Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2013-10-10

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

All times shown according to UTC.

Time Nick Message
00:04 taylorgumgum joined #salt
00:05 jimallman joined #salt
00:07 ninkotech joined #salt
00:13 fllr_ joined #salt
00:14 fllr_ Hey guys. I'm using the users module to add a user to my dev environment. Except, I wanted to be able to login to that user without any passwords since, well, this is my dev env. How do I do that? I've looked around google and the docs, and haven't been able to find anything. I'm a bit of a beginner here...
00:17 honestly fllr_: if you mean ssh login you need to copy your ssh public key to the other machine
00:17 honestly (the one you want to log in on
00:17 honestly )
00:18 fllr_ Isn't there a way to accomplish the same thing without a pub key?
00:19 honestly passwordless remote login? No, and even if there were, you shouldn't get into the habit
00:20 fllr_ I wasn't... I'm just trying to setup a dev env, is all..
00:21 mannyt joined #salt
00:23 jetblack joined #salt
00:28 danielmcbawse joined #salt
00:30 __number5__ fllr_: the initial account with pub key should be set up when you install os
00:33 troyready joined #salt
00:34 Jahkeup joined #salt
00:34 mwillhite joined #salt
00:42 justlooks joined #salt
00:43 bemehow joined #salt
00:43 justlooks anyone familiar with reactor/event system ,i have a problem here https://gist.github.com/justlooks/6898169
00:46 g4rlic left #salt
00:46 bemehow_ joined #salt
00:53 jaequery joined #salt
00:57 ipmb joined #salt
00:59 redondos joined #salt
01:01 helderco joined #salt
01:02 alunduil joined #salt
01:04 woebtz joined #salt
01:09 brimpa joined #salt
01:11 cnelsonsic joined #salt
01:16 Jahkeup joined #salt
01:18 bemehow joined #salt
01:21 Lue_4911 joined #salt
01:25 Jahkeup_ joined #salt
01:27 __number5__ justlooks: should line 12 be just: cmd.run:
01:28 justlooks __number5__:  hi ,here is doc link ,you can see the code in it ,also use cmd.cmd.run  http://docs.saltstack.com/topics/reactor/index.html
01:28 redondos joined #salt
01:28 jaequery joined #salt
01:28 redondos joined #salt
01:28 justlooks __number5__:  i just following the doc's example
01:38 fragamus joined #salt
01:40 redondos joined #salt
01:42 amahon joined #salt
01:44 krissaxton joined #salt
01:48 avienu joined #salt
01:50 brimpa What is best practice regarding upgrading the salt-minions and salt-master?
01:50 brimpa Are backwards incompatible changes between the versions published somewhere?
01:52 mannyt joined #salt
01:52 justlooks brimpa: test you code in new version in you test envionment
01:54 brimpa justlooks: I intend to but I was hoping to at least have a general understand about what changed
01:54 drags joined #salt
01:54 brimpa Or rather, what changes I'll need to make to bring my states up to (for instance) 0.17
01:54 brimpa understanding*
01:57 deepakmd_oc joined #salt
01:57 justlooks brimpa: you can check the changelog of 0.17
01:57 redondos joined #salt
01:57 redondos joined #salt
01:59 xl1 joined #salt
01:59 cshuman joined #salt
02:00 brimpa justlooks: Do you know where I can find that?  Earlier changelogs would be good too.  I found the release notes but it only discussed high-level changes
02:04 justlooks sorry , i think you should do test first ,then observer carefully ,find problem ,look for the reason,if you give a specific problem ,may be you will get more useful help
02:08 higgs001 joined #salt
02:21 cshuman joined #salt
02:29 jaequery joined #salt
02:34 mianos joined #salt
02:37 Brew joined #salt
02:38 cnelsonsic joined #salt
02:41 taylorgumgum joined #salt
02:42 xmltok joined #salt
02:42 jaequery joined #salt
02:44 krissaxton joined #salt
02:48 andrej How can I find the installed minion version from the master?
02:48 andrej Let me rephrase that: I'd like all minions to report their salt version back to me
02:50 andrej Meh
02:50 andrej salt \* test.version
02:50 andrej :)
02:52 ClausA joined #salt
02:59 forrest_ joined #salt
03:00 Marion joined #salt
03:01 Marion_ joined #salt
03:04 backjlack joined #salt
03:10 danielmcbawse joined #salt
03:11 jefimenko joined #salt
03:11 jaequery joined #salt
03:12 raadad joined #salt
03:12 raadad hey guys
03:12 raadad eyError: 'master_uri'
03:12 raadad *oops,
03:12 raadad what I meant to say is, Im getting eyError: 'master_uri'
03:13 raadad * KeyError: 'master_uri'  when i run: salt-call state.highstate on a minionless master running salt develop
03:13 raadad any help would be appreciated
03:15 raadad * I mean a masterless minion
03:15 raadad sorry I dont know whats wrong with me today :p
03:16 zwe joined #salt
03:16 raadad * i think is because i did not append --local to it
03:16 raadad all good :D
03:26 mwillhite joined #salt
03:30 andrej Is it a design-feature that the update from 0.16.4 to 0.17 culls the accepted keys on minions?
03:31 forrest_ not at as far as I know andrej
03:32 andrej I just updated 1 centos and two ubuntu machines (after updating the master), both ubuntu boxes had to have their keys accepted again
03:33 andrej While I'm just evaluating salt I'm happy w/ that; if I move salt to a production status and had to re-accept a largish number of keys every I update salt I'd be less impressed :)
03:35 forrest_ yea, I didn't have to do that moving from 0.16 to 0.16.3 to 0.16.4 on my ubuntu machine
03:35 andrej Have you moved to 0.17 yet?
03:36 andrej Just curious whether this is a local issue, or hits others, too
03:37 joehh andrej: I suspect that it is related to https://github.com/saltstack/salt/issues/7558
03:37 alrs joined #salt
03:38 andrej thanks joehh ... just stumbled upon that myself
03:41 raghavp joined #salt
03:42 Marion_ joined #salt
03:42 deepakmd_ joined #salt
03:45 krissaxton joined #salt
03:50 torandu left #salt
03:53 christopherl joined #salt
03:55 Lue_4911 joined #salt
03:55 luminous anyone else find the formatting changes on the output of state.{sls,high,highstate} diminish readability?
03:57 tinuva joined #salt
03:58 vipul joined #salt
04:03 kermit joined #salt
04:03 kermit joined #salt
04:03 christopherl trying the new version now, cannot confirm
04:03 christopherl looks badass to me
04:09 forrest_ 0.17?
04:09 christopherl yeah
04:09 joehh is badass good or bad?
04:09 forrest_ Yea it's pretty awesome, I was talking it up at devops day atlanta last week, people were pretty excited when I explained all the new stuff/changes.
04:09 forrest_ badass = good
04:09 christopherl basass == good
04:09 christopherl badass*
04:11 jslatts joined #salt
04:14 Katafalkas joined #salt
04:14 joehh good to know
04:14 bemehow joined #salt
04:15 saurabhs joined #salt
04:16 forrest_ Did anyone else go to devopsday in Atlanta? I didn't see anyone around that I knew from in here.
04:16 forrest_ Not too many other salt evangelists there other than the sales guys.
04:16 saurabhs left #salt
04:17 forrest_ Oddly enough some companies considering config management aren't aware of salt, one guy I talked to had only looked at puppet/chef.
04:18 jefimenko joined #salt
04:18 jefimenko joined #salt
04:18 andrej I'm sure it will gather momentum
04:18 andrej I'm quite fond of it
04:19 forrest_ Yea, it was just interesting to hear, the guy wanted to know a bunch, so I was more than happy to show him
04:19 andrej :)
04:19 forrest_ Some guys using puppet were at our table too and started asking questions
04:20 forrest_ Explained the master/minion setup and they were blown away by how efficient it was.
04:20 fragamus joined #salt
04:20 * luminous doesn't find "file_|-/var/www/default_|-/var/www/default_|-directory" readable output
04:20 forrest_ lol
04:21 forrest_ Did you bring that up on the mailing list luminous?
04:21 luminous nopw
04:21 luminous nope
04:21 forrest_ You should
04:21 forrest_ be like 'errors were lame before, this is unacceptable!'
04:21 luminous my guess is this is more machine parseable
04:21 JaredR joined #salt
04:21 forrest_ ehh but that's worthless
04:21 luminous meh, not really something I'm willing to make much noise about
04:22 luminous I respect what they do, and you can't make us all happy
04:22 forrest_ Well, the errors need to be approachable to all users, new and old, and those unfamiliar with python
04:22 forrest_ so if you're getting an error and it's hard to read, that's not great.
04:22 luminous I'm more curious about WHY
04:23 luminous that's not an error, that's NORMAL output from a state
04:23 luminous that's the new "id" in the state output
04:23 forrest_ hmm
04:23 luminous my guess is that the internals changed for better namespacing or something
04:25 forrest_ Yea I don't know, that reminds me that I need to write some docs for the roster system this weekend
04:26 luminous speaking of badass, I have just completed my first pass on a totally awesome script which A) installs a minion, B) configures that minion  C) compiles some state data and runs state.high  D) runs state.highstate after this  and E) runs some list of module/func
04:26 luminous what's badass?
