Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-02-03

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

All times shown according to UTC.

Time Nick Message
00:05 om2 I just ran my ps script manually in ps as admin and it works fine
00:05 om2 so weird
00:05 om2 will figure it out in asa
00:05 om2 sec
00:06 tzero joined #salt
00:09 keimlink_ joined #salt
00:10 PeterO joined #salt
00:15 alvinstarr joined #salt
00:17 brianfeister joined #salt
00:19 snicers-work I am getting 'Failed to create staging-elb ELB.' when I apply the state to my minion. Here is my sls file: https://gist.github.com/anonymous/3e3a1012e07032fc98e8
00:19 murrdoc joined #salt
00:19 yidhra joined #salt
00:20 snicers-work I know the credentials are good as I am able to create the ERB w/ listener via aws-cli on the server with the same credentials.
00:21 JTeatime joined #salt
00:21 murrdoc joined #salt
00:22 tzero joined #salt
00:27 Guest18700 joined #salt
00:31 SunPowered joined #salt
00:33 quasiben joined #salt
00:35 tzero joined #salt
00:39 snicers-work Anyone have experience with the boto_elb state? I created a python file and imported boto and was able to bring up an elb with it without issue.
00:39 snicers-work But when I try to call my state file it fails.
00:39 devopsprosiva joined #salt
00:40 devopsprosiva Hi.. If I deploy noded in multiple clouds, how can I get the cloud type? Is it stored in pillar?
00:42 jfindlay devopsprosiva: you could set a grain on the minion telling you what cloud it's in when you create it
00:43 jfindlay there might be a better way, but I can't think of anything
00:43 jfindlay snicers-work: is there anything interesting in the minion log when you run the state?
00:44 murrdoc joined #salt
00:45 PeterO joined #salt
00:47 snicers-work rtfl (read the friggin log), bad availability zone.
00:48 snicers-work thanks jfindlay, any reasons we don't elivate that message to the state response?
00:48 ageorgop joined #salt
00:48 ajw0100 joined #salt
00:48 jfindlay snicers-work: we should
00:49 devopsprosiva jfindlay: Grains make much more sense. Thanks
00:49 snicers-work lol, ok.
00:50 jfindlay snicers-work: if you want you can file an issue and I'll see what I can do about returning that info
00:50 jfindlay devopsprosiva: sure
00:50 snicers-work jfindlay, https://github.com/saltstack/salt/issues/30840
00:50 saltstackbot [#30840]title: Generic Error for SALT.STATES.BOTO_ELB | I am configuring an ELB instance on AWS and when I try to apply the state I get the following error:...
00:50 jfindlay nice, thanks, snicers-work
00:50 snicers-work you got it
00:51 snicers-work thank you
00:51 hasues joined #salt
00:51 jfindlay snicers-work: can you also post the relevant minion log, with any private info removed if necessary?
00:51 hasues left #salt
00:52 snicers-work sure
00:52 jfindlay and adding your state might help as well, https://gist.github.com/anonymous/3e3a1012e07032fc98e8
00:54 abednarik joined #salt
00:55 Guest18700 joined #salt
00:59 snicers-work jfindlay, another issue https://github.com/saltstack/salt/issues/30843
00:59 saltstackbot [#30843]title: SALT.STATES.BOTO_ELB register_instances error | my init.sls...
01:10 spuder joined #salt
01:11 shaggy_surfer joined #salt
01:12 jfindlay thanks, saltstackbot
01:12 jfindlay s/saltbot/snicers-work/
01:25 baweaver joined #salt
01:26 upb joined #salt
01:29 _JZ_ joined #salt
01:31 fuzzy_id joined #salt
01:36 bhosmer joined #salt
01:39 upb joined #salt
01:43 ajw0100 joined #salt
01:48 joe_n joined #salt
02:02 michelangelo joined #salt
02:06 brianfeister joined #salt
02:12 spuder_ joined #salt
02:14 cuonglm joined #salt
02:14 upb joined #salt
02:15 brianfeister joined #salt
02:17 zmalone joined #salt
02:19 Guest18700 joined #salt
02:23 k_sze[work] joined #salt
02:24 subsignal joined #salt
02:25 qu9 joined #salt
02:32 zmalone joined #salt
02:34 brianfeister joined #salt
02:40 aharvey joined #salt
02:46 catpig joined #salt
02:46 ITChap joined #salt
02:46 aharvey Hi foiks any salt-cloud users on?  Trying to solve a seemingly simple thing
02:54 catpigger joined #salt
02:58 voidspacexyz joined #salt
03:00 donmichelangelo joined #salt
03:02 aharvey For the openstack salt-cloud provider, must you pass it uname/passwd hardcoded into $provider.conf file?  There a preferred method of variable insertion for it?
03:10 jmastron joined #salt
03:12 viq joined #salt
03:14 jmastron FYI, looks like the 4.1.3 & 2015.8.4 el7 rpm packages have a broken dependency on python-tornado >= 4.2.1
03:15 mapu joined #salt
03:16 zmalone I think that may have been broken in earlier 2015.8 releases too
03:17 jmastron It didn't block install though
03:18 akhter joined #salt
03:21 jmastron Nevermind, my bad.  Forgot I had repo priority turned on and it was trying to pull from RHEL.  The correct version is in the Salt repos.
03:25 zmalone There's a ticket open to eventually name the salt versions of system packages with a "salt" suffix
03:25 zmalone so that there isn't conflict over whether the system python-tornado should be installed, or the salt python-tornado should be installed
03:25 zmalone sorry, issue, not ticket.
03:28 qu9 joined #salt
03:28 beardedeagle joined #salt
03:34 Guest18700 joined #salt
03:43 beardedeagle joined #salt
03:47 ITChap joined #salt
03:50 beardedeagle So for real though, during the last week I have been packaging winexe 1.1 rpm's for centos 7.2. As far as I can see though, the project is pretty much dead.
03:58 ITChap joined #salt
03:58 hightekvagabond joined #salt
04:01 beardedeagle This could present a real problem to saltstack windows provisioning, packaging is all well and good but are the dev's willing to pick up the project if no one else does.
04:03 PeterO joined #salt
04:07 favadi joined #salt
04:08 iggy is there a replacement? (winrm gets lots of mentions in here, but not having touched windows in years, I don't know)
04:09 mrmak joined #salt
04:10 ITChap joined #salt
04:12 beardedeagle pywinrm is basically it so far, but it can be flaky as hell
04:12 racooper joined #salt
04:16 MTecknology people still use windows?
04:17 joe_n joined #salt
04:18 beardedeagle especially with ssl support pywinrm can get a little quirky. and yeah unfortunately. not by choice though, I just have to support it for the dev's here.
04:20 ramteid joined #salt
04:23 evle joined #salt
04:25 beardedeagle outside of pywinrm there is txwinrm: https://github.com/zenoss/txwinrm which claim to fame is async
04:28 Guest18700 joined #salt
04:34 bbradley mrwboilers: here's a discussion about no previous release in the repo.
04:34 bbradley https://github.com/saltstack/salt/issues/27385
04:34 saltstackbot [#27385]title: repo.saltstack.com contains only current latest release, should contain previous releases, too | I see that only 2015.8.0 (the current latest release) is available in repo.saltstack.com. I would like to roll back to 2015.5.3, or at least 2015.5.5. The repo should contain at least a few prior releases.
04:38 aharvey joined #salt
04:40 jdubski joined #salt
04:40 voidspacexyz joined #salt
05:00 qu9 joined #salt
05:11 Guest18700 joined #salt
05:13 Bryson joined #salt
05:19 zsoftich1 joined #salt
05:21 anmol joined #salt
05:24 quasiben joined #salt
05:24 ITChap joined #salt
05:33 Guest18700 joined #salt
05:46 jhauser joined #salt
05:50 dayid joined #salt
05:50 dayid joined #salt
05:56 calvinh joined #salt
06:03 Guest18700 joined #salt
06:04 voidspacexyz joined #salt
06:05 malinoff joined #salt
06:06 hightekvagabond joined #salt
06:08 krymzon joined #salt
06:09 nidr0x joined #salt
06:10 favadi joined #salt
06:10 zsoftich1 joined #salt
06:14 Crazy67 joined #salt
06:16 nidr0x joined #salt
06:20 Crazy67 joined #salt
06:20 nidr0x joined #salt
06:21 baweaver joined #salt
06:22 lemur joined #salt
06:23 ITChap joined #salt
06:24 Guest18700 joined #salt
06:27 keimlink joined #salt
06:34 tehsu anyone seen this with nova driver at rackspace, Method Not Allowed (HTTP 405)
06:35 udhayprakash joined #salt
06:42 atmosx joined #salt
06:54 Guest18700 joined #salt
07:02 bhosmer_ joined #salt
07:14 dayid joined #salt
07:14 iggy bbradley: don't hold your breath
07:17 opdude joined #salt
07:22 zulgabis joined #salt
07:28 Guest18700 joined #salt
07:48 fredvd joined #salt
07:49 KermitTheFragger joined #salt
07:52 opdude joined #salt
07:53 opdude regarding winexe, I recently had to go through the pain of getting that to work. WinRM seemed to just fall flat for me, I managed to find a winexe binary that I grabbed about a year ago and placed that on my salt server that seemed to work but it's not exactly optimal :)
07:54 dkrae joined #salt
07:56 bhosmer_ joined #salt
07:57 pouldenton joined #salt
08:02 ajw0100 joined #salt
08:03 pouldenton Hello! What is the best practice to write mutually exclusive states? I need to implement simple feature: state #1 must be applied unless state #2 isn't assigned to this minion.
08:06 elsmo joined #salt
08:07 dgutu joined #salt
08:13 Bryson joined #salt
08:13 iggy pouldenton: it requires things you really shouldn't be doing
08:13 iggy how are you targetting state #2?
