please don't multiline paste into channel) | Ask with patience as we are volunteers and may not have immediate answers
02:48 anmol joined #salt
02:49 hightekvagabond joined #salt
02:50 racooper joined #salt
02:52 catpigger joined #salt
03:05 malinoff joined #salt
03:23 favadi joined #salt
03:26 anthpa joined #salt
03:33 ALLmightySPIFF joined #salt
03:36 bucky joined #salt
03:37 andrew_v joined #salt
03:39 bucky_ joined #salt
03:53 ramteid joined #salt
04:11 izrail joined #salt
04:18 iceyao_ joined #salt
04:30 bhosmer joined #salt
04:55 treaki_ joined #salt
04:56 cpowell joined #salt
05:00 viq joined #salt
05:03 FTfz joined #salt
05:04 pwalsh joined #salt
05:05 FTfz joined #salt
05:06 FTfz joined #salt
05:21 rdas joined #salt
05:25 bhosmer joined #salt
05:25 jhauser joined #salt
05:31 dean joined #salt
05:37 favadi joined #salt
05:40 iceyao joined #salt
05:46 pwalsh joined #salt
05:46 joe_n joined #salt
05:50 neilf__ joined #salt
06:02 malinoff joined #salt
06:10 aharvey joined #salt
06:11 uros joined #salt
06:13 cswang_ joined #salt
06:13 linjan__ joined #salt
06:13 vincehu joined #salt
06:17 uros Hi guys, I'm a bit new to salt stack .. I'm trying to run python script via cmd.script ..but I get this https://dpaste.de/b3JB
06:19 bhosmer joined #salt
06:19 nidr0x joined #salt
06:22 FTfz joined #salt
06:23 hightekvagabond joined #salt
06:27 mariusv joined #salt
06:34 FTfz joined #salt
06:34 anmol joined #salt
06:43 JohnnyRun joined #salt
06:43 pwalsh joined #salt
06:46 eseyman joined #salt
06:46 felskrone joined #salt
06:53 eodgooch joined #salt
06:56 LondonAppDev joined #salt
07:13 chiui joined #salt
07:14 bhosmer joined #salt
07:15 JohnnyRun joined #salt
07:24 yuhlw joined #salt
07:35 ntropy uros: not sure, try running with debug log level and paste the output
07:41 Antiarc joined #salt
07:44 jbrnds joined #salt
07:44 FTfz left #salt
07:45 opdude joined #salt
07:46 fracklen joined #salt
07:47 fracklen joined #salt
07:48 fracklen joined #salt
07:49 jhauser joined #salt
07:53 aarontc joined #salt
08:01 deniszh joined #salt
08:05 Antiarc joined #salt
08:07 cprior joined #salt
08:09 antpa joined #salt
08:10 slav0nic joined #salt
08:12 cprior joined #salt
08:16 colttt joined #salt
08:16 cprior How would I use states.portage_config.flags if I want to set =sys-apps/util-linux-2.27.1 ? I have tried numerous ways to get the > past the YAML parser. This setting is part of a bigger file. If anyone had a working version I'd be glad
08:17 cprior I am on salt 2015.8.5
08:17 joshin joined #salt
08:17 joshin joined #salt
08:18 Zachary_DuBois joined #salt
08:19 tweakism so it doesn't work as a multi-line string using | or > ?
08:20 tweakism you can use '' to make a multi-line string
08:20 tweakism I'm certain that would work.
08:20 atmosx joined #salt
08:20 tweakism it eats newlines though
08:22 tweakism and, I should mention... I don't see any >'s in '=sys-apps/util-linux-2.27.1'
08:23 colttt_ joined #salt
08:23 kawa2014 joined #salt
08:24 KermitTheFragger joined #salt
08:28 iceyao_ joined #salt
08:34 dgutu joined #salt
08:37 josuebrunel joined #salt
08:37 douardda joined #salt
08:39 dariusjs joined #salt
08:45 douardda joined #salt
08:45 elsmo joined #salt
08:46 ronnix joined #salt
08:48 cprior oops, sorry -- cop&paste error, I pasted the workaround. I want >=sys-apps/util-linux-2* :) I think I had tested all | or > variations. If it should work though then I will try harder! ;) (The workaround-enabled ebuild is compiling right now though)
08:48 douardda joined #salt
08:49 josuebrunel joined #salt
08:51 tweakism I don't know the rules for >, but inside single quotes '' everything is literal except ', and can span several lines
08:51 cprior ah, ok. Will try that first -- thanks!
08:51 tweakism the line breaks will be removed from the string though
08:51 tweakism so it makes a 1-line string
08:51 tweakism which... honestly doesn't make much sense to me but *shrug* yaml
08:52 cberndt joined #salt
08:54 bdrung_work joined #salt
08:56 krymzon joined #salt
08:57 LondonAppDev joined #salt
08:57 chiui joined #salt
09:00 krymzon joined #salt
09:02 bhosmer joined #salt
09:05 iceyao joined #salt
09:07 uros thanks, ntropy .. I figured it out .. skipped shell option and just used #!/usr/bin/env python in script
09:08 iceyao joined #salt
09:11 pooogles joined #salt
09:14 rmnuvg joined #salt
09:19 antpa joined #salt
09:20 ronnix joined #salt
09:23 GreatSnoopy joined #salt
09:23 cberndt joined #salt
09:23 Rumbles joined #salt
09:25 Xevian joined #salt
09:29 keimlink joined #salt
09:30 s_kunk joined #salt
09:35 thalleralexander joined #salt
09:35 anthpa joined #salt
09:36 dmaiocchi joined #salt
09:37 mdupont joined #salt
09:38 iceyao_ joined #salt
09:47 jimcroft joined #salt
09:49 ronnix joined #salt
09:51 bucky joined #salt
09:59 wych joined #salt
10:00 cprior tweakism: I used unicode within "" to mask the troubling >
10:00 cprior http://pastebin.com/gdDn1rUS
10:01 denys joined #salt
10:02 cprior Now I have a base to work out how much wildcarded I can make this package
10:05 is_null hi all, calling salt '*' saltutil.refresh_pillar  creates a job in the background, which means that it can cause race conditions with a state.sls call, how do you suggest to work around that ?
10:07 Sacro joined #salt
10:11 iceyao joined #salt
10:12 colttt joined #salt
10:13 Rkp joined #salt
10:14 ronnix joined #salt
10:14 tweakism cprior: so are you saying that '>=sys-apps/util-linux-2.27.1' didn't work?
10:15 tweakism that sounds suspicious.  I bet you $1 there's a security / shell quoting bug in there
10:16 tweakism cprior: what version of salt?
10:18 laidback joined #salt
10:19 laidback Hello, is it possible to merge defaults.yml with the os_family_map recursively/deep? Apparently default_settings.update does not do the job.
10:22 babilen laidback: That's a common bug in the way most formulas are implemented
10:26 JohnnyRun joined #salt
10:26 babilen You'd need https://github.com/saltstack/salt/blob/develop/salt/utils/dictupdate.py#L20
10:28 is_null we were thinking of using ansible to do a salt-call saltutil.refresh_pillar through ssh on all machines, but perhaps there is a pure-salt solution ?
10:29 babilen salt-ssh ?
10:29 Muchoz joined #salt
10:29 babilen is_null: What's the actual problem you are trying to solve?
10:30 is_null i'd like to run saltutil.refresh_pillar on all machines and then trigger a state.sls command
10:30 babilen laidback: https://docs.saltstack.com/en/develop/ref/modules/all/salt.modules.defaults.html#salt.modules.defaults.merge (cf. https://github.com/saltstack/salt/pull/28235 )
10:30 saltstackbot [#28235]title: Provide a helper function so that merging in jinja in formulas can be… | … similar to merging in pillars.
10:31 babilen is_null: Which event is being raised when the pillar refresh is done?
10:32 is_null babilen: i don't know, but it would be one event per minion, not one when they have all finished ?
