Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2017-05-04

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

All times shown according to UTC.

Time Nick Message
00:06 antpa joined #salt
00:10 rlatimore joined #salt
00:11 woodtablet left #salt
00:12 antpa joined #salt
00:14 Felgar joined #salt
00:20 keldwud joined #salt
00:20 m0nky joined #salt
00:21 Awesomecase joined #salt
00:37 ahrs joined #salt
00:37 greyeax joined #salt
00:43 nick123 joined #salt
01:03 cgiroua joined #salt
01:07 gnord joined #salt
01:09 edrocks joined #salt
01:10 druonysus_ joined #salt
01:39 ashmckenzie joined #salt
01:39 mosen joined #salt
01:49 ilbot3 joined #salt
01:49 Topic for #salt is now Welcome to #salt! <+> Latest Versions: 2016.3.6, 2016.11.4 <+> Support: https://www.saltstack.com/support/ <+> SaltStack Webinar on Carbon, Nitrogen, and Enterprise 5.1 on May 18, 2017 https://goo.gl/PvsOvQ <+> Logs: http://irclog.perlgeek.de/salt/ <+> Paste: https://gist.github.com/ <+> See also: #salt-devel, #salt-offtopic <+> We are volunteers and may not have immediate answers
01:58 ponyofdeath joined #salt
02:03 canci joined #salt
02:04 ahrs joined #salt
02:08 DEger joined #salt
02:13 jas02 joined #salt
02:14 zerocoolback joined #salt
02:18 wangofett joined #salt
02:19 renoirb joined #salt
02:24 wangofett joined #salt
02:36 keldwud joined #salt
02:38 mpanetta joined #salt
02:44 nick123 joined #salt
02:49 tkharju joined #salt
02:52 masber joined #salt
02:55 antpa joined #salt
02:56 evle joined #salt
03:02 seanacais_ joined #salt
03:03 Xenophon1 joined #salt
03:03 Zaunei_ joined #salt
03:04 tcolvin_ joined #salt
03:04 antpa joined #salt
03:06 colttt_ joined #salt
03:06 Kelsar_ joined #salt
03:06 mosen joined #salt
03:06 chadhs_ joined #salt
03:06 Straphka_ joined #salt
03:06 viq_ joined #salt
03:07 simmel_ joined #salt
03:09 Whissi_ joined #salt
03:09 watersoul_ joined #salt
03:10 Vye joined #salt
03:10 kaushal_ joined #salt
03:10 ZombieTwiglet joined #salt
03:10 pocketprotector- joined #salt
03:10 themacproguy joined #salt
03:11 shadoxx_ joined #salt
03:12 shakalaka joined #salt
03:12 Xevian joined #salt
03:12 Sketch joined #salt
03:14 cwright joined #salt
03:15 ekkelett joined #salt
03:17 dezertol joined #salt
03:17 capn-morgan joined #salt
03:18 ChubYann joined #salt
03:18 ecdhe joined #salt
03:19 mavhq joined #salt
03:19 pcn joined #salt
03:19 wendall911 joined #salt
03:21 justanotheruser joined #salt
03:30 flebel joined #salt
03:35 Praematura joined #salt
03:37 paant joined #salt
03:47 ivanjaros joined #salt
03:50 brent_ joined #salt
03:58 mavhq joined #salt
04:06 sp0097 joined #salt
04:11 masber joined #salt
04:11 edrocks joined #salt
04:17 mpanetta joined #salt
04:21 antpa joined #salt
04:25 mosen joined #salt
04:30 antpa joined #salt
04:45 nick123 joined #salt
04:49 wangofett joined #salt
04:54 nick123 joined #salt
04:56 wangofett joined #salt
05:03 wangofett joined #salt
05:09 fracklen joined #salt
05:10 paant joined #salt
05:11 evidence joined #salt
05:11 wangofett joined #salt
05:12 m0nky joined #salt
05:18 wangofett joined #salt
05:22 Bock joined #salt
05:24 nick123 joined #salt
05:26 nick123_ joined #salt
05:40 rdas joined #salt
05:52 jas02 joined #salt
05:54 juntalis joined #salt
06:00 XenophonF joined #salt
06:06 jhauser joined #salt
06:06 antpa joined #salt
06:09 mavhq joined #salt
06:17 colttt joined #salt
06:24 ryoya joined #salt
06:27 do3meli joined #salt
06:27 do3meli left #salt
06:28 aldevar joined #salt
06:31 zerocoolback joined #salt
06:46 fracklen joined #salt
06:48 Qlawy joined #salt
06:53 michaelb I have a master, syndic, a Linux minion (all ubuntu) as well as a Windows. I am regularly getting a not connected (usually syndic) / not responding (all randomly) I have to restart services to get them up again
06:53 michaelb is there something stupid I could be doing?
06:55 Ricardo1000 joined #salt
06:55 whytewolf check your network, check the logs, up your logging level and see if something else is going on, check your versions.
06:56 whytewolf thats pretty much the standard stuff to try first
07:01 o1e9 joined #salt
07:03 armyriad joined #salt
07:04 aidin joined #salt
07:06 DEger joined #salt
07:08 aboe joined #salt
07:16 zerocoolback joined #salt
07:17 mbologna joined #salt
07:20 Xenophon1 joined #salt
07:24 antpa joined #salt
07:26 paant joined #salt
07:26 fracklen joined #salt
07:29 antpa joined #salt
07:32 saltazure joined #salt
07:32 JohnnyRun joined #salt
07:33 rgrundstrom joined #salt
07:33 preludedrew joined #salt
07:37 wangofett joined #salt
07:38 rgrundstrom Good morning everyone.
07:38 antpa joined #salt
07:39 haam3r Morning
07:40 rgrundstrom I have a installed master were I have some sls files in a repos folder... What I would like to do is that use the sls files in /repo as a baseline for all servers and then a /fqdn folder that holds server specifik configuration. So settings in /fqdn is overriding /repos.. Any hints or ideas on how i can do this?
