Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-11-07

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

All times shown according to UTC.

Time Nick Message
00:08 jeddi joined #salt
00:11 jas02 joined #salt
00:16 jas02 joined #salt
00:18 mosen joined #salt
00:20 patrek joined #salt
00:21 PerilousApricot joined #salt
00:24 mosen joined #salt
00:41 johnkeates joined #salt
01:10 hasues joined #salt
01:14 hemebond Anyone know if the version of python-requests changed with 2016.3.4?
01:20 babilen joined #salt
01:23 anotherzero joined #salt
01:30 juanito joined #salt
01:31 hasues left #salt
01:38 amcorreia joined #salt
01:40 ninjada joined #salt
01:46 edrocks joined #salt
02:01 jas02 joined #salt
02:05 abonilla joined #salt
02:07 Satyajit joined #salt
02:11 netcho joined #salt
02:15 lilvim joined #salt
02:18 jas02 joined #salt
02:23 akhter joined #salt
02:25 subohm joined #salt
02:28 evle joined #salt
02:29 akhter joined #salt
02:31 DEger joined #salt
02:40 sebastian-w joined #salt
02:41 ninjada joined #salt
02:43 juanito_ joined #salt
02:48 ilbot3 joined #salt
02:48 Topic for #salt is now Welcome to #salt! | Latest Versions: 2015.8.12, 2016.3.4 | Support: https://www.saltstack.com/support/ | Logs: http://irclog.perlgeek.de/salt/ | Paste: https://gist.github.com/ (please don't multiline paste into channel) | See also: #salt-devel, #salt-offtopic | Ask with patience as we are volunteers and may not have immediate answers
02:50 abonilla joined #salt
02:51 subohm joined #salt
02:52 scoates joined #salt
03:01 abonilla joined #salt
03:02 jas02 joined #salt
03:03 fas3r in the cloud profiles, is it possible to specify if the net-id is public or private ?
03:04 hemebond net-id?
03:04 fas3r network
03:06 hemebond ssh_interface: private_ips
03:07 fas3r can I force the minion to listen on a specific interface ?
03:08 hemebond Minions don't listen.
03:08 fas3r communicate*
03:08 hemebond Uh. Not sure about that. I mean, it'll use whichever interface the route table says it can use.
03:09 fas3r that's my probleme, the minion is set with the wrong ip subnet as I have multiple private networks
03:10 hemebond You mean the VM is in the wrong subnet?
03:10 fas3r when the instance is deploy, salt is using the wront subnet.
03:11 hemebond I don't understand. salt-minion doesn't do anything like that. It just communicates out.
03:11 hemebond If your instance is in the "wrong" subnet that's a problem with your instance interfaces and route tables.
03:11 fas3r I have 10.0.0.0/8 and 192.168.0.0/24. Despite the order in the cloud profiles of the networks, salt is setting the minion on 192.168.0.x instead of 10.0.0.x
03:12 hemebond Do you mean eth0 is on the wrong subnet?
03:12 hemebond You want eth0 and eth1 reversed?
03:12 bastiand1 joined #salt
03:13 systo joined #salt
03:13 fas3r no the IP are correct on the host
03:13 fas3r it's the minion which is set on the wrong interface
03:13 hemebond salt-minion will use whichever interface the route table says to use, no?
03:14 hemebond It's not listening so it isn't bound to an interface.
03:14 mosen_ joined #salt
03:14 mpanetta joined #salt
03:28 fas3r hemebond: the problem is that salt consider one of the 2 network as public. Despite that both are privates.
03:29 hemebond Does that mean the minion can't contact the master?
03:30 fas3r yes
03:31 hemebond So what does your route table look like?
03:31 fas3r that's why I was asking if there were no extra parameter for the "networks" sections to specify if it's private or not.
03:31 hemebond I mean, regardless of what Salt thinks is a public interface, like any other network application it'll just use the routing table to get out.
03:33 fas3r the route table is fine when I connect on it.
03:33 hemebond Can you share it?
03:34 hemebond I would check the output of "netstat -anp | grep 450"
03:34 hemebond ip addr
03:34 hemebond and "route -n"
03:34 fas3r I reinstance a VM ..
03:35 fas3r can take some time ...
03:46 JPT joined #salt
03:50 fracklen joined #salt
04:00 justanotheruser joined #salt
04:03 jas02 joined #salt
04:12 netcho joined #salt
04:19 jas02 joined #salt
04:22 PerilousApricot joined #salt
04:23 PerilousApricot joined #salt
04:34 robbintt joined #salt
04:47 edrocks joined #salt
04:49 informant1 joined #salt
04:59 shbst joined #salt
05:02 Ni3mm4nd joined #salt
05:08 mosen_ joined #salt
05:20 jas02 joined #salt
05:25 mosen_ joined #salt
05:29 wt joined #salt
05:31 wt_ joined #salt
05:33 samodid joined #salt
05:39 netcho joined #salt
05:41 rdas joined #salt
05:41 jfelchner joined #salt
05:51 prashant joined #salt
06:02 MikaT joined #salt
06:02 DEger joined #salt
06:04 ivanjaros joined #salt
06:04 jas02 joined #salt
06:40 bluenemo joined #salt
06:44 eseyman joined #salt
06:49 felskrone joined #salt
06:50 edrocks joined #salt
06:52 colttt joined #salt
06:53 jas02 joined #salt
07:00 sh123124213 joined #salt
07:12 preludedrew joined #salt
07:14 wt joined #salt
07:16 mpanetta joined #salt
07:16 rylnd joined #salt
07:24 yuhlw____ joined #salt
07:25 ivanjaros joined #salt
07:36 iggy man, wouldn't it be nice if salt respected the worker_threads setting...
07:37 hemebond It doesn't? I've never played with it.
07:39 iggy well, no matter what I set it to, there are 10+ master processes (with 0 minions attached)
07:40 sh123124213 joined #salt
07:47 fracklen joined #salt
07:49 sjorge joined #salt
07:49 sjorge joined #salt
07:51 fracklen joined #salt
07:52 akhter joined #salt
07:52 jrklein joined #salt
07:52 fracklen joined #salt
07:55 samodid joined #salt
07:58 mohae_ joined #salt
08:03 mavhq joined #salt
08:14 ronnix joined #salt
08:18 abonilla joined #salt
08:24 dkrae joined #salt
08:30 infrmnt joined #salt
08:36 JohnnyRun joined #salt
08:40 sevenfourk joined #salt
08:52 om2 joined #salt
08:55 krymzon joined #salt
08:55 sh123124213 joined #salt
09:01 mavhq joined #salt
09:02 mikecmpbll joined #salt
09:06 titilambert joined #salt
09:09 froztbyte joined #salt
09:09 froztbyte joined #salt
09:14 keimlink joined #salt
09:15 jhauser joined #salt
09:20 patrek joined #salt
09:21 Rumbles joined #salt
09:22 Rumbles has anyone got any examples of how to set up users & groups in the pillar? I was looking at moving my users to use pillar (partly so I can get smething working) and I want to make sure I on't over look anything...
09:23 hemebond Rumbles: Have a look at the users formula
09:23 hemebond It's what I've just started using.
09:24 babilen One way to do it is to keep user definition in their own SLS and then target users individually. You could also mix in additional settings in their own SLS if that fits your needs.
09:25 babilen You'd then target the the "users" state to all minions on which you would like to manage users and the pillar data decides which users are created/removed
09:27 N-Mi joined #salt
09:30 martoss joined #salt
09:32 Rumbles well, at the moment I have everyone set up their own state files, but I wanted to get something to work and I can see any other way but to move them in to pillar
09:33 Rumbles if I wanted use a state to add a symlink to every users home directory (i.e. only those setup on this machine), the users are configured in my users/username.sls files currently. Is there any way I can check what users should be installed on my server and create a symlink for each on in a for loop?
09:34 Rumbles that's what I was trying to do ^^
09:35 babilen Rumbles: How would the "check" work?
09:35 netcho joined #salt
09:35 Rumbles not sure
09:35 babilen It's quite hard to tell you how to implement it then :)
09:36 Rumbles I was looking at using "sudo salt 'ds1*' file.find /home type=d maxdepth=0" and using that command in a for loop
09:36 Rumbles so if that user exists create a symlink
09:36 netcho joined #salt
09:36 Rumbles but that command returns noothing
09:36 Rumbles (although max depth 1 works fine)
09:36 babilen How do you decide which user to deploy on which box?
09:37 Rumbles (max depth 1 also returns all the folders in the users home dir)
09:37 Rumbles currently it's defined in the top file
09:37 babilen I mean, in the end you'll just have data that represents your user's configuration and you target that to the boxes you want to manage the user on
09:37 Rumbles not sure how I would move that in to the pillar
09:37 Rumbles it's one of my unanswereed questions
09:38 Rumbles I was considering doing it by group
09:38 Rumbles but still not sure how I would define it :/
09:38 Rumbles which is why I asked
09:38 rawzone joined #salt
09:38 babilen And you would target the configuration in the pillar top.sls just like you target it in your states' top.sls now
09:40 babilen Did you already take a look at the users formula?
09:40 Rumbles I'm looking now
09:41 Rumbles is there an example of what the pillar should look like?
09:41 babilen https://github.com/saltstack-formulas/users-formula/blob/master/pillar.example
09:44 fracklen joined #salt
09:44 Rumbles hmmm okay, that doesn't help me understand how I would apply a given user to a given server
09:45 sagerdearia joined #salt
09:47 AndreasLutro create one pillar .sls file for each user and specify which users belong on which servers in pillar/top.sls
09:47 babilen Rumbles: You would target the "- users" state to the server in your states' top.sls and then suitable pillar data in pillar/users/foo.sls from the pillar top.sls
09:48 Elsmorian joined #salt
09:48 babilen Rumbles: As mentioned earlier: I'd recommend to have a .sls file for each user which you can target individually
09:49 babilen You can always mix in additional configuration (e.g. sudo/files/gitconfig/...)
