Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2014-11-03

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

All times shown according to UTC.

Time Nick Message
00:02 alex-mesos joined #salt
00:11 crocket joined #salt
00:12 crocket I removed a virtual machine with 'salt-run virt.purge vm-name' and recreated a new VM with the same name. I got "[salt.master      ][ERROR   ] Authentication attempt from vm-name failed, the public keys did not match. This may be an attempt to compromise the Salt cluster."
00:13 crocket 'salt-run virt.init vm-name 2 1024 salt://vm-name.qcow2' fails to preseed the master with that error.
00:13 crocket Why?
00:13 aurynn the old key probably didn't get purged
00:13 crocket Is the previous minion public key for vm-name cached and reused?
00:13 aurynn but that's just a guess
00:13 crocket aurynn, Then, it's a problem with salt-run virt.purge
00:14 aurynn Sure?
00:15 crocket aurynn, I only suspect virt.purge at this point.
00:15 aurynn well, it's a good place to start debugging
00:18 crocket aurynn, I'm recording debug logs from master and minion simultaneously.
00:24 badon joined #salt
00:35 skullone are many people using Halite?
00:48 crocket skullone, Halite is dead.
00:49 crocket skullone, People will not use Halite.
00:49 bhosmer joined #salt
00:50 emostar is there a way to view the minion logs from the master while doing an operation like state.highstate?
00:50 aurynn I require additional attentions
00:50 aurynn sigh
00:50 aurynn ww
00:57 crocket Is gnutls-bin required by salt virt runner?
00:57 crocket I mean libvirt.keys
00:58 crocket emostar, I think you need to ssh into the minion.
01:02 skullone crocket: is there an alternative gui that is being planned?
01:02 crocket skullone, None
01:04 meylor joined #salt
01:07 aqua^mac joined #salt
01:07 rojem joined #salt
01:07 thayne joined #salt
01:15 halfss joined #salt
01:15 rojem joined #salt
01:21 MrFuzz joined #salt
01:22 bhosmer joined #salt
01:27 bezeee joined #salt
01:34 thayne joined #salt
01:48 nam_ joined #salt
01:49 aynik joined #salt
01:56 nam_ joined #salt
01:59 BossR joined #salt
02:05 ghanima joined #salt
02:05 elfixit joined #salt
02:09 gwmngilfen joined #salt
02:13 badon joined #salt
02:15 hgl joined #salt
02:17 Nexpro joined #salt
02:20 hgl left #salt
02:22 SheetiS joined #salt
02:28 TyrfingMjolnir joined #salt
02:31 rojem joined #salt
02:57 TyrfingMjolnir joined #salt
02:58 crocket How can I make salt pull a git repository on an existing directory?
02:59 aurynn what?
03:00 aurynn git.latest is probably what you want still
03:03 tinytub joined #salt
03:03 crocket aurynn, git.latest fails to execute if target is not empty.
03:03 crocket force: True deletes the target directory.
03:04 aurynn force_checkout
03:04 diehard_ joined #salt
03:04 crocket aurynn, force_checkout:True doesn't work.
03:04 aurynn what are you trying to do
03:04 aurynn ?
03:07 crocket aurynn, I want to back up and restore jenkins
03:07 crocket ubuntu jenkins packages stores configurations in /var/lib/jenkins
03:10 pdayton joined #salt
03:10 bhosmer joined #salt
03:15 crocket aurynn, https://bpaste.net/show/0721c9ac4189 did the magic for me.
03:16 thayne joined #salt
03:19 TyrfingMjolnir joined #salt
03:25 Mso150 joined #salt
03:36 jalbretsen joined #salt
03:55 thayne joined #salt
03:56 jnials joined #salt
03:59 crocket salt v2014.7.0rc7 says "State 'npm.installed' found in SLS 'selenium-node.selenium.npm-pkgs' is unavailable"
03:59 crocket However, npm.installed should be available according to http://docs.saltstack.com/en/latest/ref/states/all/salt.states.npm.html#salt.states.npm.installed
03:59 TyrfingMjolnir joined #salt
04:03 racooper joined #salt
04:04 pdayton joined #salt
04:05 srage joined #salt
04:08 smcquay joined #salt
04:20 snuffeluffegus joined #salt
04:36 kermit joined #salt
04:40 diehard_ left #salt
04:46 thayne joined #salt
04:51 ndrei joined #salt
04:51 ndrei_ joined #salt
04:59 bhosmer joined #salt
04:59 TyrfingMjolnir joined #salt
05:00 Ryan_Lane joined #salt
05:13 crocket How do I make sure a package is installed via https://dl.bintray.com/mitchellh/vagrant/vagrant_1.6.5_x86_64.deb ?
05:13 res0nat0r joined #salt
05:15 jnials joined #salt
05:33 felskrone joined #salt
05:43 MrFuzz joined #salt
05:50 ramteid joined #salt
06:03 zach joined #salt
06:07 shookees joined #salt
06:16 tmh1999 joined #salt
06:25 jchen joined #salt
06:41 ghanima joined #salt
06:44 Sypher joined #salt
06:48 Ryan_Lane joined #salt
06:48 bhosmer joined #salt
06:51 bhosmer_ joined #salt
07:00 billy_ joined #salt
07:02 monkey66 joined #salt
07:02 diehard_ joined #salt
07:03 NightMonkey joined #salt
07:03 cberndt joined #salt
07:05 billy_ Seem to be running in to a weird problem, in my reactor sls files trying to do something simple like local.cmd.run after a minion starts up.
07:05 billy_ Getting a "AttributeError: 'ReactWrap' object has no attribute 'local'" error
07:07 sdh_ joined #salt
07:08 dooshtuRabbit joined #salt
07:09 jnials joined #salt
07:13 jeddi joined #salt
07:18 tmh1999 joined #salt
07:28 oyvjel joined #salt
07:31 xsteadfastx joined #salt
07:32 gfa left #salt
07:35 yomilk joined #salt
07:37 tinuva joined #salt
07:40 snuffeluffegus joined #salt
07:41 yomilk joined #salt
07:42 mgw joined #salt
07:44 b1nar1_ joined #salt
07:45 BossR Is there a list of example salt states people can pick from / use - like a simple LAMP stack...
07:45 TheThing joined #salt
07:46 jhauser joined #salt
07:49 jhauser joined #salt
07:54 flyboy joined #salt
08:01 trikke joined #salt
08:03 tomspur joined #salt
08:07 honestly BossR: check out salt-formulas on github
08:07 honestly has a bunch of stuff, dunno if it has lamp
08:10 lcavassa joined #salt
08:16 BossR oh doesnt have to be lamp... just examples to help me build a library of things useful
08:18 honestly yeah salt-formulas is definitely good for that
08:18 BossR yeah talk about a goldmine!!
08:18 honestly you didn't wipe off saltstack after all? :P
08:20 BossR I have...
08:20 BossR no...
08:20 BossR well...
08:20 BossR umm...
08:20 honestly lol
08:21 BossR The best way to put it is this... I am willing to try and learn the fundamental design of it...
08:21 BossR and until I give that a fair shot, I am going to leave it alone for now...