07:41 babilen rgrundstrom: Have a generic state that takes its configuration from pillar and implement the "configuration override" with pillar data merging
07:42 ivanjaros joined #salt
07:49 Rumbles joined #salt
07:51 bdrung_work joined #salt
07:53 rgrundstrom ok, thsnk you
07:54 felskrone joined #salt
08:04 mikecmpbll joined #salt
08:09 pbandark joined #salt
08:16 antpa joined #salt
08:17 cswang joined #salt
08:21 wangofett joined #salt
08:23 _KaszpiR_ joined #salt
08:26 alex-zel joined #salt
08:26 alex-zel hello, I having issues restart services in Centos 7 with service.reload
08:27 alex-zel salt executes the following command: "systemctl reload nrpe", while it should be "systemctl restart nrpe"
08:28 hemebond alex-zel: Wouldn't you use service.restart if you wanted to restart the service?
08:29 alex-zel oh just sew it
08:29 alex-zel is it the same for states? just replace reload with restart?
08:29 hemebond Well, normally you just notify a service (using `watch` or something) and let it take care of the restarting/reloading.
08:31 alex-zel well now i have "reload: True" and it fails in Centos 7
08:31 hemebond Well that tells the service to reload rather than restart.
08:33 alex-zel "By default if a service is triggered to refresh due to a watch statement the service is by default restarted" sorry I missed this in the docs, now it makes sense
08:33 wangofett joined #salt
08:33 hemebond 👍
08:39 Awesomecase joined #salt
08:49 bvcelari joined #salt
08:51 m0nky joined #salt
09:06 cyborg-one joined #salt
09:17 ToeSnacks joined #salt
09:20 N-Mi__ joined #salt
09:22 Terminus joined #salt
09:23 Terminus hello. question, are there any pitfalls to installing minion on the same host as the master?
09:23 hemebond Terminus: Nope.
09:23 anthonyshaw joined #salt
09:25 Terminus hemebond: gotcha. thanks. wanted to specify the config of the master with salt as well.
09:35 nikdatrix joined #salt
09:37 moy joined #salt
09:41 hvn joined #salt
09:41 hvn joined #salt
09:49 yuhl______ joined #salt
09:50 remyd1 joined #salt
09:54 antonw joined #salt
09:54 jhauser joined #salt
09:56 remyd1 Hi. I have cloned a full salt master configuration (including pillars, states... : https://github.com/oxedions/banquise) and I would like to be able switch between my master original configuration and this one regularly. How can I achieve this easily ?
09:59 Reverend my OTG USB cable arrived
10:00 Reverend might stand a chance at getting to those texts on the demolished phone now xD
10:01 Felgar joined #salt
10:17 edrocks joined #salt
10:21 mattl joined #salt
10:22 antpa joined #salt
10:24 golodhrim|work joined #salt
10:26 BlackBishop joined #salt
10:27 BlackBishop anyone can help me debug "[ERROR   ][6539] Got insufficient arguments for grains match statement from master" ? Why I get this on my minions :/
10:27 Ricardo1000 joined #salt
10:27 Reverend because you supplied the function with insufficient arguments
10:27 aidin joined #salt
10:27 Reverend do you know what state is failing?
10:29 hvn joined #salt
10:29 hvn joined #salt
10:30 BlackBishop well, nope, I get that while doing some saltutil.sync stuff
10:30 BlackBishop one sec, lemme see how to reproduce it in my case
10:30 huddy joined #salt
10:34 ople joined #salt
10:36 BlackBishop heh, a simple sync_grains causes it :|
10:38 pbandark Hi. i have bit confusion in mine_function. I can see example given in salt documentation:
10:38 pbandark test:
10:38 pbandark mine_functions:
10:38 pbandark network.ip_addrs:
10:38 pbandark interface: eth0
10:39 pbandark What exactly it will do? Will it fetch ip address of "eth0" interface of all available minions and store it on salt master? In above example, what is "network.ip_addrs"? is it just name given to track mine_function? What is the syntax of mine_function?
10:40 anthonyshaw joined #salt
10:44 aidin joined #salt
10:45 munhitsu_ joined #salt
10:46 BlackBishop so .. when trying to get "everything" (modules,beacons,grains, etc..) up to date .. what should the order of the operations be ? clear_cache, sync_all, refresh_all ?
10:48 mat_ I'm a bit new to salt, I'm migrating from puppet, and I have a state that errors out with: "Error: Invalid formatted command, see debug log for details: <the command>" I tried running it with salt -l debug, but it does not gives any more errors
10:50 wangofett joined #salt
10:52 sp0097 joined #salt
11:05 bd__ mat_: i am also coming from puppet, do you have any resources like the "git for svn users howto" but with puppet and salt?
11:05 aidin joined #salt
11:07 Terminus hello. what's the best way to ensure that a file that contains just one line exists? i'm looking at salt.states.file but if possible, i'd rather just specify the line in the sls file itself.
11:10 antpa joined #salt
11:11 Terminus ah... i'm stupid. states.file.managed has a content parameter.
11:16 lorengordon joined #salt
11:16 mat_ bd__, nope, I'm slowly crumbling through all the docs to get things working
11:33 cyteen joined #salt
11:36 aidin joined #salt
11:36 zerocoolback joined #salt
11:37 zerocoolback joined #salt
11:37 hvn joined #salt
11:39 asyncsec joined #salt
11:52 haam3r so what's the recommended way of doing large file copy's inside a state, ala making sure and iso file is present on the minion
11:52 haam3r I'm guessing using file.managed with salt:// source is not it?
11:53 rgrundstrom haam3r: are we taling MB or GB or TB of data?
11:54 haam3r in my case windows installation iso files so GB-s of data
11:56 rgrundstrom Well if its for the occasonal windows installation I would use file.managed. Question is how often its going to copy the files.