09:50 Rumbles I think I would need to change how that works
09:50 babilen Change it how?
09:50 Rumbles well we don't set individual users
09:50 Rumbles we tend to use groups
09:50 babilen You can group user data in the same .sls
09:51 Rumbles can you include other (user) pillars in to one (group) pillar?
09:51 lubyou_ joined #salt
09:51 babilen There is no need to split that data into multiple SLS files, its just that you *can* do it if you want to be able to target users individually
09:52 edrocks joined #salt
09:53 babilen I would still recommend to keep individual users in their own SLS and adjust your targeting to match your needs/requirements
09:54 Rumbles ok, as I asked before, can I have them as individual pillar files for each user and include those in a group pillar? like you can include in normal states
09:54 Rumbles ?
09:54 babilen sure
09:54 s_kunk joined #salt
09:55 babilen But you might still be better to model this with targeting, rather than includes
09:55 Rumbles what's the difference?
09:55 AndreasLutro includes have unpredictable ordering
09:56 Rumbles can you explain what you mean by targetting?
09:56 babilen Which brings us back to the question: How do you actually decide which users get deployed on which box?
09:59 lubyou_ "Failed to authenticate SSH session: Callback returned error" - i am getting this for all my salt gitfs remotes
09:59 lubyou_ http://dpaste.com/0YBMZVB relevant parts from the config
10:00 lubyou_ cloning from the shell with the same identity file works just fine
10:00 AndreasLutro lubyou_: that's a pygit2 specific bug, I remember having to get very sepcific version of libgit2/pygit2 to make it work with salt
10:01 lubyou_ :|
10:01 AndreasLutro I use libgit2 0.23.4 and pygit2 0.23.1 but maybe more recent versions work fine as well
10:02 lubyou_ ii  libgit2-24:amd64                 0.24.1-2                          amd64        low-level Git library
10:02 lubyou_ ii  python-pygit2                    0.24.0-0ubuntu2                   amd64        bindings for libgit2 - Python 2.x
10:04 lubyou_ @AndreasLutro you dont happen to have a link to the bug report, do you?
10:05 yidhra joined #salt
10:06 AndreasLutro can't find it
10:07 lubyou_ https://github.com/libgit2/pygit2/issues/552 think this might be my problem
10:07 saltstackbot [#552][OPEN] ECDSA and ed25519 keys do not work | Hello,...
10:07 AndreasLutro you use id_rsa in your config
10:08 Rumbles can anyone say why "sudo salt '*' file.find /home type=d maxdepth=0" returns nothing and yet "sudo salt 'logstash*' file.find /home type=d maxdepth=1" returns all the folders in the /home folder and the subfolders in them
10:11 AndreasLutro lubyou_: I was wrong, nevermind what I said - I confused your error message with this one https://github.com/saltstack/salt/issues/28072
10:11 saltstackbot [#28072][MERGED] pygit 0.23.2 is not supported in Salt 2015.8.1 | I just rolled a fresh Salt master running 2015.8.1, freshly installed libgit2 from source and hit this issue:...
10:11 fracklen joined #salt
10:11 rickflare joined #salt
10:12 aaa joined #salt
10:12 catpig joined #salt
10:13 fracklen_ joined #salt
10:13 lubyou_ @AndreasLutro thanks anyway
10:14 fracklen joined #salt
10:15 fracklen joined #salt
10:18 abonilla joined #salt
10:20 lubyou__ joined #salt
10:28 POJO joined #salt
10:37 amcorreia joined #salt
10:46 IdoKaplan joined #salt
10:47 IdoKaplan Hi, Is anyone have an idea how to set "password never expire" when creating a new windows user?
10:49 felskrone1 joined #salt
10:51 Rumbles afaik that isn't supported IdoKaplan
10:51 IdoKaplan Rumbles: Thank you. Do you know why?
10:52 Rumbles no idea, I don't use Windows
10:52 Rumbles I doubt many people do
10:52 Rumbles and most who do would se a domain controller and that kind of thing would be configured there
10:57 Rumbles I've taken this straight from the documentation here: https://docs.saltstack.com/en/carbon/ref/states/all/salt.states.apache_module.html I'm getting this error: https://paste.fedoraproject.org/474966/51619914/ can anyone say what I'm doing wrong with my use of apache_module.enabled ??
10:58 samodid joined #salt
11:00 sh123124213 joined #salt
11:05 Reverend is there any way to 'retain' the current indentation when priting multiline yaml in jinja?
11:06 Reverend I'm printing like 20 lines, and only the first line has the tab indent on it
11:06 Rumbles pipe it to yaml_encode ?
11:07 Rumbles https://docs.saltstack.com/en/2015.8/ref/renderers/all/salt.renderers.jinja.html#filters
11:07 abednarik joined #salt
11:07 Reverend nopes :(
11:08 Reverend nevermind. I'll just have to lump it.
11:08 Reverend it's no big issue, just makes the nginx conf harder to read
11:08 Reverend i guess jinja isn't prepending the line with a \t. it's just using the yaml's \n
11:09 AndreasLutro Reverend: indent filter
11:09 Reverend lemme looksie
11:09 jas02_ joined #salt
11:10 Rumbles AndreasLutro, any idea why I'm getting that erro using apache_module ?
11:11 AndreasLutro Rumbles: using an old version of salt?
11:11 Reverend {{ data['nginx_server_bottom'] | indent(2, true) }} #<----- that worked AndreasLutro :) thanks
11:11 Rumbles I'm using 2015
11:12 Rumbles salt 2015.5.3 (Lithium)
11:12 AndreasLutro so yes
11:12 AndreasLutro read documentation for your version of salt instead of the most recent version
11:12 Rumbles yeah but more recent than the docs say the features were added
11:12 AndreasLutro > New in version 2016.3.0.
11:13 AndreasLutro how is that more recent than 2015.5.3 ?
11:13 AndreasLutro sorry, other way around
11:13 Rumbles the doc version selecter doesn't work and there isn't a version for 2015.5.3
11:13 AndreasLutro heh, you should report that
11:13 AndreasLutro good sign that you should upgrade anyway though
11:14 AndreasLutro anyway use the deprecated function names "disable" and "enable"
11:14 Rumbles that is the latest version for my os :/
11:14 Rumbles well apt doesn't offer an update
11:15 AndreasLutro https://repo.saltstack.com/
11:16 Rumbles is  ppa:saltstack/salt no longer maintained?
11:17 Rumbles and it's been replaced with an arbitrary shell script?
11:17 AndreasLutro just follow the instructions for ubuntu on the same page
11:17 AndreasLutro the shell script is unnecessary
11:17 Rumbles .... but what about the ppa?
11:17 mpanetta joined #salt
11:18 Rumbles that's what we use
11:18 AndreasLutro no clue, I don't use ubuntu
11:18 AndreasLutro but if it doesn't contain the latest versions I'm sure it's safe to assume it's no longer maintained
11:18 Rumbles the docs here: https://docs.saltstack.com/en/latest/topics/installation/ubuntu.html still say to use the ppa
11:20 N-Mi joined #salt
11:20 N-Mi joined #salt
11:21 AndreasLutro Packages for Ubuntu 16 (Xenial), Ubuntu 14 (Trusty), and Ubuntu 12 (Precise) are available in the SaltStack repository.
11:21 AndreasLutro Instructions are at https://repo.saltstack.com/#ubuntu.
11:21 AndreasLutro dunno
11:30 lubyou_ joined #salt
11:30 amontalban joined #salt
11:31 Rumbles that's so lame, even the salt16 ppa is broken
11:31 babilen "http://repo.saltstack.com/apt/ubuntu/16.04/amd64/latest xenial main" you mean?
11:31 Rumbles no
11:31 Rumbles https://docs.saltstack.com/en/latest/topics/installation/ubuntu.html
11:32 Rumbles the ppa's listed there
11:32 babilen Don't use the PPA
11:32 Rumbles why not?
11:32 Rumbles it's in the docs
11:32 babilen Because it is broken and not maintained
11:33 Rumbles so it's in the dcumentation still because... ?
11:33 babilen It hasn't been removed yet
11:33 babilen And people *might* want to use packages from there for one reason or another
11:34 babilen pkg-salt maintains salt within Debian, but I'm not sure if anybody still maintains the PPA
11:34 Rumbles but that page, which is the only one I knew about the tells you how to set up on ubuntu, it doesn't even mention this new way of installing on ubuntu
11:34 babilen It does
11:34 babilen It's the first section on the page
11:35 babilen "INSTALLATION FROM THE OFFICIAL SALTSTACK REPOSITORY"
11:35 babilen There are also maintained packages in Ubuntu repositories themselves, but the PPA is, AFAIK, dead
11:37 Rumbles well that's a pain in the arse
11:37 babilen Why is that?
11:38 Rumbles because I don't know if I can use that repo, if I can have to update that repo on every server, then update salt and restart master/minions
11:39 babilen You should install salt from either http://repo.saltstack.com/#ubuntu or from the Ubuntu repositories. In Debian we maintain current versions in jessie-backports and unstable (which will migrate into stretch)
11:40 babilen The Ubuntu repository versions are, naturally, not kept up-to-date
11:40 LotR ubuntu doesn't have an unstable equivalent?
