Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2015-08-06

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

All times shown according to UTC.

Time Nick Message
00:01 aw110f joined #salt
00:04 yomilk joined #salt
00:04 womble forrest: It looks like test-kitchen solves the part of the problem I've already got sorted(ish): we've already got a CI env for dev.  I'm more looking for the tools that help to specify the tests.  The Testkitchen docs reference bats, though, which I hadn't seen, so that's a good start.
00:05 VR-Jack2 forrest: something missing perhaps from cmd.run that the wrapper script needs?
00:05 VR-Jack2 pathing, environment or something
00:05 forrest VR-Jack2: I don't think so, I'm running the exact same command and it works fine, and has worked fine in the past.
00:05 forrest I mean you can actually copy the command and run it and it's fine
00:05 forrest I don't see anything in debug
00:05 forrest but it's time to be done for the day, so FUCK IT
00:05 forrest womble: Glad that helped.
00:06 VR-Jack2 that's why I was curious about pathing and environment, as running it manually would have those possible extras
00:07 VR-Jack2 as I understand it, postgresql is just a wrapper, and after it switches into the pg user, it executes initdb
00:08 druonysuse joined #salt
00:14 nzero joined #salt
00:14 otter768 joined #salt
00:22 sunkist joined #salt
00:25 pipps joined #salt
00:31 Outlander joined #salt
00:32 omegamike joined #salt
00:32 Outlander hi, my salt master is getting random returns for a set of minions, sometimes all of them, sometimes only a few will show up. What's going wrong
00:32 Outlander when i run commands against a set of minions
00:34 Diaoul joined #salt
00:36 sunkist joined #salt
00:39 gladiatr joined #salt
00:40 sunkist joined #salt
00:52 dyasny joined #salt
00:58 napsterX joined #salt
01:00 micko joined #salt
01:00 zwi joined #salt
01:02 hasues joined #salt
01:05 hasues left #salt
01:06 aw110f joined #salt
01:10 hacfi joined #salt
01:24 aw110f joined #salt
01:27 dthom91 joined #salt
01:32 dendazen joined #salt
01:36 aw110f joined #salt
01:45 nzero joined #salt
01:47 ilbot3 joined #salt
01:47 Topic for #salt is now Welcome to #salt | 2015.5.2 is the latest | Please use https://gist.github.com for code, don't paste directly into the channel | Please be patient when asking questions as we are volunteers and may not have immediate answers | Channel logs are available at http://irclog.perlgeek.de/salt/
01:53 murrdoc joined #salt
01:56 evle joined #salt
01:59 napsterX joined #salt
02:03 blu_ joined #salt
02:05 druonysus joined #salt
02:10 dangra joined #salt
02:11 Singularo joined #salt
02:15 digitalhero joined #salt
02:15 otter768 joined #salt
02:16 pppingme joined #salt
02:17 writtenoff joined #salt
02:18 MatthewsFace joined #salt
02:20 omegamike joined #salt
02:20 xDamox joined #salt
02:26 forrest joined #salt
02:28 Aidin1 joined #salt
02:32 murrdoc joined #salt
02:33 scbunn joined #salt
02:34 Aidin joined #salt
02:37 ITChap joined #salt
02:37 napsterX joined #salt
02:39 forrest joined #salt
02:39 ITChap Hi everyone
02:39 ITChap when using multiple environments
02:39 murrdoc joined #salt
02:39 ITChap do you recommend to use different branches for the pillar ?
02:40 ITChap I use branches for my states and it make sense
02:40 ITChap but for pillar it feels strange
02:42 phx it's not when you are using pillars from states, and depending on the environment it might need different result
02:43 phx like you are adjusting a state in dev, and you need to add some stuff to a pillar for it
02:43 phx if you don't have a dev env for pillars, that might get tricky
02:46 ITChap but the thing is, I can merge one of my state branch with another one after testing, when everything works
02:46 ITChap but not pillars
02:46 ITChap pillars just share the structure
02:47 spark_ joined #salt
02:51 DammitJim joined #salt
02:53 scbunn joined #salt
02:53 dthom91 joined #salt
02:56 jasonrm joined #salt
02:58 dthom911 joined #salt
03:00 yomilk joined #salt
03:07 oyermolenko joined #salt
03:10 jmreicha joined #salt
03:11 favadi joined #salt
03:13 quasiben joined #salt
03:23 dthom91 joined #salt
03:24 rhodgin joined #salt
03:27 digitalh_ joined #salt
03:29 sakaYK joined #salt
03:44 seb` joined #salt
03:47 DammitJim joined #salt
03:48 oyermolenko joined #salt
03:53 oyermolenko joined #salt
04:08 omegamike joined #salt
04:12 oyermolenko joined #salt
04:16 otter768 joined #salt
04:20 sakaYK joined #salt
04:23 clintberry joined #salt
04:25 CeBe joined #salt
04:26 scbunn joined #salt
04:26 sakaYK joined #salt
04:27 evle joined #salt
04:28 ageorgop joined #salt
04:31 clintber_ joined #salt
04:34 dthom911 joined #salt
04:35 subsignal joined #salt
04:42 seena joined #salt
04:43 seena hi Guys
04:44 seena I am trying to create an environment and getting some odd error
04:44 seena file_roots:   base:     - /salt/srv/salt   dns:      /salt/srv/salt/dns
04:44 seena this is my salt master file
04:45 seena [root@169 salt]# ll  /salt/srv/salt/dns total 4 -rw-r--r-- 1 root root 106 Aug  6 04:37 dns.sls [root@169 salt]#
04:45 seena i have dns.sls file placed in that directory
04:45 seena dns:  'role:dns':   - match: grain   - dns
04:45 seena this is my top file
04:46 seena But when I running highstate
04:46 seena No matching sls found for 'dns' in env 'dns'
04:46 seena I am getting no sls found
04:46 seena even if there is sls file
04:46 seena any help?
05:06 dopesong joined #salt
05:08 golodhrim|work joined #salt
05:08 golodhrim|work left #salt
05:13 Fiber^ joined #salt
05:18 digitalhero joined #salt
05:22 bitc01d joined #salt
05:22 catpigger joined #salt
05:23 oyermolenko joined #salt
05:32 napsterX joined #salt
05:33 rdas joined #salt
05:39 mosen joined #salt
05:46 forrest joined #salt
05:47 forrest_ joined #salt
05:47 xDamox joined #salt
05:48 berserkir joined #salt
05:53 jeddi joined #salt
05:59 berserkir joined #salt
06:05 jaybocc2 joined #salt
06:07 lazybear joined #salt
06:07 dopesong joined #salt
06:08 rdas joined #salt
06:09 malinoff joined #salt
06:09 lb1a joined #salt
06:13 MohShami joined #salt
06:13 sakaYK joined #salt
06:15 spark_ joined #salt
06:17 otter768 joined #salt
06:20 jaybocc2 joined #salt
06:21 sakaYK joined #salt
06:24 aparsons joined #salt
06:26 sakaYK joined #salt
06:27 sirex joined #salt
06:29 aparsons joined #salt
06:34 markm joined #salt
06:38 KermitTheFragger joined #salt
06:42 CeBe joined #salt
06:45 sakaYK joined #salt
06:51 AndreasLutro joined #salt
06:51 sakaYK joined #salt
06:56 jaybocc2 joined #salt
06:58 sakaYK joined #salt
07:04 sakaYK joined #salt
07:10 losh joined #salt
07:11 jaybocc2 joined #salt
07:18 aron_kexp joined #salt
07:19 eseyman joined #salt
07:23 kawa2014 joined #salt
07:24 sakaYK joined #salt
07:30 forrest joined #salt
07:30 sakaYK joined #salt
07:32 Brew joined #salt
07:32 BlackBishop joined #salt
07:32 CeBe joined #salt
07:33 BlackBishop when testing a state "salt '*' state.sls repos test=True" .. is there something I can do to see what it would change and what/from/to/who/when ..
07:33 AndreasLutro BlackBishop: no
07:34 markm_ joined #salt
07:35 BlackBishop :| how can I be sure it's doing the right thing ?
07:36 AndreasLutro test it properly in a vm or staging environment before deploying it to production
07:37 sakaYK joined #salt
07:38 subsignal joined #salt
07:43 sakaYK_ joined #salt
07:45 stoogenmeyer__ joined #salt
07:46 omegamike joined #salt
07:48 krymzon joined #salt
07:49 gnord joined #salt
07:49 sakaYK joined #salt
07:50 _JZ_ joined #salt
07:54 sakaYK joined #salt
07:57 kawa2014 joined #salt
07:57 sakaYK joined #salt
08:00 sakaYK joined #salt
08:01 ITChap joined #salt
08:02 jaybocc2 joined #salt
08:05 sakaYK joined #salt
08:06 jhauser joined #salt
08:09 ttrumm joined #salt
08:14 BlackBishop left #salt
08:15 digitalhero joined #salt
08:15 spark_ joined #salt
08:16 digitalh_ joined #salt
08:16 vincehu joined #salt
08:16 slav0nic joined #salt
08:18 jaybocc2 joined #salt
08:18 otter768 joined #salt
08:22 che-arne joined #salt
08:22 shantanoo joined #salt
08:22 shantanoo hi all
08:22 N-Mi joined #salt
08:24 shantanoo is it ok to have require and watch_in the same state in another state. e.g. i have ''jenkins-service-restart' state and 'jenkins-install-plugins' state. for 'jenkins-install-plugins', 'jenkins-service-restart' is required (require), and once plugin is installed, need to run 'jenkins-service-restart' (watch_in).