11:59 haam3r well from my testing, the changes comparison happens on the minion so on every state run the iso get's copied over even if no changes are required. I'm currently using the "replace: False" paramter so the copy would not happen every time
12:00 rgrundstrom On the other hand im more used with large scale stuff like SSD - flashdrives. or terabytes och data as fast as possible using dedicated hardware with Infiniband connections.
12:00 haam3r but that means if at some point i actually swamp out the iso files I would have to do some manual things like removing the destination iso files
12:01 aidin joined #salt
12:02 nikdatrix joined #salt
12:02 rgrundstrom I might missunderstand you here but if you change the source then the target should be owerwritten with the new data?
12:03 rgrundstrom And not have to do manuall work.
12:09 haam3r the "- replace: False" tells the minion that only copy the file from the master if it does not exist and do not check for changes. If I do not use it, the file copy happens every state run, because the comparison happens locally between the minion
12:09 haam3r s/between/on/
12:10 zerocoolback joined #salt
12:14 lorengordon joined #salt
12:14 rgrundstrom Oh ok... Hmm well if I wanted to sync files on a Windows system id use robocopy from server to client or vice versa. If its a Linux system an rsync would be prefered. But the best way to do this using Salt is unknwon for me.
12:15 impi joined #salt
12:16 amcorreia joined #salt
12:18 rgrundstrom unknown* Jesus.... I need to start spellchecking myself before i send stuff.
12:18 thinkt4nk joined #salt
12:19 Terminus rgrundstrom: i'm not sure but maybe check out https://docs.saltstack.com/en/latest/ref/states/all/salt.states.file.html#salt.states.file.recurse
12:24 bwellsnc joined #salt
12:33 asyncsec joined #salt
12:35 haam3r ok thanks rgrundstrom
12:35 hvn joined #salt
12:36 colttt joined #salt
12:40 renoirb_ joined #salt
12:40 ssplatt joined #salt
12:42 edrocks joined #salt
12:47 aidin joined #salt
12:49 numkem joined #salt
12:50 brasko XenophonF was helping me yesterday
12:50 brasko I'm trying to run N jobs on X minions where N > X
12:51 brasko I'm wondering, do I have to kick off N jobs, or is there a way to kick off just 1
12:51 brasko ?
12:51 brasko and have salt do the queuing
12:55 shoemonkey joined #salt
12:57 cgiroua joined #salt
13:00 lorengordon joined #salt
13:01 Xenophon1 oh hai
13:01 XenophonF stupid IRC client
13:01 XenophonF oh hai
13:02 brasko nice
13:02 brasko I got what you asked me working yesterday
13:02 XenophonF fantastic!
13:02 brasko sudo salt '*' cmd.script salt://orch/ls.sh template=jinja pillar='{"path": "/home"}'
13:02 brasko works just great
13:02 XenophonF wonderful
13:03 brasko so at this point, I can run 1 job, on a node -- and I could do that N times
13:03 XenophonF yup
13:03 brasko I assume salt queues those commands?
13:03 brasko and runs them when a minion is available?
13:03 XenophonF Hm, good question.
13:04 XenophonF hang on let me RTFM real quick
13:04 t0m0 joined #salt
13:04 brasko ouch! meeting i guess :( be back
13:05 XenophonF np
13:05 XenophonF brasko: when you get back, take a look at https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.schedule.html
13:06 XenophonF although now that i think about that, i'm having some problems with the scheduler on windows
13:11 wangofett joined #salt
13:15 swills joined #salt
13:17 DanyC joined #salt
13:17 wangofett joined #salt
13:19 Praematura joined #salt
13:20 golodhrim|work joined #salt
13:24 bvcelari joined #salt
13:27 aidin joined #salt
13:30 wangofett joined #salt
13:41 wangofett joined #salt
13:42 LordZedd joined #salt
13:42 SaucyElf joined #salt
13:47 canci joined #salt
13:50 greyeax joined #salt
13:51 shoemonkey joined #salt
13:57 cwright joined #salt
14:00 wangofett joined #salt
14:01 englishm_llnw joined #salt
14:07 DanyC joined #salt
14:10 brasko XenophonF: ok, looking
14:10 brasko I can run the commands from linux if that's what you are saying
14:10 speedlight joined #salt
14:13 cscf haam3r, late response, but you should add - keep: True to the file.managed; then it will cache the file in /var/cache.
14:16 dubg joined #salt
14:17 ronnix joined #salt
14:18 dubg Hi there, anyone know if i can display all timer during an state.apply (compilation, execution, ect..) ?
14:18 haam3r cscf: yeah I kinda do not want to keep a copy of the iso there
14:19 cscf haam3r, where do you want to keep it?  Only in place on the minion?  You could use rsync.syncronized for that
14:20 cscf haam3r, in fact, I think it's generally not recommended to use file.managed for large files.
14:21 renoirb joined #salt
14:21 haam3r yeah only in one place on the minion and do the copy from master operation only if there are changes...so compare the file on the minion with the file on the master and copy onlyif differences found
14:22 haam3r cscf: yeah I kinda thought that as well...thats why I'm discussing other options with people smarter than me :D
14:24 dubg i'm asking this because i'm making a comparaison between Salt and Puppet, and puppet is faster for checking, and i have to find where is the trouble
14:25 cscf dubg, you mean like performance profiling?
14:26 dubg there is an option in puppet called --summarize which show that
14:26 khaije1 Is there a way to prove where an unaccepted salt-key came from?
14:26 khaije1 or at least audit the IP where it came from?