08:21 BossR worst case scenario is I rebuild it myself using puppet
08:22 BossR so I am looking at real-world examples of salt in action / like saltstack-formulas
08:22 honestly cool
08:22 BossR then visualize how I see it should be configured... and as long as they did something to a standard... then I should be able to understand what they did and make changes myself
08:23 eseyman joined #salt
08:23 duncanmv joined #salt
08:23 BossR my concern is they set this up in some illogical pattern that only they understand and I am going to snap
08:28 BossR honestly: truth is, what they are telling me I should be able to do just seems like a lot of work to set up for a minimum node count of 10 maximum of 25
08:28 BossR I could do the same shit, with a nice web interface with Puppet Enterprise in 1/2 the time
08:30 BossR but again, I am only seeing one part of it right now, and I want to get a better picture
08:37 bhosmer joined #salt
08:39 P0bailey joined #salt
08:40 P0bailey joined #salt
08:50 PI-Lloyd joined #salt
08:53 slav0nic joined #salt
08:54 TheThing joined #salt
08:54 yomilk joined #salt
08:55 peno joined #salt
08:58 jnials joined #salt
08:59 AirOnSkin How can I get a specific part of grains info in a state? I'm looking to get the value of eth0 of ip_interfaces
09:03 intellix joined #salt
09:04 honestly in a jinja2 templae, {{ grains['ip_interfaces']['eth0'] }} should do it
09:05 jdmf joined #salt
09:10 N-Mi joined #salt
09:10 Mso150_n joined #salt
09:10 AirOnSkin honestly: Ah, that was what I was looking for. Big thanks :)
09:11 bhi joined #salt
09:22 hvn joined #salt
09:28 cliff-hm joined #salt
09:33 crocket left #salt
09:38 karimb joined #salt
09:43 alex-mesos joined #salt
09:47 tinuva joined #salt
09:48 _ale_ joined #salt
09:49 iwishiwerearobot joined #salt
09:49 oyvjel joined #salt
09:53 tillo left #salt
09:57 dunz0r How do you have your config-files setup? Like how does your tree look like?
09:58 dunz0r Looking to set up salt and I want to do this right from the start :)
09:58 arthurlutz joined #salt
09:58 arthurlutz is there a way to import between two custom outputters ?
10:00 h8 yeah, I'm fighting with the same things, setting up salt for the first time with 100+ servers, wanna do it right. Are there any scenarios? Something I should be looking out for
10:03 dunz0r I think I'm going to use git for the config management, since this will basically give me free "audit logs"
10:05 hvn left #salt
10:05 h8 for now I have the default config files, even for the minions, we have our own DNS servers and I just setup the "salt" zone to the IP of the master
10:08 cruatta joined #salt
10:08 flyboy82 joined #salt
10:09 TheThing joined #salt
10:11 Corey_ joined #salt
10:12 Ironhand I've recently upgraded to the 2014.1.13 packages for Debian Stable, but am still experiencing very regular minion timeouts on commands like test.ping (though maybe somewhat less regular than before)
10:12 hintss joined #salt
10:12 Ironhand is this issue supposed to be fixed in these package versions or will that be a later version?
10:16 iwishiwerearobot joined #salt
10:16 CycloHex joined #salt
10:17 brayn joined #salt
10:19 lothiraldan joined #salt
10:25 bhosmer joined #salt
10:33 AirOnSkin Can I require a grain?
10:34 AirOnSkin For example for VMware tools installation, I'd like to copy&extract the archive only if the the grain virtual equals VMware
10:35 AirOnSkin I know I can surround it with an if statement, but I'd find the require option more elegant
10:40 lothiraldan joined #salt
10:43 dunz0r Haha
10:43 dunz0r "Minions can be matched using Perl-compatible regular expressions (which is globbing on steroids and a ton of caffeine)."
10:47 jnials joined #salt
10:58 flyboy82 AirOnSkin why don't you just target the minions that have that grain from top.sls???
10:58 flyboy82 so no ifs and no requires
10:59 flyboy82 as for your question, i don't think it's possible to require a grain
11:11 diehard_ joined #salt
11:16 Mso150_n joined #salt
11:16 xsteadfastx joined #salt
11:18 ggoZ joined #salt
11:20 aquinas joined #salt
11:28 halfss_ joined #salt
11:31 techdragon joined #salt
11:32 viq What happens to mine data when I restart salt-master ?
11:37 AirOnSkin flyboy82: Funny, I thought of the same thing while I was at lunch. Needed the distance ;) Thanks anyway.
11:38 viq AirOnSkin: some people say it's good practice anyway to make it so, to make sure it does nothing on systems where it's not supposed to
11:39 glyf joined #salt
11:40 hobakill joined #salt
11:40 AirOnSkin viq: I'm don't quite follow. You mean double-check if the system's a virtual machine?
11:41 lopezio hi, any orchestrator user here?
11:42 CeBe joined #salt
11:43 lothiraldan joined #salt
11:44 AirOnSkin Another question: I've seen the following online in a formula: {{ pillar.get('vmware.version', '8.3.12-493255') }}
11:44 AirOnSkin Is this the right way to get pillar information in a state, and why is there still a version number in the line?
11:45 lothiraldan joined #salt
11:46 ze- AirOnSkin: the value is the default value used if the pillar is not defined/found.
11:47 viq AirOnSkin: yeah, wrap the state in an if anyway, so if someone runs it by hand on a physical machine it will still not do anything
11:47 viq What happens to mine data when I restart salt-master ?
11:48 AirOnSkin ze-: I see. But why would someone do that (in this particular case)? I only see more room for error like that...
11:49 AirOnSkin viq: Ok, sounds reasonable. Thanks :)
11:54 ze- AirOnSkin: pillar are *defined* on master anyway. *they* have a default value for hosts for which they don't specify anything in the pillar configurations.
11:54 TheThing joined #salt
11:54 ze- might be to define which version of software to install prior to having the software available.
11:56 CeBe1 joined #salt
12:02 arif-ali joined #salt
12:02 intellix joined #salt
12:04 hobakill joined #salt
12:04 ramishra joined #salt
12:06 yomilk joined #salt
12:07 scottpgallagher joined #salt
12:07 hobakill Hello Salters... has anyone done much work with upgrading salt minion clients on windows? i haven't found a good way to automate that process. any thoughts?
12:10 joehh hobakill: I know our admins do this in someway, but we've a horse race on tomorrow so I can't talk to them till wednesday...
12:11 lothiraldan joined #salt
12:11 hobakill joehh: thanks. just curious. seems to be a bit of an Achilles heel with salt.
12:12 joehh It's definitely something you need to be careful with - I've stuffed it up on debian before (as package maintainer)
12:13 joehh so I'm interested to hear how they are doing it on windows. If your around in about 36 hours or so, I should have an answer
12:13 hobakill joehh: i'm always in this room from 6AM-2PM CST
12:14 ramishra joined #salt
12:14 CeBe1 joined #salt
12:14 bhosmer joined #salt
12:15 ndrei joined #salt
12:15 ndrei_ joined #salt
12:15 joehh I'll let you know in a couple of days then :)
12:16 joehh maybe someone else can answer in the mean time
12:16 akafred joined #salt
12:17 hobakill joehh: perhaps. i've floated minion updates and general reporting to the room before without much success butwho knows. thanks again.
12:19 ramishra_ joined #salt
12:22 aqua^mac joined #salt
12:23 ramishra joined #salt
12:23 pkruithof_ joined #salt
12:24 joehh I'll send them an email - maybe they'll respond, but given today and tomorrow seem to be the least worked days of the year, I reckon it will be a couple of days before we know the answer
12:24 joehh http://en.wikipedia.org/wiki/Melbourne_Cup
12:24 pkruithof joined #salt
12:24 babilen TIL
12:25 pkruithof Anyone else got issues with pkgrepo on Ubuntu? I'm getting errors like this: "AttributeError: 'module' object has no attribute 'expand_ppa_line'"
12:25 hobakill thanks. i was rather wondering what in the world you're talking about. not much horse racing interest in my part of the world.
12:25 babilen pkruithof: Could you paste your entire state and the error?
12:26 babilen (which version of salt is this on master/minion?)
12:28 pkruithof http://pastebin.com/raw.php?i=9njHabsi
12:28 pkruithof server is running the latest Ubuntu (14.10 - utopic)
12:29 joehh yeah - its pretty ridiculous - always on a tuesday so most people take the monday off as well and I reckon there is a better chance of getting anything done on christmas or easter than cup day
12:30 ramishra joined #salt
12:31 pkruithof FTR: I've used this state for over a year now, today I created a new project with the latest Salt/Ubuntu and this happens, so it must be an update somewhere
12:31 babilen pkruithof: I don't have an Ubuntu box around, but you do, presumably, have python-software-properties installed, don't you?