08:24 lz-dylan joined #salt
08:27 markm joined #salt
08:39 subsignal joined #salt
08:43 jhauser_ joined #salt
08:44 s_kunk joined #salt
08:44 GnuLxUsr joined #salt
08:44 babilen shantanoo: Sure
08:45 babilen Could you paste your states to, say, http://refheap.com ? It is probably easier to comment on them if we can see them in toto.
08:45 shantanoo babilen: it won't give me 'Recursive requisite found' error. right?
08:45 babilen Only if you define a recursive requisite :)
08:45 babilen The dependency graph has to be acyclic
08:46 shantanoo this won't set the cyclic graph. right?
08:47 babilen I can't answer it without having seen the states in their entirety.
08:47 babilen (and it might be easier to simply test it than trying to analyse it)
08:49 GnuLxUsr joined #salt
08:50 shantanoo babilen: hold on
08:56 shantanoo babilen: https://gist.github.com/shantanoo/a8a25ef1e9a5f736d2cd
08:56 denys joined #salt
08:56 gcfhvjbkn joined #salt
08:57 babilen That won't work and is, indeed, recursive
08:57 babilen What are you trying to achieve?
08:58 babilen Are you aware of https://github.com/saltstack-formulas/jenkins-formula/ ?
08:58 shantanoo i am trying to install jenkins plugins using cli. to do this, jenkins service should be running, then using cli plugin is installed, and after that, i need to restart jenkins service.
08:59 babilen First: Restart or reload?
08:59 shantanoo using jenkins-formula to install jenkins
09:00 shantanoo but it doesn't contain installation of plugins
09:00 babilen The jenkins formula already defines service.running for you: https://github.com/saltstack-formulas/jenkins-formula/blob/master/jenkins/init.sls#L43
09:00 aw110f joined #salt
09:00 babilen I would develop against the formula and submit a PR once you have integrated the plugin installation
09:02 shantanoo somehow it is not working for me. when i run it again using "salt '...' state.sls jenkins" it is working. same code through top.sls is not working.
09:02 babilen And why don't you use "require: - service: jenkins" and "- watch_in: - service: jenkins" ?
09:02 babilen You would have to import "- jenkins"
09:03 babilen Also please define "not working"
09:03 shantanoo ok. will do require and watch_in
09:03 shantanoo not working = jenkins is installed, but service doesn't start
09:03 babilen A "- listen_in" might be even better in this case, btw
09:04 babilen Not entirely sure if salt figures out the "should be running and has to be restarted afterwards"-bit though
09:06 shantanoo babilen: thanks. let me try with listen_in.
09:07 shantanoo but i still need require to make sure jenkins is running. right?
09:10 babilen Indeed
09:11 shantanoo thanks.
09:13 Ztyx joined #salt
09:13 Ztyx left #salt
09:14 sakaYK joined #salt
09:31 denys joined #salt
09:31 rdas joined #salt
09:36 denys joined #salt
09:36 Singularo joined #salt
09:53 jaybocc2 joined #salt
09:59 katyucha joined #salt
10:02 vstoniest joined #salt
10:06 indispeq joined #salt
10:08 ]V[ joined #salt
10:08 mongoose123 joined #salt
10:08 jaybocc2 joined #salt
10:12 mongoose123 Morning. Attempting to generate and exchange ssh keys between two servers. The only sensible-ish way I can see is to capture the content of id-rsa.pub with mine and then append it into authorized_keys on the other server. Not having much luck, so simplified with this test: http://p.rig.gr/amefozuruj.sm
10:12 mongoose123 Get the feeling I'm using mine incorrectly
10:13 mongoose123 Any help appreciated
10:16 yomilk joined #salt
10:18 otter768 joined #salt
10:19 CeBe joined #salt
10:21 sakaYK_ joined #salt
10:24 Outlander joined #salt
10:31 jhauser are pillars defined in different pillar_roots supposed to be only visible in different environments
10:31 jhauser I can’t get this to work
10:32 jhauser on the minion I get always all pillar data regardless of the pillar_root where it is defined
10:41 subsignal joined #salt
10:44 jaybocc2 joined #salt
10:47 evle joined #salt
10:49 yomilk joined #salt
10:50 omegamike joined #salt
11:00 jaybocc2 joined #salt
11:02 bluenemo joined #salt
11:03 calvinh joined #salt
11:08 amcorreia joined #salt
11:14 Outlande1 joined #salt
11:20 daemonkeeper So with gitfs as a file server backend, salt pulls environments from branches. Can I override this somehow? I'd like to have two gitfs remotes, each pulling master but put them into different environments. Is that possible, or can I only achieve this over branches with gitfs?
11:21 scottpgallagher joined #salt
11:22 daemonkeeper i.e. basically something like I can do with file_foots
11:22 daemonkeeper *file_roots
11:26 ninkotech joined #salt
11:27 digitalhero joined #salt
11:27 saltuser joined #salt
11:29 shantanoo joined #salt
11:34 stoogenmeyer__ hey, i started encountering an error with salt-cloud when creating a minion using 'sudo salt-cloud -p <profile> <id>'. The very end salt-cloud runs daemons_running() and complains that salt-minion was not found running, and it says there's a profile error. Same commands worked a few days ago, any ideas?
11:35 jaybocc2 joined #salt
11:38 stoogenmeyer_ joined #salt
11:41 gcfhvjbkn joined #salt
11:41 subsignal joined #salt
11:42 fish_ joined #salt
11:47 CedNantes joined #salt
11:47 CedNantes Hello
11:48 CedNantes I've a quick question : how do i set multiple file_roots in master conf when using gitfs ? Do i have to create new branches ?
11:51 jaybocc2 joined #salt
11:51 stoogenmeyer_ so salt-cloud starts a minion but the minion won't start.. and when I ssh to the minion and try to start it myself it shows that in the config file master_type: str which is invalid for some reason
11:54 stoogenmeyer_ ok so i basically need to update salt-cloud, do I just do apt-get update salt-cloud ?
11:54 stoogenmeyer_ ill try that (-:
11:56 BlackBishop joined #salt
11:58 fredvd joined #salt
11:58 BlackBishop Any easy way to make that part of the state only be done if there is an epel repo to be changed ? "Comment: Failed to configure repo 'epel': The repo does not exist and needs to be created, but a name was not given"
11:58 BlackBishop something like a condition
11:59 yomilk joined #salt
12:01 mschiff using "- watch:\n  - file: /some/file" will trigger this state only when salt did somwthing with /some/file. How can I make salt execute this state whenever the file changes on disk?
12:04 shantanoo i want a state, which will hit given URL and retry 'r' times with interval 'i' seconds until it gets HTTP 200. Fails if unable to get HTTP 200 after attempting 'r' retries.
12:05 terinjokes joined #salt
12:05 numkem joined #salt
12:05 shantanoo babilen: even though i start the jenkins, server takes time to come up. if i try to install plugins using CLI, it fails.
12:09 snaggleb joined #salt
12:11 harkx joined #salt
12:12 mapu joined #salt
12:14 lrojas joined #salt
12:17 pdayton joined #salt
12:19 ljt joined #salt
12:19 otter768 joined #salt
12:22 quasiben joined #salt
12:26 jaybocc2 joined #salt
12:29 Furao joined #salt
12:31 FRANK_T joined #salt
12:32 shiriru joined #salt
12:32 rideh joined #salt
12:32 _mel_ joined #salt
12:33 FRANK_T I have this on my top file https://www.refheap.com/5a028a0851533462f01035444
12:34 teryx510 joined #salt
12:34 FRANK_T I am getting this error https://www.refheap.com/2f0280f53ecc463c0888e38fd
12:35 napsterX joined #salt
12:39 omegamike joined #salt
12:39 dthom911 joined #salt
12:42 kawa2014 joined #salt
12:42 jaybocc2 joined #salt
12:42 zwi joined #salt
12:42 subsignal joined #salt
12:49 ze- hey. Is there a way to add ignored comment/description to states?
12:51 gfa joined #salt
12:53 rhodgin joined #salt
12:56 subsignal joined #salt
12:58 che-arne joined #salt
12:58 pdayton joined #salt
13:00 AndreasLutro ze-: # denotes a yaml commment
13:03 dyasny joined #salt
13:05 fivmo joined #salt
13:05 fivmo joined #salt
13:05 AndreasLutro so if you start a line with # it won't be parsed
13:06 fivmo left #salt
13:06 amcorreia_ joined #salt
13:06 DammitJim joined #salt
13:09 yomilk joined #salt
13:09 babilen FRANK_T: Did you define the "apollo" nodegroup?
13:09 babilen (and your indentation is off)
13:10 Aidin joined #salt
13:11 hacfi joined #salt
13:12 dthom91 joined #salt
13:13 amcorreia__ joined #salt
13:15 FRANK_T yes
13:17 jaybocc2 joined #salt
13:17 FRANK_T https://www.refheap.com/f7d9b09425f6325217101b1db
13:18 subsignal joined #salt
13:20 racooper joined #salt
13:22 FRANK_T where?
13:23 AndreasLutro you can't see the misaligned dashes?