14:27 dubg cscf: hm something like that, to the begin (compilation of highstate) to check on the minion (follow by the report)
14:27 mpanetta joined #salt
14:27 cscf dubg, so, a normal state run will tell you how long a given state takes, I'll look into compilation
14:28 brasko XenophonF: i'm not totally sure how the link you provided helps me out
14:29 brasko I'm looking for a solution that allows me to say "Run N jobs and wait for them to complete"
14:29 dubg cscf: i need to look the things happened before the states
14:29 dubg the workflow
14:30 fracklen joined #salt
14:30 dubg but i think the thing who i need doesn't exist and i will have to look into the code
14:30 brousch__ joined #salt
14:31 _JZ_ joined #salt
14:35 zerocoolback joined #salt
14:36 tapoxi joined #salt
14:36 Brew joined #salt
14:38 jas02 joined #salt
14:38 renoirb joined #salt
14:39 navik Hello everyone, I talked about my Salt issue two days ago... To recap, a master on CentOS6.9 64bit, with working minions, also on centOS6.9 64bits. I tried to install a minion on a CentOS6.9 32bit.
14:40 navik Running "salt-minion -l debug" gives the standard output until "Re-using AsyncAuth for ('/etc/salt/pki/minion', 'test-minion_32', 'tcp://..............')", then simply stops. ps aux | grep salt does give a salt-minion process running, though
14:41 mikecmpb_ joined #salt
14:41 navik Then, on the master's side, the minion key is received and accepted correctly. But, a test.ping fails with the message " Minion did not return. [No response]"
14:42 sarcasticadmin joined #salt
14:43 navik haam3r told that an architecture difference between master and minion should in no way interfere with communication between the two; is there any official information about that ?
14:44 navik Also, the minion log file only contains "2017-05-04 10:36:29,197 [salt.utils.verify][WARNING ][4001] Insecure logging configuration detected! Sensitive data may be logged."
14:45 navik Any help concerning this issue would be really appreciated.
14:48 Tantagel joined #salt
14:51 EvaSDK navik: I run a master on regular x86_64 and have a minion on arm, it works just fine
14:51 EvaSDK out of the box
14:52 tapoxi joined #salt
14:52 jdipierro joined #salt
14:52 EvaSDK and the log warning is because you set it to debug or something which could output sensitive information, just go with info and it should disappear
14:52 jdipierro joined #salt
14:57 fracklen joined #salt
15:01 numkem joined #salt
15:01 sp0097 joined #salt
15:04 mpanetta joined #salt
15:06 evle1 joined #salt
15:07 brent_ joined #salt
15:10 CrummyGummy joined #salt
15:11 jas02 joined #salt
15:12 mikecmpbll joined #salt
15:13 Master_Book joined #salt
15:14 jas02 joined #salt
15:14 Master_Book @all Hello, you know a good book Salt Stack ?
15:14 hashwagon joined #salt
15:16 aidin joined #salt
15:16 hashwagon Hey guys, I had a salt/tomcat question I posted on stackoverflow. Wondering if anyone could chime in on it: https://stackoverflow.com/questions/43770756/saltstack-tomcat-deployment-tomcat-war-deployed-error?noredirect=1#comment74585520_43770756
15:17 JPT joined #salt
15:17 PatrolDoom joined #salt
15:17 PatrolDoom joined #salt
15:18 navik Actually, there are no error in the logs. When switching from ZeroMQ to tcp transport, an error appears in the master log  "TypeError: 'int' object is unsubscriptable". I am using the 2016.11.4 version
15:20 Praematura joined #salt
15:22 legreffier joined #salt
15:22 brasko I've got to say, I'm surprised with all the documentation, that I can't figure out how to run N jobs on X machines with salt in a simple fashion
15:23 brasko It's pretty trivial to run some jobs on all the active machines, or a subset of them
15:24 LeProvokateur joined #salt
15:27 sp0097 joined #salt
15:29 Praematura joined #salt
15:29 speedlight joined #salt
15:35 EvaSDK brasko: you mean run jobs in batch of X ?
15:35 EvaSDK X being the number of nodes running the command at the same time
15:39 censorshipwreck joined #salt
15:40 brasko I require running like 1000 jobs on 100 nodes
15:40 brasko and hopefully having salt queue, and run them on available minions
15:40 brasko so 100 run at a time, if each of the jobs are balanced
15:40 brasko can't seem to figure it out.
15:43 Hetman left #salt
15:49 Tantagel joined #salt
15:51 EvaSDK isn't the -b option good enough for your case ?
15:51 EvaSDK hum, probably not
15:52 brasko it's sort of the opposite
15:52 brasko but close
15:52 brasko I need something like -B lol
15:53 cyteen joined #salt
15:54 dyasny joined #salt
15:56 jas02 joined #salt
15:58 dezertol joined #salt
15:58 Tantagel joined #salt
16:02 ivanjaros joined #salt
16:03 Xenophon1 joined #salt
16:05 DanyC joined #salt
16:05 jas02 joined #salt
16:07 Reverend "Git remote to use. If this state needs to clone the repo, it will clone it using this value as the initial remote name. If the repository already exists, and a remote by this name is not present, one will be added."
16:07 Reverend the documentation lies
16:08 aldevar left #salt
16:08 DanyC joined #salt
16:09 DanyC_ joined #salt
16:09 jdipierro joined #salt
16:09 saltazure joined #salt
16:10 dyasny joined #salt
16:13 danielfallon joined #salt
16:13 oida joined #salt
16:23 Cidan joined #salt
16:23 prg3 joined #salt
16:24 EvaSDK if someone from salstack-formulas is around, please merge https://github.com/saltstack-formulas/nginx-formula/pull/129
16:24 saltstackbot [#129][OPEN] Added dhparam file creation | In order to improve security and ease of use, added creation/generation...
16:26 renoirb joined #salt
16:27 pipps joined #salt
16:27 pipps joined #salt
16:27 gtmanfred EvaSDK: did you read gravyboats comment? it doesn't look like you added an example to the pillar.example
16:28 wangofett joined #salt
16:28 woodtablet joined #salt
16:28 gtmanfred or that they added it
16:28 EvaSDK oh right
16:31 aphor joined #salt
16:31 jas02 joined #salt
16:33 EvaSDK gtmanfred: while at it, would you say both dh_keygen and dh_keysize make sense as separate pillar entries ?