10:33 babilen is_null: Well, find out and then use https://docs.saltstack.com/en/latest/ref/states/all/salt.states.saltmod.html#salt.states.saltmod.wait_for_event
10:33 is_null right, but again that would be one event per minion wouldn't it ?
10:33 is_null not one event for when they are all ready
10:33 babilen is_null: I take it that "just wait a little" is not an option as this is in the context of orchestration?
10:34 babilen is_null: Yes, that would be one event per minion (which is why you list all minions that you wait for)
10:34 is_null that's not the kind of engineering i'm expected to do :D
10:34 is_null wait and retry would be ok
10:35 is_null but then again how would i know that *this* refresh_pillar is over and not say a previous one ran by another orchestration process ?
10:35 malinoff joined #salt
10:36 babilen is_null: A refreshed pillar is a refreshed pillar, isn't it?
10:36 is_null well not exactly we have to refresh pillars twice and run mine.update in between
10:36 is_null because we use the mine in pillars
10:36 saltyswede joined #salt
10:36 anmol joined #salt
10:37 babilen So you are essentially trying to implement a recursive pillar compilation?
10:37 babilen (in the sense that it expands non-terminals)
10:37 saltyswede Good morning! I have a question about salt-ssh: Is it possible to authenticate with ssh-key instead of type in a password for every single node when running salt-ssh -i '*' [command] ?
10:38 is_null yeah we should perhaps refresh pillars 3 times with mine.update in between
10:38 babilen is_null: My "normal" approach in this case would be to implement the SLS and then raise an explicit and unique event at the end. The master would then use wait_for_event to synchronise runs on that event, but that is not necessarily an option as "refresh pillars" isn't a state, but a runner.
10:39 MadHatter42 joined #salt
10:39 is_null ok so you'd recommend to implement this as a runner ?
10:39 babilen is_null: That is why I suggested wait_for_event, but in the end you naturally want something explicit that ties "refresh pillar" and "run number 1213234 finished" together
10:39 sfxandy joined #salt
10:41 laidback @babilen: many thanks
10:41 is_null interresting, thanks
10:42 babilen is_null: Sorry, I don't really have a definite solution right now, but those are the "keywords" that pop into my mind ..
10:43 cprior tweakism: salt 2015.8.5 -- and it is not the > sign but only when I used * in the atom's package modifier state.highste fails. Even unicode-ing the asterisk as \u002A didn't help. I get a traceback leading to InvalidAtom http://pastebin.com/idaeNww2
10:44 babilen saltyswede: Sure .. it really just uses SSH in the background so if you have ssh-agent running (or something that provides that functionality), unlocked your key and configured SSH to use that key it should be honored
10:45 cprior tweakism: an hour ago I introduced too many changes at once, onlt now I realize it is not > but *      I will investigate further. Somehow I am sure this is valid and documented Gentoo configuration, although these atom modifiers usually get explained in the context of useflags.
10:46 saltyswede babilen: It seems that it does'nt work. I am currently using an ssh-key to log in to all my servers but when I try to run salt-ssh, it still ask for password
10:47 cprior tweakism: When I want to emerge a package that requires the ~ keyword _and_ then the same for some dependencies, I usually look at the ebuild and try to guesstimate some version scheme that might survive the next update, too. That's how I came to test these * wildcards.
10:47 babilen saltyswede: And "ssh the.same.server" works fine? (as the user you use for salt)
10:48 saltyswede babilen: It does, user and the remote server is the same
10:50 bhosmer joined #salt
10:51 babilen saltyswede: I did not experience that beheaviour, but then I'm on an older version of salt still :-/
10:51 babilen -e
10:51 rubenb Hi. In salt, is it possible to see all connected minions in a fast way?
10:51 babilen rubenb: Is "salt-run manage.up" too slow?
10:51 saltyswede babilen: I am currently running salt-ssh 2015.8.7
10:52 saltyswede I'll do some research and tell you if I find something out
10:52 rubenb babilen: Yes, it seems (I'm checking it out) to connect to each minion.
10:52 babilen saltyswede: Yeah, I'm on 2015.5 .. but I have no idea why they would have changed that behaviour, but simply can't be sure.
10:52 babilen rubenb: How else would it know if a minion is up?
10:52 rubenb babilen: I thought salt works with persistent connections to the minions.
10:53 rubenb I would think each connection not_being_timed_out counts as a connected minion :)
10:53 babilen It might have crashed in the meanwhile
10:54 babilen rubenb: Besides: manage.up is fast .. what's the problem with it?
10:55 rubenb babilen: It takes some seconds here.
10:55 babilen It takes half a second here on a setup with ~800 minions here. How many minions do you have and how many cores does your master have?
10:56 colttt joined #salt
10:56 rubenb 50 minions, 2 cores.
10:56 babilen My master has 12 cores .. you might want to give yours more.
10:58 evle1 joined #salt
10:58 amcorreia joined #salt
10:58 rubenb babilen: Do you connect via the internet to your minions?
10:59 babilen rubenb: These minions are all in the same datacentre, but not necessarily rack on a 10GbE network between them
11:00 babilen Well, bonded 10GbE actually
11:01 saltyswede babilen: Does your master really need so much power?
11:01 babilen saltyswede: No, but I don't pay for cores ...
11:02 babilen It runs on a rather idle box and I though "Well, it uses Python's multiprocessing that just *loves* cores, so why not give it 12?"
11:02 saltyswede yeah lol
11:02 Rumbles joined #salt
11:02 babilen It's probably overkill, but if the salt master isn't busy those cores don't hurt me (and it is not busy most of the times) while it runs faster when I *do* need the power.
11:03 saltyswede Mine runs on a eight core box with too much ram (I had one server that was left over)
11:03 babilen I haven't really done measurements, but there was a definite improvement when I went from 4 cores to 12.
11:03 saltyswede True
11:03 saltyswede Interesting
11:04 babilen I mean in the end salt uses multiprocessing that splits jobs based on the number of cores .. the more cores you have the smaller the respective queues are. Not sure if they play much with chunk size ..
11:05 aub joined #salt
11:05 saltyswede Interesting theory, I'll do some tests also
11:06 aub joined #salt
11:07 rubenb babilen: My minions connect via the internet, that might add some latency.
11:07 rubenb And only 1GbE.
11:08 babilen rubenb: Sure, it would
11:09 rubenb But 30 seconds for 50 minions is quite a lot.
11:09 babilen Agreed
11:09 babilen rubenb: Use manage.present then ..
11:10 ronnix joined #salt
11:12 rubenb babilen: That returns two present hosts.
11:13 babilen rubenb: Is that bad?
11:14 rubenb I would expect them to be all connected. Am looking for a way to monitor the number of minions that are connected.
11:15 babilen rubenb: So the output of manage.present and manage.up differs?
11:15 rubenb Yes. manage.status returns all hosts as up.
11:15 rubenb manage.present shows 2 hosts being up.
11:16 babilen Did you enable presence events?
11:17 babilen https://docs.saltstack.com/en/latest/ref/configuration/master.html#presence-events
11:18 rubenb babilen: No, I did not.
11:18 babilen Okay, that would somewhat explain why manage.present doesn't work ;)
11:18 babilen Could you enable them and try again? (give it a bit of time after enabling it)
11:19 rubenb Done.
11:19 babilen Maybe wait a minute or two and then run manage.present again
11:23 mavhq joined #salt
11:24 babilen rubenb: So, does it work better now?
11:24 rotbeard joined #salt
11:25 rubenb babilen: No, still the same two servers that report being present.
11:25 mavhq joined #salt
11:26 Rumbles joined #salt
11:27 babilen rubenb: Could you restart the minion processes?
11:28 babilen manage.present works perfectly here and is *really* fast, but I've enabled presence events a year ago :)
11:32 rburkholder joined #salt
11:33 rubenb I restarted one minion, which did not show up in the manage.present.