13:23 jdesilet joined #salt
13:26 FRANK_T https://www.refheap.com/0258cebb4f5803ef740d12058
13:26 Tecnico1931 joined #salt
13:26 FRANK_T same error :(
13:28 AndreasLutro FRANK_T: try running the highstate with -l debug and share the output
13:28 AndreasLutro my best guess is you misplaced the top.sls
13:29 gcfhvjbkn joined #salt
13:30 bhosmer joined #salt
13:30 FRANK_T here
13:30 FRANK_T https://www.refheap.com/202f9cd0891016773f83b6ea3
13:32 FRANK_T and I am using salt 2015.5.2 (Lithium)
13:32 FRANK_T In both master and minion
13:33 gcfhvjbkn joined #salt
13:33 jaybocc2 joined #salt
13:34 FRANK_T If I use the group to execute .sls it will work
13:34 FRANK_T the problem is with the highstate
13:34 AndreasLutro indeed
13:34 AndreasLutro where is your top.sls located?
13:34 che-arne joined #salt
13:34 FRANK_T .
13:34 FRANK_T ./srv/salt
13:35 AndreasLutro why not /srv/salt/top.sls ?
13:35 FRANK_T sorry
13:36 FRANK_T top is inside /srv/salt/
13:36 FRANK_T so it is /srv/salt/top.sls
13:36 AndreasLutro alright, ./ means the current working directory so that puzzled me a bit :)
13:36 hasues joined #salt
13:36 hasues left #salt
13:37 FRANK_T sorry.
13:37 AndreasLutro well, I can't see what the problem is, sorry
13:37 FRANK_T okay thank you.
13:37 AndreasLutro maybe check the minion's log as well
13:37 FRANK_T ok
13:37 anotherZero joined #salt
13:39 digitalhero joined #salt
13:39 FRANK_T AndresLuto
13:40 FRANK_T Here the error inside de minion
13:40 FRANK_T [ERROR   ] The Salt Master has cached the public key for this node, this salt minion will wait for 10 seconds before attempting to re-authenticate
13:40 dendazen joined #salt
13:46 FRANK_T It was a key problem
13:46 FRANK_T thank you.
13:47 Tyrm joined #salt
13:48 Tyrm joined #salt
13:48 andrew_v joined #salt
13:52 circ-user-qoxEj joined #salt
13:53 ksj if I have custom python modules for my state files in /srv/_modules, and they import other python modules, is there a way for the imported python modules to use salt dunder variables like __pillar__?
13:53 mpanetta joined #salt
13:54 ksj e.g. in the _modules/mymodule.py, it has "import anothermodule". however, anothermodule cannot see __pillar__ and salt fails on trying to execute functions from that module
13:54 cpowell joined #salt
13:55 circ-user-qoxEj in a pillar file, can I define a variable 'var1 : hello' and within the same pillar file reference to it using 'var2: {{ pillar['var1'] }}'?
13:57 jberends in a pillar file, can I define a variable 'var1 : hello' and within the same pillar file reference to it using 'var2: {{ pillar['var1'] }}'? (repost using proper nick)
13:58 wnkz joined #salt
13:59 di5dude joined #salt
13:59 dopesong_ joined #salt
13:59 babilen no
14:00 babilen jberends: Take a look at https://github.com/bbinet/pillarstack
14:03 jberends babilen: check. the first pillar file needs to be fully read before these variable can be combined. I saw something similar in the postgress saltstack formula using a map.jinja that mergers a settings with defaults.
14:04 jberends babilen: it seems that pillarstack uses the same construct
14:06 deltreey joined #salt
14:07 kaptk2 joined #salt
14:08 babilen jberends: Yeah, hence the "no", it's just that pillarstack might help in, well, stacking pillars with references
14:08 babilen (not sure what you are after, but it might help)
14:09 jberends bablien: thanks indeed wanted to be sure what the cause was of the failure in the compilation of the pillar. Now I know how to solve. Thanks.
14:10 jaybocc2 joined #salt
14:13 dangra joined #salt
14:13 AndreasLutro jberends: maybe this is good enough for your use case https://www.reddit.com/r/saltstack/comments/3edbc3/pillar_using_pillar_values_in_other_pillar_values/cteb435
14:13 saltstackbot [REDDIT] pillar: using pillar values in other pillar values. is it possible? (self.saltstack) | 6 points (100.0%) | 5 comments | Posted by SpaceJesusOnAStick | Created at 2015-07-23 - 21:10:13
14:13 babilen Yeah, that is naturally an option
14:13 babilen (which uses jinja variables)
14:15 VR-Jack joined #salt
14:15 jberends AndreasLutro, babilen: check. Did not know that this was possible.
14:17 babilen http://jinja.pocoo.org/docs/dev/templates/#assignments
14:19 ljt joined #salt
14:19 dthom91 joined #salt
14:20 otter768 joined #salt
14:21 jberends check solved it. used: {% set dir = '/webapp' }; base_dir: {{dir}}; appdir : {{dir}}/foobar
14:21 dthom91 joined #salt
14:21 jberends resolved correctly to base_dir : /webapp and appdir: /webapp/foobar
14:22 jberends Thanks for the help babilen and AndreasLutro
14:22 spark_ joined #salt
14:22 saltsa joined #salt
14:24 shantanoo joined #salt
14:26 spark__ joined #salt
14:33 ingslovak joined #salt
14:43 murrdoc joined #salt
14:44 digitalhero joined #salt
14:45 TyrfingMjolnir joined #salt
14:45 domel joined #salt
14:46 armguy joined #salt
14:51 favadi joined #salt
14:57 armguy joined #salt
14:59 slav0nic joined #salt
15:03 cachedout joined #salt
15:03 yomilk joined #salt
15:07 zwi joined #salt
15:07 dthom91 joined #salt
15:10 shantanoo joined #salt
15:10 shantanoo joined #salt
15:11 deltreey joined #salt
15:11 CeBe1 joined #salt
15:11 slav0nic_ joined #salt
15:14 green_ joined #salt
15:14 linjan_ joined #salt
15:15 sdm24 joined #salt
15:24 jondonas joined #salt
15:25 joe___ joined #salt
15:25 Brew joined #salt
15:25 scbunn joined #salt
15:28 tomspur joined #salt
15:29 MatthewsFace joined #salt
15:34 Heartsbane joined #salt
15:34 Heartsbane joined #salt
15:40 holyzhou joined #salt
15:40 viq joined #salt
15:42 holyzhou one question.  if i don't want to use cmd.run ,how can i move a dirctory to another dirctory , the file state can do this ?
15:43 deltreey btw, I am going over the docs to learn saltstack, and when I downloaded the demo environment and ran `vagrant up` it crashes.  My mac is in all kinds of bad shape (running 10gb of ram on 3 other VMs) so it's not the perfect test environment, but I can post the logs if you want
15:45 VR-Jack is it safe to just delete a key from the /etc/salt/pki/master/minions_denied without leaving something floating somewhere else?
15:46 sdm24 did you already do salt-key -d 'keyname'?
15:47 sdm24 because that should do it automatically
15:47 VR-Jack sdm24: name references 2 keys
15:47 VR-Jack don't want to delete the accepted one
15:47 DammitJim joined #salt
15:47 iggy best to use salt-key for stuff as deleting files leaves things hanging in the cachedir
15:48 VR-Jack k. guess until they fix salt-key, I'll just delete both and readd
15:48 sdm24 ah, then yes I believe thats only where key data live
15:49 ageorgop joined #salt
15:49 holyzhou hi guys , if i don't want to use cmd.run ,how can i move a dirctory to another dirctory , the file state can do this ?
15:49 sdm24 I believe file.recurse is what you want
15:50 sdm24 or file.copy
15:50 lloesche joined #salt
15:51 iggy yeah, depends whether you want to copy files from master or from one place to another on a system
15:51 sdm24 or maybe even a file.rename and rename to a different location
15:52 holyzhou from one place to another on a system
15:52 zer0def joined #salt
15:53 spark_ joined #salt
15:53 sdm24 I'm not sure if this will work, but maybe like a file.rename:  -name: /path/to/destination  - source: /original/path/file
15:54 holyzhou ok , i check this
15:54 dendazen_ joined #salt
15:55 iggy I'm thinking you'll have to cmd.run it or module.run rsync.rsync/file.copy (w/ recurse)
15:57 xDamox joined #salt
15:58 zsoftich1 joined #salt
15:58 dthom91 joined #salt
15:59 sdm24 file.rename seems to work with both files and directories
15:59 pdayton joined #salt
16:02 writtenoff joined #salt
16:06 theologian joined #salt
16:06 troyready joined #salt
16:09 Lionel_Debroux joined #salt
16:10 VR-Jack ewww. looks like 2015.5.x broke salt.wheel.key in orchestrate. :(
16:11 holyzhou yes , but must add mkdirs and force flag ,otherwise it can't be true
16:12 mejiamariano joined #salt
16:12 sdm24 gotcha
16:12 mejiamariano left #salt
16:14 holyzhou thank you ~
16:14 sdm24 no problem, I wasn't sure if it would've worked
16:15 Bryson joined #salt
16:16 tiadobatima joined #salt
16:19 UtahDave joined #salt
16:19 bhosmer joined #salt
16:20 sunkist I'm using salt to configure redmine deployments...  trusty/postgres/redmine3.0.4.
16:21 otter768 joined #salt
16:21 sunkist I can create the database with the postgres_database and postgres_user state modules.
16:22 sunkist But I'd also like to create a backup user with read-only permissions to the database.
16:22 sunkist To create the user, I added another postgres_user state.
16:22 sunkist But I can't figure out what the proper salt-method is to grant it permissions.
16:25 capricorn_1 joined #salt
16:27 iggy I'd read up on postgres perms in general, they don't really act like certain other databases that may be more familiar to some
16:30 sunkist iggy, I know what queries to run, but I don't know how to get salt to run them saltishly.