12:31 dooshtuRabbit joined #salt
12:31 babilen (if so: which version?)
12:31 pkruithof Yes I do
12:32 ramishra joined #salt
12:32 joehh does it still work on earlier versions (ie trusty/precise)
12:32 pkruithof python-software-properties: v0.94
12:33 pkruithof yes, existing boxes still work, though they also run older Salt versions
12:33 CycloHex What is manageable on windows machines as minions? I'm thinking of installing it to my windows machines as well, But I'd like ot use it to automate some gpo's deployment and rds deployment
12:33 pkruithof I'll test a previously working config with the latest Salt, hang on
12:35 babilen pkruithof: Could you log into your minion, start python, run "python" and then "import softwareproperties.ppa" followed by "dir(softwareproperties.ppa)" and paste the output to, say, http://refheap.com or http://paste.debian.net ?
12:35 babilen s/,start python//
12:36 jnials joined #salt
12:37 pkruithof http://paste.debian.net/plain/129978
12:38 babilen pkruithof: It is indeed missing. Looks as if 0.94 either removed that function or that it has been added in a later version (and that salt targets that version)
12:39 hobakilllll joined #salt
12:39 babilen pkruithof: Looks as if salt used that functionality for quite a while and I am not sure where softwareproperties is maintained (LP probably)
12:40 pkruithof I'm searching as well for the source
12:40 hobakilllll joined #salt
12:41 pkruithof http://bazaar.launchpad.net/~ubuntu-core-dev/software-properties/main/files
12:43 pkruithof here is the relevant change: http://bazaar.launchpad.net/~ubuntu-core-dev/software-properties/main/revision/904#softwareproperties/ppa.py
12:43 thayne joined #salt
12:44 babilen Okay, why should Ubuntu core devs care about backwards compatibility anyway?
12:44 babilen *sigh*
12:45 pkruithof yeah, seems like a weird move to not document this
12:45 pkruithof at the very least..
12:45 babilen It looks as if it was removed by mistake to be honest
12:45 pkruithof shall I create a issue in the Salt repo, so they can either contact the Ubuntu devs or fix it in Salt?
12:46 babilen But "no fucking clue" -- Talk to upstream and feel free to file a bug with salt (as that change obviously breaks the pkgrepo state on later Ubuntu releases)
12:46 lothiraldan joined #salt
12:46 * babilen blames Michael Vogt
12:46 pkruithof I'll do that, thanks for the help
12:47 babilen Michael is not online at the moment
12:48 babilen It looks like an arbitrary and unrelated change. My impression is still that mvo removed that function by mistake.
12:48 pkruithof if that's the case then that code is untested, which is also not good
12:49 babilen It does introduce "mange_ppa_shortcut" though which you might want to use in lieu of expand_ppa_line, but then this is *really* bad style to just remove it. They should have added deprecation warnings and left it in place (ensure that it uses the "new" functionality)
12:50 babilen Sorry, I will not dive deeper into this, but I'd take it up with mvo.
12:51 pkruithof Just created an issue for it: https://github.com/saltstack/salt/issues/17105
12:51 AirOnSkin Hmm, I'm using salt.states.archive.extracted and the state runs fine, but the extracted source isn't there. How can I debug that?
12:53 bhosmer joined #salt
12:56 AirOnSkin Nevermind that. It seems that if you're trying to extract a tar file, you need to specify - if_missing
12:57 babilen pkruithof: You really should file a bug with upstream about this as that change is, IMHO, irresponsible.
12:57 babilen You don't change your API like that
12:57 pkruithof I would but I'm really not familiar with that project, or it's version/release management
12:59 babilen pkruithof: https://help.ubuntu.com/community/ReportingBugs
12:59 pkruithof for instance, I don't know if every function in this file is considered part of the public API, or an internal function
12:59 babilen Everything that is a public function in there (i.e. "it does not start with _") should be considered public API.
13:00 pkruithof in that case I completely agree, I'll file a bug now
13:05 CycloHex How to jinja this: {% if grain == 'True' or 'foo' in grain or 'bar' in grain %}
13:05 CycloHex this always resorts to false, thus skipping the sls
13:06 CycloHex So I'm pretty sure it is a jinja-syntax error I'm making. is it possible to give multiple statements in one IF-loop?
13:08 babilen CycloHex: I don't see anything wrong with that expression
13:08 CycloHex neither do I..
13:08 babilen (and due to or it shouldn't short-circuit)
13:09 babilen CycloHex: Which means that: 'grain' is not set to 'True' (i.e. the *string* True) and that neither foo nor bar are present in the same 'grain'. Why would you have 'foo' in True?
13:09 CycloHex maybe it'd be better if I don't use the salt['grains.et'].. expr bu tinstead, first set grain = salt['grains.get'].. and then use that var in my if-statement..
13:09 CycloHex although that shouldn't make a difference
13:09 CycloHex it's not the same grain I'm evaluating
13:09 hobakill joined #salt
13:10 babilen My impression is that you simplified the actual code a bit too much so that it is impossible to spot the actual error.
13:10 CycloHex I'll pastebin it
13:10 tkharju joined #salt
13:10 babilen (anything but pastebin.com please)
13:11 diegows joined #salt
13:11 hobakill https://www.hobapolis.com/paste/ !!!!! :)
13:12 pkruithof what's bad about pastebin?
13:12 jcockhren its' not hoba's
13:12 jcockhren ;)
13:12 babilen pastebin.com mangles input, takes forever to load, often makes us enter a CAPTCHA to see your paste and fills the screen with ads.
13:12 wnkz joined #salt
13:12 hobakill thanks jcockhren :)
13:13 babilen it is also one of the ugliest pastebins and syntax highlighting and general UI is atrocious
13:13 pkruithof babilen here is the launchpad bug: https://bugs.launchpad.net/ubuntu/+source/software-properties/+bug/1388825
13:13 CycloHex babilen: http://paste.debian.net/129984/
13:13 babilen pkruithof: Perfect, thank you!
13:13 wnkz hi, how can I make a state running on a server A check/require for another state on a server B ?
13:14 babilen CycloHex: And why should that evaluate to true?
13:14 CycloHex babilen: if it evaluates to true, I run my git sls, if it's false I don't run anything.. this always evaluates as false
13:14 babilen CycloHex: Are you *sure* that it is set to the *string* 'True' and not the boolean ? There is some fancy/horrible typecasting going on in yaml.
13:14 CycloHex although I manually set the init_hs grain to True
13:15 jcockhren CycloHex: are you just trying to check if the grain is defined?
13:15 jcockhren oh I see
13:15 babilen Also I'd just use "if salt['grains.get']('init_hs')"
13:15 CycloHex jcockhren: nope, it should have the value true.. I evaluate some grains for True as well, they evaluate fine
13:15 babilen true != True
13:16 babilen != 'True'
13:16 jcockhren right ^
13:16 babilen jcockhren: You should have said "^ true" !
13:16 CycloHex ok, i'll try to omit the == 'True'
13:17 jcockhren haha
13:17 babilen And the only reason in your current setup for that to evaluate to true is because salt['grains.get']('init_hs') ? (i.e. both other conditions are wrong)
13:17 CycloHex babilen: it's true that true != True != 'True'
13:17 CycloHex I ommitted the == 'True' and it executed
13:17 CycloHex thanks
13:17 babilen you speak the true true
13:18 babilen yay
13:18 babilen I guess that you checked if the boolean is the same as the string
13:18 X86BSD joined #salt
13:18 CycloHex Yes, I did.. But I do the same thing in my code somewhere, I think.. I'm goign ot check that.. I really thought there wa sno difference in 'true', 'True' or the bool-val True
13:23 CycloHex Does this evaluate the value of the grain or whether it is set or not? { if grainname %}
13:23 cpowell joined #salt
13:23 babilen CycloHex: This isn't simplified by the fact that YAML and jinja both do implicit typecasting
13:23 babilen CycloHex: It evaluates the value of the grain (i.e. it will be false if the grain is set to False)
13:23 CycloHex ok, perfect, thanks!
