Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-02-19

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

All times shown according to UTC.

Time Nick Message
00:00 UtahDave ah, yeah.  would you mind commenting on that regarding cp.push?  I think that's a slightly different problem than the other cp functions
00:03 mbrgm I added a quick comment.
00:03 ronrib joined #salt
00:04 ajw0100 joined #salt
00:05 UtahDave thanks!
00:05 eliasp joined #salt
00:11 krymzon joined #salt
00:18 ajw0100 joined #salt
00:29 ajw0100 joined #salt
00:35 saltycharles joined #salt
00:35 urda so is multi-master mode failover supposed to be super buggy? Following the PKI instructions I can't seem to get my minions to fail over correctly at all. Removing the failover mode and letting minions connect to ALL masters at once allows all masters to issue commands.
00:37 drawsmcgraw left #salt
00:40 kevinquinnyo1 joined #salt
00:41 djgerm1 anybody use salt on F5 GTMs?
00:43 iggy basepi: /topic and 2015.8.7 release notes say "upgrade to 2015.8.4 as soon as possible"
00:43 forresta iggy, basepi is no longer a channel mod
00:43 forresta iggy, ping jfindlay, he should be able to update.
00:44 iggy whoever
00:45 iggy this is why there needs to be a "community liaison"
00:47 iggy jfindlay: and if you could kindly have whoever is doing releases now update their how to do a release doc to include changing the /topic here, that'd be great
00:47 iggy or just stop putting the version in the /topic
00:47 iggy either way
00:48 forrest Just change it to say 'UPGRADE OR DIE' then the rest of the message.
00:50 jfelchner joined #salt
00:54 Muchoz Does anyone know how I could do an unless directory is a symlink?
00:57 ingslovak joined #salt
00:58 Muchoz Ah, I can do it with ls -l | grep '^l.*nameofdir'
00:58 iceyao joined #salt
00:59 tweakism Muchoz: you probably want -d flag to ls
01:00 bucky joined #salt
01:00 krymzon joined #salt
01:01 Muchoz tweakism, good idea
01:01 Muchoz Thank you
01:01 whytewolf [ -h <file> ]
01:01 tweakism even better
01:02 whytewolf "not teted but should work as unless and and onlyif are only looking for exit output"
01:03 Muchoz whytewolf, tweakism, -h is human readable. Why does that help?
01:03 tweakism Muchoz: `[ -h $filepath]` is the whole command
01:03 tweakism Muchoz: see man test
01:04 tweakism whytewolf: if that ends up bring a problem, you could use "/usr/bin/test -h $filepath"
01:05 tweakism Muchoz: historically [ is a symlink to test (and nowadays both are shell built-ins)
01:05 Muchoz Well now you're talking about test, not ls
01:05 whytewolf Muchoz: yes,
01:05 Muchoz You didn't mention that so I was confused
01:05 Muchoz But ya, test -h does make sense
01:06 whytewolf sorry what i typed was meant to be the whole comamnd.
01:06 tweakism it confused me too for a moment.
01:06 Muchoz And that test works in salt's unless? Cause I don't see any output
01:07 whytewolf Muchoz: test outputs as exit codes
01:07 Muchoz Makes sense
01:07 whytewolf Muchoz: once you run a test use echo $?
01:07 rotbeard joined #salt
01:07 whytewolf which is all unless is really looking for
01:07 Muchoz whytewolf, yup that checks out
01:08 Muchoz I keep forgetting about $? and other combinations
01:08 iceyao joined #salt
01:08 whytewolf I can never get it out of my head. to many years twiddleing ont he CLI as an SA
01:09 Muchoz I have hard reset my repo twice now
01:09 Muchoz Because I had a symlink to my vagrant shared folder
01:09 Muchoz Don't judge my stupidity
01:10 whytewolf lol. I'm not judging. I had a boss once forget the different between dump and drop in a sysbase database. lot a newspaperw NEWS database
01:11 mdupont joined #salt
01:11 mdupont_ joined #salt
01:12 whytewolf thats when you thank your lucky stars for backups
01:13 conan_the_destro joined #salt
01:18 Muchoz unless test -h {{ path }} doesn't seem to work
01:18 Muchoz Oh no no
01:18 Muchoz No
01:18 whytewolf ?
01:18 Muchoz Ignore that, I know what's the problem
01:18 bujordan joined #salt
01:19 andrej joined #salt
01:19 andrej salt prodmon\* mine.get oob-\* grains.items
01:20 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
01:20 andrej this gives me a list of all grains in the mine
01:20 jfindlay iggy: sorry, I was commuting home
01:20 andrej how do I get a singular grain, e.g. os ?
01:20 whytewolf andrej: have a mine that just grabs that grain
01:20 jfindlay iggy: still finishing up the 2015.8.7 release paperwork
01:20 whytewolf or parse the whole tree
01:21 whytewolf your pick
01:21 andrej whytewolf : there's no way to just get a single grain? :/
01:21 iceyao joined #salt
01:21 tweakism whytewolf: my favorite personal story is the time I made a "read-only" data gathering script and ran it on hundreds of HP-UX boxes... thus setting all of their hostnames to "-s".
01:21 tweakism wasn't particularly destructive, but at the time I was prideing myself on my POSIX-compatible sh skillz
01:22 andrej lol @ tweakism
01:22 whytewolf andrej: like i said create a mine that isn't grains.items. but is grains.get [mines work on a function level and they run that function on the minion then send the data to the master]
01:22 whytewolf ohhh. HP-UX lovely little beasty that is.
01:22 whytewolf was it in secure mode?
01:23 andrej aight ... looping and parsing it is. Some of the mininons are currently turned off, so changing the way the mine is populated is not a good idea
01:23 andrej thanks for the input
01:23 whytewolf s/was it/were they
01:23 whytewolf andrej: sorry i didn't have better news for you
01:24 tweakism whytewolf: I dunno, I didn't know that was a thing.  I don't really know anything about HP-UX, it was just the legacy we were replacing w/ Linux.
01:25 whytewolf ahh. at the same paper we were running solaris, HP-UX [in secure mode] and Linux. HP-Ux was the nightmare system of the three
01:25 whytewolf but it also handled the money and paper routes
01:26 tweakism that customer did use a MAC tool that was a constant source of frustration
01:26 tweakism from CA... where old software goes to die, or rather, live on in an un-dead state and terrorize sysadmins.
01:27 whytewolf lol
01:27 whytewolf ahh CA. I have yet to meet a piece of software i like from them
01:28 whytewolf when i first started at IGT we used their cloud software. even the designers of it constently bagged on it
01:28 tweakism they make their own software?
01:28 tweakism I thought they just inherited other people's legacy products.
01:28 Muchoz whytewolf and tweakism thank you btw. Everything works great.
01:28 whytewolf oh no, they bought it from someone else
01:29 whytewolf woohoo! good news Muchoz
01:29 Muchoz It's now possible to select what branches to update *woot woot*
01:29 bujordan joined #salt
01:30 bhosmer_ joined #salt
01:30 hightekvagabond joined #salt
01:31 bujordan joined #salt
01:32 whytewolf I do hear that CA moved their backend cloud technology from what ever they were using to openstack. which might help. but as bad as they are at maintaining software i doubt it
01:36 edrocks_ joined #salt
01:36 iceyao joined #salt
01:38 Muchoz How do I hide certain states from output?
01:40 whytewolf i don't think there is a way to block some states from output, but you can change how the output is shown. https://docs.saltstack.com/en/latest/ref/configuration/master.html#state-output
01:42 Muchoz whytewolf, ya I don't want passwords to show
01:43 Muchoz Guess I'll just have no output at all on our CI
01:43 tweakism yay I got something merged into salt
01:43 Muchoz "Woops something went wrong, check the logs."
01:43 whytewolf tweakism: yay! welcome to the club
01:44 tweakism there's lots of stuff I want to do but I don't know python very well.