11:41 babilen And we also don't maintain them in {xenial,yakkety}-backports
11:42 babilen So my recommendation would be to: For Ubuntu: Use the packages from Saltstack and for Debian: Use the packages from Saltstack or from jessie-backports
11:43 babilen LotR: Ubuntu's unstable is essentially Debian :)
11:43 LotR heh
11:44 babilen https://wiki.ubuntu.com/UsingDevelopmentReleases might be of interest
11:44 * LotR would always pick distro packages over 3rd party ones, so would swap the order for the debian recommendation
11:44 babilen Oh, that wasn't an ordered list
11:44 babilen (which, one might argue, doesn't exist)
11:45 babilen I'd deploy packages from jessie-backports on jessie
11:46 babilen Also Saltstack syncs packaging updates every now and then ..
11:47 babilen The Debian packages might also differ from the Saltstack ones in that we backport/cherry-pick bugfixes every now and then
11:49 babilen Rumbles: So, does that clarify the current state?
11:55 Rumbles the docs need updating?
11:57 babilen One might want to remove the mention of the "community repo", yes
11:58 onlyanegg joined #salt
12:00 geomacy joined #salt
12:02 abonilla joined #salt
12:06 aaa joined #salt
12:11 patrek joined #salt
12:11 jas02_ joined #salt
12:14 Rumbles I've submitted a PR to update the docs
12:19 fracklen joined #salt
12:22 fracklen joined #salt
12:25 babilen Great, thanks!
12:25 babilen Rumbles: And did you manage to sort out your deployment?
12:26 Rumbles no, that's a lot of work
12:26 Rumbles actually d
12:27 Rumbles do you mean updating salt repos?
12:27 Rumbles can anyone say why "sudo salt '*' file.find /home type=d maxdepth=0" returns nothing and yet "sudo salt 'logstash*' file.find /home type=d maxdepth=1" returns all the folders in the /home folder and the subfolders in them?
12:28 babilen Rumbles: Nothing?
12:28 Rumbles I got a response
12:28 Rumbles but it's empty
12:28 babilen Rumbles: Why is it a lot of work? Did you install from the PPA on many boxes?
12:28 Rumbles yeah
12:28 Rumbles like 200
12:28 Rumbles my boss set it up that way
12:28 babilen Even from the boxes that are targeted by "logstash*" ?
12:29 Rumbles the target is irrelivant
12:29 Rumbles I get the same reponse
12:29 babilen Rumbles: Should be a simple file.replace call on /etc/apt/sources.list
12:29 fracklen joined #salt
12:29 Rumbles it not just the replacing part that's the work
12:29 Rumbles it needs apprving and testing
12:29 babilen You said that you get a reply if you target "logstash*" whereas you don't get one if you target "*"
12:30 Rumbles in my earlier query the target was incorrectly left in
12:30 Rumbles the different i the maxdepth
12:30 Rumbles difference is*
12:30 fracklen joined #salt
12:30 Rumbles if I use maxdepth 0 I get nothing, if I use maxdepth 1 I get the folders for both 0 and 1
12:30 babilen Also: My feeling is that you are off on a tangent that'll cause you pain in the future with your "find home directories" approach unless you use this for information gathering only. Even in that case there are better ways to get a user list.
12:31 netcho hi all, having issues getting pillar data to a jinja template
12:31 Rumbles I know but without moving all the users in to the pillar (a lot of work) I have no way of getting a list of installed users
12:31 Rumbles netcho, pastebin what you're trying to do
12:31 netcho Rumbles: https://paste.debian.net/893259/
12:31 babilen Rumbles: maxdepth 0 wouldn't traverse into subdirectories of /home
12:32 Rumbles try: set reg = pillar['aws']['region']
12:32 babilen s/into/to
12:32 Rumbles I was expecting it would just give me the list of folders in /home babilen
12:33 babilen netcho: No, don't ever use ['foo']['bar'] .. there is a better option with the pillar.get function in the salt pillar module. Try "set reg = salt['pillar.get']('aws:region', DEFAULTRETURN)"
12:33 babilen Rumbles: Did you try that command locally?
12:33 Rumbles what's wrong with ['foo']['bar'] ?
12:34 babilen Rumbles: It is cumbersome to write, throws a KeyError if either foo or bar is not defined in the pillar and is not as idiomatic in salt code as salt['pillar.get']('foo:bar')
12:34 Rumbles if I run "find /home -type d -maxdepth 1" on the machine I get the user folders
12:34 babilen Rumbles: Yeah, you want maxdepth 1
12:34 Rumbles idiomatic?
12:34 Rumbles but if I run that via salt it gives me the next level of folders
12:35 babilen Rumbles: idiomatic means "using, containing, or denoting expressions that are natural to a native speaker"
12:35 Rumbles i.e. /home/username/folders
12:35 babilen In this case: salt['pillar.get'] is more commonly used within salt code
12:35 Rumbles I never use it
12:35 Rumbles didn't work for me
12:39 babilen Rumbles: Try running 'salt "logstash*" cmd.run "find /home -maxdepth 0 -type d"' and compare that to maxdepth 1
12:39 babilen I could reproduce the behaviour with file.find, which looks like a bug
12:39 abonilla joined #salt
12:39 babilen Why didn't salt['pillar.get'] not work for you?
12:40 babilen s/not//
12:40 Rumbles I've no idea
12:40 Rumbles I couldn't get it to work and gave up
12:40 babilen It's used everywhere in salt code
12:40 Elsmorian joined #salt
12:40 patrek joined #salt
12:41 Rumbles iirc it just errored every time I used it
12:43 netcho LOL, sorry guys my bad :) i forgot - template: jinja in my state file :)
12:44 abonilla_ joined #salt
12:45 Rumbles babilen, https://paste.fedoraproject.org/474990/22698147/
12:47 babilen Rumbles: And now run the last command with maxdepth 1
12:48 Rumbles yeah that gets me /home and the /home/users folders
12:48 babilen There you go
12:48 Rumbles .... so why the difference between system find and salt find?
12:49 babilen "13:39:18         babilen > I could reproduce the behaviour with file.find, which looks like a bug"
12:49 Rumbles so, not sure, bug... ?
12:49 babilen This is, IMHO, a bug, yes
12:49 Rumbles I can raise that
12:49 Rumbles thanks
12:49 Rumbles wasn't sure if I was doing something wrong
12:49 Rumbles I won't use that method, but at least I know what's wrong now
12:50 babilen What do you need that for anyway? (to get back to the earlier question)
12:50 Rumbles I was going to use it in a loop
12:50 Rumbles because I wanted to test my method would work without moving all the users in to pillar
12:50 Rumbles but I guess I have to move everyone to the pillar
12:50 Rumbles I just wanted a list of installed users on a system
12:50 Rumbles so I could create a symlink in each home dir
12:51 Rumbles moving the users in to the pillar just to get a symlink working seems like a lot of work :(
12:51 babilen symlink?
12:51 fracklen joined #salt
12:52 Rumbles for duc
12:52 babilen So you don't actually want to manage users with salt?
12:52 Rumbles we do
12:52 Rumbles each user has their own sls
12:52 Rumbles (my boss set it up)
12:52 babilen And (s)he decided to hardcode the states rather than create them dynamically from pillar data?
12:52 Rumbles of course
12:52 Rumbles he didn't use pillar until I started using it
12:53 Rumbles he was new to salt when he started using it and didn't use pillar
12:53 John_Kang joined #salt
12:53 babilen Well, you can either keep on doing things that way or move to a pillar based scheme
12:53 babilen (like the one used by the users formula)
12:53 Rumbles yeah I know
12:53 Rumbles but that's a lot of work
12:54 Rumbles and I don't have tie for a symlink
12:54 babilen My recommendation would be to introduce the users formula, which naturally means a bit of work
12:54 Rumbles I don't like the users forumla
12:54 Rumbles I will probably write my own
12:54 edrocks joined #salt
12:54 babilen It should be rather trivial to write a script that creates symlinks in Bash, Python, ... which you would then run with cmd.script
12:54 babilen What don't you like about the users formula?
12:55 Rumbles I don't understand how it works
12:55 Rumbles too many parts I won't use
12:55 babilen You don't have to use parts you don't want to use. Just define data for the bits that you *do* want to define
12:55 Rumbles I would rather have something simple that makes sense than something complicated I don't understand
12:56 Rumbles which is why static users is fine
12:56 babilen I'd recommend the users formula. It is widely deployed, tested and maintained
12:56 babilen But feel free to do whatever you want to do
12:56 Rumbles thanks
12:56 Rumbles I don't like using things I don't understand
12:56 Rumbles because when it stops worng I need to be able to fix it
12:56 babilen Why do you use salt then?
12:56 babilen Or Linux?
12:57 Rumbles I meant in salt
12:57 babilen I mean that argument could be made for everything. We really understand very few things and have to accept a certain level of uncertainty all the time
12:57 Rumbles yes
12:57 Rumbles but I don't have to fix everything
12:57 Rumbles I do have to fix salt when it doesn't work
12:58 Rumbles so I don't want to start using things I didn't write myself or understand
12:58 John_Kang is there a way to sync a directory with preserving the permission and deletion option like rsync ?
12:58 babilen Just don't define bits in the users formula pillar data you don't want to define. It really doesn't do things if there is no corresponding pillar data.
12:58 babilen Rumbles: Trust me, you will face another huge transition and a high maintenance cost if you go down the "have to write everything myself" route
12:58 John_Kang I tried to use file.recurse, but not sure if it can delete the file which are not existing in source
12:59 Rumbles babilen, I don't even get how to use those states
12:59 Rumbles I don't want to use them
12:59 Rumbles I don't really want users in pillar
12:59 tkharju joined #salt
13:02 John_Kang could someone give me some advice for that ?