16:31 ekristen joined #salt
16:31 ajw0100 joined #salt
16:33 kbyrne joined #salt
16:33 jaybocc2 joined #salt
16:33 sunkist iggy, doing a cmd.run on a `psql' command seems a little hackish.
16:34 iggy little bit, yeah
16:34 sunkist I looked at bechtoldt's postgres formula, it's nice but doesn't cover my use case.
16:34 * iggy barfs
16:34 sunkist porque?
16:35 sunkist is bechtoldt a dirty word?
16:35 sunkist What I would love to be able to do is a query.run...
16:36 sunkist - name: GRANT perms to USER
16:36 iggy module.run
16:36 sunkist - unless: USER has permissions already
16:36 sunkist Let me check the modules!
16:38 sunkist salt.modules.postres.psql_query looks like the one I need.
16:42 malinoff joined #salt
16:45 amcorreia joined #salt
16:45 bbbryson joined #salt
16:48 aparsons joined #salt
16:48 bbbryson_ joined #salt
16:51 mapu joined #salt
16:52 ageorgop joined #salt
16:53 dfinn joined #salt
16:53 dfinn can you do conditionals with a salt run from the master?  something like if pkg.version == $bad_version then run $upgrade_command?
16:54 iggy state.single
16:54 iggy not really
16:54 iggy for packages, you'd have to get that package version info into grains and target that and
16:55 dfinn ok
16:55 iggy or write out a state that can do it and just state.sls it
16:55 alemeno22 joined #salt
16:58 saltsa joined #salt
16:58 Bryson joined #salt
16:59 theologian joined #salt
16:59 kawa2014 joined #salt
17:00 pipps joined #salt
17:00 spark_ joined #salt
17:00 jalbretsen joined #salt
17:00 bbbryson joined #salt
17:01 gnudon joined #salt
17:03 cpowell joined #salt
17:08 jhauser joined #salt
17:09 bhosmer joined #salt
17:13 murrdoc anyone here change the ulimits on salt-masters
17:13 iggy yes
17:13 murrdoc i am trying to put ulimit -Hn 64000
17:13 murrdoc ulimit -Sn 32000 into /etc/default/salt-master
17:13 dfinn i think it's kind of mandatory
17:13 murrdoc but it fails cos of dash
17:13 murrdoc sup foo
17:14 iggy I just have it in /etc/security/limits.d/99-salt.conf (and the default init script has a ulimit call)
17:15 iggy or it doesn't (anymore?)
17:16 murrdoc limit nofile 100000 100000
17:16 murrdoc but if u restart salt-master (it seems) from shell
17:16 murrdoc it doesnt read in the 99-salt.conf
17:18 iggy yeah, I could swear there were ulimit calls in the init script
17:18 iggy but just double checked and no
17:19 * iggy got nothing apparently
17:19 VR-Jack ahhh. evil of evil. having to do cmd.run salt-key -y -d {{ minionid }}
17:20 kawa2014 joined #salt
17:23 dfinn left #salt
17:23 pipps joined #salt
17:23 impi joined #salt
17:24 sdm24 if it works, it works
17:24 forrest joined #salt
17:26 CeBe1 joined #salt
17:28 VR-Jack yeah. annoyed that salt.wheel broke, though
17:28 shantanoo joined #salt
17:28 shantanoo joined #salt
17:31 VR-Jack not annoyed enough to fix it, though. lol
17:32 oznah I have 20 minions, of which 3 are test machines. There is zero standard on host naming. what's the best way to isolate prod minions (17) from test minions (3)?
17:34 mindscratch joined #salt
17:35 sdm24 a manual pillar list
17:35 sdm24 'test1,minion15,anothername': - match: list
17:36 mindscratch When issuing a command via the command line, is there a way I can pass a bash variable into the command? For example: needle="foobar"; salt 'myhosts*' cmd.run 'ps -eaf | grep $needle'
17:36 oznah ok
17:37 phx mindscratch, it's basic shell handling. whatever inside '' is not getting expanded, while "" gets expanded
17:37 mindscratch ok i'll try that
17:37 VR-Jack I just setup minionid as hostname-type-pri-group
17:38 VR-Jack That being said, I've got a wicket top.sls that auto-builds lists based on pillar data. iggy absolutely hates it.
17:39 iggy true story
17:40 oznah why would you do this in pillar vs node groups?
17:41 Fiber^ joined #salt
17:41 Ahlee node groups require master restarts, for starters
17:41 VR-Jack oznah: because I have jinja all over the place that checks on node type, priority, and hgroup. So it as well as the lists used in top.sls all use the same pillar data
17:42 oznah VR-Jack: that makes sense
17:42 VR-Jack For convention, I usually have those details in the node name, but it's not required because of my top.sls jinja
17:43 VR-Jack that's important, because if I add a new qualifier, I can easily build a new lists based on it.
17:43 breakingmatter joined #salt
17:44 spark_ joined #salt
17:44 breakingmatter Is there a way to restrict salt states so they can't run on certain minions?
17:44 surge__ joined #salt
17:45 digitalhero joined #salt
17:45 sdm24 {% if grains['id'] != 'yourminion
17:45 pipps joined #salt
17:45 sdm24 ' %}
17:45 iggy lol, that was going to be my suggestion
17:45 mindscratch phx: thanks, that was it
17:45 breakingmatter Is there a way to do it without wrapping the whole state in jinja? lol
17:45 sdm24 unless statements?
17:46 breakingmatter I was hoping for it to trip a failure on the state.
17:46 sdm24 you can do that too
17:46 breakingmatter i.e., "Failed: Grains do not match"
17:46 breakingmatter Can you have a require: grain?
17:46 iggy no
17:46 breakingmatter le sigh
17:46 sdm24 http://docs.saltstack.com/en/latest/ref/states/all/salt.states.test.html
17:47 murrdoc u can do a prereq state with the link sdm24 pointed out
17:47 sdm24 {% if id = 'thisguy' %} test.fail {% else %} your stuff
17:47 breakingmatter Oh that's perfect
17:47 breakingmatter I mean, not "perfect"
17:47 breakingmatter But it works lol
17:47 breakingmatter I'd love to be able to lock down states based on grains/nodegroups/pillar/etc
17:47 breakingmatter without jinja
17:48 sdm24 yeah I had to include that after I installed postfix client on our mail server X_X
17:48 sdm24 What is the saltiest way to print the current datetime?
17:48 breakingmatter Well, the first step would be to NOT run a state on a server you don't want it to run on. But good luck with that in production.
17:49 iggy sdm24: http://docs.saltstack.com/en/latest/ref/renderers/all/salt.renderers.jinja.html
17:49 sdm24 Yeah, it was part of the generic "linux" highstate group, since 99% of the minions will need postfix installed and configured
17:50 CeBe1 joined #salt
17:50 sdm24 thanks iggy
17:50 murrdoc breakingmatter:  simple fix
17:50 murrdoc write a 'validate' state
17:50 murrdoc put it as the top state in your top.sls for all matches
17:50 murrdoc and throw a failHard: true in it
17:52 breakingmatter murrdoc: But, I don't want it to fail hard if it matches a certain grain, I just don't want it to run it.
17:52 breakingmatter I want to make so if you run `salt '<target>' state.sls some.state` and it doesn't match a certain grain/group/etc it fails and tells you why it failed.
17:52 murrdoc u want it to exit the state run
17:53 murrdoc and nothing in sdm24 link helped u with that
17:53 murrdoc ?
17:53 breakingmatter For instance, we have some KVM nodes and some OpenVZ nodes. I want to make it so under NO circumstances can KVM states run on OVZ nodes and vice versa.
17:54 breakingmatter Jinja if statements would "work" (and that's what we're doing already, sort of), but I thought there might be some require: grain functionality somehwere.
17:54 digitalhero joined #salt
17:54 tmclaugh[work] joined #salt
17:54 iggy require: has one format.... <module>: <ID>
17:55 breakingmatter iggy: I didn't necessarily mean the `require:` flag itself, just some kind of require based on globbing/grain/etc
17:55 sdm24 onlyif?
17:55 JPaul I'm having an issue with file.symlink, as far as I can tell my syntax is correct, but I am getting "Directory /etc/php5/conf.d for symlink is not present" returned. Name I am passing it is "/etc/php5/conf.d/20-zendopcache.ini" Anyone have any ideas what might be wrong?
17:55 iggy top file?
17:55 breakingmatter iggy: I don't want to only secure the highstate though.
17:56 iggy then jinja is your only current option
17:56 breakingmatter iggy: I figured. I was just hoping there was something else out there.
17:56 breakingmatter Jinja with the salt.states.test just feels wrong
17:56 evidence anyone seen timeout issues on salt commands on 2015.5.3?  despite my timeout setting or trying with -t/--timeout, jobs never timeout against minions that are down
17:56 evidence didn't seen an issue on it
17:57 breakingmatter evidence: Timeouts work fine for me, fwiw
17:57 jaybocc2 evidence: timeout is working for us.
17:57 breakingmatter evidence:  Even with [No response] and [Not connected] errors.
17:57 evidence yeah, so after the bump from 2015.5.2 -> 2015.5.3, it never times out and displays either of those errors ^
17:58 evidence nothing changed except the version bump.. might be a bug related to my unique config then
17:58 jaybocc2 what does salt-run manage.status show you?
17:58 breakingmatter evidence: Did you update the minions and the master?