01:44 tweakism time to learn, I guess
01:44 whytewolf tweakism: same boat as me. I know jack all of python when i started in salt. I have learned a TON from salt. not sure if good practies. but i have learned a lot
01:44 Muchoz It's pretty easy
01:45 Muchoz I'm getting 2 other people started with Python, 3 if you count someone who I don't work with directly on a project
01:45 Muchoz Which is why I'm making the project available on a vagrant box
01:45 bujordan joined #salt
01:46 Muchoz The project requires a web server to serve files and getting them through that and through pyenv and all of the bullcrap is too much
01:47 bujordan joined #salt
01:50 bujordan joined #salt
01:53 brianfeister joined #salt
01:54 scoates joined #salt
01:57 iggy jfindlay: danke
02:02 akhter joined #salt
02:03 krymzon joined #salt
02:03 DammitJim joined #salt
02:08 zmalone joined #salt
02:15 krymzon joined #salt
02:16 dean|away joined #salt
02:19 malinoff joined #salt
02:20 brianfeister joined #salt
02:21 michelangelo joined #salt
02:25 joe_n joined #salt
02:27 colegatron_ joined #salt
02:29 rem5 joined #salt
02:38 scoates joined #salt
02:39 Tgrv joined #salt
02:47 ilbot3 joined #salt
02:47 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
02:49 snc joined #salt
02:51 evle joined #salt
02:54 JDiPierro joined #salt
02:55 catpigger joined #salt
02:58 tristianc joined #salt
03:03 Muchoz joined #salt
03:18 bhosmer joined #salt
03:20 subsignal joined #salt
03:21 hosttor joined #salt
03:22 dkrae joined #salt
03:27 scoates joined #salt
03:29 tweakism Does salt provide any way to use modules/packages other than extension_modules_dir/{type}
03:29 scoates joined #salt
03:30 tweakism like 1) can you use as an external pillar something that is in an installed python package, rather than living under extension_modules
03:31 tweakism and 2) can you install packages somewhere that salt will pick them up so can e.g. 'import blah' without installing blah globally
03:31 tweakism those are two different questions
03:32 tweakism for the latter, I was thinking something like extension_modules_dir/site-packages
03:38 iggy tweakism: no, no
03:38 tweakism damn.
03:38 iggy why?
03:39 k_sze[work] joined #salt
03:39 tweakism re: 1), if you had a complicated external pillar that you wanted to distribute separately from salt, it would be nice if you could release it as a python package instead of as files/directories + instructions for how to place them into extension_modules
03:40 tweakism re: 2), it would be nice if you could make an external pillar (or any other kind of extension module) that has dependencies beyond salt's dependencies but not have to install those dependencies globally
03:40 iggy you'd be better off writing a module and then having the pillar be a thin layer over that module
03:41 tweakism if e.g. you install salt from packages but there's no package for the dependency and you don't like to use pip to install things globally
03:41 iggy welcome to python
03:43 tweakism hrm.
03:43 onlyanegg joined #salt
03:44 joe_n joined #salt
03:50 rbjorklin joined #salt
04:00 anmol joined #salt
04:02 jfelchner joined #salt
04:03 malinoff joined #salt
04:03 jfelchner Is there a stateful way of running a command and verifying that the output contains a certain regex?
04:09 knite joined #salt
04:10 brianfeister joined #salt
04:13 bhosmer_ joined #salt
04:13 bujordan joined #salt
04:20 joe_n joined #salt
04:20 bujordan joined #salt
04:21 subsignal joined #salt
04:21 hasues joined #salt
04:21 hasues left #salt
04:29 joe_n joined #salt
04:31 bujordan joined #salt
04:34 bujordan joined #salt
04:41 spuder joined #salt
04:44 spuder_ joined #salt
04:45 bujordan joined #salt
04:46 bujordan joined #salt
04:51 joe_n joined #salt
04:57 bujordan joined #salt
04:57 racooper joined #salt
04:58 treaki joined #salt
04:58 joe_n joined #salt
04:59 viq joined #salt
04:59 bujordan joined #salt
04:59 kermit joined #salt
05:05 joe_n joined #salt
05:09 bujordan joined #salt
05:14 bujordan joined #salt
05:15 tvinson joined #salt
05:16 berto- joined #salt
05:17 karlthane joined #salt
05:18 bujordan joined #salt
05:19 bucky joined #salt
05:24 brianfeister joined #salt
05:24 garphy joined #salt
05:25 jaybocc2 joined #salt
05:39 antpa joined #salt
05:46 rdas joined #salt
05:53 joe_n joined #salt
05:56 bucky joined #salt
05:59 bucky joined #salt
06:01 bhosmer_ joined #salt
06:13 Diaoer joined #salt
06:17 onlyanegg joined #salt
06:20 bucky_ joined #salt
06:21 oravirt joined #salt
06:39 linjan joined #salt
06:50 kawa2014 joined #salt
06:51 egalano joined #salt
06:55 bhosmer_ joined #salt
07:04 felskrone joined #salt
07:13 drawsmcgraw joined #salt
07:18 malinoff joined #salt
07:27 bucky joined #salt
07:31 bucky joined #salt
07:33 keimlink joined #salt
07:37 jaybocc2 joined #salt
07:39 impi joined #salt
07:49 bhosmer_ joined #salt
07:50 akhter joined #salt
07:52 jaybocc2 joined #salt
07:59 slav0nic joined #salt
08:04 elsmo joined #salt
08:06 H2Ov4 joined #salt
08:07 _Diaoer joined #salt
08:07 jhauser joined #salt
08:13 jbrnds joined #salt
08:14 kawa2014 joined #salt
08:16 fracklen joined #salt
08:16 NV joined #salt
08:18 fracklen joined #salt
08:20 bucky joined #salt
08:20 bucky joined #salt
08:22 brianfeister joined #salt
08:22 bucky joined #salt
08:22 evle joined #salt
08:23 fredvd joined #salt
08:24 bucky joined #salt
08:26 bucky joined #salt
08:27 bucky joined #salt
08:28 patchedmonkey joined #salt
08:30 martoss joined #salt
08:30 ajw0100 joined #salt
08:31 martoss left #salt
08:31 colegatron I'm trying to up ext_pillar to read git repos but always get the same parsing error: http://pastebin.com/yJLxSMVA
08:31 colegatron I tried multiple combinations reading the docs, but with the same result :-(
08:33 colegatron could someone have a look into? thank you very much in advance
08:35 tweakism I think the : is confusing it
08:35 colegatron damn... salt 0.17.5
08:35 tweakism wrap things that are supposed to be interpreted as strings in ""
08:35 tweakism or ''
08:35 colegatron salt --version should spit something like 2015.8.whatever right?
08:36 tweakism also, you can't give a string value *and* an array like that
08:36 AndreasLutro vagrant@orch:~$ sudo salt --version
08:36 AndreasLutro salt 2015.8.7 (Beryllium)
08:36 colegatron AndreasLutro, tnx the server bootstrap setup failed installing salt :-(
08:37 colegatron tweakism, tnx anyway :)
08:37 KermitTheFragger joined #salt
08:38 tweakism your config is still invalid, regardless of your server issues
08:38 tweakism well, maybe not, they have examples like that
08:38 tweakism huh, I wonder wtf that gets parses as
08:38 tweakism *parsed
08:39 colegatron tweakism, I've been copying/modifying from the doc
08:39 patchedmonkey joined #salt
08:39 tweakism no, I take it back, the example I saw makes sense
08:40 _Diaoer left #salt
08:44 bhosmer_ joined #salt
08:44 antpa joined #salt
08:44 tweakism I'm not certain this is right either, but closer:
08:44 tweakism http://paste.debian.net/398680/
08:44 tweakism aside:  Why does yamllint say that lines > 80 chars are an error... is that true?!
08:45 west575 joined #salt
08:46 jhujhiti_ joined #salt
08:47 rubendv joined #salt
08:47 rubendv joined #salt
08:47 josuebrunel joined #salt
08:48 krymzon joined #salt
08:48 toastedpenguin joined #salt
08:49 linjan joined #salt
08:49 dgutu joined #salt
08:51 LondonAppDev joined #salt
08:53 malinoff joined #salt
08:54 LondonAppDev joined #salt
08:57 Rumbles joined #salt
08:58 dariusjs joined #salt
08:58 atmosx joined #salt
09:01 antpa joined #salt
09:11 GreatSnoopy joined #salt
09:13 dmaiocchi joined #salt
09:15 rmnuvg joined #salt
09:15 patchedmonkey joined #salt
09:20 colegatron tweakism, you were right, I've fixed the bootstrap and the error continues there. now I should go out, I'll have a look later to your paste
09:20 colegatron tnx!