13:02 ronnix joined #salt
13:03 Rumbles John_Kang, you could file.absent your top folder first, then file.managed it afterward
13:04 Rumbles so the files are all gone before you manage them again
13:04 lordcirth John_Kang, if you use file.recurse and add - clean: True, it will delete files
13:04 Rumbles are you restoring them from a tar?
13:04 Rumbles ah right
13:05 DEger joined #salt
13:05 John_Kang Rumbles: lordcirth: Thank you guys, for the information
13:05 John_Kang but it doesn't preserve the permission, do it ?
13:06 hlub I'd like to have one signing policy configuration for every minion. Now I amwondering hwo should I implement this. I could use orchestrate but I don't know how to query all of the minions within orchestration. Any ideas?
13:06 lordcirth John_Kang, as the docs say, you can specify the permissions.
13:07 Rumbles babilen, I assume you're on the latest salt, you said you reproduced that find issue
13:07 John_Kang lordcirth: I knew that i just can set the dir_mode and file_mode
13:08 lordcirth John_Kang, yep, that's how it works.  Did you need varying permissions?
13:08 babilen Rumbles: You use the state by targeting "- users" to a box
13:08 Rumbles yes
13:08 Rumbles well - user.username
13:08 Rumbles or a group with includes all the users
13:08 Rumbles which*
13:08 John_Kang some files could be different permission to normal file what is defined by file_mode
13:09 lordcirth John_Kang, I'm not sure if there's a good way to do that.
13:10 lordcirth John_Kang, May I ask what directory you are managing?  How often will you want to change the files in it?
13:10 Rumbles John_Kang, can ou pastebin your state?
13:10 babilen Rumbles: No, the *state* is "- users" -- Do not conflate that with pillar data. You can organise that in whatever way you want and it will be merged according to your configuration -- https://docs.saltstack.com/en/latest/ref/configuration/master.html#pillar-merging-options
13:11 Rumbles huh?
13:11 DEger joined #salt
13:12 scoates joined #salt
13:12 John_Kang lordcirth: Rumbles: the state is very simple like synopsis, i just want to manage to sync one directory without care of the changing permissions
13:12 Rumbles sync a directory from where?
13:12 John_Kang from salt://
13:12 lordcirth John_Kang, how often will you want to change the files on the master?  You could use tar.
13:12 John_Kang actually, we are using saltstack plugin in Jenkins
13:13 Rumbles you keep the whol directory structure in salt?
13:13 John_Kang I am creating the state and send the state to developer to use\
13:13 John_Kang which mean i don't change the source
13:13 John_Kang developer does
13:13 lordcirth I see, that's a problem
13:14 lordcirth Still, there might be a way to use archive.extracted or some other download method
13:14 John_Kang lordcirth: btw, i just ran the state with clean option, it works well
13:15 lordcirth John_Kang, yup.  Btw, why does source code need varying permissions?
13:15 John_Kang lordcirth: let me consider it
13:15 lordcirth It should all just be rw, and the binaries you're compiling there, so they'd be rwx
13:16 John_Kang lordcirth: hm... I just don't want to argu with developer about the permission he made :D
13:16 jschoolcraft joined #salt
13:17 John_Kang lordcirth: Rumbles: thanks again, guys !!!
13:17 Rumbles sorry babilen I didn't understand what you meant in your last response
13:17 Rumbles np John_Kang
13:17 John_Kang Rumbles++
13:17 John_Kang lordcirth++
13:18 babilen Rumbles: What exactly did you not understand?
13:19 Rumbles any of it?
13:19 Rumbles I don't get what you mean
13:19 Rumbles or why it's relenvant
13:20 babilen Rumbles: I replied to your "Rumbles > babilen, I don't even get how to use those states"
13:20 babilen So, basically: Target the "- users" (and other states if you want to use them) to all boxes. This won't do anything. In addition target suitable pillar data.
13:20 babilen You can organise the pillar data in whatever way you deem most advantageous
13:21 mage_ is there a difference between using "local.file.symlink" and using a "salt.function: - name: file.symlink" in the reactor ?
13:21 lordcirth John_Kang, do you know if the devs have different permissions, or are you just assuming they might?
13:21 Rumbles okay, but I asked you ".:Rumbles:. babilen, I assume you're on the latest salt, you said you reproduced that find issue"
13:21 numkem joined #salt
13:21 Rumbles I'm not sure why you were even asking me about users still
13:23 babilen I ran it on 2016.3.3
13:23 Rumbles thanks
13:24 babilen And please apologise that I answered a question you asked earlier. I will make sure to only reply to your most current utterance from now on and do not take earlier ones into account
13:24 Rumbles wtf? no need to be an arsehole
13:25 Rumbles I just idn't get why you were asking that question
13:26 babilen I replied to your messages .. and you apparently think/thought that it was no longer relevant or even understood why I might have replied to it
13:27 babilen It is also not nice to call people "arsehole", but okay
13:27 Rumbles I asked you a direct question and you replied with something from an earlier conversation
13:27 babilen Stack vs Queue
13:29 edrocks joined #salt
13:30 Rumbles wtf, I get KeyError: 'apache_module.enable' on 1 server while another server is fine with it, both running salt-minion 2015.5.3
13:33 netcho when i try to run state which is cmd.run my script it says that script has a syntax errror, but when i run that state from minion with salt-call it runs fine
13:33 Rumbles pastebin the state and the error?
13:35 netcho Rumbles: https://paste.debian.net/893277/
13:36 mage_ do you think this is correct https://gist.github.com/silenius/255138a299354ed3a823ad68a23dbb97 ?
13:36 mage_ (especially line 29-35)
13:36 netcho Rumbles: error: https://paste.debian.net/893278/
13:37 Rumbles mage_, I've never seen local.state.apply before, sorry
13:38 mage_ it's on https://docs.saltstack.com/en/latest/topics/reactor/index.html
13:38 netcho when ran from minion all is fine
13:38 babilen netcho: Are you sure you want "stateful" ?
13:38 babilen netcho: https://docs.saltstack.com/en/latest/ref/states/all/salt.states.cmd.html#using-the-stateful-argument
13:38 babilen netcho: What is the output of that script if you run it manually?
13:42 babilen netcho: It would also help if you could paste the script itself. Are you perhaps using #!/bin/sh as shebang?
13:43 toastedpenguin joined #salt
13:44 netcho babilen: somehow shebang was missing
13:44 netcho looks like that was the issue
13:44 babilen And it is working as expected once you defined it?
13:45 netcho looks like that was the issue
13:45 netcho yes
13:45 babilen Wonderful :)
13:48 JohnnyRun joined #salt
13:49 netcho also noticed smth... maybe it;s just me :)
13:50 netcho this script associates elastic ip to minion, and after that some minions become unresponsive until service is restartyd
13:50 netcho restarted
13:50 ivanjaros joined #salt
13:53 SaltyVagrant_ joined #salt
13:55 impi joined #salt
13:57 mage_ so local.state.apply should not be used with a salt.runner.orchestrate
13:57 mage_ but only in a reactor, right ?
13:58 mage_ local.state.apply in a reactor is the same as salt.state in an orchestrature runner if I understand well
13:58 mage_ a runner.state.orchestrate I mean
14:05 perfectsine joined #salt
14:07 dyasny joined #salt
14:07 Miouge joined #salt
14:09 cmbrnt joined #salt
14:09 Miouge joined #salt
14:10 akhter joined #salt
14:11 cmbrnt is it possible to do negative matching on minions, to match a rule to all hosts except one?
14:13 jas02_ joined #salt
14:13 Reverend is there a regex escape in jinja? I've tried regex_escape() but it says it doesn't like it :(
14:13 DEger joined #salt
14:13 Reverend we've got www.example.com that needs to be www\.example\.copm
14:13 Reverend com*
14:13 tuxx joined #salt
14:15 dyasny joined #salt
14:16 Cottser joined #salt
14:16 subsignal joined #salt
14:23 netcho looks i was right, when associating elastic ip to minion it becomes unresponsive
14:26 Reverend are you using a nat gateway?
14:26 netcho yes, in a VPC
14:26 Reverend if you associate an EIP to a box within a network using a NAT GW, you're gonna have a bad time
14:27 _JZ_ joined #salt
14:27 netcho naah, no NAT GTWs
14:27 Reverend oh
14:27 Reverend your master is using the internal IP for your minion though right?
14:27 Reverend not the external?
14:27 netcho yes
14:27 Reverend hmm. What are your route tables showing>?
14:28 Reverend your minion -is- accessible by yoiu though right?
14:28 Reverend just not the master?
14:28 netcho nothing specia, for local subnet tgt is local ands 1 internet gtw for 0.0.0.0.0
14:28 DEger joined #salt
14:29 Reverend can you still SSH onto it and dick aboot?
14:29 netcho yes
14:29 netcho everytnig is ok just minmion does not return nothing
14:29 Reverend can you do a `salt-call state.highstate` from the minion?
14:29 netcho \i can run states from minion
14:29 netcho but not from master
14:29 netcho not test.ping works
14:30 Reverend I've had this before... but i can't remember what it was
14:30 Reverend gimme a second. lemme check my wiki.
14:30 netcho sure, thanks
14:30 Reverend whilst I'm gone, try removing the minions key and getting a new one
14:30 netcho i restart minion and then it works
14:30 netcho but..