04:26 luminous B, C, D, and E are all completely arbitrary and based on a yaml config
04:27 luminous so you can go from a bare VM to having checked out some git repo, run highstate and some list of modules/funcs, all in one pass
04:27 mgw1 is anyone having issues with gitfs on 0.17? I have to wipe the cache every time I add a new repo to gitfs_remotes
04:27 forrest_ nice
04:27 luminous mgw1: are you restarting the master?
04:27 mgw1 otherwise new repos don't show up
04:27 mgw1 yes
04:27 mgw1 luminous: ^
04:28 luminous sounds like a reasonable bug
04:28 forrest_ Is Dan Garthwaite around under different alias?
04:28 forrest_ https://github.com/saltstack/salt/issues/7716
04:28 luminous mgw1: I pretty much stay 1 release behind because of issues like this :P
04:28 forrest_ err under a different alias
04:28 forrest_ man can't type
04:28 luminous I let everyone else find all teh problems
04:28 luminous salt-stack QA needs to improve
04:28 forrest_ mgw1, what release are you on?
04:29 luminous forrest_: mgw1 noted 0.17
04:29 mgw1 latest 0.17 tag from git
04:29 mgw1 to be precise
04:29 forrest_ oh sorry, reading comprehension isn't great tonight
04:29 fllr_ joined #salt
04:29 forrest_ that's odd, I thought they fixed that already.
04:29 luminous mgw1: maybe you aren't running what you think?
04:29 mgw1 I didn't find an issue on github about it
04:30 mgw1 but I thought I'd seen some mention in irc earlier
04:30 mgw1 before I ran into the issue
04:30 forrest_ yea it's been mentioned a bunch of times
04:30 mgw1 luminous: I'm pretty sure I'm on 0.17 tag
04:30 forrest_ I could have sworn I remember UtahDave saying they fixed it.
04:30 mgw1 maybe in develop
04:30 mgw1 I'm not running develop
04:30 forrest_ this was back on a 0.16 release
04:30 forrest_ so it should have been in 0.1
04:30 forrest_ *7
04:31 mgw1 hmm
04:31 luminous assuming you are running ubuntus, PPA's released, maybe you can install from apt and retry?
04:31 luminous or run on a new/clean host
04:31 luminous to confirm
04:31 mgw1 luminous: $ salt --version
04:31 mgw1 salt 0.17.0-82-g0f8ea74
04:32 forrest_ I'm not trying to say you're wrong mgw1, I'm just saying I thought that was fixed already
04:32 mgw1 same for salt-master
04:32 Marion Anyone familiar with : failed to preserve ownership for '/etc/salt/minion'  ???
04:32 mgw1 forrest_: yeah, I understood that :-)
04:32 forrest_ ok cool
04:32 forrest_ what about this issue: https://github.com/saltstack/salt/issues/6989
04:32 forrest_ terminalmage is saying they update every 60 seconds it looks like?
04:33 forrest_ are you waiting 60 seconds mgw1?
04:33 danielmcbawse joined #salt
04:33 mgw1 no, I'm restarting the master entirely
04:33 forrest_ can you make an update, then wait 60 seconds and see what happens?
04:33 forrest_ just to see if the functionality that exists per that issue is working?
04:33 mgw1 that is in reference to changes to an existing repo. my problem is with adding a repo in the master's config
04:34 forrest_ Hmm
04:34 forrest_ Can you still try to wait 60 seconds to see what happens?
04:35 zwe hi folks!  Just used salt-bootstrap to install 0.17 master on an ubuntu vm.  Trying to run salt-ssh = command not found.  Not seeing the executable with the others (salt, salt-run,salt-key, etc).  Convinced I'm missing something obvious but haven't found it yet.  Ideas?
04:35 forrest_ zwe, you're the second person who has mentioned this.
04:35 mgw1 forrest_: yeah, i'll try that
04:35 forrest_ mgw1, cool
04:36 joehh zwe: salt-ssh is in the salt-ssh package rather than salt-master - docs could be clearer
04:36 ninkotech joined #salt
04:37 forrest_ joehh, it looks like the docs were updated to mention this at the top of the salt SSH docs
04:37 forrest_ but yea I agree, it could be more 'out' there.
04:37 zwe Ah.  So the tidy bootstrap needs another step to install salt-ssh.  Or I need to provision that into the vm separately
04:38 forrest_ zwe, by default the bootstrap ONLY installs what is absolutely required for salt
04:38 forrest_ so no gitpython, or git (if you don't use the git option), or salt-ssh
04:38 luminous mgw1: can you reproduce on a clean/new host/vm?
04:38 idkfa joined #salt
04:38 forrest_ zwe, I've brought this up before and optional items will not be part of the bootstrap
04:39 mgw1 luminous: not right now, but I can try later
04:39 luminous I'd do that before a bug or github ticket update
04:39 forrest_ any luck waiting the 60 seconds mgw1?
04:39 mgw1 forrest_: well, removing it from the master config doesn't remove it from the cache
04:39 mgw1 and it is still available
04:40 forrest_ after the 60 seconds? hmm
04:40 mgw1 correct
04:40 mgw1 adding repos doesn't work either, even after 60s
04:40 forrest_ very strange
04:40 mgw1 I'll try to get this installed on a fresh host tomorrow
04:41 mgw1 and see if it does it there
04:41 Furao joined #salt
04:41 luminous hrm, anyway to get file.append to use a file (as the text source) from/on the host/minion?
04:41 forrest_ not as far as I'm aware luminous
04:41 luminous bah
04:41 forrest_ because Salt won't know that file exists
04:41 forrest_ unless you laid it down with salt that is :P
04:41 luminous well, I have a file.exists
04:42 forrest_ ehhhhh I don't think it reads the content when you do that
04:42 luminous using source: file:///path/to/file ought to use the text in that file for file.append
04:42 luminous at least, that is what I would expect from what the docs say
04:42 forrest_ ifnull, are you around?
04:42 jslatts joined #salt
04:43 forrest_ yea but that's a file path on the master.
04:44 luminous forrest_: yea, would seem so, but bah.
04:44 forrest_ you can't config manage what the system doesn't know about! Madness!
04:44 zwe forrest: thank you.  glad to be unstuck.
04:45 luminous forrest_: well, i'm also running that with salt-call / masterless
04:45 forrest_ zwe, it was all joehh.
04:45 forrest_ luminous, Ahh, hmm maybe it could work if you did the full filepath then? I still don't like it
04:45 zwe joehh:  thanks!
04:45 forrest_ sounds dirty
04:45 forrest_ why don't you just append the text straight in there?
04:45 luminous I tried
04:45 krissaxton joined #salt
04:45 luminous forrest_: because I don't know the text ahead of time
04:46 forrest_ and you can't use logic to determine that information? Or make the file a template?
04:46 luminous yea, this particular bit is actually really dirty, but I need it for the moment. I can remove it in another month or so when other issues are resolved
04:46 luminous nope
04:46 luminous it's either 'cat /foo/bar >> /foo/baz' or file.append
04:47 luminous I liked the idea of file.append to get rid of the need for a nasty grep with unless
04:47 forrest_ yea
04:48 joehh nw
05:13 justlooks hi,all help for this https://gist.github.com/justlooks/6898169
05:16 ninkotech joined #salt
05:17 bemehow joined #salt
05:20 forrest_ justlooks, '{"testme":"buy-me-a-beer"}'  should be '{"testme": "buy-me-a-beer"}' , missing a space there
05:20 forrest_ also I don't know about cmd.cmd.run: on line 12
05:21 forrest_ but I see that in the example below
05:21 justlooks forrest_:  i changed still no use
05:21 justlooks forrest_: :(
05:22 justlooks forrest_:   cmd.cmd.run it is LocalClient API
05:22 forrest_ hmm, can you run a debug on that call?
05:22 forrest_ yea you're right
05:23 justlooks forrest_: you mean run master on debug mode?
05:24 forrest_ justlooks, salt-call -l debug event.fire_master '{"testme": "buy-me-a-beer"}' 'demo'
05:25 justlooks forrest_: https://gist.github.com/justlooks/6913437 here is debug output ,seems normal
05:26 forrest_ yea there are no errors in there.
05:26 forrest_ can you just do a cmd.run from the commandline for CM3 to create that /tmp/demo file?
05:27 justlooks forrest_: i try no use
05:27 forrest_ so even running a cmd.run from the master targeting that minion to just create the file doesn't work?
05:27 forrest_ that doesn't seem like a reactor system problem
05:28 justlooks forrest_:  it ok if it run on master commandline use cmd.run
05:30 forrest_ Ahh ok
05:31 justlooks forrest_: but if in reactor sls file ,you should use cmd.cmd.run because it's LocalClient API call ,use cmd.run is no use
05:31 forrest_ I agree with you
05:31 forrest_ can you try to change the tgt from CM3 to CM2?