17:58 evidence salt-run manage.* commands never return, because they wait for full output/timeout, which never happens :p
17:59 evidence breakingmatter: yes all clients are now also 2015.5.3 also
17:59 jaybocc2 what is your timeout set to?
17:59 evidence 10 sec
17:59 evidence but like i said tried passing -t also with no affect
17:59 jaybocc2 anything in the logs?
17:59 * Heartsbane is sometimes surprised at the things he can do with salt. :-)
17:59 evidence nuked the master cache.. guess i'll start messing w settings :/
18:00 evidence jaybocc2: negative, logs ar eclean
18:00 jaybocc2 have you tried running in debug or verbose mode to see whats happening?
18:00 breakingmatter Anyways, thanks sdm24 and iggy and murrdoc for your help.
18:00 Gareth UtahDave: ping
18:00 sdm24 good luck breakingmatter
18:00 evidence jaybocc2: i see a massive mount of return from XXX like it should (and on commands such as test.version the output is properly shown)
18:01 evidence but then it just hangs there forever, it has to simply be a timeout bug
18:01 evidence continually spins on checking if JID is still running, which it is
18:01 jaybocc2 sounds like you might have found a bug. I'd submit your configs, logs and outputs in a bug.
18:01 jaybocc2 see if anyone can reproduce it with you
18:01 evidence yeah, going to poke around a bit, must be some interaction w my settings
18:02 Heartsbane whiteinge: UtahDave: we need to do that again, it was fun
18:03 jaybocc2 can you share a gist of your master and minion configs?
18:05 digitalhero joined #salt
18:05 xDamox joined #salt
18:06 evidence relevent settings http://pastebin.com/60QQYtna .. con_cache seems like a possible culprit
18:07 evidence if it will wait forever on minions it believes are connected and only times out on ones that is has a key for
18:07 evidence has a key for but aren't connected
18:09 UtahDave Gareth: pong!
18:09 UtahDave Heartsbane: totally!
18:12 sdm24 joined #salt
18:13 repl1cant are there any other methods other than filesystem for storing minion keys? like any kinda of db backend method?
18:14 evidence repl1cant: currently none i'm aware of.. lsycnd might fit in nicely/simply
18:15 mrwboilers joined #salt
18:16 mrwboilers Is it important to have the same version of minions throughout your environment? I have both CentOS and Ubuntu machines, and the versions of salt in their respective repositories are different.
18:17 tiadobatima joined #salt
18:17 iggy mrwboilers: not highly, but if you ask for help with a problem we're most likely going to tell you to get everything updated
18:18 troyready joined #salt
18:19 mrwboilers do most people install via bootstrap.saltstack.com as opposed to using public package repos?
18:19 dthom91 joined #salt
18:21 baweaver joined #salt
18:22 otter768 joined #salt
18:23 murrdoc nope
18:23 digitalhero joined #salt
18:23 iggy we do (from git using tagged versions)
18:23 krymzon joined #salt
18:24 iggy it has historically been easier for us to maintain versions we know are good that way
18:24 murrdoc we == I
18:24 murrdoc cos iggy is team of one
18:24 murrdoc SSSSC approved
18:24 iggy well, 1.5, but
18:24 dthom91 joined #salt
18:26 evidence mrwboilers: best bet is to just make sure no minion is newer than your master though
18:27 digitalhero joined #salt
18:27 mrwboilers Thanks.
18:28 mrwboilers When you say you get it from git, is that from an official saltstack git repo, or your own?
18:28 iggy the main repo
18:28 bhosmer joined #salt
18:29 iggy ./bootstrap.sh -U -P -p python-jinja2 -A {master} git v2015.5.3
18:29 evidence jaybocc2: lul.. setting the timeout in master.conf on 2015.5.3 causes it to be indefinite .. with some combination of settings i have at at least
18:29 iggy like so
18:29 evidence anyone else running on 2015.5.3 set timeout?
18:29 breakingmatter evidence: We have a five minute timeout
18:31 dcs_ joined #salt
18:36 gcfhvjbkn joined #salt
18:40 dthom91 joined #salt
18:41 baweaver joined #salt
18:44 funzo joined #salt
18:46 tolbeen joined #salt
18:48 tolbeen Hello, I am running into an issue reading the output of a highstate.  When I have an error like return: [ 'Pillar failed to render with the following messages:', 'Rendering SLS \'firewalld\' failed. Please see master log for details.' ], I am getting the value of success: true for the result, which seems to be inaccurate considering it failed.
18:49 jaybocc2 joined #salt
18:49 VooDooNOFX joined #salt
18:50 tolbeen Also there is no retcode available to determine whether the result failed or succeeded, only the message in the return.  Is this an omission or bug, or would it be possible to get the retcode for each return?
18:50 evidence hah.. it's the event that checks if the job is still running, it seems to reset the timeout.  if i set the timeout <= 12 it times out properly, and i see no check running jobs event on the bus
18:51 evidence so 13 on up i see the event fire every 10s and the timeout never happens
18:51 sdm24 how do I force a {% set %} variable to be a list?
18:52 sdm24 I want to make a list of down minions, but {% set list = salt['cmd.run']('salt-run manage.down') %} will give me a string
18:52 surge__ joined #salt
18:52 surge__ How can I force nginx to restart at the end of a salt state run ? In chef, this was easy because the default behavior was to notify —> restart, :delayed (meaning the end of a chef run). Salt…. not so easy, or obvious at least.
18:52 sdm24 so a {% for i in list %} {{ i }} {% endfor %} will give me each character
18:52 sdm24 service.restart, - order: last
18:53 surge__ sdm24: But won’t that restart nginx every time? I don’t want that. I want it to watch a file, but restart only if that file changes.
18:53 bhosmer joined #salt
18:54 surge__ The docs make it seem that it is an “always run” condition to use “first/last”
18:54 sdm24 http://docs.saltstack.com/en/latest/ref/states/requisites.html you probably want "watch", and watch it to a config file. If the config file changes, salt will restart nginx
18:55 _JZ_ joined #salt
18:55 surge__ But it will do that as soon as a change happens, the idea is to wait for everything else to complete before doing so.
18:55 sdm24 use order and watch?
18:56 iggy surge__: listen
18:56 druonysus joined #salt
18:56 druonysus joined #salt
18:57 surge__ sdm24: “Before using the order option, remember that the majority of state ordering should be done with a Requisite declaration, and that a requisite declaration will override an order option, so a state with order option should not require or required by other states.”
18:57 DammitJim how does one make a grain upper case or lowercase in a template?
18:57 iggy DammitJim: {{ grains.host|lower }}
18:57 iggy or |upper
18:57 iggy jinja filters
18:59 DammitJim nice! thanks!
18:59 surge__ iggy: ??
18:59 surge__ Wrong username?
19:01 iggy surge__: the functionality you are looking for is called listen (vs watch it does the action at the end of the run)
19:01 iggy http://docs.saltstack.com/en/latest/ref/states/requisites.html#listen-listen-in
19:02 surge__ Ah, perfect. This is it.
19:02 surge__ I must’ve read that page so many times, but glossed over that thinking it was the same exact thing as “Watch"
19:03 iggy you can thank ryan lane for that little bit of gold
19:04 DammitJim is there a way to undo a state?
19:04 forrest iggy: Have you run into any issues with cmd.run on the more recent releases? service postgresql-9.3 initdb doesn't work
19:04 forrest works fine manually, fails when run by salt
19:04 iggy forrest: yes, like all the stuff I had to patch aptly for?
19:05 forrest I probably spaced it, what did you have to do to make it work?
19:05 forrest iggy: Come on man, like I can remember all these PRs....
19:05 forrest ;)
19:05 iggy set environment variables (HOME)
19:05 dthom91 joined #salt
19:06 dendazen_ joined #salt
19:06 iggy I'm trying to find the issue
19:07 DammitJim what is the difference between {{ grains['fqdn'] }} and {{ grains.fqdn }}
19:07 iggy https://github.com/saltstack/salt/issues/19869
19:07 saltstackbot [#19869]title: Salt don't seem to set $HOME when issuing commands | Hi,...
19:07 iggy DammitJim: not much
19:07 DammitJim the former looks more python-like, I guess?
19:07 DammitJim nah, not really
19:08 Mate i suppose it's for compatibility with django's templating language
19:08 breakingmatter DammitJim: Personally, the shorthand dictionary style is very unpythonic and not intuitive at all.
19:09 breakingmatter I guess it makes sense though if you don't know python.
19:09 mapu joined #salt
19:09 forrest iggy: Thanks for the issue, explain to me how this hasn't been fixed by now? This shit should be priority #1
19:09 DammitJim LOL
19:10 DammitJim thanks breakingmatter
19:10 iggy "not a bug" - salt devs
19:10 breakingmatter DammitJim: But, I also prefer the pyobjects and py renderers, so there's that. heh
19:11 stoogenmeyer_ joined #salt
19:11 breakingmatter But py also uses the dictionary style lookup, but I digress.
19:12 spark_ joined #salt
19:13 bbbryson joined #salt
19:17 markm joined #salt
19:17 DammitJim what do I need to prefix my conf file template to be able to use grains?
19:19 DammitJim nevermind, I forgot to add the template option for the file
19:20 DammitJim weird... I am running a state and it returns nothing ?
19:25 dthom91 joined #salt
19:25 forrest iggy: Do you have the PR handy where you added the fix? Did you just add a cwd?
19:26 forrest oh god you had to set env and home??