14:32 Reverend yeah. I've got here in my notes to remove the minion key, and restart the master. that fixed mine.. but I think it was because I was moving servers to a NAT'd subnet
14:33 netcho i can set minion restart after this state but then my highstate will run again cuz i have reactor on /minion/smth-*/start
14:33 Reverend is that a problem? :S
14:34 netcho not sure if it'll complain that highstate is already running?
14:34 netcho or thatr is just for state.apply
14:35 Brew joined #salt
14:35 akhter joined #salt
14:36 toastedpenguin joined #salt
14:36 netcho hm
14:36 netcho looks like it returned
14:36 netcho just needed some time
14:36 Reverend sweet!
14:37 anotherzero joined #salt
14:37 netcho but it kinda slows down my process
14:38 Reverend netcho: whilst im here though, it's worth using a NAT gateway to prevent any weird internet stuff getting to your servers :)
14:38 Reverend although they don't come cheap (in the world of AWS costs). they're like $30 each / month
14:39 Reverend just means you can scale in and out as needed, and only need to worry about 3 ip addresses
14:41 netcho that would solve some issues here
14:43 netcho but that change is not an option atm
14:43 netcho cuz lots of EIPs on akamai and whitelisted around
14:43 netcho it would require a  huge reorganization
14:45 racooper joined #salt
14:48 akhter joined #salt
14:48 oida joined #salt
14:49 Reverend akami?
14:49 Reverend the CDN?
14:49 netcho yes
14:49 Reverend y u no cloudfront :P
14:49 netcho akamai wtf :D
14:49 netcho ftw lol
14:50 Reverend wtf indeed
14:50 Reverend xD
14:50 Reverend you -could- pick up your 3 EIP'
14:50 Reverend s
14:50 Reverend and add them to the whitelist as needed, and then migrate each subnet to a NATted route
14:50 netcho no expirience with cloudfront but on akamai we habe huge conf
14:50 netcho s/habe/have
14:51 netcho lots of streams/video playbacks to be cached
14:52 keimlink_ joined #salt
14:52 Reverend oh man.
14:52 netcho i think cloud fron could not cache https back then...
14:52 netcho 4 years ago
14:52 netcho or smth
14:52 Reverend CDN's work relatively the same. I've used cloudfront a couple of times... and it works pretty well. It does do HTTPS for sure.
14:53 Reverend but if you have cookies, it might play funny buggars
14:53 patarr joined #salt
14:53 netcho yeah we use cookies
14:54 netcho and more then 10 million users
14:54 netcho it will tripple now when new API comes ouit
14:54 netcho out
15:01 Reverend damn son.
15:02 Reverend I think you can whitelist what AWS uses in it's hashes... a little like varnish, you can tell it to include cookies and whatnot. We chose not to for our CDN, as the assets should be identical between users.
15:04 Tanta joined #salt
15:06 netcho for the moment i just need to autoassign EIPs whitelisted from our partners on autoscaling instance creation, before app gets there
15:07 netcho manage to do that but it's kinda slow
15:07 Reverend what I would do, is get your NAT gateway sorted. seriously
15:07 Reverend it's gonna be quicker to whitelist 3 ip's than it is to figure out how to work with that.
15:07 XenophonF anyone else experience problems with the 2016.3.4 salt-master ignoring its configuration files?
15:07 Tanta until your SPOF NAT gateway fails
15:07 Tanta then you might wish you had direct routes to the internet
15:07 Reverend Tanta: SPOF NAT? really?
15:07 babilen XenophonF: I haven't heard about it, but then .4 has just been released
15:07 ferbla joined #salt
15:08 Tanta I can't tell you how many times I've been brought on to 'work on a deployment' at aws and find that setup
15:08 Reverend Tanta: you need the same number of NAT's as you have availability zones.
15:08 Reverend so if an AZ dies, you still ahve outbound conns to each other NAT for the AZ's that are still up.
15:08 XenophonF i'm seeing this in the master log file: NameError: global name '__utils__' is not defined
15:08 felskrone joined #salt
15:08 Tanta I just assign public IPs to my instances and use a proper IP filter
15:08 XenophonF the error's coming from salt/utils/s3.py
15:09 Reverend Tanta: how do you work with IP Whitelisted API's?
15:09 netcho Tanta: yeas i have rly strict sec groups
15:09 Tanta do you mean a remote API?
15:09 netcho Reverend: u wanna hear my solution? :D
15:09 XenophonF brb gonna check the mailing list
15:09 Reverend netcho: sure man - hit me
15:09 Reverend Tanta: i mean an API that has an IP whitelist.
15:09 keltim joined #salt
15:10 netcho just to be sure $$ is not an issue here :)
15:10 Tanta most of the APIs that I work with implement a token that is used to authenticate against the API as part of the session
15:10 netcho our servers need to be whitelisted at pur partners
15:10 Tanta IP whitelisting is going to be a fragile solution since it always must be maintained
15:10 netcho i solve it like this
15:11 XenophonF i'm going to try downgrading to 2016.3.3
15:11 Reverend Tanta; that's exactly the point of NAT gateways... it's also more secure. Strict SG's are great, but DMZ subnets not onyl protect the aPI's your connecting to, but also your servers from direct incoming conns from the net.
15:11 fracklen joined #salt
15:11 netcho i serve API
15:12 netcho and only allowd connection to our ELBs are Akamai ips
15:12 Reverend netcho: does your partner have an API to manage your IP whitelist? :P haha
15:12 netcho they whitelisted like 10 of our EIPs
15:12 Tanta you could do the same thing with an HTTP proxy
15:12 netcho those 10 i habe in DNS entry
15:13 XenophonF ok yum took me back to 2015.5.10, which is what I had installed before
15:13 Reverend netcho: intriguing. I see where you're taking this. However most companies won't want to whitelist a hostname for the reasons that you are in control of it and not them.
15:13 Reverend you can always ask them though...
15:13 Reverend if they're friendly and trust you... then I can't see why they wouldn't.
15:13 netcho on boot (highstate) i query that DNS entry and get list of whitlisted IPS
15:13 babilen XenophonF: That's quite the jump, it would be interesting to see 2015.8 and 2016.3.3 though
15:14 netcho i also get list of unassociated EIPs
15:14 netcho compare 2 lists
15:14 babilen XenophonF: Had a look and haven't seen that issue yet
15:14 XenophonF babilen: i'm using s3fs. it looks like the error is somewhere in the built-in AWS client
15:14 Reverend netcho: you using boto3 for this? you could buy like 30 EIP's and get them all whitelisted, and just script something on boot to grab one of the unallocated EIP's to associated it to itself :)
15:14 netcho_ joined #salt
15:14 Reverend netcho: you using boto3 for this? you could buy like 30 EIP's and get them all whitelisted, and just script something on boot to grab one of the unallocated EIP's to associated it to itself :)
15:15 netcho_ dripped
15:15 Reverend aha
15:15 netcho_ thats what i-m doing
15:15 Reverend that will work. it sounds messy, but it will work.
15:15 XenophonF the 2015.5.10 master is working - so the downgrade worked.  that's the version in EPEL, unfortunately.
15:15 babilen XenophonF: Unfortunately I do not have much experience with s3fs and cannot comment on changes in 2015.8 / 2016.3.3 and 2016.3.4 -- But you did only ever run it on 2015.5, didn't you?
15:16 nicksloa_ joined #salt
15:16 Reverend cloudformation might be able to do it for you :)
15:16 XenophonF yeah
15:16 netcho_ Reverend: it's not that messy
15:16 XenophonF babilen: I've got a FreeBSD master running 2016.3.3 with s3fs.  I'll try upgrading it to 2016.3.4 tonight.
15:16 netcho_ around 30 lines of code, pretty formated :D
15:16 Reverend ahahaha
15:17 XenophonF If I get the same error, I'll open an issue.
15:17 babilen XenophonF: You could try http://repo.saltstack.com/2015.8.html
15:17 babilen (on your RHEL box)
15:17 Reverend netcho_: have a look at cloudformation. not sure if you can allocate EIP's from that... never tried... but it might be able to. :) it can do pretty much everything else
15:17 netcho_ naah
15:17 netcho_ i use salt for mostly everuytniog
15:17 Reverend i mean,just the creation of the boxes and whatnot
15:17 babilen XenophonF: Comparing 2015.5 to 2016.3.4 will result in a huge list of changes. If you could reliably reproduce it between 2016.3.3 and 2016.3.4 that would surely help a lot
15:18 netcho_ salt creates load balancer, autoscaling groups, instances... all that goos stuf
15:18 netcho_ s/goos/good
15:18 Reverend nice.
15:18 Reverend do you spin up clusters regularly then?
15:18 netcho_ actually devs thinks jenkins does that... :D
15:18 Reverend LOL
15:19 XenophonF babilen: https://gist.github.com/niheconomoum/fb0f14c72b9269259f50441cc7fefed0
15:19 babilen netcho_: You should tell them or they won't appreciate it! :)
15:19 netcho_ well it does, just has salt behind LD
15:19 mpanetta joined #salt
15:19 Reverend babilen is a salt evangelist :P
15:19 * Reverend gets the bread and wine
15:20 netcho_ Reverend: yea on every prod deploy i create new external and indernal load balancers, new instances and then just switch aliases on route53
15:20 XenophonF Eh, we can skip the bread. ;)
15:20 netcho_ actually new autoscaling group
15:21 lubyou_ joined #salt
15:21 lubyou_ can I pass salt:// url's to cmd.run?
15:21 Reverend netcho_: that sounds like a very good idea. that's basically the entire point of full virtualisation.