05:31 forrest_ to see what happens when we perform the reactor on the same system?
05:32 justlooks forrest_: still nothing appare
05:33 justlooks forrest_: i change it to '*' ,also nothing happen
05:33 forrest_ ok, so that makes me think that the reactor isn't getting triggered somehow
05:33 jkleckner joined #salt
05:34 justlooks forrest_: sample test,you can run it in your env ,fire from one host,create something on another host, i test it because i need this function
05:34 justlooks forrest_:  but the result make me confuse
05:34 forrest_ Yea I'm not sure either, I haven't messed with the reactor system a whole tin
05:34 forrest_ *ton
05:38 forrest_ I've gotta head out justlooks, let me know if you figure out what was causing that.
05:38 justlooks forrest_: ok
05:42 A21BC joined #salt
05:44 cshuman joined #salt
05:46 krissaxton joined #salt
05:47 jkleckner joined #salt
05:50 Furao joined #salt
05:50 A21BC left #salt
05:55 jkleckner joined #salt
05:56 druonysus joined #salt
05:59 druonysus joined #salt
05:59 druonysus joined #salt
06:01 cshuman joined #salt
06:01 scristian_ joined #salt
06:03 drags ack, almost rolled 17.0.1 to my ubunties before hitting this on the master https://github.com/saltstack/salt/issues/7558 :)
06:07 Furao 0.17.0 don't works
06:08 Furao as usual, wait until .1
06:14 falican joined #salt
06:22 bonezed joined #salt
06:25 vipul joined #salt
06:28 bonezed guys, can somebody point me at docs on dynamic states? we're looking at spinning up multiple hosts (aws) and then doing dynamic configs based on certain rules
06:33 jkleckner joined #salt
06:37 zizzler joined #salt
06:38 krissaxton joined #salt
06:38 zizzler have you had any trouble recently when updating ubuntu installs. There was some python updates coming to 13.04 release and I'm getting errors from pillar data I didnt get before.
06:40 ml_1 joined #salt
06:43 az87c joined #salt
06:43 az87c_ joined #salt
07:02 malinoff joined #salt
07:02 gildegoma joined #salt
07:05 balboah joined #salt
07:16 ckao joined #salt
07:26 zizzler how can I debug pillar data? sudo salt '*' pillar.data is empty now after updating packages
07:26 mbbs joined #salt
07:27 malinoff zizzler: salt \* pillar.items
07:28 zizzler malinoff, it is empty too
07:28 adepasquale joined #salt
07:29 malinoff zizzler: salt \* saltutil.refresh_pillar
07:29 woebtz joined #salt
07:30 zizzler it prints the hostname and None. and pillar.items is empty after that.
07:33 jkleckner joined #salt
07:34 zizzler malinoff, I get this to master log when running pillar.items http://pastebin.com/7B5xdGDw
07:35 matanya_ joined #salt
07:36 mianos joined #salt
07:38 luminous bah. how to get ubuntu ppa user/archive for something like mongodb which only publishes key?
07:40 drdran joined #salt
07:41 bud joined #salt
07:42 zizzler malinoff, I got it. in pillar top.sls I had one host specified with nothing after that. removing fixed it.
07:43 malinoff zizzler: Cool :)
07:46 zizzler luminous, not sure if it is what you're asking but I use this for all ppa stuff. https://github.com/saltstack/salt-contrib/blob/master/states/apt_repository.py
07:48 Iwirada joined #salt
07:48 zooz joined #salt
07:50 christopherl joined #salt
07:51 Furao joined #salt
07:53 drdran left #salt
07:54 pkimber joined #salt
07:59 packeteer bonezed | guys, can somebody point me at docs on dynamic states? we're looking at spinning up multiple hosts (aws) and then doing dynamic configs based on certain rules
08:04 jkleckner joined #salt
08:05 s0undt3ch joined #salt
08:06 honza joined #salt
08:06 honza is here a way to pretty print the dependency graph (what's going to be executed when) without doing the work?
08:07 luminous zizzler I'm looking for the ppa user/archive for mongo
08:07 honza also - recently upgraded to salt-minion 0.17 and now the user state is busted
08:08 felixhummel joined #salt
08:08 luminous honza: test=True?
08:09 honza luminous: where does that go? in top.sls? salt-call doesn't seem to have that arg
08:09 Katafalkas joined #salt
08:10 luminous honza: after everything else
08:10 luminous salt-call --local state.sls foo.bar test=True
08:11 Cidan oh is .17 out?
08:11 Katafalkas joined #salt
08:11 honza luminous: nice! thanks
08:11 luminous ;)
08:11 honza Cidan: it's in the ppa, yes
08:11 Cidan awesome
08:12 Cidan changelog anywhere?
08:12 luminous Cidan: release notes are in docs
08:12 luminous or you can look at git?
08:12 Cidan thanks :D
08:12 felixhummel_ joined #salt
08:12 honza Cidan: http://docs.saltstack.com/topics/releases/0.17.0.html
08:12 Cidan yay!
08:12 honza it refuses to create a user/group combo
08:14 Cidan holy jesus, auto order
08:14 Cidan thank the lord
08:14 Katafalkas joined #salt
08:16 Katafalk_ joined #salt
08:17 Katafalkas joined #salt
08:19 Cidan how do I enable halite?
08:20 Katafalkas joined #salt
08:21 mianos joined #salt
08:21 malinoff packeteer: It depends on your configuration
08:22 malinoff packeteer: But I think, Salt Mine is what you need
08:22 Katafalk_ joined #salt
08:23 packeteer ok, thanks. I'll read those docs
08:23 Katafalk_ joined #salt
08:24 samuel_ joined #salt
08:24 Katafalk_ joined #salt
08:26 Katafal__ joined #salt
08:27 Katafalkas joined #salt
08:30 Katafalkas joined #salt
08:34 mmilano left #salt
08:34 jkleckner joined #salt
08:38 ddv joined #salt
08:39 jeddi joined #salt
08:41 samuel_ Hi everybody, can anyone tell me how to show progress of salt install packages ?
08:43 nliadm joined #salt
08:49 unicoletti_ joined #salt
08:54 ebodine1 joined #salt
08:55 raghavp80 joined #salt
08:55 higgs001 joined #salt
08:57 krissaxton joined #salt
08:58 Katafalkas joined #salt
08:58 [M7] joined #salt
08:58 [M7] joined #salt
09:01 ebodine2 joined #salt
09:05 jkleckner joined #salt
09:10 honza left #salt
09:22 scott_w joined #salt
09:22 justlooks samuel_: check the debug info ,maybe
09:24 mastrolinux joined #salt
09:24 mastrolinux Hi guys I have: Comment:   Failed to confirm config of repo deb http://apt.datadoghq.com/ unstable main: repo "deb http://apt.datadoghq.com/ unstable main" was not found
09:24 mastrolinux just after 0.17 update
09:24 mastrolinux on master
09:24 mastrolinux is it a regression?
09:24 mastrolinux it was used to work
09:28 Furao mastrolinux: 0.17 is buggy
09:28 Furao i'm skipping that version
09:28 Furao well .0
09:28 ggoZ joined #salt
09:35 Katafalk_ joined #salt
09:35 jcristau joined #salt
09:35 Katafalkas joined #salt
09:36 jkleckner joined #salt
09:37 felixhummel_ hi
09:37 unicoletti_ mastrolinux: https://github.com/saltstack/salt/issues/7696
09:38 felixhummel_ is there any particular reason why http://docs.saltstack.com/topics/releases/0.17.0.html still says "Current Salt release: 0.16.4"?
09:38 felixhummel_ and on another note: how can you tell salt-cloud to use a specific minion version while bootstrapping?
09:42 mastrolinux umberto: thanks :)
09:44 pengyao joined #salt
09:44 mastrolinux unicoletti_: thanks it works removing the trailing /
09:49 pengyao does anybody have UtahDave's email?
09:57 MK_FG joined #salt
10:03 krissaxton joined #salt
10:04 joehh pengyao: I think if you search the google group/mailing list it will be there
10:08 pengyao joehh: yeah, you are right
10:08 mastrolinux joined #salt
10:12 Chocobo joined #salt
10:12 Chocobo joined #salt
10:12 viq Hello, I am having some problems with salty-vagrant, namely with the seed_master option, namely it seems to do absolutely nothing. Any hints?