13:24 babilen (or otherwise not true (e.g. [], '', ...)
13:24 babilen )
13:24 babilen Very rarely do you want to use == in Python as there typically is a better way to express it.
13:25 CycloHex Ok, thanks for the info, babilen!
13:29 agend joined #salt
13:34 micah_chatt joined #salt
13:35 aqua^mac joined #salt
13:40 Morbus joined #salt
13:40 XenophonF joined #salt
13:40 Morbus if i have salt/roots/apache/apache.sls, how do i specify that to run in a top.sls? apache/apache?
13:40 teebes joined #salt
13:41 Morbus new salt user.
13:41 XenophonF roots.apache.apache
13:41 CycloHex Morbus: if your salt base is /salt/roots apache.apache would work
13:41 Morbus why roots? i don't specify roots.common, for example.
13:41 * Morbus tries apache.apache.
13:41 Morbus thanks.
13:41 XenophonF then it'd be apache.apache
13:41 CycloHex anyway, in top file you can replace a '/' by a '.' and omit the extension (.sls)
13:42 Morbus so, i've got apache in a subdirectory cos i have config files to go along with it.
13:42 XenophonF you could rename apache.sls to init.sls, and just specify apache
13:42 CycloHex ^ yes
13:42 Morbus but is it "better" to have a master conf/ directory (allowing me to put apache.sls in the root, and not needing apache.apache)...
13:42 XenophonF i tend to follow how saltstack/salt-formulas structures things
13:42 Morbus or is it better ot have them packaged together in per-service directories?
13:42 XenophonF where i have a dir for the state, e.g., apache
13:42 XenophonF and then have an init.sls and map.jinja in there
13:42 Morbus ooOOoo.
13:42 XenophonF and then make a subdir for files named "files"
13:42 Morbus that's how init.sls works...
13:42 mgw joined #salt
13:43 Morbus so rename /srv/roots/apache/apache.sls to /srv/roots/apache/init.sls and then specify just "apache" in the top.sls?
13:43 CycloHex The map.jinja is mostly for multi-platform support, right? XenophonF
13:43 Morbus that sounds a lot better.
13:43 XenophonF CycloHex: it can be
13:43 XenophonF i also use it for multi-version support
13:43 XenophonF like RHEL 6.5 vs RHEL 7
13:43 CycloHex I follow the formula-rule as well.. bu tI have no maps.jinja yet.. since I'm not going multi-platfrom and don't rly see ay othe ruse of it
13:44 CycloHex yes, multi-version or platform seems to be the only time you'll need maps.jinja?
13:44 Morbus so does salt-formulas do the apache/init.sls thing?
13:44 XenophonF yes, i think so
13:45 Morbus huh.
13:45 XenophonF just to confuse you more :) you can also create apache.sls in the current directory and an apache subdirectory, and still refer to "apache" in top.sls
13:45 CycloHex Morbus: yes, they do.https://github.com/saltstack-formulas
13:45 Morbus well, apache/init.sls definitely works, yay, but still getting an error from vagrant/salt-call/non-zero exit status
13:45 Morbus but, new things to google.
13:45 Morbus ;)
13:46 XenophonF Morbus, I'll see if I can check my current state tree into Git
13:46 XenophonF i have an all-singing, all-dancing apache formula that you might like
13:46 XenophonF PM me your email address and I'll ZIP it and send it to you
13:47 elfixit joined #salt
13:47 Morbus first timer fiddling wtih salt, to build a new vagrant environ. fun stuff.
13:47 XenophonF awesome!
13:47 XenophonF i'm really, really digging salt
13:48 AirOnSkin Hmm, the unless in the following state doesn't work: http://hastebin.com/wubenubutu.sm
13:48 AirOnSkin Does anyone see why? I'm clueless :/
13:48 Morbus XenophonF: huh. so i shouldnt even make my own apache formula then? i should start with the salt-formula and tweak?
13:49 diehard_ joined #salt
13:51 CycloHex AirOnSkin: are you on 2014.7 both on master and minion?
13:52 AirOnSkin I'd need to check... are there known issues with unless in prior versions?
13:52 CycloHex I'm asking this because unless currently is only supported on cmd and git (may have some others supported) and form 2014.7 you can unless and onlyif on every state
13:52 AirOnSkin Ok, great. Will check that...
13:53 CycloHex AirOnSkin: http://docs.saltstack.com/en/latest/topics/releases/2014.7.0.html#global-onlyif-and-unless
13:54 Morbus wooo. first vagrant+salt build successful.
14:03 bhosmer joined #salt
14:04 XenophonF Morbus: you should definitely look at the salt-formulas repo.
14:04 gngsk joined #salt
14:04 oyvjel joined #salt
14:04 XenophonF I started there, was dissatisfied, and wrote my own Apache state.
14:04 XenophonF which at some point i'd like to contribute back
14:10 peters-tx joined #salt
14:10 xsteadfastx joined #salt
14:11 lothiraldan joined #salt
14:15 miqui joined #salt
14:16 micah_chatt joined #salt
14:18 crocket joined #salt
14:18 crocket Is network state stable on v2014.7.0rc7?
14:19 the_drow According to the docs not yet
14:19 the_drow At least not on ubuntu/debian
14:20 teebes joined #salt
14:20 racooper joined #salt
14:22 TyrfingMjolnir joined #salt
14:22 cpowell Good morning everyone
14:23 perfectsine joined #salt
14:24 jnials joined #salt
14:25 viq What happens to mine data when I restart salt-master ?
14:26 nitti joined #salt
14:26 flyboy82 I'm guessing nothing viq. the mine is actually a concatenation of files residing in the master;s cache dir
14:26 cpowell it should be rereshed
14:26 cpowell salt \* mine.update
14:27 flyboy82 so it should read them on restart
14:27 viq Well, pondering if it persists or gets blanked
14:27 TOoSmOotH joined #salt
14:27 cpowell this popped up in our internal salt mailing list and I was wondering if anyone here would have a creative take on the problem
14:27 cpowell https://gist.github.com/anonymous/3006741be1d459091824
14:28 flyboy82 it should persist, just like running jobs, since it's in the minions' subloders
14:28 flyboy82 subfolders*
14:28 viq flyboy82: ok, thank you
14:29 glyf joined #salt
14:29 Nazzy joined #salt
14:29 Nazzy joined #salt
14:33 ramishra joined #salt
14:36 faust joined #salt
14:36 dude051 joined #salt
14:37 ramishra_ joined #salt
14:38 Setsuna666 joined #salt
14:39 jnials joined #salt
14:41 thayne joined #salt
14:41 diehard_ joined #salt
14:47 gmcwhistler joined #salt
14:48 intellix joined #salt
14:49 jeddi joined #salt
14:52 desertigloo joined #salt
14:53 mgw joined #salt
14:54 fredvd joined #salt
14:55 bhosmer joined #salt
14:56 desertigloo joined #salt
14:57 micah_chatt_ joined #salt
14:57 jnials joined #salt
14:58 tk75 joined #salt
14:59 jngd joined #salt
15:00 crocket left #salt
15:00 mdasilva joined #salt
15:00 mdasilva morning all
15:01 fredvd joined #salt
15:03 kt766 joined #salt
15:03 halfss joined #salt
15:12 jaimed joined #salt
15:12 gngsk_ joined #salt
15:16 ndrei joined #salt
15:16 ndrei_ joined #salt
15:17 rojem joined #salt
15:17 jchen joined #salt
15:18 rojem joined #salt
15:20 kaptk2 joined #salt
15:24 Morbus if I have:
15:24 Morbus uploadprogress:
15:24 Morbus pecl.installed:
15:24 Morbus and then later on, i need to move a config file into place...