08:19 viq joined #salt
08:20 ruxu joined #salt
08:22 pouldenton I'm using Salt + Foreman, so i'm trying to implement all targeting through Foreman web interface (it can act instead Salt top files) and i need some kind of protection when several mutually exclusive states can be assigned to host or group of hosts in two mouse clicks.
08:27 pouldenton The simplest way to do this, just no to set such states to host, but you can find out what states are mutually exclusive only by looking through states source code, there no such information in web interface.
08:29 om joined #salt
08:33 s_kunk joined #salt
08:34 AirOnSkin pouldenton: You could define a pillar value (I call them roles) for mutually exclusive states and test for that role in your top file. Then you just need to assign these pillar values (roles) to your minions
08:36 AirOnSkin Which could look something like this in your top.sls: http://hastebin.com/ricubijeye.django
08:37 genunix joined #salt
08:38 amcorreia joined #salt
08:39 pouldenton Yes, actually i'm experimenting with auto assigned grains that should act's like simple flags that can help to determine what software are installed through salt, but i hoped there is a better way to achieve this...
08:39 Guest18700 joined #salt
08:41 AirOnSkin Well, you always need to test against some meta information. And somehow the meta information needs to be created/assigned. I don't see a "better" way. But then again, I'm not the most experienced SaltStack user... ;)
08:43 AirOnSkin Maybe you or iggy can tell me how to test with jinja if something is not (!) in a list? Because {% for port not in salt['pillar.get']('iptables_tcp', []) %} doesn't work. I'm wasting hours on this -.-
08:44 garphy joined #salt
08:48 denys joined #salt
08:48 ruxu joined #salt
08:49 kawa2014 joined #salt
08:52 AirOnSkin Never mind. I'll write a script ...
08:52 pouldenton I think this should be looking like: {% for pillar_port in salt['pillar.get']('iptables_tcp', []) %}  and then {% if pillar_port != port %}...
08:52 iggy pouldenton: sorry, no clue about foreman, in top matching you can negate with compound matchers
08:54 iggy AirOnSkin: you can subtract lists (not _entirely_ sure what you're trying to do
08:55 Guest18700 joined #salt
08:55 iggy AirOnSkin: there's also something like {% for x in list if not x.nope %}
08:56 krymzon joined #salt
08:57 AndreasLutro AirOnSkin: {% if my_var is list %}
08:57 * AndreasLutro implemented this
08:59 slav0nic joined #salt
08:59 av_ joined #salt
09:00 sroegner joined #salt
09:01 GreatSnoopy joined #salt
09:02 pouldenton iggy: actually i think this isn't problem of matching, in many cases  it will be great to know what states are going to be applyed to the minion (of was applyed previously) and react on this, but i can't find this features. It looks like feature request.
09:03 ruxu joined #salt
09:05 opdude_ joined #salt
09:07 malinoff joined #salt
09:11 abednarik joined #salt
09:11 ruxu joined #salt
09:12 kshlm joined #salt
09:12 analogbyte joined #salt
09:13 voidspacexyz joined #salt
09:13 iggy show_highstate
09:13 iggy highstate test=True
09:13 iggy are but a few
09:13 WildPikachu joined #salt
09:15 Guest18700 joined #salt
09:15 AirOnSkin Thanks for the hints. I'll look into them. :)
09:18 stooj joined #salt
09:19 pooogles joined #salt
09:26 calvinh joined #salt
09:28 pooogles joined #salt
09:30 Guest18700 joined #salt
09:31 calvinh joined #salt
09:37 TyrfingMjolnir joined #salt
09:37 impi joined #salt
09:39 Guest18700 joined #salt
09:43 malinoff joined #salt
09:43 malinoff joined #salt
09:43 cyteen joined #salt
09:44 arapaho joined #salt
09:46 cberndt joined #salt
09:50 elsmo joined #salt
09:52 Guest18700 joined #salt
09:52 chiui joined #salt
09:52 totzky joined #salt
10:03 Guest18700 joined #salt
10:10 keimlink joined #salt
10:10 abednarik joined #salt
10:10 LondonAppDev joined #salt
10:11 malinoff joined #salt
10:11 saltyswede Is it possible to push files to minions when they boot automaticly?
10:15 ggoZ joined #salt
10:17 calvinh joined #salt
10:18 babilen saltyswede: You could execute any state and/or execution module when the minion comes online via reactors or startup_states
10:18 babilen See https://docs.saltstack.com/en/develop/topics/reactor/index.html and https://docs.saltstack.com/en/latest/ref/states/startup.html for details
10:19 babilen You might be after presence events, but there are plenty of other events to react to
10:20 malinoff joined #salt
10:21 saltyswede Thank you, Ill take a look
10:21 ingslovak joined #salt
10:22 laidback joined #salt
10:24 babilen saltyswede: The idea would be to simply run a state with file.* states in it (e.g. file.managed or file.recurse)
10:24 chiui joined #salt
10:25 abednarik joined #salt
10:26 usernkey joined #salt
10:26 laidback Hello, i wonder if it is possible to separate formulas in a way, where i have e.g. a users formula with a defined behaviour and pillar. I would like to reuse it as a 'base' formula e.g. in my apache formula to setup www-data users. How could one separate the pillar in a way, where i use users-formula first to setup admin user etc. and later installing apache and calling users again with pillar for www-data defined in another pillar b
10:27 Guest18700 joined #salt
10:28 iggy just vary the pillar data and cross your fingers that it merges correctly
10:28 malinoff joined #salt
10:29 usernkey Hello, I'm trying to figure out on which release the failover functionality got broken I was able to find the rpm for the  2015.5.* but I'm not able to find all the old 2015.8.* packages.
10:29 babilen laidback: I'd document that users should use the users formula, define requisites on the user.present states and simply use appropriate pillar data
10:29 usernkey Do you have any clue where should I look for
10:29 babilen laidback: What would be nice in this context if you could "symlink/rename" pillar data, but that is not possible, so you just have multiple first level pillar keys
10:30 usernkey RHEL salt repository  store only the latest release
10:31 laidback that means when i have a bigger provisioning plan for a server where i use users formula multiple times for different other states, it is my responsibility to have a) users-formula written functional and idempotent and b) to ensure my pillar structure is clean so it merges correctly
10:35 giantlock joined #salt
10:37 laidback babilen: it would be very convenient to include e.g. users-formula from apache to setup www-data and be able to hand over specific pillar struct, e.g. the pillar users structure for www-data from the apache top level in pillars.
10:37 Guest18700 joined #salt
10:38 malinoff joined #salt
10:39 impi joined #salt
10:40 LondonAppDev joined #salt
10:41 AirOnSkin So it seems, that when I make changes to iptables rules and restart the iptables service, the salt minion won't return anymore for following salt commands. Any ideas how to work around this?
10:42 babilen laidback: So far there is no formal way to include other formulas, but you can document it that way and simply rely on the user making sure that all needed formulas are available.
10:42 babilen laidback: And your observations are correct, yes
10:44 laidback Many thanks, i was wondering if i miss something, but it seems to me that i need to structure my file layout accordingly.
10:45 catpig joined #salt
10:45 amcorreia joined #salt
10:48 malinoff joined #salt
10:48 babilen laidback: A thing that I find important about pillars is that their location in pillar_roots does not have to corelate with their content. That is: You can easily have a "user: foo ..." block in pillar_roots/web/example.com/server.sls
10:48 babilen +r
10:48 dstokes joined #salt
10:50 SunPowered joined #salt
10:50 laidback babilen: This is exactly what i want to do. The position of a users:www-data pillar data should be a direct counterpart to the apache state.
10:51 joyrida08 joined #salt
10:52 babilen aye :)
10:53 babilen And whilst you are at it: Read through https://docs.saltstack.com/en/latest/ref/configuration/master.html#pillar-source-merging-strategy
10:57 calvinh_ joined #salt
11:01 Guest18700 joined #salt
11:07 calvinh joined #salt
11:09 chiui joined #salt
11:21 malinoff joined #salt
11:24 Guest18700 joined #salt
11:25 antpa joined #salt
11:33 Guest18700 joined #salt
11:37 rotbeard joined #salt
11:38 antpa I need some help configuring the nginx-formula from https://github.com/saltstack-formulas/nginx-formula.  I wrote a custom nginx state before that works fine for servers hosting a single domain name but this limits me to a single vhosts file on the server.  I need to be able to deploy multiple vhosts to a single minion, each with a different domain name (3 small WordPress websites for a client).  I use the nginx-extras package from Dotdeb as it includes
11:38 antpa the Google PageSpeed module and I have several customizations in the vhost file.  Could someone work with me today to set this up?  I'm can pay via Paypal for the support.
11:40 SunPowered antpa: If you can share the intended vhosts output, I can help you write your pillar structure
11:40 antpa SunPowered : nice :)  I'll post a gist with the existing files that I use.
11:41 chiui joined #salt
11:43 giantlock joined #salt
11:45 Guest18700 joined #salt
11:46 antpa SunPowered: I used a period in the filename instead of a slash due to Gist filename requirements.  https://gist.github.com/anthonyscolaro/06d1191fe1c333287640
11:47 antpa SunPowered: There are some other configuration files like the SSL stuff and then state files for a couple other platforms we manage besides WordPress (Magento, WHMCS, Toran).  I left those out for now to avoid confusion and focus strictly on WordPress for now.
11:52 et0r joined #salt
11:53 antpa SunPowered: I'm not sure if I should be creating a pillar for the normal nginx or nginx.ng...  Since it's not a PPA but a custom repo, Dotdeb.