10:13 aptiko joined #salt
10:14 carmony joined #salt
10:14 dccc joined #salt
10:17 viq akoumjian: supposedly you're the person to poke about this ;)
10:35 mmilano joined #salt
10:38 MK_FG joined #salt
10:47 helderco joined #salt
11:02 giantlock joined #salt
11:05 ebodine3 joined #salt
11:06 ebodine4 joined #salt
11:06 Katafalk_ joined #salt
11:07 ebodine joined #salt
11:11 ebodine1 joined #salt
11:13 viq Ah, there we go. https://github.com/mitchellh/vagrant/pull/2359
11:14 ebodine2 joined #salt
11:15 scristian joined #salt
11:16 ebodine3 joined #salt
11:19 christopherl joined #salt
11:21 Katafalkas joined #salt
11:21 tonthon Hi
11:21 tonthon I'm encoutering some problems with salt's debian packages
11:21 tonthon for example, with the appropriate repository configured, on a debian wheezy
11:22 tonthon if I run apt-get update && apt-get upgrade
11:22 tonthon salt is not upgraded
11:22 tonthon I need to remove/re-install it to get the latest version
11:23 joehh tonthon: which version are you upgrading from
11:23 joehh ?
11:23 tonthon 0.15.1
11:23 tonthon (or 0.16.0 on another server)
11:24 Furao you don't want 0.16.0
11:24 Furao start with at least 0.16.3
11:24 joehh the dependencies changed at 0.15.3 (zeromq3 rather than zeromq) and so it will need more than just an apt-get upgrade
11:24 Furao .4 had been released a few weeks ago
11:25 tonthon joehh: the same problem occures with 0.16
11:25 tonthon sorry, with 0.16.0 it upgraded to 0.16.3
11:26 joehh tonthon: I would have thought 0.16.0 would have worked - just saw it did
11:27 joehh 0.16.3-1 had a packaging "bug" that would make upgrading via (within) salt tricky
11:27 jgelens Lately I'm getting this error the whole time for multiple IDs, any idea?
11:27 jgelens 'Detected conflicting IDs, SLS IDs need to be globally unique.\n    The conflicting ID is "whoopsie" and is found in SLS "base:core" and SLS "base:core"',
11:27 joehh tonthon: a manual apt-get upgrade should work, but within salt, it would get stuck
11:28 tonthon joehh: ok, good to know
11:28 tonthon :)
11:31 Furao jgelens: 0.17.0 bug
11:31 Furao jgelens: same problem here
11:31 jgelens Furao: ah ok, thx
11:31 jgelens I'll downgrade then :)
11:32 Furao you got plenty of include?
11:32 jgelens yes
11:32 Furao I think this happens when you have 2 .sls that include that same file
11:32 jgelens that's the reason then
11:32 Furao I didn't investigate much
11:32 felixhummel_ is there a 0.16.4-1precise package somewhere? it's not on launchpad any more :(
11:32 jgelens so it's a bug or is including a sls multiple times not valid?
11:33 Furao felixhummel_: http://archive.bit-flippers.com/mirror/salt/0.16.4/
11:33 Furao I mirror PPA :)
11:33 felixhummel_ thanks Furao
11:34 jgelens Is there any way to configure apt to pin a certain version? :/
11:37 uta joined #salt
11:37 joehh jgelens: I suspect this may be wha tyou are after: http://www.debian-administration.org/articles/67
11:37 jgelens thx joehh
11:38 uta has the runner system changed in 0.17? the upgrade broke all our runner code
11:39 uta but i can't find any mention of changes
11:40 Furao jgelens: no an APT repo can't hold multiple version of the same package
11:41 jgelens hmmm, maybe better to switch to pip install instead then
11:42 jcristau err.
11:42 joehh jgelens: what is your problem?
11:42 jcristau Furao: that's not true...
11:42 joehh is it that ubuntu is automatically upgrading packages?
11:42 jgelens joehh: 0.17.0 has bugs, so I should be able to install old versions with much hassle
11:43 jcristau just that apt will by default install the latest version
11:43 jgelens without*
11:43 joehh ubuntu or debian?
11:43 Furao jcristau: I wrote a webui for reprepro
11:43 jgelens ubuntu
11:43 Furao and I had been managing .deb for more than 10 years
11:44 Furao and everything that I knew just exploded
11:44 joehh jgelens: you could use Furao's mirror (I believe he is happy with this) - he keeps each version separate
11:44 jgelens :) ok
11:44 jgelens thanks
11:45 joehh remove the old one, change your sources.list then install again
11:45 Furao jcristau: show me that
11:45 joehh hope that helps
11:45 Furao show me a single repo with that
11:46 jcristau dpkg-scanpackages will happily keep multiple versions in a Packages file
11:46 Furao that are across multiple repos
11:46 Furao not into a single one
11:46 jcristau uh, no.
11:47 Furao this is local dpkg db
11:49 Furao http://askubuntu.com/questions/84788/create-deb-repository-with-several-versions-of-the-same-package
11:49 Furao that is a reprepro limitation
11:49 Furao and from what I see it's a limitation of many repo server
11:52 ipmb joined #salt
11:52 rgbkrk joined #salt
11:52 adepasquale joined #salt
11:53 jetblack joined #salt
12:01 Katafalkas joined #salt
12:03 bhosmer joined #salt
12:03 krissaxton joined #salt
12:05 krissaxton1 joined #salt
12:11 copelco joined #salt
12:12 blee joined #salt
12:17 arthurlutz joined #salt
12:17 wincus joined #salt
12:21 jslatts joined #salt
12:24 arthurlutz I just published a report about the salt meetup that took place in Paris last week with thatch45 http://www.logilab.org/182361
12:28 joehh Hi arthurlutz, just noting the debian wheezy backports comment - Julien has already been assisting (though he may not have been aware at the time)
12:29 joehh he recently uploaded the last remaining dependency to squeeze-backports-sloppy
12:29 joehh we just have to wait for it to go through the new queue and I'll be able to backport 0.16.4 to squeeze (already backported to wheezy)
12:30 arthurlutz joehh: I kinda knew, I'm sure he'll appreciate it (cc jcristau)
12:32 ml_1 joined #salt
12:36 az87c_ joined #salt
12:37 tyler-baker joined #salt
12:37 tyler-baker joined #salt
12:43 helderco joined #salt
12:44 ggoZ joined #salt
12:45 Nexpro joined #salt
12:48 uta joined #salt
12:49 jcristau joehh: right, feel free to ping me here or via mail if you need anything else
12:52 groovehunter joined #salt
12:52 joehh jcristau: thanks - good to know you are around and will do so if needed
12:52 joehh just a waiting game at the moment for both 0.17.0 in unstable and the backports
12:56 blee_ joined #salt
12:59 rgbkrk joined #salt
12:59 Kraln joined #salt
13:01 mwillhite joined #salt
13:01 Kraln-_ joined #salt
13:03 oz_akan_ joined #salt
13:06 simonmcc Hi - is there a better way to debug jinja template errors: https://gist.github.com/simonmcc/6e92a7acb7963b626ccb
13:06 timoguin joined #salt
13:11 juicer2 joined #salt
13:12 racooper joined #salt
13:13 helderco joined #salt
13:14 christopherl1 joined #salt
13:17 brianhicks joined #salt
13:19 jrdx joined #salt
13:20 toastedpenguin joined #salt
13:21 agend joined #salt
13:22 pengyao joined #salt
13:22 Drawsmcgraw joined #salt
13:26 lynxman joined #salt
13:28 oz_akan_ joined #salt
13:36 mike_perdide joined #salt
13:37 micah_chatt joined #salt
13:37 jslatts joined #salt
13:37 blee joined #salt
13:41 fllr_ joined #salt
13:42 jslatts joined #salt
13:46 Kholloway joined #salt
13:46 Furao jgelens: https://groups.google.com/forum/#!topic/salt-users/uEABhH7qtSI
13:46 jY if i update a file on the master is the only way to get it synced right away to the minions is force them to run a highstate?
13:48 timoguin jY, that, or you could execute a single state immediately after
13:48 timoguin but i'm sure there's a way you could make it automatic
13:49 jY ok i'll read more on doing that
13:49 Furao cronjob * * * * * * salt-call state.highstate :P
13:49 Furao maybe an extra *
13:52 pengyao scheduler
13:53 jY i do use scheduler to run a highstate every 5 minutes
13:53 timoguin neither of those things are going to trigger it when a file on the master is modified
13:53 jY just wanted to know if it could be instant.. like git pull on master then run some command on master to tell the minions to update that one file
13:53 jY rather then run a highstate on them
13:54 timoguin jY, I'm wondering if the gitfs backend has a way of firing an event when it grabs new commits
13:56 cshuman joined #salt
13:56 timoguin jY, that's the first place i'd dig into
14:00 aberant joined #salt
14:00 mannyt joined #salt
14:01 jY ok thanks
14:03 bemehow joined #salt
14:03 djn since you seem informed about the gitfs backend, do you know if there is a way to force the master to look for new commits?
14:05 Brew joined #salt
14:05 timoguin djn, I'm not sure how it checks the gitfs remotes, but when I've used it my commits got refreshed within a few minutes.
14:05 timoguin But I've never gone in to see how it actually works
14:06 gmoro joined #salt
14:07 djn timoguin: thanks, I guess I'm just not patient enough then :)
14:08 timoguin djn, my testing has been pretty limited so far. :)
14:09 timoguin it'd be ideal to be able to have a POST commit hook that hits the salt-api to tell it to refresh its cache and run highstate
14:09 timoguin should be doable
14:09 jY timoguin: the git part is easy but i can run a single state over just one sls?