19:26 forrest That sucks
19:26 iggy lol, welcome to my world
19:26 iggy I have though about writing a module that did some of the env/cmd stuff so there's not so much cmd.run in that formula
19:27 iggy there's also an api iirc
19:30 PredatorVI joined #salt
19:30 clintberry joined #salt
19:31 surge___ joined #salt
19:37 slav0nic joined #salt
19:38 TyrfingMjolnir joined #salt
19:38 digitalhero joined #salt
19:39 PredatorVI Does doing `salt 'myminion' pillar.items show all pillars that are matched based on the /srv/pillars/top.sls file?
19:39 scbunn is there a way to set a default using {% set foo = { ...}.get(grains.foograin) %} ?
19:39 beardo joined #salt
19:39 PredatorVI My pillar data isn't matching and I don't know what I've done to break it.
19:40 spark_ joined #salt
19:41 PredatorVI My pillar top.sls has `dev:  'myminion':  - myapp` where my app is located at /srv/pillar/myapp and has an init.sls file.  But when I do a pillar.items nothing is returned.
19:42 digitalhero joined #salt
19:42 PredatorVI Is there a way to debug the pillar matching process on the master?  I enabled 'debug' but I didn't see anything relevant to what env or pillars it tried matching/loading.
19:42 nzero joined #salt
19:44 iggy pillar.items takes a saltenv arg I think
19:44 iggy dunno if you need to set it though
19:44 baweaver joined #salt
19:45 * iggy is a saltenv dummy
19:45 murrdoc it do
19:45 murrdoc it do
19:46 mordonez joined #salt
19:46 mordonez joined #salt
19:46 PredatorVI The example doesn't have that as an option:  http://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.pillar.html
19:46 mordonez joined #salt
19:47 deedubs I just updated to the latest packaged salt-master / salt-minion for ubuntu 14 and I'm getting
19:47 deedubs > TypeError: render() got multiple values for keyword argument 'env'
19:47 deedubs now
19:47 iggy deedubs: you shouldn't be using env any more (it's been deprecated for quite some time)
19:48 deedubs iggy: this is the sls file thats causing it https://gist.github.com/bd4bce82f1980eb634a6
19:48 deedubs I'm not using env directly, that I'm aware :|
19:50 iggy try quoting command?
19:50 iggy not really messed with the docker states much
19:57 Ben___ joined #salt
19:58 zwi joined #salt
19:59 DammitJim how do I check if my keys() are empty?
20:00 PredatorVI Even with 'saltenv=base' I get no pillar data.
20:00 rideh joined #salt
20:02 Norrland PredatorVI: not sure if you need to have the init.sls in pillar for 'myapp'
20:02 [1]PredatorVI joined #salt
20:02 Norrland PredatorVI: just /srv/pillar/myapp.sls; and in /srv/pillar/top.sls: dev: 'myminion':  - myapp
20:03 alemeno22_ joined #salt
20:03 Norrland PredatorVI: if possible, on the minion 'salt-call pillar.items'
20:03 bbbryson joined #salt
20:03 DammitJim how do I check if an object is empty or has no keys with jinja?
20:05 Bryson joined #salt
20:06 jaybocc2 PredatorVI: what does your file_roots: look like?
20:06 jaybocc2 and your top.sls
20:07 jaybocc2 actually that looks liek it might be the problem
20:07 DammitJim mine or Norrland ?
20:07 PredatorVI jaybocc2:  file_roots or pillar_roots?
20:07 jaybocc2 PredatorVI: run salt-call pillar.items env=dev
20:08 PredatorVI jaybocc2:  salt-call pillar.items env=dev on minion just returns '---------'  Same as on master
20:08 jaybocc2 in your pillar top.sls you're defining that the dev environment, myminion gets myapp pillar data
20:08 jaybocc2 what does pillar_roots look like
20:08 zsoftich2 joined #salt
20:09 malinoff_ joined #salt
20:09 VR-Jack huh. Did 2015.5 remove the job summary?
20:09 TyrfingMjolnir_ joined #salt
20:09 Tyrm_ joined #salt
20:09 PredatorVI D'oH!  my 'base' env got lost...knew it was something dumb...thanks for walking through things with me.
20:09 clintber_ joined #salt
20:10 jaybocc2 VR-Jack: salt-run jobs.list_jobs
20:10 jaybocc2 PredatorVI: np
20:11 jalbretsen1 joined #salt
20:11 VR-Jack jaybocc2: I was looking for this:
20:11 VR-Jack Summary
20:11 VR-Jack -------------
20:11 VR-Jack Succeeded: 11 (changed=10)
20:11 VR-Jack Failed:     0
20:11 VR-Jack -------------
20:11 VR-Jack Total states run:     11
20:12 jdesilet joined #salt
20:12 hasues joined #salt
20:12 jaybocc2 VR-Jack: i get that every time i run state.highstate for each minion, on version 2015.5.3
20:13 VR-Jack hmm. on 2014.7 I got it on my orchestrate. Now I don't seem to. Doesn't show in jobs.lookup_jid either
20:13 jaybocc2 hmmmm
20:13 jaybocc2 it does for me
20:13 jaybocc2 i don't use orchestration though
20:15 baweaver joined #salt
20:15 DammitJim I am doing this: {%- for section in samba_sections.keys() %}
20:15 VR-Jack you're right on highstate.
20:15 DammitJim but in my case, there are no keys
20:15 VR-Jack that did show the success. weird
20:15 DammitJim what can I test before that line so that I don't run that?
20:17 rideh joined #salt
20:20 bhosmer joined #salt
20:22 xDamox joined #salt
20:23 otter768 joined #salt
20:23 oznah still working on custom minion list in pillar. trying to ping 'salt -I "testing" test.ping'. i get error, no minions match target
20:24 rideh joined #salt
20:24 beautivile joined #salt
20:25 oznah salt * pillar.items shows list of minions. testing: - host1 \- host2 etc.
20:25 oznah not sure how to call from command line
20:26 DammitJim what is the difference between {%- and -%}
20:26 DammitJim like when does one put the - where?
20:28 mpanetta IIRC: {%- == remove empty line before template, and -%} means after
20:28 mpanetta DammitJim: ^
20:28 DammitJim thanks
20:29 DammitJim dammit... why is this line not working??!!!{%- if samba_sections == 'None' %}
20:29 mpanetta Do you really mean to quote off None?
20:29 DammitJim doing salt '*' pillar.get samba_sections returns None for this minion!
20:29 mpanetta Yeah but it is probably a python None not the string None
20:30 DammitJim oh
20:30 mpanetta Maybe it will work without the quotes?  Donno how jinja translates that
20:30 SheetiS Yeah that is probably None as in the python NoneType
20:30 beautivile looking for help for a specific issue, I am attempting to do an if comparison on the grains id and whether that id is in a node group, is this possible in an .sls file?  only thing I found online stated {% if grains['id'] not in salt[pillar.get]('master:nodegroups:GROUPNAME', []) %}, but that failed miserably
20:30 DammitJim Oh My... that was it
20:30 DammitJim that's so silly of me!!!!
20:30 DammitJim thanks mpanetta
20:30 DammitJim someone one just needs someone looking from the outside in
20:30 mpanetta DammitJim: no prob :)  We all make those mistakes.
20:31 mpanetta Yep
20:33 DammitJim I'm very confused
20:33 forrest iggy: Still failed for me even when setting the user as /root for the home env :(
20:33 DammitJim I am making decisions in pillars but also in states and templates
20:33 DammitJim is that normal?
20:34 SheetiS you can do that, but it might be confusing to find your decisions in the long run
20:34 SheetiS I think it can be fine as long as you are consistent with the types of decisions you are making in particular places
20:35 DammitJim well, I have decisions in pillar as to the version of java to install
20:35 DammitJim but then I have decisions on the configuration of a file in the template
20:35 iggy forrest: some other env variable it's looking for maybe?
20:35 DammitJim and now I'm thinking I need to make decisions about what states need to be run for a particular server
20:35 DammitJim or is the latter supposed to go somewhere else?
20:35 iggy forrest: gist it up!
20:36 SheetiS well the configurations could go into the pillar and then the template could just pull the pillar information.  That could add some consistency there.
20:37 SheetiS I actually use an ext_pillar to decide what roles are assigned to a server, but it's basically something like that or everything in the top.sls really
20:37 DammitJim Actually, I take that back... in my templates, I am not making decisions necessarily... it's just calling pillar
20:38 subsignal joined #salt
20:38 forrest iggy: https://gist.github.com/gravyboat/a2d993b53de3dd4ec199
20:39 DammitJim where does one normally define if a state needs to be run for a minion?
20:39 ingslovak joined #salt
20:39 DammitJim like I don't want to create a folder if it is not a particular server
20:42 SheetiS DammitJim: in the end that has to be in the top file, but you can have some logic poll somewhere else for the data.  like i said, I use an ext pillar and then my top file looks like this to collect the data: https://bpaste.net/show/2513b180f033
20:42 iggy forrest: assuming your service_release is 9.3 there? (unless test has version hardcoded)
20:42 forrest yeah it is
20:43 DammitJim SheetiS, not sure how that top file defines what states go on what minion
20:43 forrest The error is literally just Initializing database: ?[60G[?[0;31mFAILED?[0;39m]
20:43 iggy salt-call -l debug ?
20:44 s_kunk joined #salt
20:44 SheetiS Each system has a set of roles it gets from my ext pillar.  http://devop.ninja/configuration%20management/2015/03/01/beyond-grains-roles-with-an-external-pillar/ describes my proof-of-concept external pillar.  You don't have to do something like this.  I just wanted a good way to do category (groups of roles) inheritence in a way that made sense to me.