11:53 SunPowered ah, i see.  My experience has been better with nginx.ng myself as it allows for the entire vhosts config to be written in a pillar data
11:54 SunPowered all it would do is render out a nginx.conf and vhosts config, so if these configs are working now then they should work the same with nginx.ng
11:55 SunPowered in general, this is not too bad with nginx.ng.  Some of the logic of your environment checks can be better managed via pillar environment
11:55 Buzer Anyone know if there is an easy way to escape something that will be fed as parameter to shell script in Jinja2? e.g. I would like to something like "check-process.rb -p {{ pattern|yaml_dquote }}" in pillar and the pattern variable would need to quoted and all quotes inside it escaped
12:00 bluenemo joined #salt
12:01 hoonetorg_ joined #salt
12:02 Guest18700 joined #salt
12:02 antpa SunPowered: Sounds good.  When I looked at the formula, I automatically leaned towards nginx.ng and I'm perfectly fine with that as long as I can convert my current state to it so we can deploy multiple vhosts for individual domain names.  I may need help creating the pillars for a couple domain names at first and possibly making sure SSL certs are working too. Can you help me set this up?
12:03 antpa SunPowered: There would be no difference between vhost files besides the domain name.
12:04 pviktori joined #salt
12:05 SunPowered sure,  multiple hosts are all just entered as extra pillar data
12:08 antpa SunPowered: You'll see that I store vhost and log files in a custom directory.  Could you help me set up the initial pillar data for a couple example domain names as well as specify file name/location within the pillar directory?
12:08 SunPowered sure thing
12:09 chiui joined #salt
12:09 LondonAppDev joined #salt
12:10 et0r left #salt
12:11 abednarik joined #salt
12:17 sfxandy joined #salt
12:20 akhter joined #salt
12:28 `chris joined #salt
12:28 impi joined #salt
12:28 ruxu joined #salt
12:29 krymzon_ joined #salt
12:30 anmol joined #salt
12:31 justanotheruser joined #salt
12:33 pooogles_ joined #salt
12:34 mrwboilers joined #salt
12:35 kawa2014 joined #salt
12:36 sarlalian joined #salt
12:38 ruxu joined #salt
12:41 abednarik joined #salt
12:41 afics left #salt
12:41 afics joined #salt
12:41 chiui joined #salt
12:46 lordl4zaruz joined #salt
12:47 lordl4zaruz Hi. I hope this is the right place for salt-cloud questions: I've setup salt-cloud to speak to my Kilo-deployment and spawning instances works just fine. Bootstrapping salt does not however and I think I've identified a problem: there's no "private_ips" returned. Any ideas?
12:47 lordl4zaruz Wouldn't surprise me if I misconfigured something...
12:53 human01d joined #salt
12:55 human01d hello, can you help with new minion, after update to new version salt-minion 2015.8.4 (Beryllium). We deployed old states, which worked on old version of minions, but with new version we got errors
12:56 human01d Source file salt://auth/ssh-keys/........ is missing
12:56 human01d in debug salt-minion writes
12:57 babilen human01d: You really shouldn't have deployed that version .. you *might* have encountered https://github.com/saltstack/salt/issues/30820
12:57 saltstackbot [#30820]title: State runs involving watch_in or extending break on 2015.8.4 | Environment: rackspace, Ubuntu 14.04, salt packages from `http://repo.saltstack.com/apt/ubuntu/14.04/amd64/latest`...
12:57 CeBe joined #salt
12:57 human01d [INFO    ] Executing state ssh_auth.present for osham-ibm-ssh-key
12:57 babilen human01d: Ah, no .. that would look differently
12:57 human01d [DEBUG   ] Could not find file from saltenv 'base', u'salt://auth/ssh-keys/osham.pub'
12:57 human01d [DEBUG   ] Could not find file from saltenv 'base', u'salt://auth/ssh-keys/osham.pub'
12:58 human01d [ERROR   ] Failed to add the ssh key. Source file salt://auth/ssh-keys/osham.pub is missing
12:58 babilen Please don't paste into the channel, but use one of http://refheap.com, http://paste.debian.net, https://gist.github.com, http://sprunge.us, …
12:58 babilen What does "salt 'theminion' cp.list_master" give you?
13:00 quasiben joined #salt
13:00 human01d babilen: it show me all files on defined environment
13:01 human01d babilen: is that correct?
13:02 babilen human01d: Does that list contain salt://auth/ssh-keys/osham.pub ?
13:02 human01d what i see, from debug that salt:// tries to load files from "base" environment but it must use another one
13:02 human01d yes
13:02 human01d - auth/ssh-keys-revoke/osham.pub
13:02 babilen ssh-keys-revoke != ssh-keys
13:03 human01d - auth/ssh-keys/osham.pub
13:03 human01d sorry :)
13:03 human01d you are right
13:03 human01d but ss-key is also threre :)
13:03 babilen Okay, so you are using environments and the minion is both in base and another one?
13:03 babilen Sorry .. it is just that those details matter ;)
13:05 human01d the full command is #salt target state.sls auth.users-ssh-key env=tuilte
13:05 human01d and i see from debug of minions [INFO    ] Fetching file from saltenv 'tuilte',
13:06 human01d but when it come to salt://
13:06 babilen A flood environment
13:06 human01d it use base environment
13:06 babilen :)
13:06 human01d it must stay with tuilte and not with base
13:07 human01d strange :\
13:07 Guest18700 joined #salt
13:08 babilen Well .. I can't really comment on that as I simply use environments to model development workflows (i.e. dev, qa, prod) on separate masters. I never agreed, liked nor understood the way many people use environment for other semantics.
13:09 babilen Could it just be an issue with processing the passing of "env=tuilte" (I take it that you have corresponding targeting in your top.sls) ?
13:09 chiui joined #salt
13:10 Buzer hmm... Is it possible that "salt 'minion' state.highstate test=True" on master and "salt-call state.highstate test=True" on minon give different results?
13:11 impi joined #salt
13:11 AndreasLutro everything is possible
13:11 AndreasLutro I wouldn't expect it though
13:11 Buzer For some reason when I execute on master, "unless" check fails on test '"$(blkid -s TYPE -o value /dev/vgname/volumename)" == "ext4"', but not on minion
13:12 Buzer s/test '/' test/
13:13 Edur joined #salt
13:14 babilen human01d: Is that on gitfs?
13:16 human01d babilen: i added it to top.sls, but have the same result. Only the - source: salt://file_name use "base" environment inspite of all other state use "tuilte" environment :(
13:16 human01d no we do not use gitfs
13:16 anmol joined #salt
13:17 rbjorklin joined #salt
13:18 human01d if i execute top on another salt-minion it works :). The difference is in the version on minions.
13:19 human01d looks like a new version of minion handles source : salt:// differently
13:19 DammitJim joined #salt
13:20 Guest18700 joined #salt
13:21 JTeatime joined #salt
13:21 babilen human01d: I know that Erik has made some changes to the way environments are handled .. You might want to look into "git log v2015.8.3..v2015.8.4" for suspicious commits
13:22 bhosmer joined #salt
13:22 human01d also in our configuration, we do not have "base" environment at all, we he considered "base" at all :)
13:22 human01d can we install older version of salt-minions?
13:22 human01d looks like you removed old rpms :(
13:22 giantlock joined #salt
13:23 Edur joined #salt
13:24 babilen I didn't do anything :)
13:24 human01d :) okay
13:24 KennethWilke joined #salt
13:25 babilen https://github.com/saltstack/salt/commit/fa6b1b30228f8324c55944c8181ff62fb41e8a92 might be it
13:25 babilen But then .. *shrug*
13:25 babilen Please file an issue on https://github.com/saltstack/salt/issues
13:26 babilen But you really don't want to deploy .4
13:26 babilen (too late now I guess)
13:26 antpa joined #salt
13:28 human01d it only one minions :), we will not continue at this moment :)
13:29 human01d i will create issue, thank you for help anyway.
13:29 human01d babilen: :}
13:30 lompik joined #salt
13:31 Guest18700 joined #salt
13:33 babilen human01d: If you like you could revert the change in that commit on the minion and see if that changes the behaviour (just edit the file directly)
13:33 babilen And mention the issue number to me ..
13:34 babilen (I'd like to subscribe)
13:34 insomniaSalt joined #salt
13:37 voidspacexyz joined #salt
13:38 bhosmer joined #salt
13:40 chiui joined #salt
13:43 Guest18700 joined #salt
13:45 AndreasLutro does anyone know whether mine.update makes the minion pull mine data from the master, pushes its mine data to the master, or both?
13:46 subsignal joined #salt
13:47 gh34 joined #salt
13:49 quasiben joined #salt
13:49 DammitJim why would I be getting a: only root can do that for using a state with mount.mounted?
13:51 dgutu joined #salt
13:53 evle1 joined #salt
13:54 impi joined #salt
13:55 numkem joined #salt
13:57 opdude_ @DammitJim did you specify - user: ?
13:57 DammitJim yeah, but isn't user the guy that owns the mount?
13:57 DammitJim not the one executing the mount command?
13:58 lordl4azaruz joined #salt
13:58 DammitJim user: The user to own the mount; this defaults to the user salt is running as on the minion
13:58 Edur joined #salt
14:00 DammitJim that's messed up
14:00 DammitJim maybe there is a different explanation
14:00 DammitJim + I must have something configured wrong
14:02 abednarik joined #salt
14:03 afics joined #salt
14:06 penguin_dan joined #salt
14:06 Edur joined #salt
14:07 babilen Ah, you found my "if user == 'DammitJim': mess_things_up()" code ;)
14:08 DammitJim dammit!