14:09 jY i guess that is my question
14:10 timoguin jY, yea you can: salt '*' state.sls core
14:10 timoguin http://docs.saltstack.com/ref/modules/all/salt.modules.state.html
14:10 timoguin so salt '*' state.sls supercoolfilewewannawatch
14:10 timoguin something like that, at least. :)
14:13 webben joined #salt
14:13 webben salt newbie here
14:13 premera_j_n joined #salt
14:14 webben trying to use salt with a win xp client and trying to set up the windows software repo
14:17 webben getting an error about "list indices must be integers not str" when running winrepo.genrepo: http://pastie.org/8392482
14:17 webben have i done something obviously wrong there?
14:18 jkleckner joined #salt
14:19 mgw joined #salt
14:21 ldlework salt runs on windows? neat.
14:21 timoguin the minion does, yea
14:21 ldlework webben: revmap[repodata['full_name']] = pkgname
14:21 ldlework either revmap or repodata is a list, not a dictionary
14:21 ldlework is what that error is complaining about
14:22 lmickh joined #salt
14:23 [M7] joined #salt
14:24 webben Yeah ... there are multiple iterations of the loop where it's a dict then BANG! it's a list
14:24 bemehow joined #salt
14:25 jams_ joined #salt
14:26 webben something to do with package majreg
14:26 jams_ Is there a known issue with successfully upgrading salt using apt-get? I can't get it to 17.0
14:30 joehh jams_: which release are you on now?
14:30 jams_ 0.16.0-2276-g72e9451
14:31 joehh are you getting that from the daily ppa?
14:31 xmltok joined #salt
14:31 krissaxton joined #salt
14:31 jams_ http://ppa.launchpad.net/saltstack/salt/ubuntu/
14:32 joehh and which os? ie raring/precise/...
14:32 jams_ precise/main i386
14:32 Furao 0.17.0 is buggy stays on 0.16.4
14:32 vespakoen joined #salt
14:33 jams_ Furao: maybe so, but I gotsta figure out the upgrade issue first.
14:34 jams_ joehh: any thoughts so far?
14:34 joehh it looks like your existing release was built from git directly - is that the case?
14:34 vespakoen Hey guys (n00by here) I have been playing with salt and watched a couple of talks and I think it it amazing! I have installed vagrant, the salty-vagrant plugin, and got my box up and running, and it seems it is actually doing stuff, but it is ignoring my pillar config. I probably  need to add more stuff to my pillar/top.sls file but I just cannot get it to work, here is what the debug spits out: http://hastebin.com/cewufajudu.cmake I hope
14:34 vespakoen someone can point out what I am doing wrong
14:34 joehh the repo appears to have 0.17.0 in it
14:35 Furao apt-get update; apt-get upgrade
14:35 jams_ hmmm. Maybe. I don't remember, it was a while ago. It might have been from bootstrap script, which was using git directly
14:36 vespakoen for example, in my system state there is this: {% set system = pillar.get('system', {}) %}  {{ system.get('timezone', 'Europe/Stockholm') }}:  How should I define the "timezone" property in my pillar/top.sls file? / what parent's does it need besides system: ?
14:36 xmltok joined #salt
14:38 Furao jams_: pip freeze | grep salt
14:38 Furao then run pip install -U $whatevercontainssalt
14:38 joehh have you tried apt-get install salt-minion
14:39 jams_ joehh: https://gist.github.com/jeremyll/da010022f6528e2e9b6c
14:39 mwillhite joined #salt
14:40 joehh jams_: ok - you have 0.17.0 installed
14:40 joehh you also have a version installed via pip/git or similar
14:40 Furao so you ended with salt 0.13, 0.16 and 0.17 installed in paralels
14:40 joehh I suspect Furao's approach is the way to go
14:40 Furao check if your pregnant too
14:40 jams_ lol
14:40 Furao clean all that
14:41 jams_ so i have a pip version installed at 13.0 and 17.0 via apt-get?
14:41 jams_ and 16.0 is floating somewhere else?
14:41 Furao that don't explains why sudo salt-minion --version report 0.16
14:41 danielmcbawse joined #salt
14:41 Furao dpkg -l |grep salt
14:41 jams_ right..
14:41 Furao dpkg -P all-packages-with-salt
14:41 Furao pip uninstall salt
14:41 Furao find / | grep salt
14:41 joehh I can only spot the 0.16, but Furao is normally more observant than me
14:41 Furao cleanup everything
14:42 Furao and install 0.16.4 from scratch
14:42 jams_ https://gist.github.com/jeremyll/da010022f6528e2e9b6c
14:42 Furao joehh: https://gist.github.com/jeremyll/da010022f6528e2e9b6c#file-gistfile1-txt-L26
14:42 Furao it look like a git sha1 sum
14:43 joehh ahahh - gist got a bit longer since I viewed it initially
14:43 aberant joined #salt
14:47 fragamus joined #salt
14:47 zooz joined #salt
14:48 premera_j_n Hello, I have multiline scalar value in my pillar file, I want to use it in a file through file.managed/template:jinja/context:pillar[], however this value is not outputed literally, instead some characters are escaped, like backslash, so instead of \ I am getting \\ in the generated file. Any advice how to avoid this escaping ?
14:49 vespakoen I have got this pillar file: http://hastebin.com/hubafucoha.yaml and when I run "sudo salt-call --local pillar.items minions" I get: local:\n----------  Does anyone know what's up?
14:50 StDiluted joined #salt
14:51 Furao premera_j_n: it's pipe "|"
14:52 Furao and you have to increase the lines
14:52 Furao increase indentation
14:54 premera_j_n yeah, I think this part works, pillar/yaml gets parsed correctly, but the output gets escaped
14:55 unicoletti left #salt
14:56 jams_ This doesn't make any sense - https://gist.github.com/jeremyll/da010022f6528e2e9b6c#file-gistfile1-txt-L68
14:59 avienu joined #salt
15:01 Furao jams_: dpkg -P salt-minion
15:01 Furao and then reinstall it
15:01 Furao dpkg -i /var/cache/apt/archives/salt-minion_0.16.4-1precise_all.deb
15:01 Furao don't waste time on 0.17.0
15:02 teskew joined #salt
15:05 vespakoen Can anyone help me with getting pillar to work?
15:05 pengyao left #salt
15:06 jalbretsen joined #salt
15:07 premera_j_n Here is how this escape problem looks like, any suggestions appreciated: http://pastie.org/private/jsklouwymffikjdb5ouqg
15:07 rgbkrk joined #salt
15:08 pdayton joined #salt
15:09 alrs joined #salt
15:09 backjlack joined #salt
15:12 vespakoen My pillar data seems to get parsed correctly (OrderedDict([('minions', OrderedDict([('server', ['nginx', 'php', 'php-fpm', 'pagespeed', 'postgresql'])]))]))  but I cannot retrieve the information in my salt states or when debuggin it with sudo salt-call --local pillar.items minions, does anyone have a clue why this doesn't work (I am using Vagrant with the salty vagrant plugin btw)
15:12 gaoyang_ joined #salt
15:13 yota joined #salt
15:14 charlton joined #salt
15:22 d0ugal joined #salt
15:22 zooz joined #salt
15:24 mgw joined #salt
15:31 Furao vespakoen: pillar.data
15:31 vespakoen it shows a whole lot but not the stuff I added in pillar/top.sls
15:32 krissaxton joined #salt
15:32 jkleckner joined #salt
15:34 vespakoen Ahhhh got it!
15:34 vespakoen I needed to add base:\n  '*':\n    - common  and add my config in common/init.sls =)
15:35 vespakoen I thought I could just add my items below there but it seems I need to include another sly to make it work
15:38 rgbkrk Anyone have suggestions about installing and using supervisor via salt?
15:38 rgbkrk Some of the salt states I've seen install it as a package then try to upgrade via pip.
15:39 rgbkrk Which causes the init script to be different than the client supervisor tools.
15:39 rgbkrk Causing mismatches...
15:39 cshuman joined #salt
15:39 vespakoen Is there a way to tell my server it is multiple "hosts" at once?
15:40 jslatts joined #salt
15:40 vespakoen In the future I might have multiple servers for my stuff, but now I want to put them all on one server, but I want my config to be future proof
15:40 higgs001 joined #salt
15:42 Marion joined #salt
15:42 bemehow joined #salt
15:44 vespakoen seems minionswarm.py is what i need
15:45 Marion_ joined #salt
15:45 JaredR joined #salt
15:48 druonysus joined #salt
15:49 druonysus joined #salt
15:49 CheKoLyN joined #salt
15:50 adepasquale joined #salt
15:54 uta joined #salt
15:55 jaequery joined #salt
15:57 pipps joined #salt
16:00 redondos joined #salt
16:00 redondos joined #salt
16:06 micah_chatt joined #salt
16:09 pipps joined #salt
16:09 seanz Greetings. I've got a state file that extends another, but the extended actions are not being called until the second highstate call. Any suggestions?