16:33 nidr0x joined #salt
16:33 EvaSDK of wouldn't it be better to just use dh_keysize as a hint that a key must be generated if dh_content isn't provided ?
16:33 jas02 joined #salt
16:35 impi joined #salt
16:38 antonw joined #salt
16:38 munhitsu_ joined #salt
16:40 mattl joined #salt
16:40 EvaSDK gtmanfred: in the meantime https://github.com/saltstack-formulas/nginx-formula/pull/153
16:40 saltstackbot [#153][OPEN] Add dhparam creation | Completed from #129.
16:42 moy joined #salt
16:42 m0nky joined #salt
16:42 Awesomecase joined #salt
16:43 ToeSnacks joined #salt
16:44 Trauma joined #salt
16:47 cyteen joined #salt
16:47 ople joined #salt
16:50 asyncsec joined #salt
16:56 Inveracity joined #salt
16:59 cyteen joined #salt
17:14 LeProvokateur joined #salt
17:14 hashwagon anyone here have tomcat war deployment experience with salt?
17:15 DammitJim joined #salt
17:17 Kelsar can somebody push me in the general direction for a scenario, basically i need to trigger a state on minion, after a state changed something on another minon
17:18 whytewolf Kelsar: orchestration
17:19 bvcelari joined #salt
17:19 tapoxi so I need to echo the ip address into a file via jinja, but grains['ipv4'] contains the loopback address. my hacky way of trying to avoid 127.0.0.1 using a jinja loop works, but it outputs with a ton of whitespace. suggestions?
17:20 whytewolf Kelsar: https://docs.saltstack.com/en/latest/topics/orchestrate/orchestrate_runner.html#orchestrate-runner
17:21 whytewolf tapoxi: a. don't use grains us the network module. which is cleaner. and b. http://jinja.pocoo.org/docs/2.9/templates/#whitespace-control
17:21 XenophonF brasko: you should take a look at the minion job scheduler, next
17:21 XenophonF i'm not sure if it runs stuff in parallel, but it might
17:22 Kelsar whytewolf: can i update mine in between?
17:22 XenophonF you can also use salt to schedule jobs for execution via salt.states.at or salt.states.cron
17:22 whytewolf Kelsar: you can do pretty much anything saltstack does. and yes you can update mine
17:23 ken_ joined #salt
17:23 tapoxi whytewolf: kick off the network exec module from my jinja?
17:25 whytewolf tapoxi: yes something like {% set known_ip = salt.network.interface('eth0') -%}
17:25 Kelsar whytewolf: ty, now i need to think about how i need to order stuff
17:25 tapoxi whytewolf: awesome thanks
17:26 brd Will the minion never refresh its pillar on its own?
17:27 whytewolf brd: sometimes it does i never wait personally.
17:27 brd I suppose I could schedule it like to do the highstate runs..
17:27 brd whytewolf: ok
17:28 whytewolf brd: generally i run this anytime i make changes https://github.com/whytewolf/salt-phase0-states/blob/master/orch/salt-update.sls
17:28 whytewolf plan is to when i get this on a live server to call it from a github hook
17:29 brd nice
17:29 brd I will go look those up
17:29 pbandark1 joined #salt
17:29 whytewolf i do need to add the runner version of saltutil.sync_all
17:30 whytewolf oh well
17:31 nixjdm joined #salt
17:31 _KaszpiR_ joined #salt
17:33 gtmanfred brd: pillars update on a highstate run
17:33 gtmanfred right here, https://github.com/saltstack/salt/blob/develop/salt/state.py#L3148
17:34 gtmanfred load_dynamic is called under call_highstate https://github.com/saltstack/salt/blob/develop/salt/state.py#L3628
17:35 * gtmanfred goes back into hiding
17:35 pocketprotector- joined #salt
17:35 brd gtmanfred: hmm, ok. thanks
17:37 amcorreia joined #salt
17:39 hashwagon Updated the question if anyone happens to have tomcat+salt experience: https://stackoverflow.com/questions/43770756/saltstack-tomcat-deployment-tomcat-war-deployed-error
17:39 ChubYann joined #salt
17:40 salty joined #salt
17:43 edrocks joined #salt
17:45 Cottser joined #salt
17:45 sjorge joined #salt
17:47 mishanti1 joined #salt
17:49 greyeax joined #salt
17:58 Reverend joined #salt
18:01 whytewolf hashwagon: make sure salt can access the tomcat manager page. that is what it uses in order to deply the war.
18:02 jas02 joined #salt
18:03 whytewolf hashwagon: basicly make sure you follow the tomcat manager part of this https://docs.saltstack.com/en/latest/ref/states/all/salt.states.tomcat.html#configuring-tomcat-manager
18:04 hashwagon whytewolf, the minion:8080/manager page is accessible. I setup the required pillar for the login and my output of salt '*' pillar.items shows my username and passwd. For some reason it's showing my password above my username in the output though.
18:06 whytewolf well ok it is accessable to the outside. that doens't mean you can get to it from localhost
18:07 hashwagon Good point. Also to add I know the credentials are working since I can login to manager.
18:09 whytewolf iirc [been a while since i touched tomcat] you may also need to enable the deploy part of the manager.
18:09 hashwagon okay, I can login with elinks 127.0.0.1:8080/manager, but not with localhost:8080/manager
18:09 fracklen joined #salt
18:10 hashwagon scratch that elinks localhost login is working too
18:11 whytewolf also just for testing run your command on the minion with salt-call -l debug <salt command>
18:11 whytewolf might give more insite into what is failing
18:18 brasko XenophonF: The minion job scheduler would mean we would have to specify the jobs per minion?