14:08 totzky joined #salt
14:10 chiui joined #salt
14:10 mapu joined #salt
14:14 Guest18700 joined #salt
14:16 malinoff joined #salt
14:18 penguin_dan joined #salt
14:18 JDiPierro joined #salt
14:21 antpa joined #salt
14:23 impi joined #salt
14:26 Guest18700 joined #salt
14:27 treaki_ joined #salt
14:29 giantlock joined #salt
14:31 jdubski joined #salt
14:31 ALLmightySPIFF joined #salt
14:35 racooper joined #salt
14:42 chiui joined #salt
14:42 _JZ_ joined #salt
14:46 denys joined #salt
14:47 JDiPierro joined #salt
14:51 Guest18700 joined #salt
14:52 Tyrm joined #salt
14:58 quasiben joined #salt
14:58 Guest18700 joined #salt
14:59 N-Mi joined #salt
15:00 mpanetta joined #salt
15:01 zerthimon joined #salt
15:03 numkem joined #salt
15:03 zerthimon how do I clean up old jobs in /var/cache/salt/master/jobs ?
15:05 zmalone joined #salt
15:07 XenophonF zerthimon: https://docs.saltstack.com/en/latest/topics/jobs/job_cache.html
15:09 rm_jorge joined #salt
15:09 XenophonF tl;dr is that it should be automatic
15:10 XenophonF anything older than 24 hours should get purged
15:10 chiui joined #salt
15:11 zerthimon XenophonF: I have a lot of files and dirs there
15:13 XenophonF you can use find's -delete and -ctime/-mtime/-atime flags to purge it manually
15:14 scoates joined #salt
15:14 zerthimon XenophonF: ok, just wanted to make sure I can safely delete stuff there
15:14 zerthimon XenophonF: thanks
15:15 XenophonF probably
15:15 Tanta joined #salt
15:16 ALLmightySPIFF joined #salt
15:17 TyrfingMjolnir joined #salt
15:17 Guest18700 joined #salt
15:18 CheKoLyN joined #salt
15:19 morissette joined #salt
15:20 spiette joined #salt
15:23 berserk joined #salt
15:26 Guest18700 joined #salt
15:29 jorrel joined #salt
15:31 perfectsine joined #salt
15:33 coderMe joined #salt
15:38 hightekvagabond joined #salt
15:40 chiui joined #salt
15:41 spuder joined #salt
15:42 Guest18700 joined #salt
15:44 perfectsine joined #salt
15:44 hasues joined #salt
15:47 rotbeard joined #salt
15:47 zsoftich1 joined #salt
15:48 huddy joined #salt
15:50 Heartsbane joined #salt
15:50 Heartsbane joined #salt
15:56 Guest18700 joined #salt
15:57 antpa joined #salt
15:57 jorr-el joined #salt
15:57 jorr joined #salt
15:58 Brew joined #salt
16:04 jorr-el joined #salt
16:05 shaggy_surfer joined #salt
16:09 Shirkdog joined #salt
16:12 LondonAppDev joined #salt
16:13 chiui joined #salt
16:21 jorr-el joined #salt
16:22 TyrfingMjolnir joined #salt
16:24 onlyanegg joined #salt
16:26 iggy zerthimon: it's actually safe to wipe the entire cache dir
16:26 Bryson joined #salt
16:27 iggy (it was sadly part of my normal work flow at $lastjob due to oddities surrounding some of the tech we were using)
16:27 iggy by cache dir, I mean /var/cache/salt/
16:28 berserk joined #salt
16:33 StolenToast dumb question: if a state is referencing a pillar value at "ntp:servers" in my pillar file will that show up as an "ntp:" header and a "servers:" definition under it?
16:33 tligda joined #salt
16:33 gtmanfred ntp:
16:33 yaryarrr joined #salt
16:33 gtmanfred servers:
16:33 gtmanfred - thing
16:34 gtmanfred salt['pillar.get']('ntp:servers')
16:34 gtmanfred if it is that, then yes
16:34 StolenToast k thanks
16:34 StolenToast I've been having a lot of pillar troubles
16:34 akhter joined #salt
16:37 adamisos joined #salt
16:39 Guest18700 joined #salt
16:39 StolenToast I can't I can't seem to use the pillar module to query for any pillar data
16:40 StolenToast states will receive pillar data for my ssh state but not for my ntp state
16:40 StolenToast and pillar.item(s) never returns anything
16:41 chiui joined #salt
16:41 snicers-work joined #salt
16:42 michelangelo joined #salt
16:42 adamisos greetings….i’m told this is the place to get some salt help
16:42 babilen It is indeed
16:42 gtmanfred sea or iodized?
16:43 snicers-work adamisos, this or github or the google forum yea
16:43 adamisos what do peeps use to test jinga logic without having to highstate everything?
16:43 babilen StolenToast: There is *very* little we can do to help without more information. You might want to consider pasting your pillar_roots content, top.sls, applicable commands and their output to one of http://refheap.com, http://paste.debian.net, https://gist.github.com, http://sprunge.us, …
16:43 gtmanfred adamisos: state.show_low_sls or state.show_highstate
16:44 babilen adamisos: I typically develop stuff on local vagrant boxes
16:44 subsignal joined #salt
16:45 adamisos i’ve used vagrant before….awesome!
16:46 adamisos but like something relatively simple like working with pillars….for example
16:46 adamisos {% set short_version = {{ pillar['db']['version'] | replace('.','') }} -%}
16:46 malinoff joined #salt
16:46 tawm04 joined #salt
16:46 adamisos i’m not sure this is a good way to strip a dot from a pillar variable to create one without the dot
16:47 babilen I'd strongly recommend to use salt['pillar.get']('db:version')
16:47 dh__ joined #salt
16:47 honestly joined #salt
16:47 honestly joined #salt
16:47 adamisos so if version=9.3, I want short_version=93
16:47 babilen That's how you would do it
16:48 catpig joined #salt
16:48 gtmanfred get rid of the {{}}
16:48 adamisos ok, cool….
16:49 adamisos so
16:49 gtmanfred {% set short_version = pillar.db.version|replace('.', '') %}
16:49 adamisos : {% set short_version = pillar['db']['version'] | replace('.','') -%}
16:49 gtmanfred yeah
16:49 adamisos ok
16:49 gtmanfred you are already escaped into the jinja shell thingy, and can reference pillar in there
16:49 gtmanfred or salt
16:49 adamisos so if I wanted to test on the cmd line
16:50 gtmanfred state.show_low_sls statename
16:50 gtmanfred then check that it fills out waht you want
16:50 cswang_ joined #salt
16:50 sifusam joined #salt
16:51 adamisos awesome! i’ll give that a shot
16:51 adamisos thanks guys!
16:51 norobot joined #salt
16:52 norobot Hi, I was looking in the docs but found nothing, is there some kind of force-reload option for the service module ? Because it appears the reload option will call service my_service reload and my_service doesn't have the reload option ...
16:53 norobot But it seems that force-reload is kind of standard (from what I understand) and will reload if it exists, otherwise restart
16:54 writtenoff joined #salt
16:55 ajw0100 joined #salt
16:56 gtmanfred it does not look like there is a force-reload
16:56 norobot ok :/
16:56 norobot just asked because it seems the doc is not always up to date
16:57 norobot thanks anyway
16:58 spuder_ joined #salt
16:58 AndreasLutro norobot: - force: true
16:58 AndreasLutro on the service.running state
16:58 StolenToast babilen: well this took me long enough that it's now lunchtime but https://gist.github.com/anonymous/605d90d3ff416a7b067e
16:59 Guest18700 joined #salt
16:59 tooth joined #salt
16:59 StolenToast I'll ask again after lunch
16:59 norobot AndreasLutro: if we put force and reload options together to true, will it trigger the force-reload option ? Or force alone is sufficient ?
17:01 gtmanfred it will not do a force-reload i don't think, i think it will force a restart?
17:01 gtmanfred it is worth testing it though
17:02 Tyrm_ joined #salt
17:02 babilen StolenToast: Where do you target pillar.sls to that host? That top.sls appears to be the *states* top.sls and not the pillar's top.sls
17:03 norobot Yes, I agree, but maybe it will try to first reload and force restart if it failed ? Juts curious on that point
17:03 gtmanfred i don't think so, but you would have to consult the code to find out
17:03 akhter joined #salt
17:04 gtmanfred actually
17:04 gtmanfred force will do a service.force_reload
17:04 gtmanfred from inside of mod_watch
17:04 gtmanfred so if you hit it using listen or watch, and you have force in service.running
17:04 AndreasLutro norobot: I think you need reload as well, not sure though
17:04 feld does anyone know what commit resolved CVE-2016-1866
17:04 gtmanfred it will do a service.force_reload
17:05 jgarr joined #salt
17:05 feld This seems very difficult to track down
17:05 gtmanfred norobot: https://github.com/saltstack/salt/blob/develop/salt/states/service.py#L521
17:05 norobot thanks i was looking for the code :P
17:05 ruxu joined #salt
17:05 norobot ok so it needs both options
17:07 norobot thanks
17:07 AndreasLutro feld: got a link for that cve?
17:08 AndreasLutro mitre just says it's reserved, so it's not been publicised yet?
17:10 abednarik joined #salt
17:11 babilen basepi: Could you push the 2015.8.5 tag please?
17:11 gtmanfred the tag is up
17:11 gtmanfred just not packaged yet
17:12 gtmanfred https://github.com/saltstack/salt/releases/tag/v2015.8.5
17:12 kevinqui3nyo how might i get a list of all minions from within a runner
17:12 kevinqui3nyo i figured there might be a method in the salt.client.LocalClient() object but i cant find it
17:13 gtmanfred do you want to know all the ones connected, or responding?
17:13 gtmanfred you could get a wheelclient and list the salt keys
17:13 kevinqui3nyo they dont need to be responding i just want a list of all of them by their minion_id name
17:13 feld AndreasLutro: it's in the release notes for 2015.8.4
17:13 feld AndreasLutro: MITM can do code execution on minions :(
17:13 feld https://docs.saltstack.com/en/latest/topics/releases/2015.8.4.html
17:14 feld basepi: ping -- which commit is relevant to the CVE?