16:09 christopherl joined #salt
16:13 DredTiger joined #salt
16:15 charlton is there any way to run salt on multiple named machines?  e.g. salt 'server1', 'server2' state.highstate?
16:16 supplicant salt -L server1,server1
16:16 supplicant ?
16:16 cachedout salt 'server[1-2]' state.highstate
16:17 jbunting joined #salt
16:17 charlton perfect, thanks!
16:17 cachedout charlton: Here's some documentation that should help as well. http://docs.saltstack.com/topics/targeting/globbing.html
16:18 KyleG joined #salt
16:18 KyleG joined #salt
16:18 charlton cachedout:  thx!
16:18 cachedout Happy to help. :]
16:19 liamc joined #salt
16:19 charlton ah yes, i went through this before and had forgotten
16:19 pipps joined #salt
16:22 pipps joined #salt
16:22 Lue_4911 joined #salt
16:24 robbyt joined #salt
16:28 avienu joined #salt
16:30 mgw1 joined #salt
16:30 pipps joined #salt
16:32 pipps joined #salt
16:33 krissaxton joined #salt
16:35 aberant joined #salt
16:36 christopherl joined #salt
16:42 Ahlee compound matching, is there a way to order?  salt -C 'G@grain1 and G@grain2 or G@grain3', i want 1 and (2 or 3)
16:51 danielmcbawse joined #salt
16:52 copelco joined #salt
16:53 mofomikes joined #salt
16:54 pdayton joined #salt
16:54 ktenney joined #salt
16:54 christopherl left #salt
16:55 Jahkeup joined #salt
16:55 seanz What's the best way to go about making a salt state available to the world?
16:56 andrewclegg freaky problem -- trying to install nodejs and npm from yum, it's somehow returning *before* the packages are installed, so later steps in my states that depend on it are failing
16:56 seanz I've done some work setting up a PHP lightweight avatar server salt state that could possibly be useful to someone else.
16:56 andrewclegg they are running after the pkg.installed for npm, but yum somehow isn't finished by then!
16:57 andrewclegg why would salt run yum in the background?
16:57 andrewclegg this is redhat 6 and salt 0.16.4
16:58 Ahlee salt's asynchronous
16:58 Ahlee look into cmd.wait
16:59 seanz Never mind, I can use salt-states, potentially.
17:01 rgbkrk joined #salt
17:02 andrewclegg Ahlee: not on an individual minion, it isn't -- it's supposed to complete one command before running the next in the chain
17:02 andrewclegg actually looking at my logs, it may be a different problem -- it may be failing to respect a "require" and running two states in the wrong order
17:03 andrewclegg i.e. I have something that has "require: - pkg: npm" but this is getting run before "npm: pkg.installed"
17:09 bhosmer_ joined #salt
17:09 Marion left #salt
17:09 uta every runner function now only returns "local:    ----------" ...including all the default runners
17:09 uta is this broken in 0.17.0?
17:13 rgbkrk How do you use a module with no arguments?
17:13 rgbkrk supervisord.reread in this case
17:13 rgbkrk reread:
17:13 rgbkrk supervisord.reread
17:13 rgbkrk ?
17:13 rgbkrk I'm getting various complaints about how I have this laid out
17:14 pkimber joined #salt
17:16 mofomikes joined #salt
17:18 seanz rgbkrk: Pastebin?
17:19 pipps joined #salt
17:19 rgbkrk I'll give you something better
17:19 rgbkrk https://github.com/rgbkrk/salt-nbviewer/blob/supervisory/supervisor/init.sls
17:20 rgbkrk seanz - I will also try not to be embarrased that I don't know what I'm doing with salt + yaml. Brain may be stuck in Chef + Berkshelf mode.
17:20 taylorgumgum joined #salt
17:20 seanz rgbkrk: It's not a problem. I'm definitely no expert, but I think I can help with this.
17:20 seanz supervisor:
17:20 rgbkrk That one won't work, clearly, but I'm not sure what the syntax is.
17:20 seanz pkg.installed
17:20 andrewclegg my bad -- pilot error
17:20 seanz That's it.
17:20 seanz Get rid of the colon.
17:20 rgbkrk Oh the install is fine.
17:20 rgbkrk https://github.com/rgbkrk/salt-nbviewer/blob/supervisory/supervisor/init.sls#L31-L33
17:21 rgbkrk I'm wondering how to do the reread and update
17:21 seanz Ah.
17:21 rgbkrk At the bottom
17:21 mastrolinux left #salt
17:21 seanz When is that supposed to happen?
17:21 rgbkrk # salt '*' supervisord.reread # This works beautifully
17:21 rgbkrk $ salt '*' supervisord.reread # This works beautifully
17:21 rgbkrk At the end
17:21 mofomikes can anyone direct me to more documentation on writing state stanzas?
17:22 seanz rgbkrk: That command doesn't exist in a state file.
17:22 rgbkrk :(
17:22 seanz rgbkrk: http://docs.saltstack.com/search.html?q=states.supervisord&check_keywords=yes&area=default
17:22 seanz Those are your options.
17:23 rgbkrk Oh
17:23 rgbkrk So that's a module
17:23 anuvrat joined #salt
17:23 ldlework That's just like, your options mann.
17:23 rgbkrk How would I automate using those?
17:23 rgbkrk Can I?
17:23 rgbkrk lol
17:23 seanz What do you mean, automate?
17:23 seanz What's your goal?
17:23 rgbkrk http://docs.saltstack.com/ref/modules/all/salt.modules.supervisord.html?highlight=supervisor#salt.modules.supervisord.reread
17:23 rgbkrk Well, automate a deployment
17:24 seanz You can ensure supervisord is running with .running
17:24 rgbkrk supervisor isn't actually rereading the config file when my conf file gets laid there
17:24 viq yaaay! https://gist.github.com/viq/6920878
17:24 seanz Perhaps use a - watch command?
17:24 seanz You can watch any other part of the state file, or other state files if they're included with '- include'
17:25 seanz "watch" accepts a list of other "function: id" parameters too, which is nice.
17:25 rgbkrk So I watch that conf file, which restarts supervisor as a service
17:25 seanz You can have the service restart if any number of things change.
17:25 seanz Yes.
17:26 rgbkrk https://github.com/rgbkrk/salt-nbviewer/blob/supervisory/supervisor/init.sls#L6-L7
17:26 rgbkrk It's doing that now, but the conf doesn't get updated.
17:26 rgbkrk In fact, it currently doesn't load any
17:27 rgbkrk I'm going to dig into the init script. I feel like I'm getting a collision between supervisor installed as a package and supervisor installed by pip.
17:27 seanz rgbkrk: Yeah, you're beyond me now. Good luck with it.
17:27 rgbkrk Which I used from https://github.com/ConsumerAffairs/salt-states/blob/master/supervisor.sls
17:27 rgbkrk Thanks for the help.
17:28 seanz rgbkrk: It should be a convention to quote the word "help" when the help *didn't* help. :)
17:28 seanz As in: Thanks for the "help".
17:28 jdenning joined #salt
17:28 seanz I'm glad to have confirmed what you already knew, though.
17:28 rgbkrk Oh I'm happy to have someone to chat with.
17:28 rgbkrk Sorry on my part really, as I don't know how to debug supervisor issues.
17:29 rgbkrk This does tell me I'm not using salt states appropriately though.
17:29 rgbkrk As in I had no access to that module.
17:30 bemehow joined #salt
17:31 seanz It seemed like it might be an issue outside of salt.
17:33 krissaxton joined #salt
17:35 liamc joined #salt
17:36 pentabular joined #salt
17:39 bemehow joined #salt
17:42 Jahkeup joined #salt
17:50 matanya_ joined #salt
17:51 bemehow joined #salt
17:54 bemehow_ joined #salt
17:56 mwillhite joined #salt
18:00 AdamSewell joined #salt
18:01 jefimenko joined #salt
18:04 brimpa joined #salt
18:05 jkleckner joined #salt
18:08 amahon joined #salt
18:10 baccenfutter left #salt
18:14 bhosmer joined #salt
18:15 bleh joined #salt
18:22 timoguin_ joined #salt
18:22 mwillhite joined #salt
18:26 jbunting joined #salt
18:29 jrb28 joined #salt
18:30 cachedout left #salt
18:30 bemehow joined #salt
18:31 bemehow joined #salt
18:32 higgs001 joined #salt
18:33 taylorgumgum joined #salt
18:34 krissaxton joined #salt
18:37 jrhorn424 joined #salt
18:39 felixhummel joined #salt
18:41 AdamSewell Is anyone familiar with salt cloud here? I installed the package but it didn't create the /etc/salt/cloud conf file. Is there an example anywhere of that file?