18:18 brasko I was more hoping I could say "I have J jobs, that do T task, and M minions, run them appropriately"
18:19 aphor I'm having issues extending community formula states
18:19 aphor https://gist.github.com/aphor/3a8a3b39b91c8c04e10dcda3b54106c8
18:19 Reverend joined #salt
18:19 aphor It seems like there is no way to exend a state in one saltenv from an sls in another saltenv
18:21 aphor I *really* want to be able to do this.
18:21 patrek_ joined #salt
18:23 brd gtmanfred: in cause you are lurking, I just opened a one character pull request ;) #41079
18:26 pipps joined #salt
18:26 hashwagon whytewolf: thanks, I've added the output to my question - https://stackoverflow.com/questions/43770756/saltstack-tomcat-deployment-tomcat-war-deployed-error
18:28 CrummyGummy joined #salt
18:29 brd gtmanfred: thnx
18:29 tercenya joined #salt
18:30 gtmanfred :)
18:30 whytewolf hashwagon: also don't forget to check the tomcat log. maybe there are clues there
18:30 cyborg-one joined #salt
18:30 whytewolf such as 404 or 401 errors
18:35 thinkt4nk what is the lsb in lsb_distrib grain keys supposed to express? Least significant byte doesn't make sense to me
18:36 thinkt4nk doesn't make sense to me in this context
18:39 MTecknology gtmanfred: sorry for the mess with that PR :(  If it helps, v2 is much prettier
18:39 whytewolf linux standard base
18:40 whytewolf thinkt4nk: https://wiki.linuxfoundation.org/lsb/start
18:40 thinkt4nk whytewolf thank you!
18:43 aphor Who loves to extend: ??
18:47 nixjdm joined #salt
18:48 XenophonF sorry brasko i'm at the limit of how well i understand your particular problem
18:49 jas02 joined #salt
18:50 brasko XenophonF: yeah :(
18:50 brasko it's oK
18:50 brasko i don't think it's possible with salt to treat salt like a job queue
18:51 brasko I'm basically trying to use salt like celery
18:51 nidr0x joined #salt
18:52 brasko https://groups.google.com/forum/#!msg/salt-users/HvZH395i958/AzLf0z3RBgAJ
18:55 censorshipwreck joined #salt
18:58 wangofett joined #salt
18:58 whytewolf aphor: your errors in that screen shot have nothing to do with the extend. it is the include that is failing. you can't include across enviroments
19:00 aphor whytewolf: OK
19:02 saltazure joined #salt
19:02 aphor I also tried include with base: apache syntax
19:03 whytewolf then an error in that would be more fitting to what the issue is
19:03 aphor let me work up a better gist..
19:03 aphor thx
19:03 lookcrabs joined #salt
19:03 wangofett joined #salt
19:07 gtmanfred brasko:might be worth while checking out how I added insert runner and process runner to the queue runner https://github.com/saltstack/salt/blob/develop/salt/runners/queue.py#L243-L305
19:07 gtmanfred you would need to make a queue module for the minion to interact with, but you could then insert lowdata for modules/states from the master, and then use your minions to process that data
19:08 gtmanfred that is how I would implement a celery type thing in salt
19:08 gtmanfred but celery will be better, because you spin up a certain number of threads and they all handle the locking on the queue better probably
19:08 brasko yeah, celery doesn't support windows!
19:08 gtmanfred and each thread automatically pulls another job when it finishes, instead of having to pull jobs in bulk
19:09 gtmanfred ahh well, yeah multiprocessing in windows is a pain, because it doesn't have os.fork
19:09 Trauma joined #salt
19:09 gtmanfred and even we are slower on windows because you have to reload all module because they can't be pickled and be available in the new minion process
19:10 whytewolf * in windows is a pain because it doesn't support $i
19:11 nidr0x joined #salt
19:11 aphor @whytewolf: https://gist.github.com/aphor/3a8a3b39b91c8c04e10dcda3b54106c8
19:12 aphor extend doesn't seem to be able to address states imported from another saltenv :(
19:12 gtmanfred https://github.com/saltstack/salt/blob/nitrogen/salt/minion.py#L1375 minion_instance here is only None on windows systems iirc
19:15 tercenya joined #salt
19:16 whytewolf aphor: you list the sls file as your extend. you are not extending the sls file you are extending a state. you need to use the state id not the sls id
19:17 tercenya joined #salt
19:17 aphor doh!
19:18 whytewolf although not sure if it will take into account the env still.
19:18 aphor let me try that
19:19 aphor MEGO
19:19 aphor need coffee too
19:20 Xenophon1 joined #salt
19:20 whytewolf hey gtmanfred just saw they are tagging 2016.11.5 soon. was there a regression? seems kinda soon for another release of 2016.11
19:22 DammitJim joined #salt
19:31 aphor @whytewolf that worked.
19:32 whytewolf kewlieo
19:32 felskrone joined #salt
19:33 ivanjaros joined #salt
19:33 gtmanfred whytewolf: https://github.com/saltstack/salt/issues/40889
19:33 saltstackbot [#40889][OPEN] salt-master 2016.11.4 crashes on CentOS 7 with error "RSA key format is not supported" | Description of Issue/Question...
19:33 gtmanfred whytewolf: https://github.com/saltstack/salt/issues/40845
19:33 saltstackbot [#40845][MERGED] /jobs call is broken in 2016.11.4 | Description of Issue/Question...
19:34 gtmanfred those two are the main ones, but there were some other fixes that we wanted to get in
19:37 pipps joined #salt
19:37 pbandark joined #salt
19:40 whytewolf ahhh yeah that RSA key format one is a big one. broke backwords compatibility iirc
19:41 gtmanfred the jobs was the biggest one, we have a lot of customers that use that
19:43 whytewolf i bet. that is one of the biggest and best features of salt-api
19:43 jas02 joined #salt
19:47 nixjdm joined #salt
19:50 PatrolDoom joined #salt
19:54 skatz joined #salt
19:56 fracklen joined #salt
19:59 edrocks joined #salt
19:59 hashwagon whytewolf: regarding tomcat, nothing off was found in the logs. I'm able to manually deploy my war through http://localhost:8080/manager. For anyone else interested I've written a question on stackoverflow - https://stackoverflow.com/questions/43770756/saltstack-tomcat-deployment-tomcat-war-deployed-error
19:59 hashwagon The catalina logs, to be more specific.