17:14 kevinqui3nyo thanks gtmanfred
17:14 zmalone feld: one moment, and I'll pull it up
17:14 feld zmalone: thanks
17:15 malinoff joined #salt
17:15 zmalone https://github.com/saltstack/salt/pull/30613/files
17:15 babilen gtmanfred: I guess my refspecs weren't in order. I have it now, ta!
17:15 saltstackbot [#30613]title: Fix minion/syndic clearfuncs |
17:17 feld zmalone: perfect; i've pushed the vuln entry into the FreeBSD vuxml now.
17:17 zmalone although, depending on where you got your 2015.8.4 release announcement, 2015.8.5 is coming soon.
17:17 Guest18700 joined #salt
17:17 zmalone 2015.8.4 is broken
17:17 feld zmalone: yeah, i'm waiting for 2015.8.5 before i push to the stable package branch
17:18 babilen gtmanfred: Or rather I needed an explicit "fetch --tags" as the tag is not on any branch yet.
17:18 malinoff joined #salt
17:20 gtmanfred ahh
17:20 brianfeister joined #salt
17:21 giantlock joined #salt
17:25 sjorge joined #salt
17:25 sjorge joined #salt
17:26 kevinqui3nyo gtmanfred: any idea why i'm getting None with this: print wheel.cmd('key.list_all')
17:28 gtmanfred nope
17:32 gordonm joined #salt
17:33 JDiPierro joined #salt
17:36 Guest18700 joined #salt
17:38 kevinquinnyo gtmanfred: weird.  for some reason it works fine if i call the function directly instead of through wheel.cmd (which i believe is a loader for the various classes with functions)
17:38 kevinquinnyo wheel.functions.key.list_all()  works as expected for instance
17:39 gtmanfred that is odd :/
17:39 SpX joined #salt
17:39 gtmanfred yeah that should be what it is doing
17:39 malinoff joined #salt
17:41 jim__ joined #salt
17:44 baweaver joined #salt
17:45 Guest18700 joined #salt
17:47 basepi feld: are you looking for fix or breakage?
17:50 feld basepi: i was looking for the commit that fixed it, which I've now found thanks to zmalone
17:51 StolenToast babilen: ah, a dumb mistake, I think that's it
17:51 StolenToast thanks
17:53 sroegner joined #salt
17:54 eriko_ joined #salt
17:54 jim__ anyone ever tried to use salt-cloud to provision EC2 instances with multiple security groups attached?
17:56 voidspacexyz joined #salt
17:59 horus_plex jim__: yeah. gist coming
18:00 jcockhren ^
18:00 basepi feld: great, glad you got help. :)
18:00 babilen StolenToast: You are most welcome
18:01 Tyrm joined #salt
18:02 v0rtex joined #salt
18:02 JDiPierro joined #salt
18:02 andrew_v joined #salt
18:06 Guest18700 joined #salt
18:09 chiui joined #salt
18:11 jcockhren jim__: I'll do one better. Multiple security security groups and extending to set the subnet
18:11 jcockhren jim__: https://gist.github.com/jcockhren/c23394f1b9e8efd075de
18:11 jcockhren jim__: it assumes you're launching in VPC
18:16 antpa joined #salt
18:17 snicers-work Is there a way to have salt cloud install a specific version of the salt-minion?
18:17 lumtnman joined #salt
18:18 lumtnman joined #salt
18:21 akhter joined #salt
18:21 dizzythinks joined #salt
18:22 antpa joined #salt
18:23 Dev0n_ hey, is it not possible to specify which Dockerfile to use (-f) when using the dockerng build image state?
18:24 hightekvagabond joined #salt
18:25 jcockhren snicers-work: see my gist right about your comment
18:26 cpowell joined #salt
18:28 shaggy_surfer joined #salt
18:29 dizzythinks joined #salt
18:30 snicers-work jcockhren, I don't currenrly specify `script: bootstrap-salt`, is that needed?
18:31 jcockhren snicers-work: yeah if you want to use the script the pulls code from github opposed to the minion's distribution repo
18:31 GreatSnoopy joined #salt
18:32 snicers-work I figured the minion repo would mirror that of the master, is this not the case?
18:32 jcockhren snicers-work: salt repos don't store older versions of packaged salt releases
18:32 snicers-work http://repo.saltstack.com/apt/ubuntu/14.04/amd64/archive/2015.8.3 <-
18:32 chiui joined #salt
18:33 snicers-work I am using that on my  master.
18:33 Lionel_Debroux_ joined #salt
18:33 jcockhren snicers-work: nah. it'll install whatever it latest at the time of provision
18:33 jcockhren snicers-work: so if you wait long enough, new minions could have different versions installed
18:33 snicers-work guh
18:33 snicers-work ok
18:34 jcockhren snicers-work: also avoid 2015.8.3 -> https://docs.saltstack.com/en/latest/topics/releases/2015.8.4.html
18:34 snicers-work Seems like you would always want your minions on the same version.
18:34 snicers-work jcockhren, 2015.8.4 has a regression in it that prevents aws elb from being brought up with boto_elb.
18:34 jcockhren snicers-work: yeah. in order to have more controll over that. I just use tags from github oppose to relying on salt repos or out of date distro repos
18:35 jcockhren snicers-work: 2015.8.5 should be getting tagged today
18:35 TyrfingMjolnir joined #salt
18:35 jcockhren snicers-work: already tagged -> https://github.com/saltstack/salt/tree/v2015.8.5
18:35 snicers-work Sure, but I need to figure th is out anyway in case of issues like this moving forward.
18:36 jcockhren it's MUCH easier to handle upgrades when using the script to pull tagged versions from github
18:36 jcockhren and rollback
18:36 jcockhren +s
18:37 snicers-work Doesn't look like 2016.8.5 fixes my issue.
18:37 Borromini joined #salt
18:37 snicers-work Not yet at least.
18:41 zmalone jcockhren: https://repo.saltstack.com/apt/ubuntu/ubuntu14/archive/
18:41 zmalone the main repo doesn't, but they keep other versions in different archive/<version> repos
18:42 jcockhren zmalone: ah
18:42 jcockhren nice
18:42 hightekvagabond I have a dependency to make sure a particular version glibc is on some machines, whats the best way to approach that?
18:42 zmalone I'd be surprised if any fixes except https://github.com/saltstack/salt/issues/30820 make it into 2015.8.5
18:42 saltstackbot [#30820]title: State runs involving watch_in or extending break on 2015.8.4 | Environment: rackspace, Ubuntu 14.04, salt packages from `http://repo.saltstack.com/apt/ubuntu/14.04/amd64/latest`...
18:43 zmalone that was the reason 2015.8.5 got tagged the day of 2015.8.4's announcement
18:43 babilen zmalone: That's the *only* change in 2015.8.5
18:43 babilen (as you suspected)
18:43 babilen hightekvagabond: Only deploy on distribution releases with that glibc version?!
18:44 hightekvagabond I meant to manage it with sls :P
18:45 babilen hightekvagabond: I would strongly recommend to *not* tinker with the glibc version on your system and managing that with salt doesn't change that. If you require a specific version then make sure that your minion is on the respective distribution release.
18:46 Borromini hi guys. I'm setting up mysql through salt but it looks like salt's choking on the special characters on the mysql root password (contains ? ! and #). If I specify the password in plaintext in the state file (or in /etc/salt/minion.d/), it fails to connect; if I tell it to use the hash, it works fine, but since there is a mismatch between the hash and what salt translates the plaintext password to, it rese
18:46 andrew_v_ joined #salt
18:46 Guest18700 joined #salt
18:46 Borromini ts the root user password. I've tried using yaml_encode in the pillar file containing the password as well, it still breaks... How can I work around this short of changing my password? My configs and mysql state test output: https://paste.debian.net/378388/
18:46 babilen hightekvagabond: In the end: Think about how you'd do it "manually" and then, once you got that working, you'd use the same method with salt. Given that this is a horrible idea, you shouldn't do it with salt either.
18:46 zmalone babilen/hightekvagabond: Or install it in a completely non-standard location and use chroots, library paths, or something like that to point only the thing which depends on it to your special glibc
18:47 akhter joined #salt
18:47 zmalone sometimes that's the proper solution to running commercial software after it's expiration date
18:47 babilen zmalone: aye, but then .. if you have software that requires a specific glibc then just run the corresponding distribution.
18:47 perfectsine joined #salt
18:47 hightekvagabond you assume I get to define the distro :P
18:47 babilen But I'm sure that this is some proprietary binary FUBAR that's just horrible
18:48 zmalone Might not be an option if you are talking about an old glibc, and so you just run the single piece against it.
18:48 berserk joined #salt
18:48 zmalone and hope for the best
18:48 babilen hightekvagabond: Well, why don't you start with the *actual* problem you are trying to solve? (e.g. "install FOO on BAR")
18:49 snicers-work jcockhren, /tmp/.saltcloud-de15e49e-853f-481e-a23d-94f9e048f7f1\'"\'"\' git v2015.8.3\'' failed. Exit code: 1
18:50 mpanetta_ joined #salt
18:51 jcockhren snicers-work: is that from your salt-cloud call?
18:52 snicers-work yea.
18:53 snicers-work jcockhren, ^
18:53 Guest18700 joined #salt
18:54 beardedeagle joined #salt
18:54 jcockhren snicers-work: firstly, -> https://docs.saltstack.com/en/latest/topics/releases/2015.8.4.html#security-fix
18:55 jcockhren snicers-work: also, I haven't vetted 2015.8.x as of yet. I'm still 2015.5.x
18:55 jcockhren so I might not be the best person to advise you at this time
18:55 snicers-work understood, this is not in production yet but I need to move passed this ELB issue. I can upgrade my stack later as it is all on docker.