11:34 mortis i enabled presence_events just now, and it works very well
11:34 babilen rubenb: Do you receive presence events? (→ run "salt-run state.event pretty=True")
11:34 mortis 2015.8.3
11:34 babilen mortis: Yeah, it should
11:36 evle1 joined #salt
11:36 Yoda-BZH joined #salt
11:36 Yoda-BZH joined #salt
11:38 laidback hello, i try to resort my formula template and am not sure about the defaults.yml deep merge. Should it happen in the jinja.map or in the e.g. init.sls
11:39 laidback using: https://github.com/saltstack/salt/blob/develop/salt/utils/dictupdate.py#L20
11:39 babilen laidback: map.jinja
11:39 laidback could i do something like {% do salt['defaults.merge'](dest, upd) %} ?
11:39 ujjain joined #salt
11:39 ujjain joined #salt
11:39 babilen laidback: That would be the idea, I guess
11:40 laidback babilen: as i have seen i need the develop branch for that, right?
11:40 jrklein_ joined #salt
11:41 rubenb babilen: I receive them from two servers, as far as I can see.
11:41 rubenb I do receive minion_ping events from other servers
11:41 favadi joined #salt
11:41 Kelsar joined #salt
11:41 linovia joined #salt
11:42 bbhoss joined #salt
11:42 babilen laidback: I'd just place the defaults.py in _modules
11:42 Laogeodritt joined #salt
11:42 babilen rubenb: Anything special about the minions that do not send them?
11:42 laidback ok, i try to get that done. Thanks for help.
11:43 babilen laidback: Don't forget to sync modules ..
11:43 * babilen → lunch
11:43 wiqd joined #salt
11:43 MadHatter42 joined #salt
11:43 copelco joined #salt
11:44 m0nky_ joined #salt
11:45 bhosmer joined #salt
11:45 bstaz joined #salt
11:45 skrobul joined #salt
11:46 [vaelen] joined #salt
11:46 Twiglet joined #salt
11:46 akitada joined #salt
11:46 aboe joined #salt
11:47 rubenb babilen: They are windows minions.
11:47 JonGretar joined #salt
11:47 eodgooch joined #salt
11:52 skarn joined #salt
11:54 Niamkik_ joined #salt
11:54 packeteer joined #salt
11:54 sfz- joined #salt
11:57 froztbyte # salt-run jobs.term_job 20160222135307829872
11:57 froztbyte 'jobs.term_job' is not available.
11:58 froztbyte uh.
11:58 froztbyte did this change?
11:59 froztbyte ah, if I target a minion and use saltutil.term_job, it works
11:59 froztbyte that's a bit unclear in the docs tbh
11:59 froztbyte would a PR clarifying that fly?
12:00 froztbyte and/or extending the jobs runner for it, I guess
12:19 iceyao joined #salt
12:23 felskrone joined #salt
12:30 Rkp joined #salt
12:32 jaybocc2 joined #salt
12:32 joe_n joined #salt
12:34 rotbeard joined #salt
12:34 favadi joined #salt
12:39 bhosmer joined #salt
12:45 MadHatter42 joined #salt
12:48 EvaSDK hello there
12:49 EvaSDK anyone seen anything like this yet ? https://github.com/saltstack/salt/issues/31403
12:49 saltstackbot [#31403]title: Upgrade to 2015.5.9 breaks git ext_pillar | After upgrading from debian jessie-backports ``2015.5.3+ds-1~bpo+1`` to repo.saltstack.com ``2015.5.9+ds-2`` (after #30696 got solved), I cannot apply highstate anymore....
12:52 lompik joined #salt
12:54 denys joined #salt
12:56 AndreasLutro EvaSDK: check your salt master log for errors and add those to the issue as well
12:57 rem5 joined #salt
12:58 babilen :(
13:00 EvaSDK AndreasLutro: should I use level debug ?
13:00 EvaSDK level info adds nothing, the error messages are there at level critical
13:02 bucky joined #salt
13:04 numkem joined #salt
13:05 numkem joined #salt
13:07 AndreasLutro no need, as long as there are log lines relevant to the ext pillar stuff in there
13:09 EvaSDK k
13:12 cyborg-one joined #salt
13:12 josuebrunel joined #salt
13:21 jaybocc2 joined #salt
13:24 rsharma joined #salt
13:25 akhter joined #salt
13:25 ronnix joined #salt
13:27 fredvd joined #salt
13:29 west575 joined #salt
13:33 bhosmer joined #salt
13:33 jimcroft joined #salt
13:34 EvaSDK AndreasLutro: added
13:35 Yoda-BZH joined #salt
13:35 Yoda-BZH joined #salt
13:35 gh34 joined #salt
13:37 shorty_mu joined #salt
13:40 shorty_mu Hi, we just updated from 2014.7.5 to the current 2015.8.7 and get the following error: [salt.loaded.int.returner.local_cache] An extra return was detected from minion salt.my.domain , please verify the minion, this could be a replay attack
13:40 shorty_mu Any idea what this means...?
13:41 shorty_mu Hmm, found https://github.com/saltstack/salt/issues/30146. Which isn't too helpful.
13:41 saltstackbot [#30146]title: salt minion's returner local_cache error of extra return detected from  minion 2015.8.3 on the master log   | salt master is version of 2015.8.3 , minion is 2015.8.3...
13:44 tpaul joined #salt
13:48 colttt joined #salt
13:50 DammitJim joined #salt
13:51 drawsmcgraw joined #salt
13:52 Andre-B joined #salt
13:52 saltyswede I have a question about salt-ssh: Is it possible to authenticate with ssh-key instead of type in a password for every single node when running salt-ssh -i '*' [command] ? Im asking again in case someone know the answer
13:54 colttt joined #salt
13:55 dkrae joined #salt
13:55 akhter joined #salt
13:55 AndreasLutro saltyswede: you can specify private keys in the roster file
13:56 saltyswede with
13:56 babilen (or in your ~/.ssh/config)
13:56 saltyswede priv: path-to-key
13:56 saltyswede wont work,, it's strange thou
13:57 AndreasLutro if it doesn't work you probably did it wrong
13:57 babilen You can also pass "--priv=/path/to/id_file" on the command line. That might make testing easier.
13:57 babilen It would help if you could show us some debug output
13:58 babilen But as I said earlier: It respects your SSH configuration and should just work™ if you use ~/.ssh/config and ssh-agent already.
14:00 ronnix joined #salt
14:01 teryx510 joined #salt
14:02 numkem joined #salt
14:02 subsignal joined #salt
14:03 mapu joined #salt
14:04 bucky joined #salt
14:07 edrocks joined #salt
14:12 saltyswede exit
14:12 babilen So, working now?
14:13 saltyswede ╭─jonathan@7010 ~/.ssh   ╰─➤  salt-ssh server5 --priv=/home/jonathan/.ssh/id_rsa.pub cmd.run 'cat /etc/debian_version' [WARNING ] Failed to open log file, do you have permission to write to /var/log/salt/master? [ERROR   ] An un-handled exception was caught by salt's global exception handler: IOError: [Errno 13] Permission denied: '/var/cache/salt/master/roots/mtime_map' Traceback (most recent call last):   File "/usr/bin/
14:13 babilen http://refheap.com, http://paste.debian.net, https://gist.github.com, http://sprunge.us, …
14:13 anmol joined #salt
14:13 rotbeard joined #salt
14:14 AndreasLutro well that's entirely unrelated to your ssh keys
14:14 tweakism you provided the path to a pubkey, fyi
14:14 saltyswede sorry.. i feel newbie
14:14 saltyswede https://www.refheap.com/115058
14:15 tweakism also it says you have a permissions issue
14:15 saltyswede https://www.refheap.com/115059
14:15 babilen tweakism: You might want to set cachedir, log_file and pki_dir in the master config to a file writable by your user.