18:42 pentabular joined #salt
18:48 alrs joined #salt
18:49 giantlock joined #salt
18:53 jkleckner joined #salt
18:53 ldlework AdamSewell: I don't think its supposed to generate anything by default
18:53 ldlework http://salt-cloud.readthedocs.org/en/latest/topics/config.html
18:55 AdamSewell ldlework, ah thanks
18:56 zooz joined #salt
19:00 mua joined #salt
19:02 mofomikes joined #salt
19:03 copelco joined #salt
19:04 george_torwell joined #salt
19:06 taylorgumgum joined #salt
19:09 mapu joined #salt
19:09 troyready joined #salt
19:10 pipps joined #salt
19:12 snuffeluffegus joined #salt
19:14 adepasquale joined #salt
19:18 zwe joined #salt
19:19 mwillhite joined #salt
19:21 Brew joined #salt
19:22 bhosmer joined #salt
19:23 dave_den joined #salt
19:24 streetmapp joined #salt
19:27 jbunting joined #salt
19:32 bemehow joined #salt
19:34 krissaxton joined #salt
19:39 fragamus joined #salt
19:40 copelco joined #salt
19:42 honestly joined #salt
19:44 mofomikes joined #salt
19:46 ggoZ joined #salt
19:47 jonwincus joined #salt
19:49 bhosmer joined #salt
20:00 bemehow joined #salt
20:03 jimallman joined #salt
20:05 jefimenko joined #salt
20:06 copelco joined #salt
20:08 higgs001 joined #salt
20:09 mofomikes joined #salt
20:12 jrhorn424_ joined #salt
20:13 jpcw joined #salt
20:13 aberant joined #salt
20:17 A21BC joined #salt
20:18 jdenning_ joined #salt
20:18 Marion_ joined #salt
20:20 andrej Hmmm ...
20:20 cro joined #salt
20:21 jrhorn424 left #salt
20:21 jrhorn424 joined #salt
20:21 leo_33 joined #salt
20:21 leo_33 if you go to an interview and there is no response after 3 weeks is it a no for sure?
20:22 timoguin_ well it's definitely not a hopeful anymore, i don't think
20:22 rgbkrk Wowsers. Who did you interview with leo_33 that they haven't written back?
20:22 rgbkrk Usually they have the courtesy to let you know quickly.
20:22 andrej One of the negative side-effects of the 0.16.4 -> 0.17 upgrade that comes from the minions changing their IDs is a resulting slow-down of commands run against '*' ... the two old IDs delay things till they time out
20:23 leo_33 nokia siemens networks
20:23 Marion_ leo_33: nope.  How big is the company?  Many times, bigger companies have more red tape and so the process is slower.  Also, is it possible the gov't shutdown is affecting it?  There are many unobvious ways companies are affected by this.
20:24 leo_33 it is very big. also they are recruiting for many positions and they said they wont move to the next step until all interviews are over
20:24 Marion_ leo_33: that's a big company.  I would send an email saying you are just touching base.  That you haven't heard back and are wondering if a decision has been made or if they are still working things.
20:24 leo_33 what does size have to do with it?
20:24 leo_33 i asked the recruiter
20:24 leo_33 i asked if many applied and he answered this:
20:25 leo_33 many people but not so many suitable. all people from me were highly suitable of course ;) but they hiring a lot of people so that means they have quite a heavy load with these interviews
20:25 leo_33 (not just for these positions but also others)
20:25 timoguin_ leo_33, the larger the size of the company, the slower things move typically
20:25 rgbkrk It sounds like they like you but are waiting on an offer they gave out to someone.
20:26 timoguin_ and/or the recruiter is just keeping him happy
20:26 Marion_ They just have more red-tape.  More management involved.  More times higher up decisions that may put a hold on hiring process.
20:26 leo_33 so as a reserve?
20:26 timoguin_ maybe
20:26 Marion_ you may or may not be a reserve.  But, really, it doesn't matter.
20:26 leo_33 but the recruiter has to keep me happy right?
20:28 pipps joined #salt
20:29 rgbkrk You're probably fine, but I would ask the recruiter about a status.
20:29 mapu Good afternoon- I am having some difficulty with alt formulas (mysql, apache)
20:30 mapu I keep getting Rendering SLS apache failed errors. Specifically,
20:30 mapu TypeError: filter_by() got an unexpected keyword argument 'merge'
20:30 rgbkrk Are you looking other places too leo_33? I should probably move this discussion out of #salt
20:31 taylorgumgum joined #salt
20:31 Marion_ I have a question that may or may not be salt.  I have my master up, have my minion (vagrant box) up.  I can ssh in.  salt smu.com test.ping doesn't return anything.  What does that mean?
20:32 honestly is the minion accepted?
20:32 honestly is "smu.com" the minion id?
20:32 Marion_ salt-key -L says it's accepted
20:33 Marion_ where is the minion id set?  in the config file called minion?
20:33 honestly yes
20:34 Marion_ no, I do not have a "id: " in my minion file.  I will add.
20:35 krissaxton joined #salt
20:35 alrs joined #salt
20:36 Marion_ after adding this, I only have to do a vagrant provision, right?
20:36 honestly the id autoconfig is a fallback and doesn't always behave as expected
20:37 honestly no idea how vagrant ties into that
20:37 Marion_ k
20:37 msheiny joined #salt
20:38 agtilden joined #salt
20:53 intchanter joined #salt
20:54 alunduil joined #salt
20:55 bhosmer joined #salt
20:56 mofomikes joined #salt
20:57 thehyperadvisor joined #salt
21:03 Kraln-__ joined #salt
21:03 Kraln-___ joined #salt
21:05 Kraln-____ joined #salt
21:07 Kraln joined #salt
21:08 bhosmer_ joined #salt
21:09 Kraln-_ joined #salt
21:10 cachedout joined #salt
21:11 Kraln-__ joined #salt
21:12 Kraln-___ joined #salt
21:13 pdayton joined #salt
21:14 Kraln joined #salt
21:15 xmltok joined #salt
21:19 Kraln- joined #salt
21:21 Kraln-__ joined #salt
21:23 djn any ideas how to share a password (for bacula btw) between two hosts? would you use the pillar to do that, or are there any other, more convenient alternatives?
21:24 cedwards djn: probably pillar
21:24 jbunting joined #salt
21:26 mwillhite joined #salt
21:27 Kraln- joined #salt
21:27 robertkeizer joined #salt
21:27 zwevans joined #salt
21:28 Marion_ joined #salt
21:29 jarias joined #salt
21:32 djn cedwards: is there a way to interact with the pillar from a state file? or more explicit: what exactly does pillar.ext do? can I use it to populate the pillar dynamically (from pwgen)?
21:33 cedwards you can refer to pillar values inside your state files, yes
21:34 cedwards i don't know about calling pwgen directly. pillar.ext is simply an external key value store
21:36 krissaxton joined #salt
21:36 Kraln joined #salt
21:38 djn yes, it's clear I can read them, but I actually want to populate it.. I could prepopulate it for all minions though, which should work without issue
21:41 Thiggy joined #salt
21:42 danielmcbawse joined #salt
21:44 pipps joined #salt
21:45 Kraln- joined #salt
21:46 cachedout left #salt
21:49 mofomikes joined #salt
21:50 jarias_ joined #salt
21:52 CheKoLyN joined #salt
21:52 zwevans hey folks.  Wondering if anyone has run into issues running salt-ssh from ubuntu master 0.17?  I'm seeing NameError: global name 'thintar' is not defined.
21:52 zwevans If I view pymodules/salt/utils/thin.py there seem to be some lines missing from the gen_thin method that set up the thintar variable.  If I view that same method on github develop or 0.17 tag... method looks correct.  Hopefully I'm missing something obvious..
21:57 A21BC joined #salt
21:58 jbunting joined #salt
22:01 bhosmer joined #salt
22:02 pentabular left #salt
22:03 rgbkrk What's the right way to setup environments?
22:03 rgbkrk Set them per minion?
22:03 rgbkrk In this case, I'm hoping to group them for a gitfs backed deployment.
22:07 jkleckner1 joined #salt
22:08 jkleckner1 left #salt
22:08 mmmbbbb joined #salt
22:09 mmmbbbb Hello there, anyone around for a Newbie question?
22:09 kermit joined #salt
22:09 diegows joined #salt
22:10 mianos joined #salt
22:10 mmmbbbb I have a salt-minion on a VM. When I save the state of the vm and bring it back up, the minion no longer talks to the master node, until I bounce the service.  Any ideas?
22:12 lemao joined #salt
22:13 copelco1 joined #salt
22:13 zwe joined #salt
22:14 zwe joined #salt
22:19 bhosmer joined #salt
22:19 rgbkrk mmmbbbb - Are you changing networks when you bring it back up?
22:19 rgbkrk I usually have trouble with vagrant/virtualbox if I've moved my laptop from my work network to my home network
22:21 bhosmer joined #salt
22:23 rgbkrk Anyone have tips for using gitfs + environments?
22:23 rgbkrk The docs aren't giving me much detail.
22:23 Corey rgbkrk: Hmm.