15:21 netcho_ yes
15:21 netcho_ so i have 0 donwtime deployment
15:21 XenophonF that's how it's done
15:21 Reverend granted, it seems a little long winded... but sounds like a good plan :()
15:22 babilen XenophonF: Doesn't seem to have been reported (yet)
15:22 netcho_ well it could be twice faster if i'd had prebaked AMIs
15:22 akhter joined #salt
15:22 netcho_ i just install salt-minion via cloud-init and then magic happens
15:23 babilen XenophonF: I found https://github.com/saltstack/salt/issues/34780, but that doesn't look related to your issue
15:23 saltstackbot [#34780][OPEN] S3fs broken in 2016.3.1 | Description of Issue/Question...
15:25 sarasfox joined #salt
15:25 ozux joined #salt
15:26 sarasfox ok i need help getting salt.modules.win_update to work
15:26 Ni3mm4nd joined #salt
15:27 fracklen joined #salt
15:27 fracklen_ joined #salt
15:29 Renich joined #salt
15:30 edrocks joined #salt
15:31 Reverend netcho_: prebuilding AMI's is the way if you need infrastructure OTF
15:31 netcho_ OTF?
15:33 Reverend on-the-fly
15:33 Reverend we serve web traffic here, and if we get a spike, we need a server running in like 60 seconds...
15:34 hasues joined #salt
15:34 hasues left #salt
15:34 Reverend there's no way to do that with 30+ packages to install when they can be prebuilt in an ami
15:34 demize- joined #salt
15:34 Reverend just means more maintenance when vulners are released.
15:34 netcho_ yes, and baking new on whe upgrading
15:34 netcho_ packages
15:35 preludedrew joined #salt
15:36 Reverend netcho_: yeah - like the dirtycow kernel vulnerability. basically rebuilt a bunch of AMI's for use in Autoscaling groups again
15:36 Reverend then jsut cycle them out of rotation
15:36 planner joined #salt
15:36 netcho_ Reverend: u have idea on how to change minion id via cloud-inint? so i can have minion preinstalled
15:37 Reverend no clues. I don't use cloud-init.
15:37 Reverend sorry chap
15:37 netcho_ or any other init..
15:37 netcho_ >D
15:37 Reverend nope ;P
15:37 netcho_ u just need it to change on boot, to smht-$instance_id
15:38 MaZ- joined #salt
15:41 writteno1 joined #salt
15:41 strobelight joined #salt
15:42 yetanotherzero joined #salt
15:42 Reverend dat netsplit doe
15:44 mrueg joined #salt
15:44 demize joined #salt
15:44 moy joined #salt
15:44 Kelsar joined #salt
15:44 smakar_ joined #salt
15:44 samkottler joined #salt
15:45 simonmcc joined #salt
15:45 liviudm joined #salt
15:45 WKNiGHT joined #salt
15:45 Nightcinder joined #salt
15:46 bVector joined #salt
15:46 tcolvin joined #salt
15:46 mihait joined #salt
15:46 wiqd joined #salt
15:46 m0nky joined #salt
15:46 leev joined #salt
15:47 abele joined #salt
15:47 mikepea joined #salt
15:47 supermike_ joined #salt
15:47 twodayslate joined #salt
15:47 nahkiss joined #salt
15:47 alexlist joined #salt
15:48 d3c4f joined #salt
15:48 JamieH joined #salt
15:48 lubyou_ is it possible to have a local _modules directory on a minion for module testing? if so, what would be the path?
15:49 kutenai joined #salt
15:49 antonw joined #salt
15:49 ramblinpeck joined #salt
15:49 Eugene joined #salt
15:49 mattl joined #salt
15:49 Twiglet joined #salt
15:50 tercenya joined #salt
15:50 arapaho joined #salt
15:50 dijit joined #salt
15:50 kidneb joined #salt
15:50 ALLmightySPIFF joined #salt
15:51 ALLmightySPIFF joined #salt
15:51 dlloyd joined #salt
15:52 phtes joined #salt
15:52 czchen joined #salt
15:53 Rebus joined #salt
15:53 johnkeates joined #salt
15:53 haam3r joined #salt
15:54 hillna joined #salt
15:55 KanyeWest2020 joined #salt
15:55 KanyeWest2020 mornin'
15:55 KanyeWest2020 (I like to think of ridiculous names)
15:55 KanyeWest2020 super quick question. Can you use ordering in a for loop?
15:55 jas02 joined #salt
15:55 xet7 joined #salt
15:56 Miouge joined #salt
15:56 Renich joined #salt
15:56 cmbrnt left #salt
15:57 cscf KanyeWest2020, what sort of ordering?
15:57 tapoxi joined #salt
15:58 keltim joined #salt
15:59 KanyeWest2020 within an sls file. I know it's *supposed* to order from top to bottom, but I'm putting a file on disc then executing that file and salt appears to be trying to execute the file before it's created
15:59 KanyeWest2020 so I figured I'd just add - order: 12 and - order: 13
15:59 KanyeWest2020 or whatever the numbers may be
15:59 rherna joined #salt
15:59 netcho_ KanyeWest2020: use watch or require
16:00 KanyeWest2020 yeah, require didn't work. I'll try watch
16:00 KanyeWest2020 thanks so much!
16:03 Brew joined #salt
16:03 Tanta do onchanges: previous_state
16:04 tiwula joined #salt
16:04 Tanta that only triggers if the contingent state produced a change
16:04 matth joined #salt
16:04 matth Hello
16:05 matth is it possible from the reactor file to get the id of the minion which is making the call ?
16:07 babilen data['id'] ?
16:07 Sylvain31 joined #salt
16:08 babilen matth: ^
16:09 bmccormick joined #salt
16:10 matth babilen: ok, can I pass this ID to a state to execute on the salt-master from this reactor ?
16:10 babilen matth: There are various examples of data['id'] in use on https://docs.saltstack.com/en/latest/topics/reactor/ -- If you want to pass it over you will have to explicitly pass it
16:10 SaltyVagrant joined #salt
16:11 babilen (e.g. as pillar data)
16:11 matth yes with "- pillar" ?
16:11 babilen /srv/salt/haproxy/react_new_minion.sls exemplifies that
16:11 babilen Adjust to the execution module function you are calling
16:12 babilen You could pass it in directly also
16:12 babilen But that depends on the signature of the function you are calling. If you are calling state.apply you'd pass it as kwarg
16:12 debian112 joined #salt
16:13 subsignal joined #salt
16:13 netcho_ ok so looks like just replacing id: minion in minion config should work? with sed or smth?
16:13 matth basically the reactor is trigger when the instance VM is created. and I would like to execute a states on the salt-master from this reactor. This states needs to know the id of the minion
16:14 jas02 joined #salt
16:14 babilen matth: Have you read through those examples? I believe they show a similar workflow quite nicely
16:15 babilen netcho_: Many people use the salt formula to manage the salt configuration. Also make sure that you remove /etc/salt/minion_id if it is present.
16:15 haam3r joined #salt
16:15 DEger joined #salt
16:15 sarcasticadmin joined #salt
16:16 babilen matth: And you wouldn't execute states on the master, but on the minion that's running on the same box as the salt-master (if that makes sense)
16:16 netcho_ babilen: i wanna have minion preinstalled on image and then change id on boot
16:16 Sylvain31 Hi, I got a sls file which expands code to json as argument to a script on the mininon: http://paste.debian.net/893343/ line 14, I've a dual problem of quoting and mixing quotes and also using json, any other way to pass json to my scirpt?
16:17 sarasfox sudo salt bob cp.get_url salt://home/ubuntu/NessusAgent.msi c:\\ the minion is windows
16:17 sarasfox I keep getting false back
16:17 matth babilen:  here what I've done : http://pastebin.com/PtBf0ES1
16:18 Reverend `UnicodeDecodeError: 'ascii' codec can't decode byte 0xc2 in position 11871: ordinal not in range(128)`
16:18 Reverend wtf
16:19 abednarik joined #salt
16:20 sarasfox how do i move file from ubuntu master to windows minion
16:23 cscf sarasfox, try only 1 backslash: C:\path\to\file
16:24 sarasfox still getting false
16:25 Sylvain31 sarasfox: did you try empty, the doc says: "The default behaviour is to write the fetched file to the given destination path. If this parameter is omitted or set as empty string (''), the function places the remote file on the local filesystem inside the Minion cache directory and returns the path to that file«
16:26 Sylvain31 sarasfox: https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.cp.html#salt.modules.cp.push
16:27 sarasfox ok i will try push
16:28 jas02 joined #salt
16:32 matth babilen: following : https://docs.saltstack.com/en/latest/topics/reactor/#advanced-state-system-capabilities, should I be able to get data.id in the "/srv/salt/orch/do_complex_thing.sls" file ?
16:32 sarasfox Still geting false
16:35 Sylvain31 sarasfox: is your master an minion too, try to it self too, or any linux box
16:37 sarasfox one master ubuntu and one minion a windows 2016 both on aws
16:38 subsigna_ joined #salt
16:39 haam3r joined #salt
16:43 sarlalia1 joined #salt
16:46 DammitJim joined #salt
16:47 Sylvain31 sarasfox: salt-cp -E dns /etc/motd /tmp
16:47 Sylvain31 this one works, the other gives me an error…
16:47 Sylvain31 salt -E dns0 cp.push /etc/motd
16:48 Rumbles is there a command for the apache module that allows you to do a configtest like you can with nginx? I know how to do it on the server, but I want to use that in a state
16:48 Sylvain31 Rumbles: apache2ctl configtest (debian)
16:48 Rumbles yeah that works on the server
16:49 Rumbles but how would I use that in a state?
16:49 Sylvain31 cmd.run ?