20:45 funzo joined #salt
20:45 SheetiS others have had luck using the reclass ext pillar included by default.
20:46 SheetiS it can do some similar things.
20:46 iggy forrest: are you sure there's a version on the postgresql service? (just checked normal install and pgdg install and they aren't versioned
20:46 VR-Jack Now to waiat for grumbling on why I care about a summary report on orchestrate. hehe
20:47 forrest iggy: Yes this is a custom state
20:47 forrest and it's on centos
20:47 forrest you have to have a version
20:47 forrest iggy: It works if you run that exact command manually ;)
20:48 forrest with -9.3 as the lookup value
20:48 DammitJim can one do a substring with jinja?
20:48 iggy oh, I thought only debian did initdb stuff in the init script
20:48 DammitJim like I want the grain.domain but without the .com
20:49 forrest iggy: Nope
20:49 forrest iggy: I wish ;)
20:49 iggy did you try running salt-call -l debug?
20:49 VR-Jack also had to complain in ticket about the new orchestrate reordering. heh. They sorted based on the new orchestrate id instead of run number.
20:50 SheetiS DammitJim: you can use jinja replace to strip a pattern from some text.
20:50 DammitJim lol... that's what I was just reading
20:50 shaggy_surfer joined #salt
20:50 DammitJim can I do a double filter?
20:50 X67r joined #salt
20:51 jaybocc2 dammitjim, you can probably do string[:-4]
20:51 SheetiS you could replace twice if you wanted like {{ foo|replace('bar', '')|replace('baz', '') }}
20:51 DammitJim I'll try that
20:51 jaybocc2 or string.rstrip('.com')
20:51 napsterX joined #salt
20:51 DammitJim but the replace worked
20:51 SheetiS yeah rstrip is probably better for .com
20:53 shaggy_surfer does anyone have a solution for the salt-cloud bootstrap issue here:  https://github.com/saltstack/salt/issues/25886  I tried different solutions but didn't work for me.  I tried using git solution but fails also.  I am running on ubuntu 14.04.3 LTS and it appear they removed the old versions from the repo's.
20:53 saltstackbot [#25886]title: Salt-cloud unable to bootstrap minions using COPR repo | This just started this morning using salt-cloud to bootstrap AWS minions:...
20:54 shaggy_surfer I use it to bootstrap my minions in AWS and it has broken my automated deployment now
20:58 shaggy_surfer this was the version I was using prior to it breaking:  Version: 2014.7.0+ds-2trusty1
20:58 shaggy_surfer now it installs 2015.5.3+ds-1trusty1, which doesn't work.
21:01 VR-Jack shaggy_surfer: according to the bug you listed, it has a fix destined for 2015.5.4 or 2015.8.0
21:01 iggy copr is a redhat thing, I don't think that bug is related
21:01 dthom91 joined #salt
21:01 VR-Jack and the bug listed is due to master_type: str failing minion start
21:02 shaggy_surfer correct, but it has not been pushed pulled into ubuntu repo it appears
21:02 shaggy_surfer they fixed but hasn't been released I guess
21:02 digitalhero joined #salt
21:03 shaggy_surfer I was looking to use a workaround in the meantime.
21:03 jaybocc2 shaggy_surfer: you can downgrade and lock to the 2014.7 repo
21:03 jaybocc2 https://launchpad.net/~saltstack/+archive/ubuntu/salt2014-7
21:03 shaggy_surfer how do I do that w/ bootstrap?
21:04 jaybocc2 with bootstrap you just specify you want 2014.7 i think
21:04 jaybocc2 idk i don't use bootstrap
21:04 shaggy_surfer right but I guess I would have to use git
21:04 shaggy_surfer as it's missing from the repo now
21:05 shaggy_surfer are the versions that same naming convenstion in git as in ubuntu repo
21:05 shaggy_surfer in other words:  git -v2014.7
21:05 shaggy_surfer will that work?
21:06 jaybocc2 just do stable 2014.7
21:06 jaybocc2 i think
21:06 jaybocc2 yeah
21:06 jaybocc2 ${__ScriptName} stable 2014.7
21:06 shaggy_surfer I tried that, but it didn't work
21:06 jaybocc2 line 222
21:06 jaybocc2 https://github.com/saltstack/salt-bootstrap/blob/develop/bootstrap-salt.sh
21:06 quasiben joined #salt
21:07 jaybocc2 in the script it seems to do the right thing
21:07 jaybocc2 about line 1830
21:08 peters-tx joined #salt
21:08 napsterX joined #salt
21:09 timoguin joined #salt
21:09 shaggy_surfer how do I tell salt-cloud to use a different version of the bootstrap-salt.sh script, it might be because I am using version 2014.7 (helium) on my salt-master maybe?
21:10 forrest iggy: Sorry had to get some food, I already ran debug, but I'll throw it through again in case I missed something
21:11 shaggy_surfer can i do this:
21:11 shaggy_surfer script: bootstrap-salt
21:11 shaggy_surfer script_args: git v2014.7
21:11 Brew joined #salt
21:11 shaggy_surfer it my salt-cloud profile?
21:11 iggy forrest: run env and salt 'foo' cmd.run env and copy over everything that looks relevant?
21:12 X67r joined #salt
21:14 Brew joined #salt
21:17 alemeno22_ joined #salt
21:17 shaggy_surfer I got it to work w/ git, thanks guys
21:17 shaggy_surfer I did the above
21:19 forrest iggy: Will do between putting out fires, a bit busy today
21:20 iggy forrest: you could also try unsetting things in the bash shell until the initdb breaks (if it does)
21:20 forrest iggy: Yeah good point, this is a real pain in the ass
21:20 forrest it should just be inheriting the shell :\
21:20 forrest like come on
21:20 forrest what's the point of running a command as a user (or even as root) if it doesn't use any of that shiiiiit.
21:20 forrest tempted to just install an older release of salt where I know it works correctlyh
21:21 iggy I think there's somewhere else where I was much more "adversarial" about this change
21:21 forrest iggy: It's just confusing to new users
21:21 iggy highly, good luck tracking it down, if you need more help, let me know
21:21 bhosmer joined #salt
21:21 baweaver joined #salt
21:22 RedundancyD joined #salt
21:31 Heartsbane Has anyone ever come accross this:
21:31 Heartsbane Permission denied (signaler.cpp:124)
21:33 VR-Jack Heartsbane: appears to be a zeromq error.
21:33 VR-Jack http://stackoverflow.com/questions/13684164/zeromq-subscriber-fails-to-initialize-using-1000-publishers
21:35 VR-Jack if you didn't get more details, you might file a bug report. salt should catch errors like that and make them make more sense.
21:37 Heartsbane thanks
21:40 VR-Jack general idea is either firewall, selinux, or FD issue, but I'm just guessing
21:49 rodrigc joined #salt
21:49 rodrigc left #salt
21:49 nzero joined #salt
21:49 rodrigc joined #salt
21:51 rodrigc are there any saltstack formulas for monitoring a Jenkins server?
21:51 rodrigc for example does saltstack.com have any formulas for monitoring: https://jenkins.saltstack.com/ ?
21:55 Heartsbane Oh I have narrowed it to SELinux
22:00 VR-Jack Heartsbane: Gotta love SELinux
22:01 shpoont joined #salt
22:02 Heartsbane Meh
22:02 Heartsbane Not today I don't
22:04 yomilk joined #salt
22:09 herlo joined #salt
22:09 napsterX joined #salt
22:09 * herlo is starting to dig travis-ci. Why did you guys never force me to use it previously?
22:10 Heartsbane herlo: we use it
22:10 herlo Heartsbane: who's we?
22:10 Heartsbane Developers love it
22:10 murrdoc <3
22:10 herlo Heartsbane: lies!
22:10 Heartsbane Fine
22:11 herlo lol
22:12 scbunn joined #salt
22:12 rodrigc is it possible to use travis-ci outside of github?
22:12 herlo rodrigc: yes
22:13 * herlo is using it with github, but it's just a hook.
22:13 rodrigc does travis-ci work on anything besides Linux and MacOS X these days?
22:13 herlo I saw something about it working Winders, but I don't know.
22:13 rodrigc not that important, but once in a while it is handy to have something work on oddball platforms
22:14 herlo plan9!
22:14 XenophonF joined #salt
22:14 hoonetorg joined #salt
22:18 VR-Jack I'm gonna go out on a limb and say salt doesn't follow CI values. heh
22:19 herlo VR-Jack: whatdya mean?
22:19 VR-Jack or the build tests are seriously lacking
22:19 herlo travis is for unit testing generally.
22:19 hasues left #salt
22:19 VR-Jack unit testing in general
22:19 herlo I don't think you can do things like UAT in there.
22:19 iggy "build tests" in a python project
22:20 VR-Jack iggy: it would be nice to notice when the api has broken. :)
22:20 XenophonF you need a functional test for that
22:21 iggy that was more of a "python isn't built" joke
22:21 VR-Jack I know. lol
22:22 VR-Jack it would be huge numbers of hours to write all the functionality tests that we should have, but it sure would improve stability.
22:24 otter768 joined #salt
22:25 herlo VR-Jack: never a better time to start helping out. :)
22:27 zsoftich2 joined #salt
22:28 theologian joined #salt
22:28 VR-Jack herlo: swamped. my business partner has been abscent so it's just me.