09:20 Muchoz joined #salt
09:21 pooogles joined #salt
09:23 Xevian joined #salt
09:23 pooogles_ joined #salt
09:26 jimcroft joined #salt
09:28 irctc372 joined #salt
09:32 roock joined #salt
09:34 ronnix joined #salt
09:38 bhosmer_ joined #salt
09:42 s_kunk joined #salt
09:42 impi joined #salt
09:43 sjorge joined #salt
09:43 sjorge joined #salt
09:55 oida joined #salt
10:00 toanju joined #salt
10:00 chiui joined #salt
10:01 jakwas joined #salt
10:04 jaybocc2 joined #salt
10:09 dmaiocchi salt '*' test.ping; echo $?
10:09 dmaiocchi Minion did not return. [No response]
10:09 dmaiocchi 0
10:10 dmaiocchi but it return 0. i have to add a flag for make the return compatible with bash?
10:11 dunz0r dmaiocchi: It's difficult to decide what's "good/bad". What if some of my minions answer but some don't?
10:11 brianfeister joined #salt
10:11 dmaiocchi but if i have only 1 minion
10:11 bucky joined #salt
10:12 dmaiocchi dunz0r, i agree with you, i would but return 1 for if  minions fail
10:12 jacksontj joined #salt
10:19 jaybocc2 joined #salt
10:33 anmol joined #salt
10:34 brianfeister joined #salt
10:38 bhosmer joined #salt
10:47 peters-tx joined #salt
10:57 sjorge joined #salt
10:57 sjorge joined #salt
11:13 akhter joined #salt
11:14 bucky joined #salt
11:25 tracphil joined #salt
11:29 antpa joined #salt
11:31 amcorreia joined #salt
11:39 abele joined #salt
11:40 abele How should I approach it 1. Using states 2. Using jinja macros?
11:40 abele 3. there is some other way.
11:43 AndreasLutro abele: approach what?
11:43 abele (sorry, first message got lost) Hei! I have a bunch of static web pages and uwsgi application that i would like to deploy using salt-stack.
11:45 AndreasLutro I'd say there's no specific "way" to do that
11:45 toanju joined #salt
11:49 MadHatter42 joined #salt
11:49 tracphil joined #salt
11:53 abele I'm very suggestible and wold appreciate others opinion. ;)
11:54 dunz0r abele: Hmm. Have the static pages as file.recursive, deploy to specific hosts?
11:54 abele dunz0r: no, git.latest for both static and uwsgi apps.
11:55 dunz0r Ok, so make states that does that for each page?
11:57 abele That sounds as cleanest way to do things. I don't want to overuse jinja magic even it's easy to convert existing states to macros.
11:59 dunz0r Or a for-loop if it's a lot of pages but they all share the same settings
12:09 AndreasLutro make a state for uwsgi apps, a state for nginx vhosts, a state for deploying code via git
12:09 AndreasLutro then add data for each site into pillars and make the states iterate through the pillars
12:12 abele And I use pillars for config so I get different apps on different hosts without changing states?
12:17 AndreasLutro yes
12:22 cpowell joined #salt
12:23 cpowell joined #salt
12:25 catpig joined #salt
12:26 jbrnds joined #salt
12:31 jaybocc2 joined #salt
12:38 bhosmer joined #salt
12:41 jbrnds2 joined #salt
12:43 tracphil joined #salt
12:44 benji_ joined #salt
12:45 evle joined #salt
12:46 jaybocc2 joined #salt
12:47 mohae joined #salt
12:48 iceyao joined #salt
12:53 nyx_ joined #salt
13:10 akhter joined #salt
13:11 toanju joined #salt
13:12 Cignacio hi...
13:12 Cignacio quick question
13:12 Cignacio does https://docs.saltstack.com/en/latest/ref/states/all/salt.states.iptables.html support tcp-flags?
13:13 Cignacio for a rule like IPTABLES -A INPUT -p tcp -m tcp --tcp-flags FIN,SYN FIN,SYN -j DENY
13:16 bucky joined #salt
13:17 MadHatter42 joined #salt
13:17 akhter joined #salt
13:20 rem5 joined #salt
13:27 srinivas joined #salt
13:28 JDiPierro joined #salt
13:28 numkem joined #salt
13:29 jbrnds joined #salt
13:32 tpaul joined #salt
13:33 jbrnds2 joined #salt
13:34 hax404 joined #salt
13:36 akhter joined #salt
13:39 chiui joined #salt
13:42 Ahlee joined #salt
13:43 Ahlee joined #salt
13:45 DammitJim joined #salt
13:49 zerthimon joined #salt
13:51 subsignal joined #salt
13:51 Hetman Hello :-) Howto refresh env variables during salt-call ? I've go a problem that I'm trying to install oracle-client and later cx_Oracle python module ... but after adding RPM's for oracle-client and setting up env (profile.d file and LD_CONFIG_PATH) salt is still using old variables so pip crying: NO ORACLE INSALLATION. Any idea howto do this?
13:52 cyborgIone joined #salt
13:57 mapu joined #salt
13:58 gh34 joined #salt
14:00 cpowell joined #salt
14:00 subsigna_ joined #salt
14:01 pppingme joined #salt
14:01 pppingme joined #salt
14:04 toddnni joined #salt
14:05 nyx_ joined #salt
14:06 tracphil joined #salt
14:09 jbrnds joined #salt
14:09 racooper joined #salt
14:11 ferbla joined #salt
14:17 lompik joined #salt
14:17 bucky joined #salt
14:23 morissette joined #salt
14:23 spiette joined #salt
14:26 JDiPierro joined #salt
14:26 netcho__ joined #salt
14:26 JohnnyRun joined #salt
14:27 bhosmer joined #salt
14:27 KennethWilke joined #salt
14:28 zsoftich4 joined #salt
14:35 dunz0r Hetman: Hmm, can't you add a "source <placewithnewvars> somewhere?
14:35 akhter joined #salt
14:35 dunz0r Or create the vars beforehand
14:36 Hetman dunz0r: environ.set don it like that
14:37 andrew_v joined #salt
14:44 winsalt joined #salt
14:46 felskrone does anyone knof if the ext_pillar options blocks the master until the enc-data is retrieved successfully?
14:48 nyx_ joined #salt
14:54 malinoff joined #salt
14:55 JDiPierro joined #salt
14:58 zmalone joined #salt
14:59 perfectsine joined #salt
15:04 Cignacio hi... quick question. does https://docs.saltstack.com/en/latest/ref/states/all/salt.states.iptables.html support tcp-flags?
15:04 Cignacio <Cignacio> for a rule like IPTABLES -A INPUT -p tcp -m tcp --tcp-flags FIN,SYN FIN,SYN -j DENY
15:05 tracphil_ joined #salt
15:13 sfxandy joined #salt
15:13 fracklen joined #salt
15:15 armguy joined #salt
15:16 sfxandy have a question regarding Salt mine.  i;ve got the network.ip_addr [eth0] as one of my mine functions, all working well.  however some new VM's that have been provisioned have t he interfaces defined as bond0 and bond1.  is there a "neat" way of implementing some for mof conditional logic or aliasing so that I can just reference a single mind function without having to qualify the interface?
15:16 sfxandy mine*
15:17 fracklen_ joined #salt
15:17 gtmanfred yes
15:18 gtmanfred sfxandy: https://docs.saltstack.com/en/latest/topics/mine/#mine-functions-aliases
15:18 sfxandy am reading that as we speak, but struggling to see the example given
15:18 sfxandy or to understand the example given, more accurately
15:19 gtmanfred so
15:19 gtmanfred you specify which mine function
15:19 gtmanfred and the args to pass to it
15:19 gtmanfred and then you can specify what you want to call in the mine
15:19 gtmanfred so
15:19 gtmanfred you would do mine.get \* ip_list
15:19 * sfxandy watches intently
15:19 gtmanfred ip_list:
15:19 gtmanfred - mine_function: grains.get
15:19 gtmanfred - ip_interfaces
15:20 gtmanfred or internal_ip_list
15:20 gtmanfred internal_ip_addrs:
15:20 gtmanfred mine_function: network.ip_addrs
15:20 gtmanfred cidr: 192.168.0.0/16
15:20 gtmanfred which does basically what you are wanting
15:20 gtmanfred mine_function: network.ip_addrs
15:20 gtmanfred and interface: bond0
15:20 gtmanfred and name it bond0_ip
15:21 sfxandy ok id be lying if I said that made sense
15:22 gtmanfred sfxandy: http://ix.io/otR
15:22 gtmanfred make that your mine_function
15:22 gtmanfred then call
15:22 gtmanfred salt-call mine.get \* bond0_ip
15:22 gtmanfred and it will run the alias
15:23 sfxandy ah ok
15:23 sfxandy that looks more clear
15:27 disbound joined #salt
15:28 perfectsine joined #salt
15:32 mfournier joined #salt
15:35 jhujhiti joined #salt
15:37 disbound I'm trying to write a git pre-commit hook to check syntax of sls formulas and jinja templates, but google isn't being much of help. Any suggestions?