16:49 Rumbles similar to: http://stackoverflow.com/questions/37009293/reload-nginx-config-from-salt-state-only-if-the-configtest-passes
16:49 Rumbles I use module.wait to check the configtest result in my example
16:49 Rumbles can I do the same with cmd.run ?
16:52 patarr joined #salt
16:52 gabid joined #salt
16:55 ekristen joined #salt
16:56 DammitJim what is the best way to view returned info from a salt module
16:56 DammitJim I am trying to "view" what servers have what version of salt-minion
16:56 Rumbles does that look right Sylvain31 https://paste.fedoraproject.org/475090/ ?
16:56 nicksloan joined #salt
16:56 Rumbles salt '*' cmd.run 'salt-minion -version' DammitJim
16:57 Rumbles not what you asked but should get you a response in the console
16:57 DammitJim Yeah, I already did that, but it's hard to get a status of the servers when you have 100 minions
16:57 DammitJim I probably need to use a returner?
16:58 Rumbles no idea, haven't used returners
17:04 Reverend is there any way for salt to create me git repo
17:04 Reverend repos* ?
17:05 Reverend looking at the saltstack git docs... i can't see any way to simply check if a repo exists, and if it doesn't, make it.
17:07 patarr_ joined #salt
17:10 mpanetta joined #salt
17:10 Sylvain31 Rumbles: does it work for apache2ctl ? what appends if it fail?
17:10 Tanta Reverend: test -f ~/.git && grep "whatever" ~/.git/config
17:11 Tanta if that fails, do a clone
17:11 Reverend Nice one Tanta :)
17:11 sh123124213 joined #salt
17:11 sh123124213 joined #salt
17:12 Rumbles it works on the server Sylvain31 , not sure what you mean about appending if it fails... ?
17:13 Sylvain31 yaml quote Mix: found for http://paste.debian.net/893343/ becomes:     - name: './pdns_zone.sh create ''{{ domain }}'' ''{{ data | json }}'''
17:14 Sylvain31 Rumbles: does cmd.wait check the correct return failure value if any bad syntaxe is detected?
17:18 keltim joined #salt
17:21 bltmiller joined #salt
17:22 samodid joined #salt
17:22 honestly the state will fail if it is executed but the returncode of the command isn't 0
17:24 Rumbles if it fails it returns 1
17:25 pipps joined #salt
17:25 Rumbles well $? == 1
17:25 ivanjaros joined #salt
17:25 Rumbles and if it passes it's return code is 0
17:26 Sylvain31 Rumbles: yet does states fails too as expected?
17:26 Sylvain31 yet=yep
17:27 edrocks joined #salt
17:30 ronnix joined #salt
17:30 Rumbles I haven't tried yet
17:30 Rumbles I will test and let you know
17:36 Rumbles probably tomrrow now, thanks though Sylvain31
17:37 Sylvain31 no problemo
17:41 akhter joined #salt
17:41 ahammond In the context of a packer file, I'm using the salt-masterless provisioner. I need to run saltutil.sync_modules before I compile the highstate. Is there any reasonable way to do this?
17:44 nicksloan joined #salt
17:48 Trauma joined #salt
17:48 Edgan joined #salt
17:48 akhter_1 joined #salt
17:49 deei joined #salt
17:49 nicksloan joined #salt
17:50 pcn Is there a way to get the IP addresses of all of the minions attached to a master from  within a state?
17:50 pcn (on the master, of course)
17:51 Elsmorian joined #salt
17:51 CTaylor joined #salt
17:51 Sylvain31 ahammond: orchestration ?
17:52 deei hi, in jinja as far as I can tell you can't get the date - it must be passed in via e.g. context. how can I achieve this via salt? I'm trying to create a datestamp for a bind config template
17:55 gtmanfred ahammond: sync_modules is run as part of a highstate...
17:55 gtmanfred it should be running a saltutil.sync_all as part of going into the highstate
17:56 Sylvain31 deei: https://docs.saltstack.com/en/latest/topics/jinja/#filters
17:56 sp0097 joined #salt
17:56 ahammond @gtmanfred ah, I see. It's missing the pillar. Sigh.
17:57 * ahammond thinks that maybe he shouldn't reuse names between different things
17:58 gtmanfred ahammond: it shouldn't be, sync_all should do a sync on all the different things that goes to a minion, including custom external pillars and custom modules and grains etc
17:59 ahammond @gtmanfred no, this is user error. I haven't provided the pillar it's looking for.
17:59 gtmanfred ahh ok :)
18:05 pipps joined #salt
18:05 Miouge joined #salt
18:05 deei Sylvain31: thank, but I think you misunderstood. I know you can format a date, but you can't actually get the current date without using an external python call. unfortunately I'm using salt-ssh and using external functions in a _modules dir doesn't work
18:09 pcn deei you mean something more external than the jinja rendering of None | strptime?
18:09 pcn Ah,  yes you do.  Interesting, i didn't think there was something about salt-ssh that precluded that
18:10 deei pcn: yeah, it's really annoying. we're moving from a salt setup to salt-ssh, and none of the custom modules work
18:10 deei for getting the bind serial number I need a date stamp. unfortunately, that can't be done in pure jinja
18:10 klaas joined #salt
18:11 Sylvain31 deei: dont know those limitation, could you put the jinja filter in context: ?
18:12 Sylvain31 deei: can you call {{salt['
18:12 deei sure, but it doesn't get around the problem. jinja itself has no concept of time and cannot have
18:13 Sylvain31 deei: can you call {{salt['cmd.run']("date '+%Y%m'")}} (not tested)
18:13 deei yeah, that might work
18:13 deei thanks, I'll give it a go
18:14 deei works perfectly. thanks. should have thought of that approach
18:15 Sylvain31 deei: sysadmin trick, the door is closed, try the window, next the cave, the roof, etc. ;)
18:15 Sylvain31 yep, cant be ugly…
18:15 Sylvain31 yep, can be ugly…
18:15 jas02 joined #salt
18:18 jas02_ joined #salt
18:18 Trauma joined #salt
18:21 wendall911 joined #salt
18:26 multiscan joined #salt
18:30 akhter joined #salt
18:34 akhter joined #salt
18:34 sp0097 left #salt
18:36 xbglowx joined #salt
18:39 Miouge joined #salt
18:42 Miouge joined #salt
18:42 fracklen joined #salt
18:46 DammitJim what is the appropriate way to set grains after one successfully runs a state?
18:47 sh123124213 joined #salt
18:48 sh123124213 joined #salt
18:53 fracklen joined #salt
18:54 xet7 joined #salt
18:55 misconfig joined #salt
18:59 Brew joined #salt
19:00 s_kunk joined #salt
19:02 samodid joined #salt
19:02 XenophonF deei: you can write your own execution module that exposes a function that returns a timestamp
19:03 XenophonF that's been my approach instead of shelling out in the middle of rendering
19:06 iggy it's easier than that
19:06 patarr joined #salt
19:07 deei XenophonF: as explained, unfortunately that doesn't work with salt-ssh
19:07 edrocks joined #salt
19:08 iggy extra_filerefs ?
19:10 eightyeight on the commandline, how can i target a grain numerical value to be less than a number?
19:11 eightyeight IE: $ salt -G 'osrelease_info[0] < 16' test.ping
19:11 iggy deei: {{ ''.strftime('%Y-%m-%d
19:11 iggy deei: {{ ''.strftime('%Y-%m-%d') }} try something like that
19:11 iggy eightyeight: you can't
19:11 eightyeight le sigh
19:11 eightyeight iggy: thx
19:13 iggy eightyeight: you can do: salt -C 'G@osrelease_info:0:12 or G@osrelease_info:0:14'
19:13 iggy (I think)
19:13 eightyeight note to self: don't tab-complete arguments to salt(1) on the master
19:14 eightyeight iggy: when this stops hanging, i'll give that a try. thx.
19:14 robbintt joined #salt
19:15 iggy ctrl-c
19:16 eightyeight heh. yeah...
19:16 eightyeight "The minions may not have all finished running and any remaining minions will return upon completion."
19:16 * eightyeight waits for the terminal to come back
19:22 erewok joined #salt
19:23 erewok question: I'm trying to write a salt-state module where I want to create a variable based on the result of running some other command.
19:24 erewok Here's my usecase: I need to pull a branch from a git repo
19:24 erewok but I don't know what the branch is until I pull it out of a service where those are stored
19:24 iggy variable to be used where?
19:24 erewok so I need to do something like:
19:24 erewok {% set git_branch = $(result of some shell script)$ %}
19:24 iggy won't work
19:24 patarr joined #salt
19:24 erewok yeah, that's what I Thought
19:25 iggy the jinja gets rendered before the git clone/checkout is done
19:25 iggy so the variable won't be right
19:25 erewok hmm.
19:25 erewok do you have an idea how I can dynamically do something like:
19:26 iggy where is the variable being used?
19:27 iggy in the past I've just written states that did all the work
19:27 erewok rev: {{ git_branch }}` <-- where this value can only be determined at runtime?
19:27 erewok it's inside a call to this module:
19:27 Elsmorian joined #salt
19:27 erewok https://docs.saltstack.com/en/latest/ref/states/all/salt.states.git.html#salt.states.git.latest
19:28 Elsmorian joined #salt
19:28 erewok I think  I just need to be more creative.
19:29 erewok or I need to scrap using states.git and write my own script that does what I want.
19:29 iggy that's probably your best bet
19:29 iggy you can probably just copy that function and massage it to do your bidding
19:29 iggy (I've done that more than once)
19:30 erewok that's a pretty good suggestion, actually
19:30 erewok thanks.