20:01 skatz Hey all - I'm running Salt locally on a Windows 2016 box and am getting a "'psget.avail_modules' is not available" error: https://gist.github.com/phsteve/23e56c0676bd0a90af1fb17a4fab9378 despite the psget Powershell module already being successfully installed. any ideas?
20:01 mikecmpbll joined #salt
20:04 whytewolf skatz: try  C:\salt\salt-call.bat --local -l debug psget.avail_modules
20:04 skatz OK thanks, gist coming shortly
20:05 whytewolf also add the output of salt-call test.versions
20:06 skatz whytewolf here's both: https://gist.github.com/phsteve/73d5fac7979087412937a3039cd2677c
20:08 _KaszpiR_ joined #salt
20:08 whytewolf odd
20:09 skatz Full disclaimer: I'm kind of a noob to Windows and Salt so I may be missing something obvious.
20:09 skatz When I search for commands in powershell with "psget" in them, it lists "Get-PsGetModuleHash" and "Get-PsGetModuleInfo"
20:11 aneeshusa joined #salt
20:11 whytewolf it is acting like the module isn't even reconized.
20:11 whytewolf as if it has some kind of python error. but i don't see a python error in your debug output
20:11 pipps joined #salt
20:12 hashwagon whytewolf: Thanks, I've updated it: https://stackoverflow.com/questions/43770756/saltstack-tomcat-deployment-tomcat-war-deployed-error
20:12 pipps joined #salt
20:12 whytewolf and honestly. to be fair i know next to nothing about windows. I'm generally a linux guy. i'm just looking at this from a salt perspevtive
20:12 skatz right same here :)
20:13 skatz other salt commands seem to work fine, like chocolatey and file etc, if that's helpful
20:14 whytewolf skatz: try another psget function like list_modules
20:14 Trauma joined #salt
20:14 asyncsec joined #salt
20:14 skatz Same error with that
20:15 whytewolf also you might want to think about updating to .4 there were some tweeks to that module between .1 and .4 that might help shead light
20:15 brasko left #salt
20:15 skatz OK I can give that a shot
20:18 whytewolf hashwagon: i honestly don't know. step up to -l trace see if you can get the comamnd that salt is trying to hit tomcat with. also with out knowing the stateus of the attempts in tomcat. this is a difficult problem when you only see one side of it
20:21 hashwagon whytewolf: thanks, I understand and I appreciate your feedback. I'll update the channel if I find anything useful.
20:22 haam3r joined #salt
20:25 skatz whytewolf lots more info after the upgrade: https://gist.github.com/phsteve/9cc0ce14e713de41c7d59c544b9dca00 something about a NuGet issue? I can try to upgrade chocolatey to address that perhaps
20:27 whytewolf that might be the best coarse of action
20:27 whytewolf also see if there is a NuGet provider installed. i wonder if that was a change in psget
20:28 skatz upgraded chocolatey from 0.10.5 to 0.10.6-beta, same error
20:29 maysarahina joined #salt
20:29 whytewolf have you tried the command it is tring?
20:29 whytewolf Powershell -NonInteractive -NoProfile "Find-Module | ConvertTo-Json -Depth 2"
20:30 Praematura joined #salt
20:30 skatz Same nuget error
20:30 skatz let me see if i can upgrade nuget too
20:31 whytewolf alright. yeah once that command is able to run salt should be able to run the module
20:34 pipps joined #salt
20:41 hemphill joined #salt
20:47 nixjdm joined #salt
21:05 druonysus joined #salt
21:17 gmoro_ joined #salt
21:23 snarfy^ joined #salt
21:25 snarfy^ hey dudes. i was in here before and someone had a version of grains['ip_interfaces']['eth0'][0]
21:25 snarfy^ that wasn't eth0-specific
21:25 snarfy^ i ended up using that convention an awful lot..
21:25 wangofett joined #salt
21:32 nikdatrix joined #salt
21:32 whytewolf you mean something like salt.network.ipaddrs(cird='10.0.0.0/80')? or salt.network.interface('eth0') ect.
21:32 wangofett joined #salt
21:34 snarfy^ yeah. im just realizing tho that i'm gonna get the docker lxcbr0 at 10.0.3.0/24
21:35 snarfy^ hrm
21:36 snarfy^ ooh but type=private gets me what i want
21:36 snarfy^ muy excellente
21:38 DEger joined #salt
21:40 quay joined #salt
21:41 MajObviousman snarfy^: summarize for me what you're going to do?
21:41 MajObviousman because I have need of such an idiom as well
21:43 snarfy^ tell u in a sec when i get it workin
21:44 Edgan joined #salt
21:44 sp0097 joined #salt
21:47 nixjdm joined #salt
21:47 exegesis joined #salt
21:49 snarfy^ MajObviousman, not sure, still. if i do {% set ip = salt['network.ip_addrs'](cidr='10.0.0.0/8') %} I get all 10.* networks ips
21:49 snarfy^ but our network has too many subnets
21:50 snarfy^ seems like I need to know a certain leve of CIDR before i can narrow down the ip i want
21:50 snarfy^ but that feels like knowing the ip before i know the ip
21:50 whytewolf not really. you should at least know the subnet
21:51 snarfy^ well say i have 10.1 10.10 and 10.11
21:51 snarfy^ and docker installations use a 10.0.1
21:51 whytewolf 10.11.0.0/16
21:51 snarfy^ i want to grab the 10.1.0.0/16 10.11.0.0/16 etc
21:51 snarfy^ but how will I know if the minion is on 10.1 or 10.11?