15:38 gtmanfred salt-call state.show_low_sls
15:38 jayne joined #salt
15:38 gtmanfred is how I woudl do it, but then you would need a minion setup to watch that directory
15:39 DammitJim joined #salt
15:39 gtmanfred i am not familiar with a sls/jinja linter for salt
15:39 gtmanfred oh
15:39 gtmanfred https://github.com/johanek/salt-lint
15:39 gtmanfred neat
15:39 gtmanfred worth a shot at least
15:40 disbound Thanks for the suggestions :D
15:41 Remid joined #salt
15:42 Remid Hi, I have a weird problemn with require: The following requisites were not found: require: pkg: elasticsearch
15:43 hairytoez joined #salt
15:43 Remid When you specify the pkgs: package: {{ version }} the require should use the name with {{ version }} as well?
15:46 _JZ_ joined #salt
15:50 basepi forresta: iggy: I actually still have op power, just not auto op
15:50 spuder joined #salt
15:50 perfectsine joined #salt
15:52 MadHatter42 joined #salt
15:54 nZac joined #salt
15:55 akhter joined #salt
15:57 tracphil joined #salt
16:00 ronnix joined #salt
16:01 krymzon_ joined #salt
16:03 sfxandy gtmanfred, am having trouble getting the mine function to actually return anything when using a mine alias
16:04 sfxandy have done the usual refresh of pillar and mine.update
16:04 sfxandy and I can see the mine_functions in Pillar
16:04 sfxandy but issuing the mine.get <alias> gives me nothing
16:07 DammitJim if I define in my top.sls the following: myapps[2,4,9](0|1)1*.com':
16:07 DammitJim that would mean that a server with the host myapps201test.example.com would run the states below it, right?
16:08 hasues joined #salt
16:09 buhman DammitJim: why don't you just use grains?
16:09 DammitJim or do I need to do something like: myapps[2,4,9](0|1)1*.*.com ?
16:09 DammitJim grains in what way?
16:09 gtmanfred sfxandy: https://gist.github.com/gtmanfred/2dc0bf44763d97b59064
16:09 gtmanfred works here :/
16:09 UICTamale joined #salt
16:10 buhman DammitJim: you'd give all 'myapps' servers some arbitrary grain that describes probably the group of states you're wanting to apply
16:10 buhman DammitJim: then you reference that grain in your top.sls instead of making weird hostname matches
16:10 fracklen joined #salt
16:10 DammitJim oh
16:10 DammitJim interesting
16:11 DammitJim so, I do that where?
16:11 djgerm1 Am I correct in my understanding that when salt-cloud provisions a new instance, it then sshs into the new instance, installs salt-minion, points it to the master, and then the minion starts connecting to master on 4505/4506, and no other ports are needed to be open between the two?
16:11 DammitJim my pillar's top.sls?
16:11 buhman no
16:11 buhman https://docs.saltstack.com/en/latest/topics/targeting/grains.html#grains-in-etc-salt-grains
16:12 buhman one pattern is to do like "roles:myapps" or something
16:12 DammitJim grains in the minion?
16:12 DammitJim how do I manage those grains?
16:12 buhman eh?
16:12 DammitJim I'll need a state to manage the grains on the minions?
16:12 buhman no
16:12 buhman they exist in the grains file on each minion
16:13 Remid it's frustrating The following requisites were not found: require: pkg: elasticsearch doesn't help at all, 2 lines above I have - pkgs: - elasticsearch: {{ elasticsearch.version }}
16:13 buhman there's also https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.grains.html
16:13 buhman so you can do salt 'myappsblah' grains.append roles myapps
16:13 Remid Why is it complaining about requisites not found I installed that package first
16:14 Netwizard joined #salt
16:15 buhman Remid: "pkgs: - elasticsearch" won't define the "pkg: elasticsearch" state
16:15 Remid Why that?
16:15 buhman because that isn't a pkg: elasticsearch definition
16:15 bhosmer joined #salt
16:16 buhman did you mean to include that state from elsewhere?
16:16 Remid No
16:16 hasues left #salt
16:16 Remid I install elasticsearch on top of the state and then require on that package on some files after
16:16 Remid There is something I don't understand :/
16:17 buhman does your pkg state look something like https://docs.saltstack.com/en/latest/ref/states/all/salt.states.pkg.html ?
16:18 Remid It looks like that: http://pastebin.com/x4PMLPZe
16:18 buhman ahh
16:19 buhman based on line 1, line 15 should be: '- pkg: install elasticsearch package'
16:19 srinivas joined #salt
16:19 Remid I see, It's a bit confusing
16:19 jbrnds joined #salt
16:20 Remid doesn't require pkg name tells execute that if that pkg is installed?
16:20 buhman no
16:20 buhman require references state names, not something-that's-not-a-state names
16:20 sjorge joined #salt
16:20 sjorge joined #salt
16:24 sfxandy gtmanfred, i have very similar to what you have but no joy :(
16:25 kim0_ joined #salt
16:28 gtmanfred :( can't know
16:28 DammitJim buhman, I"m a little confused
16:28 DammitJim I'll have to digest this because it's going to require a lot of work on my part to re-do it
16:28 DammitJim I'll have to do this sooner than later, I guess...
16:29 kim0_ joined #salt
16:30 Fiber^ joined #salt
16:35 writtenoff joined #salt
16:36 eodgooch joined #salt
16:42 sfxandy gtmanfred .... ok interestingly it worked after restarting the minion....
16:43 gtmanfred heh, can't know
16:43 sfxandy f***ing annoying!!
16:52 anmol joined #salt
16:58 colegatron I feel dumb with the gitfs and ext_pillar and the documentation
16:58 bucky joined #salt
16:59 colegatron after lots of tries I seems I have the master well configured, there is no errors on /var/log/salt/{master,minion} (and if I intentionally introduce some invalid argument on git_remote I can see the errors in the logs)
16:59 colegatron but... I can't see the gitfs copying the repos to /srv/{salt,pillar}
17:00 colegatron so a 'salt state.highstate whatever' fails with a "No minions matched the target. No command was sent, no jid was assigned.\nERROR: No return received"
17:01 shaggy_surfer joined #salt
17:01 colegatron Maybe I should not expect the repos specified on the git_remotes to be automaticaly cloned  into /srv/{salt,pillar} ??
17:01 Ch3LL colegatron: do you see anything in /var/cache/salt/master/pillar_gitfs/
17:03 colegatron Ch3LL, yes, I can see a long randomnly named (empty) folder
17:04 kevinquinnyo The Salt Master has rejected this minion's public key!
17:04 kevinquinnyo To repair this issue, delete the public key for this minion on the Salt Master and restart this minion.
17:04 kevinquinnyo i can't find the public key for this minion anywhere on the master
17:04 colegatron same happens with gitfs/other_expected_fodler
17:04 colegatron same happens with gitfs/other_expected_folders
17:04 kevinquinnyo it's not in /etc/salt/master/pki in any directory
17:04 buhman kevinquinnyo: what does salt-key say?
17:04 kevinquinnyo it doesn't identify this minion at all
17:04 kevinquinnyo salt-key -L
17:04 colegatron kevinquinnyo, salt-key -L
17:04 kevinquinnyo yes
17:05 kevinquinnyo does not see it
17:05 Ch3LL colegatron: hmmm thats where gitfs clones the repo. so its weird that you are not seeing anything
17:05 buhman kevinquinnyo: you might start the minion in the foreground with -l debug
17:05 Ch3LL colegatron: can you try 'salt-run fileserver.clear_cache backend=git' to delete your git cache and restat your master and see if you see in the logs the master connecting to the git repo
17:05 colegatron Ch3LL, I am seeing the expected folders. I used - name arguments for the gitfs repos, and they exists, but empty, like if master were not cloning the repos
17:06 kevinquinnyo i've also tried restarting the salt-minion service and the salt-master, i've also tried that ^ buhman  and i've deleted the salt-minion's key pair and restarted to force it to generate a new one
17:06 buhman ok
17:07 fracklen joined #salt
17:08 Ch3LL colegatron: you should see something in your master debug logs about connecting to the gitfs. i can't remember the exact syntax.