14:16 babilen s/tweakism/saltyswede
14:16 saltyswede The files are writable thou
14:17 saltyswede I ise the exact same keyfile to log in to my other servers,
14:17 saltyswede use*
14:18 babilen saltyswede: No, your user can't write to /var/cache/salt/master/
14:19 babilen I typically have a ~/salt/master in which I specify cachedir, log_file and pki_dir to reside in ~/salt/{cached,log/master,pki} respectively
14:19 babilen And then pass it via "-c ~/salt"
14:20 * babilen wonders what happend to his bug to use an XDG directory compliant scheme for that
14:20 saltyswede I have set the permission to 777 on /var/cache/salt recursively, and still got the same message
14:21 saltyswede https://www.refheap.com/115060
14:23 AndreasLutro I trust error messages more than I trust you!
14:24 saltyswede Explain why it can't write to a catalogue with 777 permission then
14:24 shorty_mu left #salt
14:24 AndreasLutro without sshing into your machine, I can only guess: it's probably not 777
14:25 tweakism the thing you pasted didn't show the permissions of /var/cache/salt/master
14:26 saltyswede https://www.refheap.com/115061
14:26 tweakism btw, chmod has a handy trick if you want to set execute bit for only things that are directories or already have execute bit:  chmod -R ugo=rwX /var/cache/salt/master
14:26 tweakism however, you should really *never* make things world-writable
14:27 saltyswede https://www.refheap.com/115062
14:27 toastedpenguin joined #salt
14:28 bhosmer joined #salt
14:28 saltyswede So, have I proved myself enough?
14:28 AndreasLutro no
14:29 dynamicudpate joined #salt
14:29 AndreasLutro there are 2 parts of the path you're getting a permission error on after /var/cache/salt/master, you should check those
14:30 babilen .oO( And why not simply focus on the SSH issue for now, set the aforementioned directories to something in your $HOME (where it belongs IMHO!) and then think about other ways to organise permissions and directories)
14:31 babilen I mean If you keep it in /var/cache you'd mix stuff for all users
14:31 babilen (on that box)
14:31 AndreasLutro ^
14:31 tweakism does he even have an actual ssh issue?
14:31 babilen I don't think so
14:31 tweakism if he would just do that it would work.
14:32 babilen But I'd like to see that :)
14:32 babilen I think the problem is that salt can't write to pki_dir
14:32 saltyswede brb, gotta smack my head a bit..
14:32 * babilen offers a soft(ish) wall
14:32 tweakism well it says directly "Permission denied: '/var/cache/salt/master/roots/mtime_map'" and he never linked permissions there
14:33 tweakism unless he has a diff error and I missed a post
14:33 saltyswede Well, that directory did'nt even exist at all. But I granted full permission on /var/log/salt
14:34 saltyswede I still can't auth with my priv key
14:34 tweakism you're not even getting that far, my friend
14:34 babilen saltyswede: https://docs.saltstack.com/en/latest/ref/configuration/nonroot.html might be of interest
14:34 tweakism just for kicks, please:  ls -ld /var/cache/salt/master{/roots,/roots/mtime_map}
14:35 babilen But then: Just *trust* me and set cachedir, log_file and pki_dir to reside somewhere in your $HOME and be done with it.
14:35 tweakism and after, do as babilen suggests
14:35 babilen You could go for a XDG directory compliant scheme or keep it all in ~/salt/ as mentioned earlier
14:36 saltyswede https://www.refheap.com/115063
14:37 kawa2014 joined #salt
14:38 saltyswede and still:
14:38 saltyswede https://www.refheap.com/115064
14:38 akhter joined #salt
14:38 babilen "no such identity: /home/jonathan/.shh/id_rsa: No such file or directory"
14:39 babilen That's a *very* different issue
14:39 tweakism you wrote .ssh
14:39 tweakism er .shh
14:39 babilen Note .ssh vs .shh
14:39 tweakism heh I can't even type it wrong when I try
14:40 racooper joined #salt
14:40 racooper_ joined #salt
14:40 racooper__ joined #salt
14:41 babilen saltyswede: It also appears as if you changed/tinkered with permissions in the interim as you are no longer getting the errors we've seen before
14:41 saltyswede So even if I run salt-ssh server5 --priv=/home/jonathan/.ssh/id_rsa cmd.run 'cat /etc/debian_version'
14:41 saltyswede I does not override the privkey set in the roster
14:41 tweakism babilen: I haven't used salt-ssh yet; does it connect to root@ by default, or like ssh same as user?
14:41 rickflare does anyone have any experience using salt-cloud and openstack?
14:42 saltyswede anyway, I am not sure if I should do this anymore, or if I am too tired
14:42 saltyswede Thank you for your patience
14:42 babilen tweakism: It behaves *exactly* as SSH does
14:42 tweakism kk
14:42 babilen I typically set a specific user and "sudo: True" in my roster
14:42 aub joined #salt
14:42 tweakism so you do NOPASSWD: for sudo?
14:43 babilen salt-call
14:43 babilen And yeah
14:43 tweakism word.
14:43 saltyswede It works now, the problem was a typo in roster. And here I was blaming everyone lol
14:43 * babilen chuckles
14:43 babilen That is why you paste your entire configuration, commands and their output
14:44 anmol joined #salt
14:44 favadi joined #salt
14:44 saltyswede It works now, the problem was a typo in roster. And here I was blaming everyone lol
14:48 perfectsine joined #salt
14:49 aub joined #salt
14:50 impi joined #salt
14:50 toastedpenguin joined #salt
14:51 numkem joined #salt
14:51 aub joined #salt
14:52 aub joined #salt
14:58 aub joined #salt
14:59 toastedpenguin joined #salt
14:59 quasiben joined #salt
14:59 spiette joined #salt
15:00 sr4f1 joined #salt
15:00 sr4f1 Hi, I was wondering if someone could point to me what I'm doing wrong here: http://pastebin.com/raw/qjf8QMDp
15:00 sr4f1 Why is file.managed /etc/fail2ban/filter.d/asterisk.conf not updated like the rest of the files?
15:02 ssplatt joined #salt
15:02 scoates joined #salt
15:02 buhman sr4f1: because 'grains['os'] != 'CentOS'' was probably true
15:02 buhman sr4f1: so that entire block was skipped
15:03 favadi joined #salt
15:04 sr4f1 buhman: no, grains['os'] is Gentoo
15:04 anthpa joined #salt
15:05 buhman sr4f1: do you understand what 'elif' does?
15:07 sr4f1 if not centos; else if not ubutnu; require file
15:08 buhman so, grain['os'] is not centos
15:08 buhman we go to the first block which is nothing
15:08 buhman now we're done with that entire block
15:08 buhman s|block|if/else|
15:09 buhman I suggest re-thinking how you want to match grains['os']
15:10 sr4f1 understood! thanks
15:10 misconfig joined #salt
15:10 ronnix joined #salt
15:12 zmalone joined #salt
15:12 tracphil joined #salt
15:14 perfectsine joined #salt
15:17 toastedpenguin joined #salt
15:18 favadi joined #salt
15:23 zsoftich4 joined #salt
15:24 hightekvagabond joined #salt
15:26 tonthon left #salt
15:29 ronnix joined #salt
15:31 KennethWilke joined #salt
15:32 jaybocc2 joined #salt
15:32 higuita joined #salt
15:32 higuita {% set isdb=False %} {% if "-db-" in role %} {{  role }} {%   set isdb=True %} {%  endif %} {{ isdb }}
15:33 higuita returns main-db-read False
15:33 higuita i don't understand why this jinja doesn't work, anybody know why?