21:52 whytewolf you don't know what your subnets are?
21:52 snarfy^ I do..
21:52 snarfy^ sorry i know i must be being obtuse.
21:53 exegesis joined #salt
21:53 snarfy^ but if i write a salt state that says grab the ip of 10.1.0.0./16 and then it runs on 10.11 it won't get the IP?
21:54 whytewolf that is correct. because 10.1.0.0/16 would only grab ips from 10.1.0.0 to 10.1.254.254
21:54 snarfy^ right so how do I write a state that accounts for all these goddawful 10-something subnets we have :)
21:54 snarfy^ i'm goin DNS configs
21:55 whytewolf well 10.0.0.0/8 grabs all 10-something ips
21:55 snarfy^ so I sorta want to be able to have a server in every region if not subnet
21:55 snarfy^ right exactly :) btu we use docker and i don't want the 10.0.3.1 ip ;)
21:55 snarfy^ sorry i'll stop. im sure i can figure it out
21:56 whytewolf okay so you want to start at 10.1.0.0 and 10.254.0.0 i wonder if there is a number between 16 and 8 that would do that ...
21:56 michaelb I'm trying to understand orchestration! I seem to be missing something - I have a state that runs on the master and prepares some files into a salt:// path - I also have a state on another machine that does a file.recurse to download them I can't seem to figure out how to write the require
21:56 whytewolf michaelb: can to share what you have so far? see if i can help
21:57 michaelb give me a mo...
21:57 whytewolf there is tricks to learning the context levels that it takes to get things like this going
21:59 jas02 joined #salt
21:59 michaelb https://gist.github.com/theMichaelB/0cb30103c093d67d47b144ffd9c766d2
22:00 whytewolf snarfy^: part of the problem is the way your subnets are divided btw. 10.0 will most likely be in the same subnet as 10.1 with anything higher then 16
22:00 michaelb I imagine it won't be the last daft question I ask :)
22:00 whytewolf michaelb: what is that? doesn't look like orchestration
22:01 whytewolf is that just the 2 states?
22:01 MajObviousman snarfy^: I think I see what you are saying
22:01 michaelb have I really missed something then!
22:02 michaelb (ahh I see what I did - the documentation link threw me)
22:03 michaelb I'll try again, and likely be back :)
22:03 whytewolf michaelb: alright. most likely will be here.
22:03 whytewolf unless it is after my bedtime... then i will probley still be here or dead
22:04 pipps joined #salt
22:05 michaelb well, don't die just yet, I'll likely need help at some point ;)
22:08 whytewolf snarfy^: yeah even for /15 10.0 and 10.1 are in the same subnet. although 10.10 and 10.11 can both be grouped together with /15
22:08 snarfy^ heh thanks :)
22:08 snarfy^ i might already be overthinking it
22:09 cyteen joined #salt
22:09 * snarfy^ notices his AWS xenial VMs are using eth0 anyway
22:09 whytewolf could just loop through the results and check if they are in 10.1.0.0/16 and remove them if they are
22:10 * whytewolf wishes his hardware was using ethX still
22:10 whytewolf was so much easier back then
22:10 snarfy^ simpler times :D
22:10 whytewolf although i would still most likely move to subnet based tracking.
22:11 whytewolf as i didn't put my instnce ips even close to the same subnet as my main hardware subnets
22:11 whytewolf I use 10.x.x.x for hardware and 172.x.x.x for virtual hardware
22:12 whytewolf and good ol 192.168.x.x for my living room
22:13 snarfy^ we got a lil of everything sprinkled all over..
22:13 snarfy^ makes it nice and challenging
22:13 whytewolf someone should shoot the network guy out of a cannon
22:13 whytewolf there should be clear distintion between subnets
22:13 Trauma joined #salt
22:14 snarfy^ yeah well.
22:14 * snarfy^ starting to answer those recruiter emails ;)
22:14 whytewolf lol
22:14 whytewolf i know the feeling
22:19 michaelb and shockingly, when I read the right documentation it all works - and makes sense - who knew :)
22:21 whytewolf lol michaelb it happens
22:22 michaelb thanks for the nudge in that direction though :)
22:25 anthonyshaw joined #salt
22:26 ConnorCG Anyone using gitfs with their salt master behind a proxy, with no route to the internet?
22:27 ConnorCG I configured corkscrew and can git clone with ssh links from the shell, but I get connection timed out when running salt-run fileserver.update
22:31 Praematura joined #salt
22:35 XenophonF joined #salt
22:36 pipps joined #salt
22:39 VR-Jack-H joined #salt
22:40 justanotheruser joined #salt
22:41 Trauma joined #salt
22:53 DEger joined #salt
22:53 DEger joined #salt
22:59 s4t0r1 joined #salt
23:05 quay hi.  i'm having issue upgrading from salt 2015.5.10 to 2016.11.4.  In 2015.5.10, we have a external grain that tries to determine the grain for the datacenter.  However, after the upgrade, it can't determine it
23:06 quay manually running the script, works fine.  sample output {'datacenter': 'iad1', 'datacenter_name': 'yeast'}
23:07 quay is there anything special that salt does with python?
23:13 edrocks joined #salt
23:14 ssplatt joined #salt
23:15 hemebond Nope.
23:15 hemebond If it's a custom grain then try running the script from salt to see what it does.
23:16 whytewolf or at the very least try syncing the grain again
23:22 DEger joined #salt
23:27 quay :)  looks like we have multiple grain
23:27 quay found the issue
23:28 hemebond 👍
23:29 jdipierro joined #salt
23:32 nikdatrix joined #salt
23:36 Xenophon1 joined #salt
23:49 michaelb If I have a variable in Powershell, running on a minion, how can I get that fed back into a grain / mine etc?
23:50 wangofett joined #salt
23:51 michaelb (the script is initiated as part of a state)

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