17:08 Ch3LL if you don't see that maybe posting your master sanitized config to a gist or something might help
17:08 linjan joined #salt
17:09 lumtnman joined #salt
17:09 colegatron Ch3LL,  http://pastebin.com/C1xtYQAp
17:11 colegatron Ch3LL, as you can see, it gets some tags from the repos, but there is no files in the cache, only .git folder. maybe it is normal? or the files on the branch should be there?
17:12 conan_the_destro joined #salt
17:12 xfs joined #salt
17:13 lumtnman joined #salt
17:15 ronnix joined #salt
17:17 Ch3LL colegatron: well i'm not an expert at gitfs, have only set it up a few times but that log looks good to me.
17:17 Ch3LL Are you using the master branch on your git repo? and do you use the base environment in salt?
17:18 edrocks joined #salt
17:18 colegatron let me sanitize the master and pasteit
17:18 Ch3LL k cool
17:19 perfectsine joined #salt
17:20 akhter joined #salt
17:21 colegatron only a question; should I expect salt to clone the repos, checkout right right branches and copy the resulting files into /srv/salt or the files from gitfs are never copied to /srv/salt ?
17:21 colegatron master: http://pastebin.com/knPa6wzU
17:21 whytewolf gitfs are not kept in /srv/salt they are kept in a cache
17:22 colegatron whytewolf, uhmmm, ok. tnx for the clarification.
17:22 whytewolf colegatron: https://docs.saltstack.com/en/latest/ref/runners/all/salt.runners.fileserver.html#module-salt.runners.fileserver is useful for seeing what salt sees
17:22 knite joined #salt
17:23 akhter joined #salt
17:23 rvankleeck joined #salt
17:23 rvankleeck joined #salt
17:24 rvankleeck joined #salt
17:25 spuder joined #salt
17:26 colegatron mmm trying to understand.... it is my impression or everybody using salt are python programmers? :)
17:27 hightekvagabond joined #salt
17:27 ffredrikk joined #salt
17:27 whytewolf I'm not a python dev. least i wasn't when i started.
17:27 oida joined #salt
17:27 whytewolf I come from the admin/operations side of things.
17:28 rvankleeck_ joined #salt
17:28 rvankleeck Has anyone experienced an issue where dockerng can't actually do anything if it is run at the same time as a state that installs docker?
17:29 rvankleeck I just opened this issue about it: https://github.com/saltstack/salt/issues/31361
17:29 saltstackbot [#31361]title: dockerng No such file or directory on initial run | I'm trying to get dockerng to pull an image and start a container, but it seems that dockerng cannot do anything if a state that also installs docker is run at the same time. However, running a another highstate pulls the image and starts the container correctly. Am I missing something?...
17:30 drawsmcgraw joined #salt
17:30 whytewolf rvankleeck: https://docs.saltstack.com/en/latest/ref/states/#reloading-modules
17:30 KennethWilke rvankleeck, i've been in that spot before, where i want to use the docker python module and it's not installed
17:30 gtmanfred rvankleeck: you may need to run a refresh_modules to reload them after the docker python module is installed
17:31 colegatron whytewolf, then it is me being so dumb  :) but your link opened a new horizon: salt-run fileserver.dir_list shows a list of all my states
17:31 KennethWilke ooo that's handy, don't think that was a feature the last time i used salt and docker together
17:31 ffredrikk hi, i was wondering ... how come this ( http://pastebin.com/zf9P7Dai ) does not install quicktime properly -- but this ( http://pastebin.com/pEJP3KSc ) does install quicktime properly ... ?
17:31 rvankleeck KennethWilke: whytewolf: gtmanfred: thanks. I have never run into this edge case. I’ll try it with that.
17:34 scoates joined #salt
17:35 josue1 joined #salt
17:40 toastedpenguin joined #salt
17:42 ffredrikk joined #salt
17:43 shaggy_surfer joined #salt
17:45 frew hey guys, I have an ext_pillar that does some validation on it's data and I'd like, if that validation fails, to stop the salt run entirely
17:46 frew I tried throwing a ValueError but that doesn't do the trick
17:46 frew is there a way to do what I want?
17:47 frew I know that I can do it by putting in a "Validation state" instead of doing it at ext_pillar time, but having all the data in the same place for this seems like something worthwhile
17:49 dlam joined #salt
17:49 frew I gotta go afk for a bit, but if someone leaves ideas here I'll see them
17:50 colegatron whytewolf, how is this possible? http://pastebin.com/BwUnAM0C
17:50 colegatron a file_list says the state is there, but a state.sls says it is not :-(
17:51 whytewolf that i don't know
17:52 * whytewolf doesn't use gitfs.
17:52 colegatron whytewolf, tnx anyway.
17:53 zmalone So what's the issue with 2015.8.5?  It didn't get mentioned in the release notes
17:53 zmalone https://docs.saltstack.com/en/latest/topics/releases/2015.8.7.html
17:53 colegatron I'm turning crazy with gitfs, but I need to set it up and understand it.
17:54 gtmanfred zmalone: boto_ regressions
17:54 AndreasLutro there are several zmalone
17:54 gtmanfred it got mentioned in the email of on salt-announce
17:54 zmalone Someone should probably let Meggie Woodfield know that the google groups mailing lists are not just a web forum
17:55 zmalone I'm on salt-users/salt-announce, and they just say to review the release notes
17:55 zmalone She mailed out a half dozen very similarly worded emails
17:58 grumm_servire joined #salt
18:00 bucky joined #salt
18:01 akhter joined #salt
18:03 bhosmer joined #salt
18:04 chiui joined #salt
18:08 JDiPierro joined #salt
18:08 bucky joined #salt
18:09 hightekvagabond joined #salt
18:09 bhosmer joined #salt
18:13 hal58th_ joined #salt
18:16 treaki joined #salt
18:18 Muchoz joined #salt
18:20 bucky joined #salt
18:20 bucky joined #salt
18:21 fracklen joined #salt
18:24 numkem joined #salt
18:25 hal58th_ I'f im reading the notes correctly, the only two major fixes are for npm and rpm???
18:25 numkem joined #salt
18:25 tedski joined #salt
18:25 __alex joined #salt
18:25 ndim joined #salt
18:25 asoc joined #salt
18:25 Shirkdog joined #salt
18:25 StevenGFX joined #salt
18:25 Sketch joined #salt
18:25 Vye joined #salt
18:25 tooth joined #salt
18:25 nicksloan joined #salt
18:26 Eureka703 joined #salt
18:26 justanotheruser joined #salt
18:26 hexa- joined #salt
18:26 __number5__ joined #salt
18:26 JPT joined #salt
18:26 jinkyu joined #salt
18:26 jesusaur joined #salt
18:26 tzero joined #salt
18:26 iggy there was a security fix in 1 version but they've had to hotfix that release a couple times
18:26 ecdhe joined #salt
18:26 Antiarc joined #salt
18:26 NightMonkey joined #salt
18:26 kuromagi joined #salt
18:26 sinonick joined #salt
18:27 avozza joined #salt
18:27 SubOracle joined #salt
18:27 ekkelett joined #salt
18:27 ekkelett joined #salt
18:27 trent joined #salt
18:27 TyrfingMjolnir joined #salt
18:27 ahammond joined #salt
18:27 cswang__ joined #salt
18:27 smkelly joined #salt
18:27 giany joined #salt
18:27 ashmckenzie joined #salt
18:27 FreeSpencer joined #salt
18:27 FreeSpencer joined #salt
18:27 B1nny joined #salt
18:27 nobrak joined #salt
18:27 nobrak joined #salt
18:27 llua` joined #salt
18:27 is_null joined #salt
18:27 penguinpowernz joined #salt
18:27 TTimo joined #salt
18:27 mr_chris joined #salt
18:27 nledez joined #salt
18:27 izibi joined #salt
18:27 SVQTQ joined #salt
18:27 rome_390 joined #salt
18:27 CrummyGummy joined #salt
18:27 rawzone joined #salt
18:27 Qwazerty joined #salt
18:27 davisj joined #salt
18:27 duckfez joined #salt
18:27 sixninetynine joined #salt
18:27 bonzibuddy joined #salt
18:27 buhman joined #salt
18:27 masterkorp joined #salt
18:27 jcockhren joined #salt
18:28 sauvin joined #salt
18:28 honestly joined #salt
18:28 babilen joined #salt
18:28 kalessin joined #salt
18:28 alias_ joined #salt
18:28 rickflare joined #salt
18:28 iggy joined #salt
18:28 stickmack joined #salt
18:28 unusedPhD joined #salt
18:29 sauvin joined #salt
18:30 AndreasLutro you're not reading it correctly
18:30 klocek joined #salt
18:30 ntropy joined #salt
18:30 AndreasLutro the changes aren't listed in order of importance
18:31 AndreasLutro nor should they be considered hierarchial
18:32 eodgooch joined #salt
18:35 twork i need a pointer. our admin accounts are backed by AD; i'd like salt to be aware of those accounts (when it creates directories, sets perms, etc.) without managing those accounts itself. the LDAP-aware areas of Salt seem to talk about governing Salt itself, rather than Salt handling other areas in an LDAP-aware way. ...right?