22:23 Corey Not offhand.
22:23 Marion_ honestly: after playing around with things (killing and restarting salt-master, halting my VM, reprovisioning my vm, etc), I don't have any keys when sudo salt-key -L.  Any suggestions?
22:23 Corey rgbkrk: And your vagrant issues should go away if you cut over to VMware instead of vbox.
22:24 Corey Marion_: Have your minions dump their keys and reregister.
22:24 rgbkrk K. I've got 5 minions, connected to master.
22:24 rgbkrk I'm just not sure how I'm supposed to set environments.
22:24 Marion_ Corey:  how do I do have my minions dump their keys and reregister?
22:25 Marion_ Corey:  How easy is it to switch from vbox to VMware?  Does a provisioner need to be purchased from VMware?
22:25 Corey Marion_: rm -rf /etc/salt/pki/*; restart the minion.
22:25 rgbkrk The branches get mapped to environments I see, http://docs.saltstack.com/topics/tutorials/gitfs.html?highlight=gitfs#branches-environments-and-top-sls-files
22:25 Corey Marion_: Once you pay your money it should be transparent. I've just had shite luck with vbox's networking.
22:26 rgbkrk But I'm not sure how to set environemtns
22:26 pipps joined #salt
22:27 Marion_ Corey:  I take it … yes, a provisioner has to be purchased from VMware.
22:27 Marion_ I already own VMware Fusion and use it to have a Windows env on my mac.
22:31 redondos joined #salt
22:31 redondos joined #salt
22:34 druonysuse joined #salt
22:34 druonysuse joined #salt
22:35 Marion_ I have used vagrant-puppet.  I was thrown into it a bit like I was thrown into vagrant-salt.  By starting to work on a project that was configured that way.  I did not have near the issues with vagrant-puppet that I have had with vagrant-salt.  I am truly frustrated.  I feel like I am really close, but … just not there.  And close doesn't count in VM configuration!
22:36 jcockhren Marion_: you're back!
22:36 jcockhren \o/
22:36 rgbkrk Ok, posted my question to StackOverflow
22:36 rgbkrk http://stackoverflow.com/questions/19307410/setting-up-environments-with-saltstack
22:36 krissaxton joined #salt
22:37 Marion_ yes, jcockhren:  I've been back off & on over the last few days.  I'm understanding more.  I downgraded to vagrant 1.3.2
22:38 bhosmer_ joined #salt
22:38 pentabular joined #salt
22:38 alunduil joined #salt
22:39 jcockhren Marion_: what part isnt working for you
22:39 jcockhren ?
22:40 Marion_ I've stopped everything.  vagrant destroy.  starting from the beginning.  Will tell you when I get there.
22:40 jcockhren ok cool. I'm heading home. I'll read the scrollback in a few
22:41 Marion_ thx!
22:43 andrej One of the negative side-effects of the 0.16.4 -> 0.17 upgrade that comes from the minions changing their IDs is a resulting slow-down of commands run against '*' ... the two old IDs delay things till they time out
22:43 andrej I wonder if closing the related report on github means that they consider it a feature?
22:44 jkleckner joined #salt
22:44 drags andrej: this report? https://github.com/saltstack/salt/issues/7558
22:44 jaequery joined #salt
22:44 andrej yah
22:44 drags that report is them fixing the ID generation stuff (fixing, aka reverting to the original way)
22:44 drags it just hasn't made it to release yet
22:45 andrej Ah OK
22:45 drags (though they also add ID caching, which I added a comment to argue against :))
22:45 andrej I read that
22:45 andrej And I second that motion
22:45 drags I would really like to see the master keep a hash table lookup of keys, and be able to notice when a minions ID has changed without requiring re-keying
22:45 andrej In general I don't think that the hostname (or anything mutable) should be used in identifying the minion
22:45 drags I want to put an issue in for that, but I wanted to have some code to go with it first
22:47 andrej Well ... I'll keep playing w/ 0.17 in the limited current set-up ...
22:47 drags andrej: I think for new users, and folks using Salt for it's excellent parallel job running capabilities, that targeting by current hostame (and having that be mutable if hostname changes) is very good
22:47 andrej Still so much to learn :)
22:47 drags andrej: also you can restore your minion IDs by adding an id: line to the /etc/salt/minion config
22:48 drags (HN=`hostname --fqdn`; echo "id: $HN" >> /etc/salt/minion)
22:48 andrej drags : that's good to know, and feasible while I only play w/ a few of them ... w/ a huge number it would become slightly tedious ;)
22:48 drags andrej: indeed
22:49 drags I almost did it so I could roll .17 to my minions last night (thankfully I tried to run test.ping against the master after I upgraded it and noticed the ID changing issue :), but I really want mutable minion IDs
22:49 andrej you do? elaborate?
22:49 drags I don't have anywhere on my roadmap to convert from puppet to salt for a long long time, so I primarily use salt for doing quick jobs in parallel, and having the hostnames and minion IDs drift from each other is jarring as a user experience
22:49 drags andrej: yeah it's due to how I'm primarily using salt
22:50 andrej Aight
22:50 drags I have clusters of hosts that recently went from beta to production, and all were renamed to remove 'scary-' from the front of their hostname
22:50 andrej We currently have a small environment, but do expect future growth
22:50 drags and I had a group of hosts with inconsistent hostnames that I changed to all be consistent
22:51 andrej and are currently NOT using any config management tools ... for quick hacks I use cssh (only started here a few month ago)
22:51 drags it was jarring to have to restart the minions and re-add the keys and remove  the old keys just to get `salt` working normally again
22:51 andrej I hear yah
22:53 jetblack joined #salt
22:54 mofomikes joined #salt
22:55 pdayton joined #salt
22:56 cbloss if I set a variable in one saltstate and include another, how can I make sure that same variable gets passed in the include without using pillars?
23:02 jslatts joined #salt
23:04 druonysus joined #salt
23:04 druonysus joined #salt
23:04 danielmcbawse joined #salt
23:15 jrhorn424 left #salt
23:16 pipps joined #salt
23:16 Ryan_Lane I have a weird problem
23:16 Ryan_Lane deployment_config = __pillar__.get('deployment_config')
23:17 Ryan_Lane that returns None
23:17 Ryan_Lane salt-call pillar.data deployment_config
23:17 Ryan_Lane that returns a hash
23:17 Ryan_Lane a similar hash returns perfectly fine in the same code block
23:18 Ryan_Lane ugh. restarting the salt minion seems to have fixed it
23:18 Ryan_Lane something must have been cached poorly
23:28 berto- joined #salt
23:33 terminalmage joined #salt
23:37 krissaxton joined #salt
23:37 berto- i have a dynamic pillar being created with an API call to an external service.  running `pillar.data` shows me the data just fine, but making a compound selection, `salt -C 'I@server_cluster:default'` doesn't work.  any ideas?
23:44 mgw1 berto-: even after running refresh_pillar?
23:45 agtilden joined #salt
23:47 berto- mgw1: saltutil.refresh_pillar magic worked!  what does that do behind the scenes?
23:47 berto- does that have to run every time pillar data is changed?
23:48 xmltok_ joined #salt
23:48 mgw1 berto-: it gets run on a highstate, but otherwise, I believe so
23:49 berto- mgw1: hmm, there might be a bug in there somewhere.  i've run highstate on these systems plenty of times and the compound lookup didn't work until i refreshed as you suggested.
23:50 mgw1 berto-: hmmm, maybe highstate pulls data for the state run, but doesn't cache it unless you call refresh_pillar. I'm not 100% sure what the intended behavior is.
23:51 mgw1 Maybe basepi or one of the other salt experts on here can answer that.
23:51 mgw1 Or you could post to the mailing list.
23:52 berto- mgw1: i'll ask here in the morning.
23:52 basepi I know terminalmage was looking into a bug related to pillar/grains not syncing on highstate. They should be syncing, but don't always. Don't know if it got resolved, can't remember.
23:53 berto- mgw1: but, thanks!  you solved my problem.  :)
23:53 mgw1 basepi, do you have a ticket#?
23:53 berto- basepi: i'll +1 the bug report.
23:53 berto- it would be great if salt just auto-cached on high state.
23:54 basepi mgw1: not ATM. On my phone.
23:54 mgw1 basepi: ok, thanks
23:55 berto- this might be it: https://github.com/saltstack/salt/issues/5716
23:55 mgw1 https://github.com/saltstack/salt/issues/5716
23:55 mgw1 yeah
23:56 mgw1 I just found it too
23:56 mgw1 I think so
23:58 basepi Yep, that looks right.
23:58 mgw1 so it looks like it's fixed in develop, I can't tell if it's been cherry-picked for 0.17.1, but it seems like it should be
23:58 berto- lots of discussion and the fix is merged into develop: https://github.com/saltstack/salt/pull/7630
23:59 berto- when's 0.17.1's release date?
23:59 mgw1 berto-: so you probably had a broken pillar at some point
23:59 mgw1 in your development

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