18:55 snicers-work past*
18:57 snicers-work Wonder why this is failing.
18:58 baweaver joined #salt
18:59 numkem joined #salt
19:00 hasues left #salt
19:02 woodtablet joined #salt
19:02 woodtablet hi guys
19:03 aw110f joined #salt
19:04 woodtablet so i am in a weird place. i have started installing salt-minion on a bunch of hosts and found that people want the host name rather than the fqdn for the minion_id. is there a way to configure the /etc/salt/minion #id line to use `hostname` rather than the python - socket.getfqdn()  ? i want to do this programatically for hundreds of nodes and not set it manually
19:04 denys joined #salt
19:08 Guest18700 joined #salt
19:10 kevinquinnyo woodtablet: i know that it's defined in /etc/salt/minion_id -- but manually updating that is probably not ideal, and even if you did i assume you'd have to restart the minions afterwards
19:10 kevinquinnyo restart the salt-minion service rather
19:12 chiui joined #salt
19:13 lumtnman joined #salt
19:13 woodtablet keveinquinnyo - thats why I was thinking updating the /etc/salt/minion file before the minion_id was generated
19:13 r05c03 joined #salt
19:13 woodtablet kevinquinnyo - since uncommenting #id should override the minion_id file
19:14 chiui_ joined #salt
19:14 ViciousLove joined #salt
19:14 mavhq joined #salt
19:16 kevinquinnyo woodtablet: not sure
19:17 TyrfingMjolnir joined #salt
19:18 baweaver joined #salt
19:18 dh joined #salt
19:19 kevinquinnyo https://gist.github.com/kevinquinnyo/1997f6f5f362777e5e8f  anyone know if my syntax is wrong here for calling a "runner" from a "reactor"  (the code in the gist is a reactor)
19:20 kevinquinnyo i get res[name]['__sls__'] = fn_ line 58 in render_reaction /usr/lib/python2.7/dist-packages/salt/utils/reactor.py
19:21 kevinquinnyo TypeError: 'str' object does not support item assignment
19:21 kevinquinnyo .. so either res, or res[name] is a str instead of a dict i guess
19:21 brianfeister joined #salt
19:23 woodtablet Anyone know if I should check the master's key ? Or if the minion can check the master's  key ? Like any config management system this is a giant backdoor, and I want to do some sanity checking (for my own sanity).
19:24 hightekvagabond joined #salt
19:24 woodtablet i think a person would either need to have root on the minion to change the master, and in that case I am screwed, or hack my dns to change the master.
19:28 kevinquinnyo https://docs.saltstack.com/en/latest/topics/reactor/index.html#calling-runner-modules-and-wheel-modules  does this actually work?  No matter what runner i try to call from a reactor, it results in the error i pasted above
19:29 hightekvagabond woodtablet, can you put the fqdn in the hosts file on the minions?
19:30 TyrfingMjolnir joined #salt
19:30 woodtablet hightekvagabond: its already in the host file on some of them, it looks like it does the socket.fqdn() look up first
19:31 dlam joined #salt
19:31 gtmanfred basepi: yo, yall working on packaging 2015.8.5?
19:32 woodtablet hightekvagabond: ohh you might be right i have both names in the host file
19:33 hightekvagabond yeah, I was going to be very surprised by that
19:34 ViciousLove 2015.4 Just came out yesterday, they putting out 2015.8.5 already?
19:34 zmalone There was a show stopping bug in 2015.8.4
19:34 ViciousLove 2015.8.4*
19:34 zmalone but 2015.8.4 included a security announcement that needed to be fixed immediately
19:35 ViciousLove Why even put it out then hahah
19:35 babilen ViciousLove: There was a rather pesky regression: https://docs.saltstack.com/en/latest/topics/releases/2015.8.5.html
19:35 kevinquinnyo ha, i see my problem, i had a stray return statement at the top of the function that i was previously using to debug something.  Oops.
19:35 jav joined #salt
19:35 zmalone The bug was found after release
19:35 abednarik joined #salt
19:36 djgerm does anybody have advice on automating updating forward/reverse DNS as new minions are created/destroyed?
19:36 ViciousLove such a sad, easy to find bug
19:37 kevinquinnyo the "easy to find" ones are also easy to miss at least for me apparently
19:38 jcockhren you don't find the bugs, they find you
19:38 jcockhren you = all of us
19:38 zmalone or you write tests, but I think that was mentioned on the salt-user's mailing list already
19:38 kevinquinnyo djgerm: i would love to help you out but i havent gotten there yet -- it's on my list though :/
19:39 jcockhren djgerm: I do.
19:39 jcockhren djgerm: use libcloud dns drivers
19:39 Borromini guys anyone have a clue about special chars and salt? salt doesn't seem to like my mysql password
19:39 jcockhren djgerm: write a custom execution module and matching state. then set as a reactor
19:40 ViciousLove I should really test the dev branch more often. I just don't have an easy way of doing it.
19:40 Guest18700 joined #salt
19:41 djgerm customer execution module… yikes. that sounds like a lot of work. the ddns salt module doesn't work well?
19:41 djgerm *custom
19:41 jcockhren djgerm: custom modules are crazy easy
19:42 giantlock joined #salt
19:42 djgerm @_@
19:42 djgerm I dont believe you :)
19:43 jcockhren djgerm: I have one for DNSimple
19:43 jcockhren I can prove it. gist incoming
19:44 djgerm *excited*
19:44 perfectsine joined #salt
19:45 zmalone Borromini: it can get scary, so pick passwords that avoid special characters that might be interpreted as yaml?
19:46 Borromini zmalone: i'll dig into what yaml doesn't like either
19:46 zmalone it's worse if you do gpg encrypted pillars
19:47 Borromini lol i won't go there
19:47 zmalone "': are the main ones I'd avoid
19:47 Borromini i have ! ? and #
19:47 jcockhren djgerm: https://gist.github.com/jcockhren/506c2e5a0079c813b724
19:48 djgerm thanks! checking it out
19:48 Borromini yaml_encode doesn't seem to like ! ? and # (or either of those)
19:49 jcockhren djgerm: change the provider to use other libcloud.dns drivers
19:49 jhauser joined #salt
19:50 djgerm i'll have to read up on those libcloud.dns drivers.
19:50 Edgan_ joined #salt
19:51 forrest joined #salt
19:52 strepsil joined #salt
19:57 ageorgop joined #salt
19:57 djgerm your definition of easy and mine are different jcockhren :)
20:00 basepi gtmanfred: yep, frantically.
20:00 jcockhren djgerm: these lines are the money makers -> https://gist.github.com/jcockhren/506c2e5a0079c813b724#file-dnsimple-py-L44-L51\
20:00 jcockhren djgerm: these lines are the money makers -> https://gist.github.com/jcockhren/506c2e5a0079c813b724#file-dnsimple-py-L44-L51
20:00 gtmanfred basepi: awesome, thanks
20:00 jcockhren 6 lines of code
20:01 forrest iggy, you around?
20:01 iggy si
20:02 forrest iggy, https://github.com/saltstack-formulas/git-formula/pull/23 could you please merge that?
20:02 saltstackbot [#23]title: Revert "Update map.jinja for CentOS support" | Reverts saltstack-formulas/git-formula#22
20:02 forrest centos/redhat is not using git2u, I can confirm on the latest release of centos as I just deployed a new system two days ago.
20:03 iggy done!
20:03 iggy go forth and break shit
20:03 djgerm I know I didnt mention it, but I am using bind, so it looks like I'll need something like this https://github.com/Kami/python-libcloud-dns-to-bind-zone to glue it all together huh jcockhren
20:03 forrest iggy, Thanks.
20:03 gtmanfred forrest: nice
20:04 forrest gtmanfred, /shrug, aboe pointed it out
20:04 gtmanfred yall can get git2 on redhat/centos
20:05 jcockhren djgerm: looks like it
20:05 gtmanfred https://ius.io/
20:05 gtmanfred forrest: i actually, misread what you were doing :/
20:05 thetoolsmith joined #salt
20:05 djgerm or… just not use bind.
20:06 forrest gtmanfred, ?
20:07 thetoolsmith when using git remote repos exclusively for state, how long before new branch commits become available as state to avoid No matching sls found for ‘xxxx’ in env ‘xxxx’ error
20:07 thetoolsmith i can add to local file_roots and it's found right away, waiting very long time when using git remote only. thanks.
20:07 gtmanfred forrest: nevermind
20:08 forrest gtmanfred, okay. I just double checked both the EPEL repo, as well as the main centos repo for both el6 and el7, no git2 in there, and no git2u whatever that is
20:08 gtmanfred git2u is from IUS
20:08 gtmanfred it stands for inline with upstream stable
20:08 antpa joined #salt
20:08 djgerm jcockhren: what do you use for internal DNS server? do you automate it with your amazing executor?
20:08 forrest Yeah, which is an even less standard repo, lol
20:08 gtmanfred and they package the newest versions of packages for centos and rhel
20:09 forrest Right, but there's no addition of this repo in that PR
20:09 gtmanfred not that less standard, carl and ben work a lot with the centos teams to make sure they follow their guidelines
20:09 gtmanfred yeah
20:09 baweaver joined #salt
20:09 gtmanfred forrest: i missread and thought you were adding the repo to it
20:09 forrest gtmanfred, Lol fuck no, I do enough unpaid work, I'm not going to start killing myself packaging shit.
20:10 gtmanfred :P
20:10 gtmanfred you wouldn't need to package it, just enable the ius repo for centos in that formula, and then use git2u :P
20:10 forrest Used to be our main package mantainer at an old job, would not recommend.