18:35 jfindlay zmalone: there were a few regressions between 2015.8.5 and .7, for which the fixes are discussed in the first section the 2015.8.7 release notes
18:36 twork for instance, https://docs.saltstack.com/en/latest/topics/eauth/index.html
18:39 twork whenever i've tried listing my own account in a pillar backed by the 'users' formula, i've hit snages.
18:41 twork and/or snags.
18:48 Muchoz joined #salt
18:53 scoates joined #salt
18:55 msheiny joined #salt
18:58 colegatron I've just downgraded to 2015.8.5 and removed the minion key with 'salt-key -d minion'. minion went into a endless loop "The Salt Master has cached the public key for this node, this salt minion will wait for 10 seconds before attempting to re-authenticate"
18:59 colegatron there is someway to regenerate the key?
18:59 rem5 joined #salt
19:00 colegatron btw, the minion is the same salt-master server
19:01 rem5 joined #salt
19:01 KennethWilke joined #salt
19:01 calculon joined #salt
19:02 jaybocc2 joined #salt
19:03 DammitJim so, grains are managed at the minion level and pillars at the master level?
19:03 DammitJim am I getting that right?
19:04 twork left #salt
19:06 slav0nic joined #salt
19:07 ajw0100 joined #salt
19:07 jorr-el joined #salt
19:07 zmalone colegatron: Downgraded from 2015.8.7, or from the development branch?
19:08 zmalone (I don't have an answer to your question, I'm just curious about what you downgraded from and why)
19:08 hal58th joined #salt
19:08 colegatron 2015.8.7 to .5
19:09 baweaver joined #salt
19:09 morissette joined #salt
19:09 colegatron I was having issues with gitfs and ch3ll recommended me to downgrade just in case there is some bug because .7 is so new
19:09 colegatron now Im having issues to restart .5 because the minion keys :-(
19:11 KyleG joined #salt
19:11 KyleG joined #salt
19:13 ajw0100 joined #salt
19:14 mapu joined #salt
19:14 jfindlay colegatron: first, stop the minion and then use salt-key to delete all keys from that minion
19:17 mariusv joined #salt
19:18 rburkholder joined #salt
19:18 colegatron jfindlay, I did it. but not worked. I had to uninstall minion purging the config and reinstall it
19:18 akhter joined #salt
19:19 rburkholder if I have a state in a .sls file named default-bridge:, how do I do a '- require:' for that state, so that that state is run first?  of is that not possible?
19:20 gtmanfred you can do a require on that
19:20 gtmanfred sls: default-bridge
19:20 gtmanfred but you will need to do an include on that sls file, or make sure you it always runs anytime you do a state.sls with the state requireing it
19:21 dmaiocchi joined #salt
19:22 bucky joined #salt
19:22 rburkholder the two states are in the same file, do I do the include anyway?  when I do 'sls: default-bridge', the comment is that the following requisite was not found
19:22 hightekvagabond joined #salt
19:24 gtmanfred oh, if they are in the same file
19:24 whytewolf rburkholder: in that insteance you want module: <state ID>
19:24 gtmanfred just require the type: name
19:24 traph_ joined #salt
19:26 whytewolf basicly if the state is a file.managed you can use file: default-bridge if it is a cmd.run it would be cmd: default-bridge service.running would be service: default-bridgeect. ect. ect
19:27 ajw0100 joined #salt
19:28 nyx_ joined #salt
19:30 rburkholder ah, ok, good guessing with incomplete info, thank you.  cmd: default-bridge was the solution as default-bridge has a cmd.run
19:30 writtenoff joined #salt
19:33 bucky joined #salt
19:35 hal58th_ joined #salt
19:38 zsoftich4 joined #salt
19:40 srinivas joined #salt
19:40 akhter joined #salt
19:48 lungaro joined #salt
19:49 lungaro Is it possible to apply salt states to a node from a git checkout (salt states are checked out into a local file system directory) ?
19:49 lungaro sorry, total newb here
19:50 akhter joined #salt
19:50 jfindlay you need to make sure your local git repo is in `file_roots` in order to access the sls files in it with salt
19:51 lungaro gotcha, i'll give it a whirl. Just starting up.
19:51 jfindlay great, don't be afraid to ask more questions :)
19:53 winsalt is it possible to have multiple pkgs under a folder in the winrepo? something like this https://bpaste.net/show/0739b33b0e8f?
19:54 chiui joined #salt
19:54 jfindlay I've never tried that before; I'm not sure
19:54 scarcry joined #salt
19:55 aljosa joined #salt
19:57 aljosa is there any support for aws ecs?
19:57 Benji_ joined #salt
19:58 moufwash joined #salt
20:01 moufwash left #salt
20:02 llua joined #salt
20:03 gtmanfred maybe? what is ecs?
20:04 gtmanfred containers, i do not believe those are in yet, but I haven't looked at the aws stuff much
20:06 snc joined #salt
20:06 bhosmer_ joined #salt
20:06 scarcry joined #salt
20:09 scoates joined #salt
20:10 GreatSnoopy joined #salt
20:10 bhosmer__ joined #salt
20:13 Muchoz joined #salt
20:13 alvinstarr joined #salt
20:14 LonelyDeveloper joined #salt
20:15 LonelyDeveloper Hey.... looking to write unit tests for my runners/modules... having a hell of a time parsing the docs about how to approach this. Are there any supplementary docs/tutorials online for testing of this nature?
20:15 taz__ joined #salt
20:16 jfindlay LonelyDeveloper: which runners/modules?
20:16 taz__ anyone using salt-cloud with openstack?
20:16 LonelyDeveloper My own runners/modules in my salt states.
20:17 rickflare i have taz
20:17 rickflare its wicked eazy
20:17 jfindlay LonelyDeveloper: also, I've edited those docs before, so I'm interested in what you think is confusing.  I'm probably going to agree with you if you say 'most of it' :)
20:18 jfindlay LonelyDeveloper: what do you have so far?
20:21 ajw0100 joined #salt
20:22 Heartsbane jfindlay: is there a new bug with having to restart a salt-minion to set a grain.
20:23 rickflare guys
20:23 * Heartsbane is looking and doesn't want to file a potential duplicate
20:23 rickflare anyone know much about salt and openstack
20:23 jfindlay Heartsbane: not that I know of
20:23 Heartsbane k
20:23 * Heartsbane keeps looking.
20:24 jfindlay Heartsbane: how are you setting the grain?
20:25 Heartsbane grains.present:
20:25 Heartsbane - name:
20:25 jfindlay rickflare: I know little about salt+openstack, but what do you want to know?
20:26 taz__ jfindlay: do you have any example configs? I can't seem to get any of the examples to work
20:27 taz__ they all just wait for networking to return private_ips or public_ips, but in debug mode, i see the addresses return in the json
20:28 jfindlay taz__: how are you interfacing salt and openstack?
20:29 taz__ jfindlay: using salt-cloud with the nova driver
20:29 taz__ http://pastebin.com/xzjhH8Zr is what it repeats for 10 min then gives up
20:29 LeProvokateur joined #salt
20:30 LonelyDeveloper Hey jfindlay, got lost in work. What I have so far is a very basic test just testing 1 function in a redis cluster module I have. My biggest beef is with the test runner. There's two different test runners listed in two different docs, but both seem to be completely tied to only running tests for the main salt project?
20:30 jfindlay Heartsbane: that works for me without restart on 2015.8.3
20:31 jfindlay LonelyDeveloper: you mean integration tests vs unit tests?
20:33 Heartsbane jfindlay: 2015.8.5
20:33 LonelyDeveloper The salt test repo seems to be for integration testing, right?
20:33 LonelyDeveloper While the tests in the main salt repo are for unit tests?