15:34 higuita the if is applying, it returns the correct role, yet the set is not changing the isdb variable
15:40 hairytoez joined #salt
15:40 Fiber^ joined #salt
15:42 AdamSewell joined #salt
15:45 spuder joined #salt
15:46 AdamSewell I keep getting an error in my master's log IOError: [Errno 21] Is a directory: '/etc/salt/pki/master/minions_pre/minion_name/ - the minion referenced here is an old minion that we no longer have a key for
15:47 andrew_v joined #salt
15:47 ronnix joined #salt
15:48 jaybocc2 joined #salt
15:49 higuita AdamSewell: you can probably remove the '/etc/salt/pki/master/minions_pre/minion_name/ and this should disapear
15:49 nkuttler has somebody here set up jenkins or travis? which one do you think is "easier" to install?
15:49 higuita probably a removal that crashed
15:50 malinoff joined #salt
15:50 AdamSewell higuita, I would if it were there
15:51 higuita on the salt master? do don't have it?
15:51 AdamSewell higuita, yea it's not there on the master
15:52 higuita restart the salt-master service and try again then
15:52 AdamSewell already have sir
15:52 higuita maybe it in his namespace
15:52 higuita then out of ideas :)
15:52 AdamSewell higuita, i guess i should open an issue on github?
15:52 higuita strace it and check why/what is it checking and failing :D
15:53 joelyuk joined #salt
15:54 FreeSpencer joined #salt
15:54 FreeSpencer joined #salt
15:55 mpanetta joined #salt
15:55 Rumbles joined #salt
15:57 thejrose1984 joined #salt
15:57 mpanetta joined #salt
15:59 cpowell joined #salt
15:59 mpanetta joined #salt
15:59 cpowell joined #salt
16:00 srinivas joined #salt
16:02 mpanetta joined #salt
16:02 andrew_v joined #salt
16:05 ronnix joined #salt
16:05 bucky joined #salt
16:06 grumm_servire joined #salt
16:09 ifattah I've Instantiated python client and I'm able to do remote command execution ...however, which argument to salt.client.LocalClient sets the timeout?
16:09 ifattah https://docs.saltstack.com/en/latest/ref/clients/index.html#client-apis
16:10 ifattah client = salt.client.LocalClient()
16:10 ifattah client.cmd(hosts, cmd, timeout=30) ?
16:10 pooogles Any reason salt-bootstrap totally fails on new ubuntu versions?
16:10 gtmanfred pooogles: it should give you an error about where it failed
16:11 gtmanfred can also run it with -D i believe
16:11 fracklen joined #salt
16:11 amcorreia joined #salt
16:12 pooogles Mmm 1 sec checking with a fresh install.
16:15 pooogles Hitting https://github.com/saltstack/salt-bootstrap/issues/686 with 11.09
16:15 saltstackbot [#686]title: Bootstrap error while installing on Ubuntu 14.04 (lxc template) | Seeing an error when trying to install salt via salt bootstrap within an lxc container...
16:16 bhosmer joined #salt
16:19 pooogles Heh, seemingly you need to use the development branch for it to install on Wily.
16:20 akhter joined #salt
16:22 gtmanfred oh yeah
16:24 ronnix joined #salt
16:25 DammitJim joined #salt
16:25 ALLmightySPIFF joined #salt
16:30 tligda joined #salt
16:32 catpig joined #salt
16:33 jbrnds2 joined #salt
16:34 aharvey joined #salt
16:34 cpowell joined #salt
16:37 kevinquinnyo1 joined #salt
16:37 srinivas joined #salt
16:43 ronp_usa joined #salt
16:44 numkem joined #salt
16:44 ronnix_ joined #salt
16:46 numkem joined #salt
16:48 beneggett joined #salt
16:49 numkem joined #salt
16:49 beneggett Syntax question: I have a conditional like this
16:49 beneggett {%- if "api"  in grains.get('roles', []) %}
16:49 beneggett ...
16:49 beneggett in a file, how can i do multiple matchers
16:49 beneggett something like
16:49 beneggett {%- if "api" OR "web"  in grains.get('roles', []) %}
16:50 beneggett to run it on either of those grains roles conditions
16:50 numkem joined #salt
16:51 numkem joined #salt
16:51 teryx510 joined #salt
16:51 Andre-B joined #salt
16:53 fracklen joined #salt
16:55 andrew_v joined #salt
16:57 pprkut I'm having troubles with git ext_pillar configuration. I have multiple git repos, all of them are synced by the master, but only one of them ends up in the rendered pillar data
16:57 pprkut anyone an idea what could be wrong?
16:59 beardedeagle joined #salt
17:02 rm_jorge joined #salt
17:03 impi joined #salt
17:04 spuder_ joined #salt
17:07 bucky joined #salt
17:10 bhosmer joined #salt
17:12 chitown_ joined #salt
17:12 andrew_v joined #salt
17:12 pprkut is every ext_pillar repo supposed to have a top.sls file?
17:13 bluenemo joined #salt
17:14 cpowell joined #salt
17:17 ChloeD joined #salt
17:18 aharvey joined #salt
17:20 writtenoff joined #salt
17:20 ALLmightySPIFF joined #salt
17:20 pprkut ok, so it seems if I have multiple repos, each with a top.sls file, they overwrite each other. But I need to have them, since the top.sls in the filesystem pillar can't include ext_pillar files
17:20 pprkut how am I supposed to fix that?
17:23 bash1235123 joined #salt
17:23 bash1235123 How do I manage the hosts file of the syndic from the syndics master which should have the minions that are connected to the syndic ? :)
17:28 joelyuk_ joined #salt
17:29 andrew_v joined #salt
17:29 ry joined #salt
17:29 DammitJim if I want to configure my users with salt for mysql
17:29 Eugene Simple: don't use a hosts file, use a protocol designed for name resolution and dynamic updates. Like DNS!
17:29 DammitJim how do I get the hash for the password?
17:29 liskl joined #salt
17:29 DammitJim DNS FTW until DNS goes down FTL
17:29 DammitJim :D
17:30 Eugene If DNS is down then you need better DNS
17:30 Eugene DammitJim - I believe it just uses mysql's hash format. http://dev.mysql.com/doc/refman/5.7/en/password-hashing.html
17:30 pprkut DammitJim: execute "SELECT PASSWORD('My new password');" on a mysql server
17:31 cpowell joined #salt
17:32 akhter joined #salt
17:32 alwaysatthenoc joined #salt
17:34 DammitJim oh wow
17:35 DammitJim hhhmmmmm
17:35 DammitJim so, I would manually do that and then set the password hash in salt?
17:35 pprkut yes
17:38 cpowell joined #salt
17:38 bucky joined #salt
17:38 bash1235123 Eugene: you trying to be a smart ass ?
17:39 bucky joined #salt
17:39 Eugene No; you are. I'm presenting a time-tested solution that works.
17:40 andrew_v joined #salt
17:41 aub joined #salt
17:41 cpowell joined #salt
17:42 mpanetta joined #salt
17:42 brianfeister joined #salt
17:46 baweaver joined #salt
17:46 pprkut turns out, my issue was a bug. Updated to 2015.8.7 and now everything works as expected :)
17:49 fracklen joined #salt
17:51 tiadobatima joined #salt
17:51 shaggy_surfer joined #salt
17:51 anmol joined #salt
17:53 rem5 joined #salt
17:55 subsigna_ joined #salt
17:55 shaggy_surfer joined #salt
17:58 jfindlay pprkut: does your issue look (exactly) like https://github.com/saltstack/salt/issues/31403?
17:58 saltstackbot [#31403]title: Upgrade to 2015.5.9 breaks git ext_pillar | After upgrading from debian jessie-backports ``2015.5.3+ds-1~bpo+1`` to repo.saltstack.com ``2015.5.9+ds-2`` (after #30696 got solved), I cannot apply highstate anymore....
17:59 jfindlay just realized it's 2015.9 on there
17:59 jfindlay 2015.5.9
17:59 * jfindlay can't IRC
17:59 jaybocc2 joined #salt
18:03 andrew_v joined #salt
18:03 mpanetta joined #salt
18:04 bhosmer joined #salt
18:05 amcorreia joined #salt
18:05 zzzirk joined #salt
18:06 dyasny joined #salt
18:11 hightekvagabond joined #salt
18:11 shaggy_surfer joined #salt
18:12 ronnix joined #salt
18:13 Heartsbane IRC is over rated
18:14 * Heartsbane blames robawt.