20:10 forrest Yeah no
20:10 forrest I'm not enabling random repos, if people want to do that they can fork it
20:10 gtmanfred :P
20:12 murrdoc joined #salt
20:12 dlam hey whats the best way to iterate through like a list of pillar data?   like i wannt collect all the ips of my app servers  a la  this pillar   https://gist.github.com/dlam26/b14da005556095490e26
20:13 dlam i just wanna do something like  ','.join([ip_addy1, ip_addy2]) etc
20:15 gtmanfred dlam: {{ [x['ip'] for x in salt['pillar.get']('hosts:app')]|join(',') }}
20:15 forrest but please don't use x
20:15 dlam ooo use jinja string join filter  ...didnt know that existed!
20:16 gtmanfred forrest: https://www.youtube.com/watch?v=YEwlW5sHQ4Q
20:16 forrest Heh, thankfully I'm not reviewing the PR where you do that either ;P
20:17 gtmanfred :P
20:17 jcockhren gtmanfred: nice list comprehension
20:18 totzky joined #salt
20:18 murrdoc that is never true
20:18 murrdoc 'nice list comprehension' is same as 'nice perl'
20:18 murrdoc it looks nice now
20:19 jcockhren murrdoc: ;)
20:21 jkennedy_ joined #salt
20:21 Guest18700 joined #salt
20:21 hightekvagabond joined #salt
20:26 lorengordon joined #salt
20:26 antpa joined #salt
20:28 dlam haha thanks jcockhren
20:28 andrew_v joined #salt
20:30 deniszh joined #salt
20:32 kevinqui3nyo hm i need a way to get a list of all minions in a reactor -- earlier i asked how to do it in a runner
20:33 kevinqui3nyo left #salt
20:33 kevinquinnyo sorry, i was in the channel twice somehow ^
20:34 ericsysmin joined #salt
20:35 ericsysmin question how can I make this {% if grains['osrelease'] == '6.6' or grains['osrelease'] == '6.7' or grains['osrelease'][0] == '7' %} do this instead
20:35 ericsysmin if grains osrelease >= 6.6
20:35 ericsysmin i've read it's actually treated like a string and not a number
20:35 kevinquinnyo you can try casting it to an int
20:35 kevinquinnyo int(grains['osrelease'])
20:36 kevinquinnyo orr
20:36 kevinquinnyo wait that may erase the decimal which would defeat the purpose
20:36 ericsysmin yea, exactly, i need a float
20:36 kevinquinnyo float()
20:36 ericsysmin so float(grains....)
20:37 kevinquinnyo yep.  if float(grains['osrelease']) ...
20:37 kevinquinnyo give that a shot
20:38 totzky joined #salt
20:38 cberndt joined #salt
20:38 abednarik joined #salt
20:38 Borromini zmalone: it turns out the hash sign was the problem
20:39 ericsysmin should i keep the '' around the 6.6
20:40 ericsysmin or since im comparing numbers can i remove it
20:41 ajw0100 joined #salt
20:42 zmalone Was it getting interpreted as the start of a comment?
20:42 gordonm joined #salt
20:42 zmalone And did you change the password, or find a good way to quote the whole thing?
20:43 Guest18700 joined #salt
20:43 Borromini no i just started removing special chars one by one
20:43 Borromini had it single quoted, double quoted, no difference
20:45 ericsysmin1 joined #salt
20:45 numkem joined #salt
20:50 zmalone try the double / single combo
20:51 zmalone salt interprets yaml twice, or something like that, so if you do "'value'" it will strip one set on the first run, and the second on the next interpretation
20:51 Borromini ok i'll give that a shot
20:51 zmalone I've got no idea if that impacts your environment, but it's one thing to try
20:51 zmalone if you use two doubles (""value"") or two singles, it'll remove them both in one interpretation
20:57 antpa joined #salt
20:57 Guest18700 joined #salt
21:00 kevinquinnyo is there a way to just print vars inside a reactor and be able to view them somewhere?  they arent showing up in the debug log for the salt master
21:00 kevinquinnyo or rather:  How do i debug code in a reactor?
21:01 Borromini zmalone: just tried with the hash again and double/single, single/double (you never know), doesn't work either
21:01 Borromini i'll just change my mysql passwords
21:02 Borromini will be easier
21:05 aharvey joined #salt
21:06 zmalone joined #salt
21:07 hasues joined #salt
21:08 antpa joined #salt
21:10 kevinquinnyo good lord.  So grains and pillar are not available in a reactor, but they are available in a runner, but a runner can't run sls on targets, but a reactor can
21:12 totzky joined #salt
21:13 brianfeister joined #salt
21:14 djgerm is there a linux was to use https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.pip.html  to install a python module?
21:14 jcockhren kevinquinnyo: yeah. going to have to use a custom state that'll pull wat you need from pillar & grains
21:14 jesusaurus joined #salt
21:16 kevinquinnyo jcockhren: can you clarify a bit?  I need to execute a salt.git state on a remote minion if an event occurs but i also need access to pillar data.  So i start at the reactor, and i have my payload data, then what?
21:16 ponpanderer joined #salt
21:17 ponpanderer Hello, in salt-cloud is the saltify driver only for Linux minions or is the Windows support available for clouds part of saltify as well?
21:17 jfindlay kevinquinnyo: that's definitely annoying
21:17 kevinquinnyo and i need *all pillar data* for all minions
21:19 amcorreia joined #salt
21:20 hightekvagabond joined #salt
21:21 jkennedy_ left #salt
21:22 baweaver joined #salt
21:27 WildPikachu left #salt
21:27 akhter joined #salt
21:29 snicers-work Ok, so now that my ELB is online (whew!) I want to dynamically register my ec2 instances, how do I access their Instance ID's from saltstack?
21:30 linjan joined #salt
21:30 Borromini thanks for the help guys, i'm out
21:30 Borromini left #salt
21:35 hasues left #salt
21:36 aharvey joined #salt
21:37 Guest18700 joined #salt
21:43 sfxandy joined #salt
21:43 kevinforbes joined #salt
21:47 rem5 joined #salt
21:49 shaggy_surfer joined #salt
21:49 pcn snicers-work: if you're using the ec2 grain, salt-call grains.get ec2:instance-id
21:51 pcn snicers-work: salt-call grains.get ec2:instance_id
21:51 pcn (underscore)
21:52 snicers-work I just found the ec2 grain module.
21:52 snicers-work would that end up in `/srv/_modules`?
21:52 snicers-work or `/srv/salt/_modules`?
21:53 CheKoLyN joined #salt
21:54 kevinforbes I’m having trouble provisioning minions with salt-cloud: [ERROR   ] Invalid keyword 'standard' for variable 'master_type’.
21:54 kevinforbes salt 2015.8..3, salt-cloud 2015.8.3
21:54 snicers-work what command are you running kevinforbes?
21:54 kevinforbes The problem used to occur when ‘master_type’ was ‘str’
21:55 kevinforbes the salt bootstrap script
21:55 pcn _grains
21:55 kevinforbes ssh -t -t -oStrictHostKeyChecking=no -oUserKnownHostsFile=/dev/null -oControlPath=none -oPasswordAuthentication=no -oChallengeResponseAuthentication=no -oPubkeyAuthentication=yes -oKbdInteractiveAuthentication=no -i /home/ubuntu/.ssh/keypair-salt-cloud.pem -p 22 ubuntu@xxx.xx.xx.xx 'sudo /tmp/.saltcloud-d510a740-3a58-4712-abea-f4986cadcfc1/deploy.sh -c /tmp/.saltcloud-d510a740-3a58-4712-abea-f4986cadcfc1 -P -D git v2015.8.3'
21:56 subsignal joined #salt
21:58 pcn snicers-work: put it in _grains
21:59 snicers-work this one? https://raw.githubusercontent.com/saltstack/salt-contrib/master/grains/ec2_info.py
22:00 feld how do i use file.line and ensure there are 4 different lines in the file?
22:00 JDiPierro joined #salt
22:05 Guest18700 joined #salt
22:05 PsionTheory joined #salt
22:09 baweaver joined #salt
22:11 hoonetorg joined #salt
22:15 Guest18700 joined #salt
22:16 lemur joined #salt
22:16 bhosmer joined #salt
22:18 brianfeister joined #salt
22:21 The_Loeki joined #salt
22:22 ronrib joined #salt
22:26 nebuchadnezzar joined #salt
22:27 baweaver joined #salt
22:28 jfindlay kevinquinnyo: The Reactor is not minion-specific. It does not run on behalf of a minion directly. If you want to _send_ grains or pillar to the Reactor from a minion look at the `with_grains` and `with_pillar` args to `event.send`.
22:28 jfindlay from whiteinge
22:29 The_Loeki hi pplz, I've been working on https://github.com/saltstack/salt-bootstrap/pull/685 and discussing with @dmurphy18 about the repo structure.
22:29 saltstackbot [#685]title: Add support for pegged versions on YUM based OS'ses through repo.saltstack.com | This admittedly hacky patch (blame your own repo infrastructure for that ;) ) adds version pegging support to repo.saltstack.com repos for any YUM based OS that properly supports glob matching in `includepkgs/exclude=`....
22:30 The_Loeki look around the docs all now seems to suggest to just install 'latest' which will always resolve to, well, the latest major version of Salt
22:32 The_Loeki I'm seriously doubting the wisdom of this decision; salt moves very fast, and each major upgrade has required updates & fixes to our code. Not that that's a problem mind you, but I fear you're setting yourselves up for a lot of problems if these issues start 'randomly' occurring as soon as some minion starts upgrading to a different major than it's master/file_roots
22:33 KennethWilke joined #salt
22:35 totzky joined #salt
22:35 Guest18700 joined #salt
22:36 Crazy67 joined #salt
22:42 zenlot1 joined #salt
22:43 jhauser joined #salt
22:45 Guest18700 joined #salt
22:45 jfindlay The_Loeki: in my opinion, moving releases into an archive repo nullifies many of the benefits of having an upstream repo
22:46 jfindlay unless you never run more than 1 release back :)
22:46 JTeatime Why *wouldn't* you keep all historical releases in your repo
22:47 JTeatime It's not like you have the disk space requirements of e.g. Debian project
22:47 jfindlay I don't recall the details, but I remember there being some difficulty with having multiple versions of the same packages in the same repo
22:48 JTeatime uh
22:48 The_Loeki that's not the point; the point is if I'm running a 2015.5 master, I want 2015.5 minions connecting to it, not 'latest' ones
22:48 JTeatime I'm not aware of any.