15:24 Morbus how do I "require" a confirmation that the above was installed?
15:24 Morbus would pkg: uploadprogress work?
15:25 mpanetta joined #salt
15:25 diehard_ joined #salt
15:25 SheetiS joined #salt
15:26 rojem joined #salt
15:27 viq Morbus: I belive that would be pecl: uploadprogress
15:28 felixhummel joined #salt
15:28 felixhummel hi!
15:28 viq <module>: <name>
15:28 Morbus aah. ok.
15:28 * Morbus tries.
15:28 viq so you have state with name uploadprogress, and it uses module pecl (and from it function install, but that's not relevant at this moment)
15:30 conan_the_destro joined #salt
15:31 mpanetta_ joined #salt
15:31 bezeee joined #salt
15:34 b0fh joined #salt
15:35 DenkBrettl joined #salt
15:35 ollins joined #salt
15:36 mackstick joined #salt
15:39 sirtaj joined #salt
15:39 gngsk joined #salt
15:40 asmaps joined #salt
15:40 glyf joined #salt
15:41 borgstrom joined #salt
15:41 mschiff joined #salt
15:41 mschiff joined #salt
15:41 Ironhand joined #salt
15:42 __alex joined #salt
15:42 dabb joined #salt
15:42 Jarus joined #salt
15:43 alice|wl joined #salt
15:44 sdh_ joined #salt
15:45 mdasilva joined #salt
15:46 pkruithof joined #salt
15:47 jalbretsen joined #salt
15:49 lothiraldan joined #salt
15:51 borgstrom joined #salt
15:54 rbaum joined #salt
15:55 Deevolution joined #salt
15:55 patarr joined #salt
15:55 patarr joined #salt
15:57 seydu joined #salt
15:59 schristensen joined #salt
15:59 mdasilva_ joined #salt
16:00 hobakill joined #salt
16:01 TjackNoire joined #salt
16:02 bezeee joined #salt
16:02 housl joined #salt
16:03 borgstrom joined #salt
16:03 rojem joined #salt
16:04 bhosmer joined #salt
16:04 thayne joined #salt
16:05 rostam joined #salt
16:09 hellerbarde joined #salt
16:10 mitsuhiko joined #salt
16:11 StDiluted joined #salt
16:12 yano joined #salt
16:12 manchot joined #salt
16:13 manchot hi everyone!
16:14 Topic for #salt is now Welcome to #salt | SaltConf 2015 Call for Speakers is open! http://saltconf.com/call-for-speakers/ | 2014.1.13 is the latest | Help us test the 2014.7 RC! http://bit.ly/salt-rc | Please be patient when asking questions as we are volunteers and may not have immediate answers | Channel logs are available at http://irclog.perlgeek.de/salt/
16:17 Ahrotahntee publicized logs, interesting
16:17 Ahrotahntee for a heck of a long time too
16:17 CycloHex hmm, weird. Salt-cloud gives me an error stating that __str__ returned non-string (type SysCallError), this renders me unable to create or delete cloud-vms, is this salt-cloud related or is it probably my cloud-provider? This worked flawlessly before :s
16:21 CycloHex Here is the stacktrace, for anyone interested
16:21 CycloHex https://gist.github.com/Cyclohex/28de7d5e3c83070b0322
16:23 ndrei joined #salt
16:24 ndrei_ joined #salt
16:28 pmcg joined #salt
16:28 thedodd joined #salt
16:29 jasonrm joined #salt
16:29 jasonrm joined #salt
16:30 manchot anyone know how i can include the grain: network.ip_addrs cidr="192.168.1.0/24" into my template file?  I want to get the ip address but don't want to go by eth# because some servers have different interface numbers.
16:31 ekristen joined #salt
16:34 timoguin manchot: should be something like {% set foo = grains.get('network:ip_addrs')[0] %}
16:35 Ahrotahntee I have a vague memory of being able to invoke network.ip_addrs as a function via salt['network.ip_addrs']()
16:35 Ahrotahntee which could probably take cidr= as a parameter with some testing
16:37 glyf joined #salt
16:39 Morbus can i have one state handle multiple file.managed?
16:39 manchot thx for answer, i'm trying to get the same output as : salt 'ssh-server' network.ip_addrs cidr="192.168.1.0/24"
16:40 Morbus nevermind.
16:44 bezeee joined #salt
16:44 moodperson is any way to disable parallel execution on the state ? or i can specify in which order do commands ?
16:45 mgw moodperson: http://docs.saltstack.com/en/latest/ref/states/ordering.html
16:46 bhosmer_ joined #salt
16:46 viq moodperson: and linked from there more at http://docs.saltstack.com/en/latest/ref/states/requisites.html
16:46 Ozack1 joined #salt
16:46 moodperson mgw, yep I sow it but it works only packages state
16:47 viq moodperson: no, you can require any state
16:47 moodperson oh cool
16:47 CycloHex but order might not, viq .. I for instance never succeeded in using order :d
16:47 moodperson will try
16:47 viq CycloHex: I believe order is discouraged and kind of last resort
16:47 mgw I use order: 0 for global requisites
16:47 manchot i've tried listenaddress: {{ salt['network.ip_addrs']()['cidr="192.168.78.0/24"'][0] }} but doesn't work. Only thing worked is listenaddress: {{ salt['network.ip_addrs']()[0] }}
16:47 CycloHex yes, it is
16:47 mgw but other than that, I don't really use it
16:48 CycloHex manchot: and does the last command give you a right output?
16:48 CycloHex right==correct
16:49 mgw manchot: try putting cidr= in the parens
16:49 mgw so (cidr="192...")
16:54 manchot no, the last command give me the last ip address of my server (eth1), i'll try with parens
16:55 TheGerm joined #salt
16:56 tligda joined #salt
16:56 wendall911 joined #salt
16:56 TheGerm left #salt
16:56 TheGerm joined #salt
16:57 beneggett joined #salt
16:57 manchot works with {{ salt['network.ip_addrs'](cidr="192.168.78.0/24")[0] }}      {{ salt['network.ip_addrs'](cidr="172.17.0.0/24")[0] }}   thx a lot!
16:58 mgw manchot: np!
16:59 alex-mesos joined #salt
17:00 cpowell joined #salt
17:03 KyleG joined #salt
17:03 KyleG joined #salt
17:03 dalexander joined #salt
17:03 b1nar1 joined #salt
17:05 bhosmer_ joined #salt
17:05 shepax joined #salt
17:05 shepax hi
17:05 beneggett joined #salt
17:06 shepax i'm writing a module for sql server but i have some trouble with the pillar definition.
17:06 Blacklite joined #salt
17:07 shepax here is the github where i have the code https://github.com/shepax/sqlserver_saltstack
17:08 ajolo joined #salt
17:10 beneggett joined #salt
17:20 flyboy82 joined #salt
17:24 iggy so... should mine functions be available in pillars? or are pillars instantiated before the mine?
17:27 ndrei joined #salt
17:27 ndrei_ joined #salt
17:27 glyf joined #salt
17:28 X86BSD joined #salt
17:29 bhosmer_ joined #salt
17:30 shepax ??