18:15 jaybocc2 joined #salt
18:15 denys joined #salt
18:21 robawt haha
18:21 cpowell joined #salt
18:22 ajw0100 joined #salt
18:23 ageorgop joined #salt
18:24 numkem joined #salt
18:25 numkem joined #salt
18:26 numkem joined #salt
18:27 numkem joined #salt
18:29 rhand joined #salt
18:30 fxhp joined #salt
18:30 cpowell joined #salt
18:30 andrew_v joined #salt
18:33 rem5 joined #salt
18:40 wendall911 joined #salt
18:40 The_Loeki joined #salt
18:41 The_Loeki hi all! Does anybody know the current 'canonical' way to make one module depend on the availability of another?
18:43 bucky_ joined #salt
18:43 fracklen joined #salt
18:43 pprkut jfindlay: nope, not that bug. For me not even pillar.ls was showing the data
18:43 buhman The_Loeki: ImportError would cause that module to not be available
18:44 josuebrunel joined #salt
18:44 dynamicudpate joined #salt
18:44 bucky_ joined #salt
18:45 bucky_ joined #salt
18:45 bucky_ joined #salt
18:46 bucky_ joined #salt
18:47 akhter joined #salt
18:48 totte left #salt
18:48 ajw0100 joined #salt
18:48 totte joined #salt
18:49 bucky_ joined #salt
18:49 bucky_ joined #salt
18:51 The_Loeki @buhman: Perhaps, but that won't take __virtual__() into accoun
18:51 DammitJim how do I get pillar data from a list of paired items?
18:56 aharvey joined #salt
18:57 kahiru joined #salt
18:59 bhosmer joined #salt
18:59 mapu joined #salt
18:59 bucky_ joined #salt
19:00 bucky_ joined #salt
19:01 spuder joined #salt
19:01 bucky_ joined #salt
19:02 mapu joined #salt
19:02 bucky_ joined #salt
19:04 bucky_ joined #salt
19:04 bucky_ joined #salt
19:05 mpanetta joined #salt
19:08 ALLmightySPIFF joined #salt
19:08 GreatSnoopy joined #salt
19:12 bucky_ joined #salt
19:13 bucky_ joined #salt
19:13 The_Loeki @jfindlay: Any pointers here? I'll make it well worth your while ;)
19:13 bucky_ joined #salt
19:17 baweaver joined #salt
19:18 bucky_ joined #salt
19:21 eodgooch joined #salt
19:23 _JZ_ joined #salt
19:23 cpowell joined #salt
19:23 hightekvagabond joined #salt
19:25 penguin_dan joined #salt
19:26 ekristen joined #salt
19:31 chiui joined #salt
19:32 fracklen joined #salt
19:32 DammitJim what do I need to do on salt to be able to edit users?
19:34 onlyanegg joined #salt
19:36 ajw0100 joined #salt
19:36 jhauser joined #salt
19:42 dlam joined #salt
19:43 bucky_ joined #salt
19:43 notnotpeter joined #salt
19:43 bash1235123 joined #salt
19:45 bucky_ joined #salt
19:45 numkem joined #salt
19:46 jaybocc2 joined #salt
19:46 numkem joined #salt
19:47 mapu joined #salt
19:47 numkem joined #salt
19:48 numkem joined #salt
19:49 numkem joined #salt
19:49 XenophonF DammitJim: you mean, like unix user accounts?
19:49 forrest joined #salt
19:49 cberndt joined #salt
19:50 DammitJim XenophonF, I found a great doc about how to set up salt for managing users in mysql
19:50 XenophonF DammitJim: you really want to check out https://github.com/saltstack-formulas/mysql-formula
19:50 numkem joined #salt
19:51 XenophonF i provision from the salt master because my mysql instances are all in amazon rds
19:51 XenophonF otherwise, i'd feed the root credentials to the minion running on the database server itself
19:51 DammitJim XenophonF, how do you configure the password for the mysql user to manage through salt?
19:52 XenophonF well, like i said, i use rds, so the admin password gets set when the rds instance gets deployed
19:52 cliluw joined #salt
19:52 acclsw joined #salt
19:52 DammitJim the admin password for mysql?
19:52 DammitJim so, where do you store it?
19:53 XenophonF in pillar
19:53 mpanetta joined #salt
19:53 bhosmer joined #salt
19:53 DammitJim in pillar in clear text?
19:53 XenophonF mysql-formula looks for a pillar key named mysql:server
19:53 DammitJim I'm already freaking out setting the hash password in pillar
19:53 XenophonF ah
19:53 XenophonF can't help you there
19:53 cpowell joined #salt
19:53 XenophonF it's stored in the clear
19:53 DammitJim no problem
19:54 DammitJim just trying to gauge the risk
19:54 DammitJim so, I better start making some changes to this salt master
19:55 akhter joined #salt
19:55 xfs joined #salt
19:57 asmaps joined #salt
19:59 asmaps hey, can somebody tell me what's the best way to install salt 2015.8.x on ubuntu wily (15.10)? via apt I only get 2015.5
19:59 rudi_s left #salt
20:00 fracklen joined #salt
20:02 mpanetta joined #salt
20:02 ALLmightySPIFF joined #salt
20:03 DammitJim joined #salt
20:07 deniszh joined #salt
20:08 andrew_v joined #salt
20:08 sfxandy joined #salt
20:09 aharvey joined #salt
20:09 akhter joined #salt
20:10 fullstop asmaps: doesn't the salt website have instructions where the deb file is pulled from salt's servers instead of ubuntu?
20:11 asmaps fullstop: I couldn't find any newer debs than 2015.5
20:11 asmaps on the instructions website it says to use the ppa, but there's also just the 2015.5.3
20:12 fracklen joined #salt
20:12 gtmanfred asmaps: use repo.saltstack.com not the ppa
20:13 gtmanfred http://repo.saltstack.com/#ubuntu
20:13 gtmanfred http://repo.saltstack.com/apt/ubuntu/14.04/amd64/
20:13 gtmanfred i don't see anything in there for wily though, so
20:13 gtmanfred ¯\(°_o)/¯
20:13 fullstop he wants to use a non-lts
20:13 gtmanfred yeah, the ppa isn't getting updated anymore
20:14 babilen Why would storing data in the pillar be risky?