19:30 CampusD joined #salt
19:36 lattenkiste[m] joined #salt
19:41 Miouge joined #salt
19:42 akhter joined #salt
19:43 patarr joined #salt
19:44 nawwmz joined #salt
19:44 nawwmz is anyone else getting errors installing salt-master because of python2-crypto.x86_64 ?
19:44 Deliant joined #salt
19:48 akhter joined #salt
19:48 netcho_ joined #salt
19:52 Rubin ugh why does file.comment use .bak backup naming scheme instead of the backup_mode like file.managed !?!
19:52 bltmiller joined #salt
19:53 Sketch salt states seem to be horribly inconsistent sometimes, it's like each one was written by a different person with no regard for what was already there
19:53 Rubin just created a .bak file in all my nodes logrotate.d/ directory which logrotate does NOT ignore!
19:53 Rubin friggin annoying and unexpected
19:57 patarr joined #salt
20:00 Renich joined #salt
20:04 bltmiller joined #salt
20:04 jas02 joined #salt
20:06 amontalb1n joined #salt
20:07 cscf Rubin, that is strange. Perhaps a hidden file will be ignored?
20:07 POJO joined #salt
20:07 cscf Rubin, perhaps you should open an issue about the differing behaviour
20:11 patarr joined #salt
20:16 bltmiller joined #salt
20:17 jas02_ joined #salt
20:21 codeape joined #salt
20:21 austin joined #salt
20:24 Guest70716 if using a syndic master, should syndic show as "up"? https://gist.github.com/austinpapp/a09e89edb2a802e7d9ca36e8b422678a
20:26 pipps joined #salt
20:27 austin_ sorry, had to change my handle
20:29 pipps joined #salt
20:31 PeterO joined #salt
20:39 akhter joined #salt
20:43 akhter joined #salt
20:44 ferbla joined #salt
20:47 patarr joined #salt
20:50 CampusD is there a way to capture every execution and state module run against a minion on the minion
20:50 CampusD I've tried returners with local_cache
20:51 CampusD if there is returned data I see it in the return.p file
20:51 CampusD but actual command that triggered the return is not there unless I run a high state, then a out.p file is created with "highstate" on it
20:51 CampusD any thoughts?
20:53 pipps joined #salt
20:53 Elsmorian joined #salt
20:55 Ahlee CampusD: run in debug mode, pull the log?
20:55 Ahlee we run all minions in debug, and that's one of the reasons
20:56 CampusD is there a flag, configuration item in the minion config to enable that by default? I guess I could look it up
20:57 Ahlee CampusD: https://github.com/saltstack/salt/blob/develop/conf/minion#L665
20:57 Ahlee whoops, scratch that. it's line 671
20:58 CampusD Ahlee, I'll give that a try, thanks so much!
20:59 jas02 joined #salt
20:59 patarr joined #salt
20:59 Edgan I am trying to use pyobjects, but I need to import a map.jinja that isn't in pyobjects. It tries to parse it as pyobjects, not jinja.
21:03 akhter joined #salt
21:03 Tanta so use a YAML parser
21:05 Edgan Tanta: You mean don't use pyobjects? or activate the yaml parser some how?
21:06 whitenoise joined #salt
21:07 Tanta Pyobjects provides shortcut functions for calling pillar.get, grains.get, mine.get & config.get on the __salt__ object. This helps maintain the readability of your state files.
21:07 Tanta Each type of data can be access by a function of the same name: pillar(), grains(), mine() and config().
21:08 Tanta it seems like you can still target the regular YAML grains and pillars even if you use the pyobjects renderer
21:08 Tanta maybe target your yaml to the host with the top.sls file
21:08 Edgan Tanta: I need a map.jinja, I am already using the pillar function.
21:09 Tanta https://docs.saltstack.com/en/latest/ref/renderers/all/salt.renderers.pyobjects.html#map-data
21:09 Tanta this appears to be your answer
21:09 Edgan Tanta: I can't rewrite this in pyobject style
21:09 Edgan Tanta: I already saw that
21:10 Edgan Tanta: I am trying to add just a little pyobjects to a jinja|yaml formula
21:13 Tanta you could serialize the data into an intermediate format with a 3rd party parser
21:13 Tanta I don't see another solution
21:17 Edgan Tanta: Found an easier way in yaml. There are so many poorly talked about features.
21:25 akhter joined #salt
21:27 tmkerr joined #salt
21:28 akhter joined #salt
21:34 tapoxi so reactors are just engines that listen for events on the bus and run states?
21:35 hemebond yip
21:35 tapoxi ok. kewl.
21:35 hemebond not actual engines
21:35 hemebond Because engines are a thing
21:35 tapoxi right
21:35 tapoxi but they're implemented as engines
21:35 tapoxi right
21:36 hemebond Mmm, no engines are things like the SQS engine
21:36 tapoxi I thought it was any long-running process that performed events on the bus
21:36 hemebond reactors are just scripts run when an event occurs
21:36 hemebond oh, no
21:36 tapoxi right, but the reactor and thorium itself
21:37 tapoxi not things I write for reactor and thorium
21:37 hemebond Engines are long-running processes.
21:38 hemebond Reactors are literally just scripts that are run when an event occurs.
21:58 iggy but reactors (the thing that watches events and runs the sls files) are implemented internally as engines now
21:58 hemebond Is there more than one reactor engine?
22:02 iggy no clue (what you are talking about or the answer)
22:02 hemebond Well, to be a reactor is just a script/sls referenced by a line in the master config
22:02 hemebond *to me
22:03 pipps99 joined #salt
22:03 hemebond Whereas an engine is something like the SQS engine, a long-running process that polls or processes things.
22:03 iggy yes, but the bit of code that does the waiting for events and running sls files is an engine
22:03 pipps_ joined #salt
22:04 hemebond Right, just one engine.
22:04 hemebond That's what I was confused about.
22:04 iggy it used to be built into the master process
22:04 edrocks joined #salt
22:05 patarr joined #salt
22:07 Elsmorian joined #salt
22:07 Elsmorian joined #salt
22:08 Elsmorian joined #salt
22:09 Elsmorian joined #salt
22:10 Elsmorian joined #salt
22:11 RandyT joined #salt
22:18 pipps joined #salt
22:19 jas02 joined #salt
22:21 RandyT joined #salt
22:27 akhter joined #salt
22:29 abednarik joined #salt
22:35 patarr joined #salt
22:35 sh123124213 joined #salt
22:36 abonilla joined #salt
22:43 pipps joined #salt
22:45 babilen matth: You have to pass it in as pillar data (kward of state.apply) and you'd access it with pillar.get then
22:46 tiaz joined #salt
22:46 tiaz joined #salt
22:48 matth babilen: yes I got it working.
22:50 matth as the reactor can not access the pillar/grains, is there a way, based on the ID to get the grains for a specific ID/host ?
22:51 iggy write it as a state and target that state at the ID
22:53 pipps joined #salt
22:54 matth iggy: yes basically I provide the id/name of my instance from the reactor to a states, but from that states, as I have the ID of the minion, can I access his pillar/grains ?
22:54 iggy if you run the state on the minion, it has access to it's own pillar/grain data
22:54 matth that's the thing, I run it on the master.
22:54 iggy why?
22:55 matth it's a runner
22:55 iggy a runner can can call LocalClient on the minion to pull grains data? but that sounds like it's getting kind of dirty
22:58 matth iggy: I need to check if the state that execute the master ( via the runner ) from which region is located the minion.
22:58 matth this is located in the pillar of the minion.
22:59 iggy is this inside a runner or a state that's run on the master from a runner?
22:59 matth inside a states running on the master, from the runner
23:00 iggy publish or mine is your best bet probably
23:00 akhter_1 joined #salt
23:00 Klas joined #salt
23:01 matth basically what I do in this state is to refresh the pillar and the grains of the new minions. And trying to get the region :)
23:02 iggy then mine isn't going to work...
23:02 pipps joined #salt
23:04 iggy {% set region = salt['publish.publish'](minion_id, 'grains.get', 'region') %}
23:04 iggy but that's probably super fragile
23:04 iggy it gets rendered at the start of the state so if you are updating pillar/grains in that same state, that variable won't be right
23:05 matth iggy: it's ok I can do an other way.
23:06 iggy you should likely be using the reactor and custom events that are fired at each stage of whatever it is you are doing
23:09 matth iggy: can I set a reactor for a specific env ?
23:10 iggy that I don't know... I avoid salt environments like the plague
23:13 matth iggy: I found, actually when instanciate a new VM, there is a variable "profile", I will just pass it to with my runner to my master and I will be able to apply the proper states.
23:16 greyeax joined #salt
23:22 justanotheruser joined #salt
23:22 ws2k3 joined #salt
23:23 mosen joined #salt
23:25 mosen_ joined #salt
23:27 sp0097 joined #salt
23:34 sp0097 joined #salt
23:36 mikecmpbll joined #salt
23:41 akhter joined #salt
23:43 IgorK__ joined #salt
23:43 IgorK__ Hello, can anybody help with error in saltstack environments?
23:44 Renich_ joined #salt
23:47 IgorK__ here is [ERROR   ] Data passed to highstate outputter is not a valid highstate return: {'minion1': ["No matching salt environment for environment 'dev' found", "No matching sls found for 'dev.backend' in env 'dev'"]}
23:49 pipps99 joined #salt
23:52 IgorK__ shit
23:53 LostSoul joined #salt
23:56 rem5 joined #salt
23:57 akhter joined #salt
23:58 rem5 joined #salt
23:59 hemebond IgorK__: Figured it out?

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