17:32 petri_ joined #salt
17:33 meylor joined #salt
17:33 cliff-hm joined #salt
17:34 viq_ joined #salt
17:34 dalibro joined #salt
17:34 viq_ joined #salt
17:35 lothiraldan joined #salt
17:35 murrdoc joined #salt
17:35 mdasilva joined #salt
17:36 saffe hi - any ideas why no response to salt commands from Windows 7 minion despite key showing (& having been accepted) ok? master is debian, from salt package repo
17:36 murrdoc monring
17:37 smcquay joined #salt
17:37 saffe the funny thing is, if I instead use osx master, installed via  "pip install salt", it works fine
17:38 saffe (ports are accessible just fine)
17:38 shepax hi im wirting a sqlserver module and i have some trouble with the pillar some one can help me? i don't understand how to write an get them.
17:38 saffe no changes to default config files (apart from setting salt master IP of course)
17:39 shepax saffe did you set the path to python?
17:40 cberndt joined #salt
17:40 MrFuzz joined #salt
17:40 saffe in master or minion? the salt commands execute fine though
17:40 viq saffe: I've seen you're using py renderer, here's an example I know of that does that https://github.com/saltstack-formulas/reverse-users-formula/blob/master/users/init.sls
17:40 shepax at the minion
17:41 saffe hmm no I did not - but it's the regular salt installer, and it works fine with the osx master...
17:41 shepax If u install a minion on windows u have to set the path of python.
17:44 joehoyle joined #salt
17:45 murrdoc joined #salt
17:46 troyready joined #salt
17:46 timoguin I don't think you have to set the path for the windows minion. The installer has python bundled with it.
17:48 saffe that's what I thought, too (thanks for suggestion shepax, though)
17:48 Corey Pillar is first match, right?
17:49 blackjid joined #salt
17:49 viq Corey: what do you mean?
17:49 shepax I'm writing this code https://github.com/shepax/sqlserver_saltstack
17:50 shepax some one can help me with the pillar writing :)
17:50 A||SySt3msG0 joined #salt
17:51 Corey viq: I define a pillar item twice. Does the first value stick, or the second? :-)
17:51 viq I think the second one will overwrite the second
17:51 viq Though I'm not sure it's entirely deterministic
17:51 viq I could very well be mistaken though
17:52 shepax im trying to get the user, host and passw from pillars the idea is to set a db.sls-> sqlserver: dbname: {user: user name , passw: password , host: hosting}
17:52 bhosmer_ joined #salt
17:52 psidox joined #salt
17:52 timoguin Corey: pretty sure the second one will overwrite
17:53 bezeee joined #salt
17:53 timoguin although 2014.7 has added some configurable pillar merging methods
17:53 psidox Hi guys, I recently posted a topic here on salt-users: https://groups.google.com/forum/#!topic/salt-users/xlSIdsUZ1CQ
17:53 psidox Anyone have any idea what I am doing wrong?
17:54 timoguin shepax: what you have in your run_query function should fetch the pillar data just fine, assuming the pillar data is configured properly
17:56 shepax thats the problem im not sure if im configuring the pillar the right way
17:57 viq erm, I highlighted the wrong person...
17:57 saffe hm very strange I reinstalled the minion on win7 and it started working... should've known, it's windows after all... reinstall, reboot, repeat... :(
17:57 viq shepax: I've seen you're using py renderer, here's an example I know of that does that https://github.com/saltstack-formulas/reverse-users-formula/blob/master/users/init.sls
17:58 shepax timoguin at the code is an example of how im writing the pillar.
18:00 shepax Viq adn timoguin thanks
18:00 timoguin shepax: test the pillar data by running salt-call pillar.get 'sqlserver:dbname:host'
18:01 shepax ok
18:01 timoguin and post your pillar top.sls and db.sls
18:02 shepax which is the difference between defining the pillar with a " - " at the front?
18:02 Ryan_Lane joined #salt
18:02 timoguin read up on YAML. items with - are elements of a list
18:02 the_drow Ryan_Lane: ping?
18:05 rojem joined #salt
18:07 psidox anyone: ping?
18:09 Corey I'm screwing up my Pillar syntax: https://gist.github.com/QuinnyPig/2d5cc0ed33923951f398
18:09 timoguin psidox: use cmd.cmd.run, not local
18:10 murrdoc Corey:  pillar[vlan:network] that works too ?
18:10 murrdoc das cool
18:10 saffe_ joined #salt
18:10 murrdoc i normally stick with the . syntax
18:10 ajolo joined #salt
18:10 dalibro joined #salt
18:10 psidox Why cmd? Should I have seen this in documentation somewhere?
18:11 timoguin pretty sure that's how it is in the reactor docs.
18:11 psidox I’ve been copy and pasting from reactor docs
18:11 timoguin ah, i see where you got local
18:12 timoguin I'm not sure what's going on in those docs. The format may have changed.
18:12 psidox ahh
18:12 timoguin Those docs are for the develop branch
18:12 Gareth morning morning
18:13 psidox Thanks again, you’ve solved a couple hours of bashing my head against the wall!
18:16 honza left #salt
18:18 timoguin psidox: here is some more info about the change: http://docs.saltstack.com/en/latest/topics/reactor/#understanding-the-structure-of-reactor-formulas
18:19 timoguin looks like the format was changed to be more consistent in 2014.7
18:20 psidox timoguin: I be running 2014.7.0rc4, according to docs it should be “local” now?
18:21 timoguin psidox: yea according to those docs
18:21 timoguin so what you had should work?
18:21 wcbzero joined #salt
18:21 psidox Yep, hmm
18:21 timoguin but the error very clearly mean it didn't know wtf 'local' meant
18:22 Ryan_Lane the_drow: ?
18:24 wcbzero joined #salt
18:26 spookah joined #salt
18:26 Corey https://gist.github.com/QuinnyPig/2d5cc0ed33923951f398 <-- Updated gist. Still not quite there with my jinja / yaml...
18:29 Gareth Corey: don't think you want to use pillar[vlan.netmask], just vlan.netmask...same for the others.
18:30 Corey Gareth: Testing.
18:32 mpanetta Anyone know how to set grains when using the cloud.profile module?
18:32 Corey Gareth: You are a wizard, and I owe you a beer for that. Huzzah.
18:32 Gareth Corey: :)
18:34 timoguin mpanetta: you can use the 'minion' dict in the profile. minion:\n  grains:\n    foo: bar
18:34 mpanetta Is it possible to do it without using a profile?
18:34 iggy mpanetta: or set it in the profile config
18:35 mpanetta I thought I found a way, but it only seems to work on the command line.  I am trying to do wit using the API and cmd.async()
18:37 mpanetta Here is the commandline that I think worked: salt -v dex cloud.profile ubuntu-x86_64-2G mng00-base "{grains: {cluster: mng}}"
18:37 mpanetta I'm trying it again now...
18:38 dalibro joined #salt
18:38 mpanetta Assuming that works, how would I translate that to using LocalClient cmd_async?
18:39 timoguin pass grains as a kwarg maybe?
18:40 quist joined #salt
18:40 mpanetta timoguin: Yeah I think that is how I do it, but apparently I don't know how to pass that as a kwarg heh
18:40 iggy mpanetta: looks like the vm_overrides kwarg should do it
18:40 mpanetta so just vm_overrides = { grains: { blah: blat }} ?
18:41 iggy I'm not sure on the exact format of that dict
18:41 mpanetta Time for a little experimentation! :)
18:42 MugginsM joined #salt
18:42 mpanetta Ok so the command line I gave worked to add the grain
18:42 mpanetta Now I just need to figure out how to put that in a format cmd_async likes...
18:46 BossR joined #salt
18:47 stephen-mw joined #salt
18:48 stephen-mw Is saltstack written in python 2.7 or 3?
18:48 timoguin stephen-mw: 2.7 due to need to be deployed on a lot of machines without python 3
18:48 meylor joined #salt
18:49 ajolo joined #salt
18:52 mpanetta Ahh!  Figured it out
18:52 mpanetta Watching the event bus helped... But it was a thinko on my part... I just needed to put the grains thing in the arg list.