20:14 gtmanfred and it doesn't look like they really support non lts in there
20:14 gtmanfred should ask @jfindley if he shows up today
20:14 gtmanfred jfindlay: ^^
20:14 babilen (re DammitJim's problem earlier)
20:14 gtmanfred that
20:14 * gtmanfred has gotten too used to @ in slack
20:18 asmaps gtmanfred: ty, I used http://repo.saltstack.com/apt/ubuntu/latest/ it seems to work
20:20 Muchoz joined #salt
20:20 garphy joined #salt
20:20 mrMute joined #salt
20:20 garthk joined #salt
20:20 Mate joined #salt
20:20 Mate joined #salt
20:20 Karunamon joined #salt
20:20 lionel joined #salt
20:20 twiedenbein joined #salt
20:20 leev joined #salt
20:20 Bucciarati joined #salt
20:20 mortis joined #salt
20:20 futuredale joined #salt
20:20 Ashald joined #salt
20:20 adrienr joined #salt
20:20 Hetman joined #salt
20:20 pprkut joined #salt
20:20 hlub joined #salt
20:20 asmaps at least salt --version says 2015.8.7 now
20:20 stbenjam joined #salt
20:20 elektrix joined #salt
20:20 LostSoul joined #salt
20:20 foundatron joined #salt
20:20 fhh joined #salt
20:20 LotR joined #salt
20:20 koomi joined #salt
20:20 gmoro joined #salt
20:20 pezus joined #salt
20:20 stooj joined #salt
20:20 zsoftich1 joined #salt
20:20 Pixionus joined #salt
20:20 beardo_ joined #salt
20:20 rsFF joined #salt
20:20 bharper joined #salt
20:20 TomJepp joined #salt
20:20 ws2k3_ joined #salt
20:20 dandelo joined #salt
20:20 MikaT joined #salt
20:20 tru_tru joined #salt
20:20 jnilsson joined #salt
20:20 Pixionus joined #salt
20:21 kidneb joined #salt
20:21 spaceSub joined #salt
20:21 skullone_ joined #salt
20:21 davidbanham joined #salt
20:21 nomad_fr joined #salt
20:21 gnord joined #salt
20:21 gadams joined #salt
20:21 harkx joined #salt
20:21 MajObviousman joined #salt
20:21 Ssquidly joined #salt
20:21 darix joined #salt
20:21 dmaiocchi joined #salt
20:21 Hydrosine joined #salt
20:21 artemz joined #salt
20:21 dabb joined #salt
20:21 NaPs joined #salt
20:21 eichiro joined #salt
20:21 keekz joined #salt
20:21 hypnosb joined #salt
20:21 lempa joined #salt
20:21 okfine joined #salt
20:21 debian112 joined #salt
20:21 borgstrom joined #salt
20:21 aurynn joined #salt
20:21 colinstgeorge joined #salt
20:21 LeProvokateur joined #salt
20:21 tmmt joined #salt
20:22 ajw0100 joined #salt
20:22 hax404 joined #salt
20:22 Hazelesque joined #salt
20:22 gtmanfred be aware that latest is 14.04
20:22 devtea joined #salt
20:23 andrew_v joined #salt
20:23 gtmanfred so versions of stuff might not be 100% correct
20:23 dober joined #salt
20:23 dober joined #salt
20:25 daschatten joined #salt
20:25 Edur joined #salt
20:25 Azid joined #salt
20:25 roock joined #salt
20:25 upb joined #salt
20:25 rofl____ joined #salt
20:25 bastiandg joined #salt
20:25 rnts joined #salt
20:25 thehaven joined #salt
20:26 KyleG joined #salt
20:26 KyleG joined #salt
20:28 hightekvagabond joined #salt
20:29 darix joined #salt
20:29 MindDrive joined #salt
20:34 baweaver joined #salt
20:36 baweaver joined #salt
20:39 akhter joined #salt
20:41 baweaver joined #salt
20:41 fracklen joined #salt
20:47 DammitJim joined #salt
20:47 bhosmer_ joined #salt
20:51 Bloodshot joined #salt
20:53 Bloodshot I can't get salt reactors working. I'm using salt-master 2014.1.13. salt-master -l debug does not mention anything about reactors or rendering the sls file.
20:53 Bloodshot My /etc/salt/master, excluding comments and empty lines: http://pastebin.com/raw/1FUm4eTM
20:54 bucky_ joined #salt
20:55 bucky_ joined #salt
20:59 Ahlee joined #salt
20:59 Ahlee joined #salt
21:00 fracklen joined #salt
21:02 aharvey joined #salt
21:03 Muchoz joined #salt
21:04 mpanetta joined #salt
21:05 jfindlay the traditional way to predicate the loading of one module on another is to check if one of the other's functions has already been loaded: https://docs.saltstack.com/en/latest/ref/modules/index.html#returning-error-information-from-virtual
21:05 shaggy_surfer joined #salt
21:06 jfindlay salt support on ubuntu 15.10 is not 'official'.  We will likely have official packages for 16.04 though
21:13 mpanetta joined #salt
21:14 zsoftich4 joined #salt
21:21 bucky_ joined #salt
21:21 bucky_ joined #salt
21:21 hax404 joined #salt
21:24 hlub joined #salt
21:27 workthrick joined #salt
21:28 akhter joined #salt
21:30 keimlink joined #salt
21:30 Bloodshot Is 2014.1.13 too old? Have there been changes to the way reactors are specified in /etc/salt/master?
21:31 zmalone Briefly looking at the reactor docs, it sounds like changes happened in 2014.7, but I don't know if they effect you.  It's probably a good idea to move to a more recent release either way
21:32 gtmanfred highly recommend moving to 2015.5
21:32 Bloodshot I'll copy my original question
21:32 gtmanfred 2014.7 is out of support in May i think
21:32 Bloodshot I can't get salt reactors working. I'm using salt-master 2014.1.13. salt-master -l debug does not mention anything about reactors or rendering the sls file.
21:32 Bloodshot My /etc/salt/master, excluding comments and empty lines: http://pastebin.com/raw/1FUm4eTM
21:33 Bloodshot I'm using salt from the Debian repositories.
21:33 Bloodshot When Stretch releases as Debian stable, and we migrate, we'll be using a newer version
21:34 Bloodshot That is, unless I can't get reactors to work with this version at all
21:34 deniszh joined #salt
21:34 bluefox65 joined #salt
21:34 gtmanfred https://repo.saltstack.com
21:35 zmalone There are security concerns up through the most recent versions of Salt, I'm not sure I'd consider it stable enough to use the same version for multiple years.
21:35 bluefox65 hi, is there a howto to install minions without connection to the Internet. E.g. can I replicate the repository to an internal one ?
21:36 fracklen joined #salt
21:37 ageorgop1 joined #salt
21:39 gtmanfred bluefox65: uhhhh, wget -R ?
21:39 gtmanfred or... --recursive
21:40 gtmanfred ahh, -r
21:40 tweakism you're better off to use a tool intended for mirroring repos
21:40 gtmanfred sure
21:40 tweakism there are several available, for both apt and yum/rpm
21:41 gtmanfred i would still contend that wget --recursive is the fastest to do it with :P
21:41 gtmanfred maybe not the most correct
21:41 hightekvagabond joined #salt
21:41 bhosmer joined #salt
21:41 tweakism the tool can avoid grabbing crap like directory listings, and you needn't be careful to tell it not to follow off-site links
21:42 tweakism and it can check that the packages are all consistent w/ each other
21:42 gtmanfred yar
21:42 forrest tweakism, I'm pretty sure gtmanfred was just being a smart ass
21:42 gtmanfred probably
21:42 zmalone I think you'll be better off just wgetting repo.saltstack.com
21:42 tweakism 'cause you don't want to mirror a repo that someone is in the process of uploading new versions to
21:42 fracklen joined #salt
21:42 zmalone people who have tried to mirror it with normal repo management tools usually complain that there are enough broken metadata files that it doesn't work
21:43 tweakism that doesn't surprise me at all.
21:43 zmalone I don't think those github issues have been addressed yet
21:43 bluefox65 thanks for the hints.
21:43 forrest zmalone, I would just pull the latest release and drop it on my internal package server. If you're using internal systems like that the chance you don't need other packages is almost 0
21:45 kermit joined #salt
21:47 krymzon joined #salt
21:50 DammitJim so, what command will print the states that are going to run for a minion?
21:53 deniszh1 joined #salt
21:53 ageorgop joined #salt
21:53 kahiru doesn't state.show_highstate targetted on that minion do something like that?
21:59 tpaul Is there an ETA for Boron?  I just started to write a custom execution module and realized the feature I need is already in Boron.  So good news overall :)
22:01 DammitJim oh, I keep forgetting show_highstate!