20:35 jfindlay taz__: I thought one of the builtin cloud drivers used nova, but (on 2016.3 at least) I can't find any
20:35 jfindlay LonelyDeveloper: salt-testing is for utils used by the test suite, whereas the tests dir in the main salt repo is for the actual tests
20:36 LonelyDeveloper Okay.
20:37 jfindlay the tests dir contains both integration and unit tests.  Both are run with unittest in a programmatic fashion
20:37 LonelyDeveloper The tutorial seems written for people who either have a fork of the salt repo or are working on the main project. In my case I have my own standalone states and pillars repos, and my modules/runners are in my states repo. So I don't have a lot of things the test runner assumes. I could write my own little helper, but the tutorials seem weird with that regard.
20:38 jfindlay yeah, that's something the test suite does not really account for yet
20:38 taz__ jfindlay: which should I use for openstack? when i use the openstack driver it says its depriciated and use nova instead
20:38 jfindlay (tests for custom salt modules)
20:38 jfindlay taz__: let me ask someone more knowledgeable
20:39 taz__ ok thanks
20:41 LonelyDeveloper Okay as long as I'm not crazy.
20:41 LonelyDeveloper It seems I'll be framing some of my own stuff.
20:42 forrest joined #salt
20:43 kevinquinnyo i know this should be simple, but how do you ensure a service is running, but also restart the service if 'onchanges' requisite of another state returns true
20:44 cberndt joined #salt
20:44 jfindlay taz__: also, gtmanfred might know :)
20:44 LonelyDeveloper Kevinguinyo: Does watch not work for your situation? https://docs.saltstack.com/en/latest/ref/states/requisites.html
20:45 sfxandy joined #salt
20:45 kevinquinnyo LonelyDeveloper: I'm sure it does
20:45 jfindlay LonelyDeveloper: sure, and don't be afraid to push anything upstream if you can/want
20:45 kevinquinnyo but i dont know how to use it properly
20:46 Muchoz joined #salt
20:46 kevinquinnyo i wish there were a service.restart method, then i could put the onchanges requisite on that one only, and have service.running be a state that runs regardless
20:47 LonelyDeveloper Watch is good for "check this config file, if it changes, restart the service"
20:47 LonelyDeveloper Look at the nptd example here: https://docs.saltstack.com/en/latest/ref/states/requisites.html#watch
20:48 LonelyDeveloper In that example if /etc/ntp.conf changes, ntpd is restarted.
20:49 LonelyDeveloper I'm unfamiliar with "onchanges" requisite of states, as you mentioned. Another way of doing something in reaction to another event is to use reactor and custom events (kick off an event when another state is completed, have reactor restart the service?)
20:49 kevinquinnyo so service.running just *knows* to do a specific thing if there is a watch key?  and in its case that specific thing is to restart or reload the service?
20:49 kevinquinnyo LonelyDeveloper: I just read about both, you're right i think that i need watch instead of onchanges
20:49 forrest kevinquinnyo, just use watch, or watch_in
20:49 LonelyDeveloper Yes, I'm not sure about differentiating between reload and restart (I believe they're arguments to watch)
20:49 whytewolf kevinquinnyo: yes
20:50 kevinquinnyo whytewolf: do we just have to look at the source code to know what things a module does when it is given a watch arg?
20:50 kevinquinnyo becuase it seems specific to each module what action it would take
20:50 LonelyDeveloper Documentation covers it mostly. Make sure you are on the docs for your current version.
20:50 LonelyDeveloper A lot of salt things will feel automagical 'til you see the states->execution states linkage. I personally don't like the terminology in salt.
20:51 whytewolf kevinquinnyo: most of the documentation says what reaction a watch has on it's code. [if watch does anything]
20:51 kevinquinnyo hmm where is the correct documentation docs.saltstack.com ?
20:52 LonelyDeveloper "The watch statement does everything the require statement does, but with a little more. The watch statement looks into the state modules for a function called mod_watch. If this function is not available in the corresponding state module, then watch does the same thing as require. If the mod_watch function is in the state module, then the watched state is checked to see if it made any changes to the system, if it has, then mo
20:52 ajw0100 joined #salt
20:52 LonelyDeveloper that line is from: https://salt.readthedocs.org/en/v0.17.0/ref/states/requisites.html#watch
20:52 LonelyDeveloper I personally find googling "saltstack" and then your question is your best bet, I have a hard time navigating the salt docs on the site.
20:52 Reilly joined #salt
20:52 forrest kevinquinnyo, https://gist.github.com/gravyboat/9d6eef9aadd688eb9f68
20:52 forrest LonelyDeveloper, Don't use readthedocs
20:53 forrest unless you're seriously on 0.17.0
20:53 forrest in which case, please upgrade.
20:53 LonelyDeveloper oh my
20:53 LonelyDeveloper lol
20:53 LonelyDeveloper didn't notice that.
20:53 kevinquinnyo if i use watch to watch a file that another state may or may not change with file.replace, do i still need to use the 'require' requisite in addition to the 'watch' requisite to ensure that the file.replace command runs first?
20:53 forrest use docs.saltstack.com, the readthedocs site shouldn't even work any longer.
20:54 whytewolf kevinquinnyo: you need to watch both [watch uses states not actual files]
20:54 kevinquinnyo yes
20:54 forrest kevinquinnyo, You don't HAVE to use require IF things are staggered in order within the file, I do a lot of the time because it creates a visual chain in my brain, and used to be a requirement before the feature was added, but you should watch both as whytewolf said
20:54 kevinquinnyo my mistake
20:55 kevinquinnyo what do you mean by watch both
20:55 whytewolf watch: - file: file manged state id - file: file replace state id
20:56 kevinquinnyo i have two states, one is service.running 'whatever-service', the other is file.replace 'replace-something-inservice-config'
20:56 LonelyDeveloper ^ assuming you're doing a file managed, and a file replace.
20:56 LonelyDeveloper You could just be doing a file replace (replacing the default)
20:56 forrest kevinquinnyo, can you gist it? I think there is a disconnect between what you are explaining, and what you are trying to do
20:57 whytewolf agreed
20:57 forrest or at least that whytewolf guy is getting super confused by it, I'M NOT CONFUSED AT ALL THOUGH!
20:57 forrest ;)
20:57 amcorreia joined #salt
20:57 whytewolf lol
20:58 rem5 joined #salt
20:58 baweaver joined #salt
20:59 kevinquinnyo https://gist.github.com/kevinquinnyo/3497e977ce15ef35cabe
21:00 whytewolf okay. yeah looks right
21:00 whytewolf although th require is not needed
21:01 LeProvokateur I am looking for a good and exhausting explanation of pillars. Any recommendations?
21:01 forrest agreed with whytewolf, looks good to me
21:01 kevinquinnyo ok thanks everyone
21:01 whytewolf LeProvokateur: https://docs.saltstack.com/en/latest/topics/pillar/
21:03 numkem joined #salt
21:03 teryx510 joined #salt
21:04 bhosmer_ joined #salt
21:05 bucky salt newbie here - trying to use the mysql-formula with ubuntu 12 minion on vagrant. Install goes fine, but other than root, users and db's aren't created. I'm using the example pillar pretty much as is.
21:05 jim__ joined #salt
21:06 bucky The only error in the minion salt log is foo.schema isn't found - would that prevent the users and db's from being created?
21:06 forrest bucky, Yes, the foo.schema reference in the pillar is simply a placeholder, that file doesn't actually exist at all, so you need to drop one into the files directory, and reference it in the pillar
21:06 forrest also note this bug: https://github.com/saltstack-formulas/mysql-formula/issues/106
21:06 saltstackbot [#106]title: mysql:server:root_user inconsistently used on Debian/Ubuntu | For Debian OS family, debconf is used to set root user password, which is fine, but root user cannot be specified this way thus it is set to root for all Debian systems....
21:07 _JZ_ joined #salt
21:07 baweaver joined #salt
21:08 bucky forrest - yes I saw that bug - I changed the admin user to root and it sets up the password I have in the pillar. Here's my pillar & top.sls: https://gist.github.com/bujordan/49d2637e92e69dcadb55
21:08 forrest bucky, Yeah you need to fix https://gist.github.com/bujordan/49d2637e92e69dcadb55#file-mysql-sls-L48-L60
21:09 forrest bucky, There's no foo, or baz schema files, they're just placeholders so people can drop their own in.
21:09 LeProvokateur whytewolf: Thanks. Are you aware of an explanation with more code examples and some example output
21:12 akhter joined #salt
21:12 bucky forrest: thanks - I'll comment those out and see if that works.