18:52 rap424 joined #salt
18:55 snuffeluffegus joined #salt
18:55 hotbox joined #salt
19:04 rojem joined #salt
19:11 thedodd joined #salt
19:11 DenkBrettl joined #salt
19:11 aparsons joined #salt
19:15 rojem joined #salt
19:19 badon_ joined #salt
19:22 ckao joined #salt
19:23 psidox joined #salt
19:25 mdasilva joined #salt
19:26 cwright we have a cmd.script job that is running in batches of 5 on our entire instrastructure and is taking a lot longer than expected.
19:26 paha joined #salt
19:26 cwright I want to kill all these running/pending jobs at once. what is the recommended way of doing this?
19:27 jmac joined #salt
19:36 wcbzero joined #salt
19:37 jeffspeff how can i use states.file.missing in an if statement?
19:37 glyf joined #salt
19:38 notpeter_ joined #salt
19:39 timoguin jeffspeff: what are you trying to do?
19:39 timoguin the file_missing state will remove a file if it exists
19:39 timoguin guessing you want to check for existence?
19:39 mdasilva_ joined #salt
19:39 jeffspeff http://docs.saltstack.com/en/latest/ref/states/all/salt.states.file.html#salt.states.file.missing
19:40 jeffspeff but does not remove the file if it is present.
19:40 timoguin so you want to go into the if block and do something if the file is missing?
19:40 jeffspeff yes
19:41 timoguin I'm not sure about the state, but this is how it'd be done with the normal execution module:
19:41 timoguin {% if salt['file.file_exists']('filepath') %}
19:41 timoguin 'if not' for missing
19:42 jeffspeff ok, thank you
19:43 timoguin I think the state module would be something like {% if salt['state.single']('file.missing', 'filepath') %}
19:47 jeffspeff timoguin, why did you say "file.file_exists" instead of "file.exists"? is that the correct syntax? I'm pretty new to salt
19:48 eriko joined #salt
19:48 timoguin jeffspeff: yea that's the name of the module
19:48 jeffspeff ok, thanks for clarifying
19:49 timoguin IMO file.exists would be much cleaner and consistent though
19:50 hasues joined #salt
19:50 hasues left #salt
19:50 wcbzero joined #salt
19:51 rojem joined #salt
19:51 timoguin ah, it's named that way because there's also file.directory_exists()
19:52 ndrei joined #salt
19:52 ndrei_ joined #salt
19:53 monokrome joined #salt
19:54 XenophonF left #salt
19:55 ingwaem joined #salt
19:55 rsanchez joined #salt
19:57 Mso150 joined #salt
19:57 housl joined #salt
19:58 meylor joined #salt
20:04 pmullin joined #salt
20:07 micah_chatt joined #salt
20:07 psidox Got a question for anyone who can help me, in documentation here: http://docs.saltstack.com/en/latest/topics/reactor/#understanding-the-structure-of-reactor-formulas there is a state file example refresh_pool.sls”
20:08 psidox “refresh_pool.sls” What format is this file, it looks like it is not yaml format
20:08 psidox But somehow its a state file?
20:08 _prime_ joined #salt
20:09 iggy that would be better named refresh_pool.jinja maybe
20:10 iggy it's a fragment of a nginx config file
20:10 timoguin yea it's YAML with Jinja
20:10 timoguin for templating
20:10 iggy not yaml
20:10 psidox ah, but how does this templating get back to nginx?
20:10 timoguin iggy: oh you're right
20:11 iggy psidox: I would guess with an include (similar to how map files are handled in the formulas)
20:11 psidox It looks like it is executed via, local.state.sls:  Which seems to suggest it would be expecting this to be a state file?
20:11 iggy honestly, I'd file an issue about that example... it's really dodgy
20:11 timoguin psidox: the state system provides context data to the template. if you look above that example you'll see where the reactor sls passes the pillar kwarg to the template
20:12 psidox ahh, thanks for clearing that up
20:12 timoguin and then refresh_pool.sls fetches pillar data for new_minion.
20:13 timoguin but... that example looks like it might not even work
20:13 cliff-hm joined #salt
20:13 iggy even if it would, it's very unclear how
20:13 jhauser joined #salt
20:13 iggy psidox: if you don't mind, file an issue about it
20:13 psidox Yessir, will do.
20:13 timoguin well it's passing in a minion_id, which the SLS then grabs the pillar data for. but, yea, it's a bit convoluted
20:14 timoguin reactor needs lots more real-world examples.
20:15 iggy ^
20:15 iggy I'm hoping to share what I've got done if I can ever get it to work
20:15 timoguin same, i've got plenty a blog post brewing
20:16 iggy I'm actually kind of hoping to do a talk at saltconf on some of the more interesting things we're doing
20:16 mpanetta Only thing I am doing with reactor at the moment is running highstate on minon start
20:17 timoguin I'm using it to trigger deploys
20:17 mpanetta Oh sweet
20:17 mpanetta Would love to see how you did that
20:17 mpanetta I've been playing with the mongodb ext_pillar and returner lately...
20:18 iggy I'm (attempting to) use it to signal graphite/riemann when events happen
20:18 mpanetta Saving the initial grains in mongo as well.
20:29 nitti joined #salt
20:30 MugginsO joined #salt
20:30 claytonk joined #salt
20:45 thedodd joined #salt
20:45 mdasilva joined #salt
20:48 nitti_ joined #salt
20:49 nitti joined #salt
20:50 Ryan_Lane1 joined #salt
20:53 glyf joined #salt
20:54 karimb joined #salt
20:55 forrest joined #salt
20:56 seydu joined #salt
20:56 seydu Hi
20:57 forrest joined #salt
21:00 seydu I found a file named .giosaveCXSZOX in my state directory. I contained addtionnal states definition (22). Do Salt behave this way (adding states states to update itself) ?
21:00 forrest seydu: It shouldn't be.
21:01 iggy seydu: possibly something left over from your editor?
21:01 iggy gvim, gedit, etc...
21:01 iggy I think it's a gvfs temp file
21:02 seydu I deleted this file even before a I see what was inside.
21:02 rojem joined #salt
21:02 seydu iggy: you write. i used gvfs to ease editing and viewing the state files !
21:03 seydu I used gedit
21:03 seydu Thanks
21:05 rsanchez joined #salt
21:06 babilen uuh, ooh ... helium :D
21:06 bemehow joined #salt
21:07 jalaziz joined #salt
21:07 Mso150_z joined #salt
21:07 meylor joined #salt
21:07 timoguin babilen: eh?
21:08 babilen Just getting excited for what's to come soon now
21:08 timoguin ah, thought i might have missed an announcement
21:08 mpanetta The RC's have been nice so far :)
21:10 ajolo joined #salt
21:10 iggy sadly, we just missed the cutoff for our next release, so we won't get to start using it until January or April
21:11 babilen Same for jessie
21:12 babilen But I guess it is okay to release jessie with 2014.1.13
21:12 manfred 2014.7.0 was just tagged
21:13 babilen (hence my excitement)
21:13 manfred 2 hours ago
21:13 manfred https://github.com/saltstack/salt/releases/tag/v2014.7.0https://github.com/saltstack/salt/releases/tag/v2014.7.0https://github.com/saltstack/salt/releases#
21:13 LGSilva joined #salt
21:13 manfred already packagd in the archlinux community-testing repo
21:13 iggy I doubt anybody is going to use the debian repo of salt for a little while to come
21:13 iggy it's still way too much of a moving target
21:14 babilen iggy: True, but I'd like to see proper packages in *-backports (that can be upgraded whenever they have proven themselves)
21:14 babilen Just saying that .13 is the right release for Debian stable
21:14 LGSilva hello. Is there such thing in salt for a state to run before other state? I know we can use require but the require would be a long list (I'm using a jinja loop) vs a
21:15 LGSilva before inside the loop
21:15 timoguin LGSilva: SLS files are evaluated top-down by default.