22:01 DammitJim thanks kahiru
22:02 edrocks joined #salt
22:02 fracklen joined #salt
22:07 gtmanfred tpaul: kind of? it is tagged as 2016.3, so, unless 2015.2 happens again, which it sounds like it wont
22:07 gtmanfred should be earlyish march
22:08 gtmanfred tpaul: but as with all opensource... the true answer is `when it is ready`
22:08 ALLmightySPIFF joined #salt
22:11 tpaul yep makes sense
22:11 gtmanfred :)
22:11 gtmanfred but yeah, i know they are planning on early march
22:11 gimpy2938 joined #salt
22:11 i90rr joined #salt
22:14 deniszh joined #salt
22:14 onlyanegg joined #salt
22:15 gimpy2938 Odd issue on 2015.8.5, pkgrepo.managed fails saying "State 'pkgrepo.managed' was not found in SLS ..."  Any clue what could cause this?  https://gist.github.com/jwhite530/e0d3ae5e0ca296e2ee03
22:21 iggy gimpy2938: try salt 'node' sys.doc pkg.mod_repo
22:22 gimpy2938 iggy: didn't seem to do anything and exited with status 0
22:23 iggy that's your problem
22:23 baweaver joined #salt
22:23 iggy if pkg.mod_repo isn't available pkgrepo won't load
22:23 mpanetta joined #salt
22:24 ageorgop joined #salt
22:25 gimpy2938 iggy: shouldn't that be in the default install?  This was installed via Saltstack's hosted repo for CentOS 7
22:25 bucky_ joined #salt
22:25 bucky_ joined #salt
22:26 iggy gimpy2938: try salt 'node' sys.doc pkg.latest_version
22:27 iggy or maybe sys.doc pkg
22:27 gimpy2938 iggy: same result ... also updated to 2015.8.7-1
22:28 iggy same result for sys.doc pkg?
22:29 onlyanegg joined #salt
22:29 gimpy2938 iggy: yup, though it seems to work with another minion
22:30 iggy I'd say that server has a local problem
22:30 ronp_usa can someone point me to a good doc on firing up a new minion that'll be connected into the master? (essentiall create minion)
22:33 gimpy2938 iggy: Sounds like a problem with the RPMs ... either it didn't pull in a dependency it needed or it failed to install correctly in the first place (I did install via yum's chroot feature - into a root fs image)
22:36 bhosmer_ joined #salt
22:37 fracklen joined #salt
22:38 Muchoz joined #salt
22:38 bucky_ joined #salt
22:39 bucky_ joined #salt
22:40 gimpy2938 iggy: this is confusing, things look correct as far as I can tell ... salt and salt-minion packages are installed, what else should I be looking for?
22:40 bucky_ joined #salt
22:40 krymzon joined #salt
22:41 bucky_ joined #salt
22:43 zenlot1 joined #salt
22:51 iggy I don't use rpm based distros, so I'm not sure how much help I'll be
22:52 iggy but if you don't have a pkg module loading, I'd say start over on the system
22:52 gimpy2938 iggy: Yea, this is definately a problem with the RPMs being provided by SaltStack ... I completely wiped the packages from the system and added them back, same behaviour
22:53 gimpy2938 This machine is fairly "light" so I suspect there's a dependency the RPMs are not pulling in ... if I can find what file it needs it should be easy to find the package and create a bug to have th eRPMs fixed
22:54 gimpy2938 for the pkg.latest_version and stuff you mentioned, what file would that be in?  Other than poking around code idk what I can do
22:56 Muchoz joined #salt
22:56 tampakrap joined #salt
22:58 stooj joined #salt
23:01 cpowell joined #salt
23:02 west575 joined #salt
23:04 andrew_v joined #salt
23:05 gimpy2938 does "pkg.mod_repo" mean I should have a pkg.py with a mod_repo() in it?
23:07 onlyanegg joined #salt
23:07 tampakrap joined #salt
23:08 ALLmightySPIFF joined #salt
23:09 fracklen joined #salt
23:12 jfindlay gimpy2938: depending on your platform, it means that you should have `yumpkg.mod_repo` or `aptpkg.mod_repo` as appropriate
23:13 hightekvagabond joined #salt
23:14 gimpy2938 @jfindlay: I do have a yumpkg.py but `salt 'sn2' sys.doc yumpkg.mod_repo` gives the same results, which is no results ... I don't get it, Salt sees to have its stuff but doesn't give a usefull error
23:16 jfindlay gimpy2938: do you see anything in the minion debug log?
23:20 gimpy2938 @jfindlay: I see "Could not LazyLoad pkgrepo.managed" but nott sure why ... here is debug output: https://gist.github.com/jwhite530/c2b5c12794bb4c92e12b
23:23 onlyanegg joined #salt
23:23 jfindlay gimpy2938: and does `salt 'node' sys.doc pkg` return anything?
23:24 jfindlay or rather `salt-call sys.doc pkg`
23:24 gimpy2938 @jfindlay: Nope, mast gives no output and salt-call gives empty output
23:28 jfindlay the problem is that the yumpkg.py module is not loading, so you should verify that it's installed and that the conditions here are met: https://github.com/saltstack/salt/blob/v2015.8.7/salt/modules/yumpkg.py#L63-L79
23:28 amcorreia joined #salt
23:29 _JZ__ joined #salt
23:30 bhosmer joined #salt
23:32 gimpy2938 @jfindlay: think I found it ... the "os" grain is being set to "dracut" instead of "CentOS" ,,, any clue in which what program grains are found and set?
23:37 Aleks3Y joined #salt
23:37 jfindlay gimpy2938: would you mind filing a bug on that?
23:38 gimpy2938 @jfindlay: yea, poking around code (guessing this is in os_data() in salt/grains/core.py) at the moment to see if I can make a patch
23:39 jfindlay right
23:39 jfindlay thanks
23:44 joe_n joined #salt
23:44 sfxandy joined #salt
23:45 ash1 joined #salt
23:46 ash1 left #salt
23:46 gimpy2938 @jfindlay: Not sure if this is a Salt bug or CentOS bug ... found dumb garbage in /etc/os-release which CentOS shouldn't be shipping
23:46 bucky joined #salt
23:47 gimpy2938 ... including 'NAME="dracut"' like Salt says
23:48 babilen Can't really blame salt then
23:48 gimpy2938 I can still blame salt for being extremely unhelpful and confusing though
23:49 tweakism gimpy2938: do you have centos-release installed?
23:49 tweakism or whatever they call it
23:50 eodgooch joined #salt
23:50 gimpy2938 tweakism: I do
23:50 __number5__ what are my best choice to manage a linux box installed on client site which normally have access to Internet but behind their firewall?
23:50 ageorgop joined #salt
23:51 __number5__ the master will be in an AWS VPC
23:51 tweakism __number5__: are the hesistant to forward ports for you?
23:51 __number5__ tweakism: yep, the client reluctant to do ports forward or similar
23:52 tweakism __number5__: get at least the ability to ssh to it. even better, get them to open something you can tunnel over, like openvpn or ipsec, and have full access to it that way.  or join their VPN yourself.  but if you absolutely had to, you can connect minion/master via ssh port forwarding (i.e., w/o having to use salt-ssh.  which is yet another option.)
23:52 ash1 joined #salt
23:53 tweakism oh, actually, you don't need incoming ports for salt minion anyway
23:53 tweakism just outbound
23:53 tweakism I'm sure you still want SSH
23:53 keimlink joined #salt
23:53 gimpy2938 yea, I would be pretty hesitant to have a Salt minion listening out on the Internet
23:53 __number5__ minion connect to master tunnelling via ssh?
23:54 tweakism well salt minions don't listen, masters do.
23:54 tweakism but yes, don't run a master listening on the internet anyway.
23:54 tweakism __number5__: yeah, it's a very handy feature of SSH
23:54 tweakism __number5__: salt-ssh may be more convenient for you
23:54 __number5__ hmm, I think I can setup vpn client on my box, vpn to the master vpc
23:55 tweakism __number5__: but if there's some killer feature you need from the master, or in general it's super useful to know for other stuff:  man ssh, look at -L and -R
23:55 tweakism __number5__: true enough also.
23:56 __number5__ tweakism: for salt-ssh, I need to able to directly ssh to my minion right?
23:56 tweakism yeah
23:56 tweakism but if you don't have that anyway, it will suck
23:57 shaggy_surfer joined #salt

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