22:48 zmalone JTeatime: I think the metadata for those repos has issues, but it kind of works when you have a single copy of Salt and the dependencies in there
22:48 zmalone so it's a work around
22:49 The_Loeki running a pkg.upgrade on a (potentially quite large) set of minions would result in them upgrading to say Boron, disconnecting from the master, borking on modules & grains etc.
22:49 zmalone I started to look at the salt builder formulas that were published a month or two back to figure it out, but they are messy.
22:49 JTeatime it has to be something you're doing wrong, and not a limitation of apt repos in general.
22:49 CheKoLyN joined #salt
22:49 zmalone JTeatime: that's my understanding
22:50 The_Loeki the salt docs explicily warn against that in the upgrading salt section
22:51 Crazy67 joined #salt
22:51 The_Loeki and you're now directing swaths of quite likely beginner users to ignore those warnings basically
22:51 hoonetorg joined #salt
22:52 hoonetorg joined #salt
22:55 jfindlay I have not been directly involved with the project, but I'm well aware of the problems there :-)
22:55 hoonetorg somebody of you know that lv.present broken on 2015.8.4
22:55 hoonetorg just upgraded from 2015.5.8
22:56 hoonetorg debug log says it checks for the /dev/dm-xx device instead of the name
22:56 hoonetorg don't know if it was the case in 2015.5.8
22:57 Guest18700 joined #salt
22:58 hoonetorg s/lv.present/lvm.vg_present/
22:58 The_Loeki @jfindlay I'll leave you guys to discuss the various merits then ;) I would like that PR off my plate though, it's sat there since last Nov
22:59 jfindlay The_Loeki: that would be my fault, as I've taken over bootstrap
23:00 hoonetorg the /dev/dm-xx device is the pv and is a luks device with name /dev/mapper/crypt_blub
23:00 jfindlay hoonetorg: can you paste your configs and error?
23:00 The_Loeki that's fine; once the repo layout crystallizes out a bit more I'll probably help you work it into the bootstrap script better anyways; it's just that it doesn't feel right sitting for months on a PR ;)
23:02 jfindlay totally, :embarrassed:, although for the record I didn't take over until December :)
23:04 The_Loeki gheghe don't sweat it. It's in need of some TLC w/regards to the new repo's though, and my other 2 PR's there are actually borking bugs
23:05 hoonetorg jfindlay: 1st i want to check what changed between 2015.5 and 2015.8 in states/lvm.py
23:06 hoonetorg jfindlay: mmhhh nothing changed
23:06 hoonetorg but i updated the box from centos 7.1 to centos 7.2
23:07 hoonetorg and i know there were changes in lvm
23:07 jfindlay also, udev tends to churn between releases in my experience
23:08 hoonetorg yep
23:08 hoonetorg i will debug a bit
23:08 djgerm anybody get the ddns module working? I am SO close. I can do a ddns update manually just fine, but when I try to use the https://docs.saltstack.com/en/latest/ref/states/all/salt.states.ddns.html I get that I am not permitted to perform the update.
23:10 Guest18700 joined #salt
23:11 djgerm I think it has something to do with the keyfile keyname but I cant tell from the docs what I am doing wrong
23:15 adamisos joined #salt
23:16 hoonetorg jfindlay: https://gist.github.com/hoonetorg/377eeb30511d5618d417
23:18 zmalone joined #salt
23:21 jfindlay hoonetorg: can you verify that the vg is/is not present?
23:21 hoonetorg it is present
23:21 dlam hey whats the opposite of state.sls ?  like i wanna remove some stuff
23:22 dlam is it `state.low` maybe?
23:23 The_Loeki no
23:23 The_Loeki it's still state.sls
23:23 ALLmightySPIFF joined #salt
23:23 The_Loeki but rather than file.managed, it'd contain file.removed
23:24 dlam ahh ok, actually i like ran manually `salt-call state.sls  database`  ...but i did it on an app server which shouldnt have postgresql on it,  so i wanna like "revert" it
23:25 dlam (so i wanna revert a state, not remove a file  :D)
23:25 zmalone joined #salt
23:25 jfindlay hoonetorg: there may be some logic that needs to be fixed up in that state
23:25 jfindlay would you mind filing a bug report?
23:25 The_Loeki that can only be done by building it yourself
23:26 The_Loeki so you'll haven an database/install.sls and a database/gtfo.sls
23:26 hoonetorg jfindlay: i updated the gist with vgs and vgdisplay -v ...
23:26 dlam ahhh ok ok thanks the_loeki
23:26 The_Loeki one with the file.managed (and pkg.installed and whatever) and one with the equal opposites (file.absent, pkg.absent et al)
23:26 hoonetorg yep, if you like, i do a bug report
23:28 hoonetorg jfindlay: and i believe it worked in centos 7.1, probably the output of [pv|vg|lv]display changed from 7.1 to 7.2 and the parsing of the output is broken
23:32 jfindlay hoonetorg: seems likely
23:34 hoonetorg jfindlay: https://github.com/saltstack/salt/issues/30882
23:34 saltstackbot [#30882]title: state lvm.vg_present broken with luks - pv on centos 7.2 | Hi...
23:34 jfindlay thanks
23:35 hoonetorg any additional info required???
23:35 Crazy67 joined #salt
23:35 joe_n joined #salt
23:36 Guest18700 joined #salt
23:37 jfindlay hoonetorg: I think that should be enough
23:37 jfindlay it has the command and the output
23:38 hoonetorg thx
23:40 djgerm what is the different between a module: https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.ddns.html#module-salt.modules.ddns and a state: https://docs.saltstack.com/en/latest/ref/states/all/salt.states.ddns.html ?
23:41 djgerm the state references the module right? is there a way I can read the module ddns?
23:41 akhter joined #salt
23:42 djgerm i found it :)
23:42 The_Loeki @djgermo states are separate functions which mostly utilize execution modules to achieve a desired state
23:42 The_Loeki state.sls is actually 'module state, function sls'
23:42 jfindlay djgerm: a module interacts with external tools, platforms, and APIs, whereas a state uses modules to conform the system/device/service to the desired configuration
23:43 djgerm so I think I might have a bug for the ddns module or state? I specify the keyfile, but it's not using it?
23:43 djgerm i dont read python too good… but I am looking at /usr/lib/python2.7/dist-packages/salt/modules/ddns.py
23:44 jhauser joined #salt
23:44 lompik joined #salt
23:44 jfindlay what line?
23:45 jfindlay lines
23:46 The_Loeki the state just calls that module and passes the args directly on; if it works as a module it should work as a state
23:46 djgerm well I dont know the lines in the module, but I know what I am seeing. It looks like the module isn't picking up the keyfile in the state.
23:46 djgerm let me gist my state...
23:48 djgerm http://pastebin.com/QVfHEH0y
23:49 The_Loeki @djgermo is that a valid keyfile for ddns?
23:49 djgerm from the module docs it says to do it like that
23:49 djgerm {"keyname.": "keycontent"}
23:50 djgerm topsecret is obviously a redacted key :)
23:51 djgerm and I know it's not sending the key because the named logs say so.
23:51 The_Loeki have you tried dit without that
23:51 The_Loeki '.'
23:51 The_Loeki ?
23:51 hoonetorg jfindlay: after checking yum.log i found out, that i updated from 2015.5.2 to 2015.8.4 and there is a difference between states/lvm.py
23:52 djgerm I think so.  The_Loeki but I'll double check. just a second.
23:52 djgerm still fail.
23:53 hoonetorg jfindlay: how can i create a link to the comparison of the 2 files (a diff behind tag v2015.5.2 and v2015.8.4 is too large - 1890 files)
23:53 The_Loeki have you added ddns options in the minion and/or master config?
23:53 djgerm i dont know what you mean.
23:53 djgerm in bind config?
23:54 The_Loeki no like /etc/salt/minion.d/ddns.conf
23:54 The_Loeki or there somewhere
23:54 djgerm oh. no…. what uhm… I have no idea what that would be or what that would look like
23:54 djgerm :)
23:54 djgerm i see the ddns attempt to update
23:55 djgerm but the named logs show the ddns client not presenting the key
23:55 rem5 joined #salt
23:55 djgerm when I run the salt '*' state.sls ddns
23:55 The_Loeki if you put ddns.keyname: ddns ddns.keyfile: bla should have same result
23:55 The_Loeki also keyalgorithm OK?
23:55 The_Loeki default is 'HMAC-MD5.SIG-ALG.REG.INT'
23:55 djgerm yup. the key works manually
23:56 The_Loeki salt-call file.exists /path/to/key ?
23:56 jfindlay hoonetorg: git diff v2015.5.2:salt/states/lvm.py..v2015.8.4:salt/states/lvm.py
23:58 djgerm The_Loeki: http://pastebin.com/81xAEYNF file.exists is not available?
23:58 hoonetorg jfindlay: thx, i know this
23:58 hoonetorg i thought it's possible in github directly
23:59 wryfi left #salt
23:59 The_Loeki my bad
23:59 The_Loeki file.file_exists
23:59 jfindlay hoonetorg: oh, sorry, I'm not sure
23:59 djgerm ah ok. I get a local: True on that

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