21:12 whytewolf LeProvokateur: not really
21:13 forrest bucky, You'll need to drop in some schema stuff so the DBs can provision, are you just trying to install the client and not the server portion?
21:13 forrest bucky, I haven't looked at the whole thing, but I THINK you have to provide even just a blank schema for it to provision (not 100% sure though, I don't look at this formula that often)
21:13 hightekvagabond joined #salt
21:15 viq joined #salt
21:16 bucky forrest: no I'll eventually want schemas was just testing. I'll try it with both real quick. I thought it was just "supposed to work out of the box".
21:17 forrest bucky, Yeah most formulas do, you might want to look at the states to see where the schema gets dropped in, and see if you can just exclude that one
21:17 forrest it MIGHT provision the database and all that without it, I'm just not familiar enough with the formula to know sadly
21:18 rem5 joined #salt
21:18 bucky forrest: yeah would be nice if they just included a really simple schema with a test table just so it works on first run for people. trying with schemas commented out now.
21:18 josuebrunel joined #salt
21:18 forrest bucky, Yeah for sure, if you want create an issue on the repo to add that (if one does not already exist)
21:19 forrest Or if you really want feel free to make a PR that adds it :)
21:19 forresta joined #salt
21:20 bucky forrest: seems simple enough let me give it a go locally first and see if that fixes it. do I need an issue to go with a PR?
21:20 forresta bucky, Nah, just open a PR.
21:21 forrest Ugh there we go, the joys of getting dropped
21:22 forresta joined #salt
21:22 bucky forrest: 'k. so tried with schemas commented out using sudo salt 'minion1' state.sls mysql from the master
21:22 gtmanfred jfindlay: that problem he had is fixed in https://github.com/gtmanfred/salt/commit/47ecb7a150f88c77e082cf6541a40388acd8ea93
21:23 bucky 6 succeed, 12 fail, first fail is ID: mysqld, Function: service.running Name: mysql-service, result: false, failed to start service
21:23 gtmanfred https://github.com/saltstack/salt/pull/31120
21:23 saltstackbot [#31120]title: Clean up some bugs in the nova driver | I am pretty sure this will fix the following issues...
21:24 bucky however mysql is certainly running on the virt (I can log into it)
21:24 forrest bucky, Might be an issue if the service is set to restart? I'm not sure, the mysql formula is pretty big and complicated.
21:25 jfindlay thanks, gtmanfred
21:28 baweaver joined #salt
21:28 bucky forrest: service.start mysql fails, service.status is true, service.restart is true. hm. perhaps the formula doesn't correctly detect if the service is running?
21:29 forrest bucky, It might not, I would say to start from a clean state if you can, and just drop a really simple schema in, see what that does.
21:31 bucky forrest: yeah clean slate is fine right now since it's just a vagrant vm. I just need to see updates to mysql working with salt or it's kinda pointless for production use - I can't just blow away a prod db server every time I want to make a change. I'll give the simple schema a go.
21:31 forrest bucky, For sure, I think there are safeguards in place for that with how it works, but I really just don't know.
21:33 gtmanfred np
21:34 jhauser joined #salt
21:35 bucky forrest: thanks for the help. I'll give the simple schema thing a try and am happy to submit a PR if it works. As for changes, maybe I need to try creating my own simpler MySQL states since I wouldn't need all the multi-os logic in there (and I'd probably use MariaDB instead anyhow)
21:36 forrest bucky, For sure. I don't think there is a mariadb formula already, but this person has some: https://github.com/GeoffMontee/mariadb-saltstack-formula
21:39 numkem joined #salt
21:39 alainv_ joined #salt
21:40 Muchoz joined #salt
21:40 alainv Hey all - I'm seeing an old minion (removed with `salt-key -d`) still being considered active when starting jobs such as by wildcard globs -`salt '*'
21:41 alainv This is in 2015.8.5, and seems incorrect?
21:44 alainv I just saw 2015.8.7 was out and updated, that appears to fix it. Interesting, none of the release notes appear relevant
21:45 stupidnic joined #salt
21:46 bucky joined #salt
21:52 Ch3LL alainv: i just tried to replicate that issue and I could not with 8.5. before you upgraded did you trying restarting the master? or clearing the cache
21:53 alainv Ch3LL: I did try restarting the master, but not cache clearin
21:53 alainv g*
21:54 Ch3LL yeah not sure why that was happening then.
21:56 akhter joined #salt
21:59 bhosmer_ joined #salt
22:05 writtenoff joined #salt
22:07 chiui joined #salt
22:09 bucky joined #salt
22:09 ageorgop joined #salt
22:10 bucky joined #salt
22:14 lorengordon joined #salt
22:15 subsignal joined #salt
22:30 viq joined #salt
22:32 jankmcjanker joined #salt
22:37 JDiPierro joined #salt
22:38 JDiPierro joined #salt
22:42 garphy joined #salt
22:42 zenlot2 joined #salt
22:43 jankmcjanker hello, is there a salt state that can add to the PATH variable of a linux server? I'm trying to globally install composer packages but it doesn't look like the composer salt state supports global installation, so i will just add /vagrant/vendor/bin to the PATH variable
22:45 iggy jankmcjanker: there are environ states
22:47 jankmcjanker ah i see. salt-call environ.get PATH returns the PATH variable
22:53 rem5 joined #salt
22:53 bhosmer_ joined #salt
22:53 knite joined #salt
22:58 eodgooch joined #salt
22:59 LondonAppDev joined #salt
23:01 amcorreia joined #salt
23:06 daemonkeeper joined #salt
23:08 Eureka703 Hey. Im having an issue using git.clone under module.run. I keep getting this error: Module function git.clone threw an exception. Exception: 'module' object has no attribute 'add_http_basic_auth'
23:08 Eureka703 Any thoughts?
23:08 baweaver joined #salt
23:16 Ch3LL Eureka703: whats your sls look like?
23:17 rem5 joined #salt
23:18 Eureka703 Ch3LL: Here is a gist of it. https://gist.github.com/anonymous/676d282e17e15c906971
23:19 Eureka703 Ch3LL: I also have rewritten it to just use git.latest without running under module but I get the same error still.
23:23 bucky forrest: tried a simple foo & bar schema - now getting source file salt://mysql/files/foo.schema not found. I put them under mysql-formula/mysql/files
23:23 Muchoz joined #salt
23:23 ingslovak joined #salt
23:24 forrest bucky, path is wrong right? Should be salt://mysql-formula/mysql/files/foo.schema
23:24 whatapain joined #salt
23:26 bucky forrest: hum. good point I'll try that. Thought for some reason if mysql-formula was mounted as gitfs it was mounted into salt's root, but then I guess you'd get lots of name conflicts.
23:26 whatapain so, the nagios module has a static variable called "PLUGINDIR" that is pointing to /usr/lib/nagios/plugins/ but on centos boxes the PLUGINDIR should point to /usr/lib64/nagios/plugins/.  how can i override a static variable in a module and point it to the correct directory?
23:29 whatapain i see, an issue is already created for this https://github.com/saltstack/salt/issues/10796
23:29 saltstackbot [#10796]title: nagios module needs to be able to change PLUGINDIR | The plugindir variable cant' be modified at the moment but this needs to be:...
23:29 whatapain this seems to fix the issue, ln -s /usr/lib64/nagios /usr/lib/nagios
23:29 JDiPierro joined #salt
23:32 kevinquinnyo1 joined #salt
23:35 keimlink joined #salt
23:36 jaybocc2_ joined #salt
23:39 cyborg-one joined #salt
23:43 jankmcjanker hello, i'm having difficulty updating my PATH variable using the environ state module. I was able to update it using sudo salt-call environ.setval PATH (( path )) and it seems to have worked, but when i use an SLS file and use state.highstate, salt logs show that the state ran, but when i echo $PATH it does not seem to be updated. my sls looks like http://hastebin.com/iqacuhupah.yml . Am I doing something wrong? what can
23:47 jankmcjanker Additonally salt-call environ.items doesn't show that the PATH was updated either
23:47 bhosmer_ joined #salt
23:50 jankmcjanker well i suppose i can cheat and just do cmd.run and set path that way
23:54 cyborg-one joined #salt
23:54 robawt jankmcjanker: are you setting the path perminently in the user you're verifying with?
23:54 robawt a lot of distros define the user's $PATH in .bashrc or .bash_profile that would overwrite what you set in a system wide conf in /etc
23:57 swa_work joined #salt

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