22:32 XenophonF writing good system tests is harder that implementing the thing being tested, sometimes
22:34 VR-Jack especially in client/server models where you sometimes need to check from both sides
22:34 gcfhvjbkn joined #salt
22:35 mpanetta joined #salt
22:35 gcfhvjbkn joined #salt
22:35 gcfhvjbkn joined #salt
22:35 mpanetta joined #salt
22:36 gcfhvjbkn joined #salt
22:36 bbbryson joined #salt
22:37 gcfhvjbkn joined #salt
22:37 bbbryson_ joined #salt
22:37 gcfhvjbkn joined #salt
22:38 gcfhvjbkn joined #salt
22:38 gcfhvjbkn joined #salt
22:39 gcfhvjbkn joined #salt
22:40 gcfhvjbkn joined #salt
22:40 gcfhvjbkn joined #salt
22:41 gcfhvjbkn joined #salt
22:41 gcfhvjbkn joined #salt
22:42 gcfhvjbkn joined #salt
22:43 Bryson joined #salt
22:43 gcfhvjbkn joined #salt
22:43 gcfhvjbkn joined #salt
22:44 gcfhvjbkn joined #salt
22:45 gcfhvjbkn joined #salt
22:45 gcfhvjbkn joined #salt
22:46 druonysus joined #salt
22:46 druonysus joined #salt
22:46 gcfhvjbkn joined #salt
22:47 bbbryson joined #salt
22:47 gcfhvjbkn joined #salt
22:47 pipps joined #salt
22:48 gcfhvjbkn joined #salt
22:48 gcfhvjbkn joined #salt
22:49 gcfhvjbkn joined #salt
22:49 gcfhvjbkn joined #salt
22:51 gcfhvjbkn joined #salt
22:52 breakingmatter joined #salt
22:52 gcfhvjbkn joined #salt
22:53 gcfhvjbkn joined #salt
22:53 gcfhvjbkn joined #salt
22:54 gcfhvjbkn joined #salt
22:54 gcfhvjbkn joined #salt
22:55 gcfhvjbkn joined #salt
22:55 funzo joined #salt
22:56 gcfhvjbkn joined #salt
22:56 gcfhvjbkn joined #salt
22:57 Bryson joined #salt
22:57 gcfhvjbkn joined #salt
22:58 gcfhvjbkn joined #salt
22:58 funzo_ joined #salt
22:58 gcfhvjbkn joined #salt
22:59 Shirkdog anyone have a great guide for automating the checkin for minions?
22:59 gcfhvjbkn joined #salt
23:00 gcfhvjbkn joined #salt
23:00 bbbryson joined #salt
23:01 gcfhvjbkn joined #salt
23:01 pipps joined #salt
23:01 gcfhvjbkn joined #salt
23:02 gcfhvjbkn joined #salt
23:02 cliluw When should I use state.highstate and when should I use state.apply?
23:03 gcfhvjbkn joined #salt
23:03 gcfhvjbkn joined #salt
23:04 gcfhvjbkn joined #salt
23:04 gcfhvjbkn joined #salt
23:05 VR-Jack cliluw: doesn't matter. state.apply is a wrapper I believe.
23:05 gcfhvjbkn joined #salt
23:05 Bryson joined #salt
23:06 gcfhvjbkn joined #salt
23:06 gcfhvjbkn joined #salt
23:07 gcfhvjbkn joined #salt
23:07 kermit joined #salt
23:08 gcfhvjbkn joined #salt
23:08 Aidin joined #salt
23:09 gcfhvjbkn joined #salt
23:09 gcfhvjbkn joined #salt
23:10 napsterX joined #salt
23:10 gcfhvjbkn joined #salt
23:10 gcfhvjbkn joined #salt
23:11 gcfhvjbkn joined #salt
23:12 gcfhvjbkn joined #salt
23:12 gcfhvjbkn joined #salt
23:13 gcfhvjbkn joined #salt
23:14 gcfhvjbkn joined #salt
23:14 timoguin joined #salt
23:14 gcfhvjbkn joined #salt
23:15 gcfhvjbkn joined #salt
23:15 Bryson joined #salt
23:15 gcfhvjbkn joined #salt
23:16 gcfhvjbkn joined #salt
23:17 gcfhvjbkn joined #salt
23:17 PredatorVI Trying to write a state that will install a custom Upstart script and process on ubuntu.  However it errors out stating the named service is not available which is true because it didn't install the script yet.  The 'service.running' module has the 'onchanges' which I thought meant it would run until the onchanges state was done.  How do ti get it to not try to check the service until after it has installed the scripts?
23:17 gcfhvjbkn joined #salt
23:18 PredatorVI *it wouldn't run until after the onchanges state was done.*
23:18 gcfhvjbkn joined #salt
23:18 Outlander joined #salt
23:18 gcfhvjbkn joined #salt
23:19 gcfhvjbkn joined #salt
23:20 gcfhvjbkn joined #salt
23:21 gcfhvjbkn joined #salt
23:21 gcfhvjbkn joined #salt
23:22 gcfhvjbkn joined #salt
23:23 gcfhvjbkn joined #salt
23:23 gcfhvjbkn joined #salt
23:23 Ztyx1 joined #salt
23:24 gcfhvjbkn joined #salt
23:25 gcfhvjbkn joined #salt
23:25 Bryson joined #salt
23:25 mackstick joined #salt
23:26 GothAck joined #salt
23:26 gcfhvjbkn joined #salt
23:26 bbbryson joined #salt
23:26 gcfhvjbkn joined #salt
23:26 omegamike joined #salt
23:27 stooj joined #salt
23:28 baweaver joined #salt
23:28 rhodgin joined #salt
23:28 kidneb joined #salt
23:28 VR-Jack PredatorVI: Would you just do a require?
23:29 gcfhvjbkn joined #salt
23:30 gcfhvjbkn joined #salt
23:30 gcfhvjbkn joined #salt
23:31 gcfhvjbkn joined #salt
23:31 gcfhvjbkn joined #salt
23:32 gcfhvjbkn joined #salt
23:33 gcfhvjbkn joined #salt
23:34 Tyrm joined #salt
23:34 gcfhvjbkn joined #salt
23:35 gcfhvjbkn joined #salt
23:35 gcfhvjbkn joined #salt
23:36 gcfhvjbkn joined #salt
23:36 gcfhvjbkn joined #salt
23:37 gcfhvjbkn joined #salt
23:38 gcfhvjbkn joined #salt
23:38 gcfhvjbkn joined #salt
23:38 digitalhero joined #salt
23:39 gcfhvjbkn joined #salt
23:39 gcfhvjbkn joined #salt
23:40 gcfhvjbkn joined #salt
23:41 gcfhvjbkn joined #salt
23:41 gcfhvjbkn joined #salt
23:42 gcfhvjbkn joined #salt
23:43 gcfhvjbkn joined #salt
23:44 PredatorVI VR-Jack: I tried that.  I thought you didn't need 'require' if you had 'onchanges' but either way I still get an error that the named service is not available.
23:44 gcfhvjbkn joined #salt
23:45 gcfhvjbkn joined #salt
23:45 herpoderp joined #salt
23:46 gcfhvjbkn joined #salt
23:46 gcfhvjbkn joined #salt
23:47 gcfhvjbkn joined #salt
23:47 PredatorVI here is the gist of the whole thing.
23:47 PredatorVI https://gist.github.com/PredatorVI/938e74f4537807cc26f3
23:48 gcfhvjbkn joined #salt
23:49 gcfhvjbkn joined #salt
23:49 gcfhvjbkn joined #salt
23:50 xDamox joined #salt
23:50 gcfhvjbkn joined #salt
23:50 xDamox Hello, has anyone run Salt minions and Salt master in a virtual environment
23:50 gcfhvjbkn joined #salt
23:51 VR-Jack PredatorVI: I don't think you can match on your own id with a require. Match on the name
23:51 VR-Jack or separate them into 2 different ids
23:51 gcfhvjbkn joined #salt
23:51 PredatorVI I have a bunch of other states that match on the id
23:51 VR-Jack their own id?
23:52 gcfhvjbkn joined #salt
23:52 PredatorVI yep...here is one https://gist.github.com/PredatorVI/e29f2bee24f78fb3a02d
23:52 gcfhvjbkn joined #salt
23:53 gcfhvjbkn joined #salt
23:53 PredatorVI It does work...
23:53 gcfhvjbkn joined #salt
23:54 gcfhvjbkn joined #salt
23:55 gcfhvjbkn joined #salt
23:55 PredatorVI xDamox:  I'd be everyone that's used salt has done it virtually, unless I'm misunderstanding your question.  my master and all 100+ minions are virtual.
23:55 PredatorVI *bet*
23:55 gcfhvjbkn joined #salt
23:55 xDamox python virtualenv
23:56 Aidin left #salt
23:56 gcfhvjbkn joined #salt
23:57 gcfhvjbkn joined #salt
23:57 scbunn losing my mind... when using http://pastebin.com/NrwYRsRv I'm getting back a string of either g1 or g2 instead of the list that I expect to get back.
23:57 gcfhvjbkn joined #salt
23:58 gcfhvjbkn joined #salt
23:58 PredatorVI xDamox:  Got me...i definitely don't use python virtualenv that I'm aware of :/
23:58 * PredatorVI instructs a pet piranha to munch on PredatorVI's shoulder blades
23:58 gcfhvjbkn joined #salt
23:59 gcfhvjbkn joined #salt
23:59 funzo joined #salt
23:59 xDamox Ok, I am have a dependancy hell :(

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