21:15 wcbzero joined #salt
21:16 LGSilva so on my top.sls just move the state with the loop before the one I want to execute after the loop?
21:16 iggy LGSilva: in 2014.7 you can require a whole state file (vs a single state)
21:17 LGSilva thanks guys, I will test both suggestions
21:17 babilen Doesn't that already work in 2014.1 ?
21:17 balltongu joined #salt
21:17 mpanetta Ooo nice!
21:17 iggy I thought it was added in 2014.7
21:17 babilen require: - sls: name
21:17 iggy could be wrong
21:17 adilroot joined #salt
21:17 babilen I think I used that before in .1 (I don't need it often)
21:17 balltongu joined #salt
21:18 iggy yeah, docs say it was added in 0.16
21:18 iggy so... go wild
21:18 babilen One typically doesn't require an entire SLS file. I would consider that a bit of a code smell.
21:18 iggy if you're using something older than 0.16, shoot yourself, then upgrade, then go wild
21:18 babilen heh
21:18 scryptic1 joined #salt
21:19 balltongu joined #salt
21:20 scryptic1 Hello, quick question for anyone that may know... What is the salt module equivalent to running "sudo apt-get update" on a linux system?
21:21 babilen scryptic1: You can use pkg.list_upgrades or pkg.upgrade with refresh=True -- What are you trying to do exactly?
21:22 murrdoc joined #salt
21:22 scryptic1 when I use salt-cloud to spin up a new instance, I want to make sure everything is up to date before deploying code.
21:22 MindDrive joined #salt
21:22 iggy scryptic1: how are you deploying the minion?
21:23 babilen scryptic1: You wouldn't use "apt-get update" to do that (as that wouldn't upgrade anything), but "apt-get update ; apt-get dist-upgrade". In salt that would be "pkg.upgrade refresh=True"
21:23 iggy salt-bootstrap has an option to update everything before installing the salt bits
21:23 mpanetta I thought it did that by default?
21:23 murrdoc joined #salt
21:24 babilen scryptic1: I would, however, strongly advise against making upgrades part of the highstate as it will make it tricky. If you install packages those packages will already be the newest candidate at that time.
21:24 mpanetta I know it is doing something, because even when I tell the bootstrap to use git to install it is still doing something with apt.  I never really payed attn to it tho
21:24 mpanetta I guess I just assumed it was doing an update.
21:24 babilen scryptic1: "pkg.upgrade refresh=True dist_upgrade=True" btw
21:25 scryptic1 I am deploying the code base to the newly provisioned node using file.recurse on the latest build...its not the fastest thing ever.
21:25 iggy mpanetta: yeah, even when it installs from git, it still does the deps from apt
21:25 babilen rightfully so
21:26 mpanetta Ah!
21:26 mpanetta Ok that makes sense
21:26 ipmb joined #salt
21:28 iggy it would likely take a while to install all the deps from source
21:29 mpanetta I was thinking pip actually
21:29 mpanetta But I guess at that point pip may not even exist
21:29 balltongu joined #salt
21:30 balltongu joined #salt
21:30 scryptic1 have you guyes ever had issues with the salt bootstrap script failing after salt-cloud finishes provisioning?
21:30 iggy mpanetta: well, zeromq, etc. isn't installable via pip
21:30 mpanetta True
21:34 kt766 joined #salt
21:37 jalaziz_ joined #salt
21:41 Vye Has anyone here ever taken SaltStack Enterprise Training?
21:42 jalaziz joined #salt
21:43 baconbeckons joined #salt
21:45 rsanchez joined #salt
21:46 cnelsonsic_ joined #salt
21:51 conan_the_destro joined #salt
21:53 TheGerm left #salt
21:56 baconbeckons does pillar inheritance exist like file system inheritance? http://docs.saltstack.com/en/latest/topics/tutorials/states_pt4.html#salt-fileserver-path-inheritance
21:58 duncanmv joined #salt
22:02 mosen joined #salt
22:08 murrdoc yes
22:10 iggy sort of
22:11 kballou joined #salt
22:13 murrdoc i might have misread, pillar inheritance is similar to state inheritance between environments
22:13 murrdoc does the pillar top.sls allow for grain based logic ?
22:13 murrdoc probably not
22:14 murrdoc like if i had pillars that were different by os
22:14 murrdoc that should probably go in the pillar file itself
22:21 iggy the docs say that inheritance is non-deterministic
22:21 iggy which is definitely what I've seen
22:21 iggy we had some stuff working for 3 months and then all of a sudden pillars started returning completely different
22:24 StDiluted joined #salt
22:24 thedodd joined #salt
22:27 murrdoc gonna have to test this out with salt —pillar
22:29 I2PNinja joined #salt
22:29 I2PNinja Hey!
22:29 I2PNinja Anyone here?
22:30 aurynn Yes
22:30 racooper just a room full of bots
22:30 I2PNinja Okay, hi! Can you answer one of my questions about i2p?
22:30 iggy murrdoc: the problem we had was mostly merging and/or overwriting keys/values previously specified
22:30 murrdoc iggy:  yeah thats a problem
22:30 aurynn I have no idea, you haven't asked the question yet
22:31 I2PNinja xD
22:32 I2PNinja aurynn: it's in regards to i2p-dev (i tried the irc's and they're dead for days). How can I establish a success HTTP (i2p) communication with a remote server using their protocol & send a request with sockets in C++, PHP, or libcurl?
22:32 I2PNinja I am looking to create an irc-encrypted-communication application using i2p
22:32 iggy do you realize what channel you are actually on?
22:32 I2PNinja Given that they have no docs & no support in regards to it, its difficult
22:32 aurynn I2PNinja, how does this relate to saltstack?
22:33 I2PNinja I have no idea, my friend pointed me here! I thought I would get an answer.
22:33 murrdoc irc needs a whereami
22:33 I2PNinja I'm really stuck though, and I thought I could get some assistance here.
22:34 murrdoc I2PNinja:  /j #i2p-dev
22:34 racooper there might bee a channel on freenode that can help you, but it's not this one
22:34 murrdoc try that
22:34 I2PNinja I looked at their i2pd c++ dev package but it has no source code with relevance to the HTTP_REQUEST stack
22:34 I2PNinja murrdoc: I'll try that, thanks!
22:37 tfield joined #salt
22:40 h8 joined #salt
22:41 murrdoc you guys thinkg v2014.7 will come out of rc ?
22:41 iggy 2 weeks
22:41 murrdoc HAH
22:42 manfred murrdoc:  they tagged .0 today, just packaging and internal testing now
22:44 murrdoc https://github.com/saltstack/salt/compare/v2014.7.0...v2014.7.0rc7
22:44 iggy so like I said... 2 weeks
22:44 iggy ;)
22:47 murrdoc manfred:  looks primising
22:50 tfield_ joined #salt
22:53 jalaziz joined #salt
22:53 diehard_ joined #salt
22:53 Ryan_Lane joined #salt
23:03 MrFuzz joined #salt
23:04 cpowell joined #salt
23:04 tk75 joined #salt
23:06 tligda1 joined #salt
23:24 tmh1999 joined #salt
23:25 murrdoc joined #salt
23:27 meylor why is the default state location /srv/salt and not /etc/salt/states ?
23:28 iggy I honestly have no clue, but if I were to guess it's because /srv/ is expected to be a larger volume than / usually is
23:29 murrdoc cos you can mount /srv
23:29 murrdoc *shrug*
23:31 wcbzero joined #salt
23:31 wcbzero left #salt
23:31 nkuttler meylor: probably depends on the distro.. check the fhs though
23:34 TyrfingMjolnir joined #salt
23:36 rojem joined #salt
23:37 rsanchez joined #salt
23:39 aparsons joined #salt
23:42 aquinas joined #salt
23:52 dooshtuRabbit joined #salt
23:54 bhosmer_ joined #salt

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