Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2013-08-23

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

All times shown according to UTC.

Time Nick Message
00:08 middleman_ joined #salt
00:13 teebes joined #salt
00:13 VertigoRay UtahDave: I think I've got my config figured out.  One thing, I'm not sure about is how pillar would be set-up.
00:13 VertigoRay oh ... dang.  you're afk  ;)
00:14 littleidea joined #salt
00:14 jaequery joined #salt
00:17 intchanter joined #salt
00:18 nrub joined #salt
00:27 viq joined #salt
00:29 cewood joined #salt
00:30 middleman_ joined #salt
00:30 jaequery joined #salt
00:30 blee joined #salt
00:30 elbaschid joined #salt
00:33 racooper joined #salt
00:46 xmj joined #salt
00:50 dthom91 joined #salt
01:00 middleman_ joined #salt
01:08 Gifflen joined #salt
01:09 kleinishere joined #salt
01:10 liuyq joined #salt
01:12 zwevans joined #salt
01:13 nrub joined #salt
01:13 liuyq joined #salt
01:16 m_george|away joined #salt
01:17 m_george left #salt
01:19 Nexpro1 joined #salt
01:24 LucasCozy joined #salt
01:26 mmilano joined #salt
01:28 KyleG joined #salt
01:28 KyleG joined #salt
01:29 juanlittledevil question for you guys… how do I go about using info from network.interfaces in particular the address associated with say bond0 in a jinja template for a file? Is that something I can use just like a grain or pillar in the state file?
01:33 UtahDave juanlittledevil: well, that info usually is in the grains.
01:33 juanlittledevil but the grains won't tell me what interface the ip address is associated with.
01:33 juanlittledevil as far as I can tell at least.
01:33 oojacoboo joined #salt
01:35 UtahDave what version of Salt are you on?
01:36 oojacoboo left #salt
01:36 juanlittledevil 0.16.0-1
01:36 UtahDave try this:  sudo salt \* grains.item ip_interfaces
01:37 juanlittledevil yup this totally works!
01:37 juanlittledevil thanks Dave!
01:37 middleman_ joined #salt
01:39 jaequery joined #salt
01:40 Jahkeup joined #salt
01:42 akoumjian UtahDave: I've got one for you. Trying to test key preseeding on salty-vagrant. Public key is correctly getting sent to /etc/salt/pki/master/minions/minion_id Also showing up correctly in salt-key -L as accepted.
01:42 akoumjian UtahDave: However salt-run manage.up says the minion is down, and the minion log says that the master has rejected the public key.
01:42 akoumjian Master log says the keys don't match, but running a diff on /etc/salt/pki/master/minions/minion_id and /etc/salt/pki/minion/minion.pub produces 0 differences
01:43 UtahDave hmm Is this on latest from git develop?
01:43 middleman_ joined #salt
01:44 akoumjian UtahDave: no, 16.3
01:45 akoumjian But let me test git real quick too
01:45 UtahDave cool
01:46 ipmb joined #salt
01:48 mgw joined #salt
01:50 juanlittledevil is it safe to have a state for the salt-minion package an ensure it's always at latest?
01:52 UtahDave It just depends if you like living on the edge!  :)  Some sysadmins would smack you for doing that. Others wouldn't mind so much.
01:53 akoumjian UtahDave: Experiencing same issue on latest develop
01:53 juanlittledevil I wouldn't call it leving on the edge if the package comes from a repo like say epel
01:53 juanlittledevil UtahDave: but what I mean is more like can the minion update itself safely.
01:53 juanlittledevil ?
01:55 UtahDave It should be fine.  We do have reports on occasion where the minion has a hard time restarting itself.
01:55 UtahDave I think it depends on the OS's service manager
01:55 jeddi juanlittledevil: i've got my stuff configured like that, but a) it's not massively mission-critical, and b) I don't cron or automatically run any salt highstates - so i notice if it's about to make a change before it happens.
01:55 UtahDave akoumjian: I'm not sure what's going on there. I think you ought to open an issue so we can get to that.
01:58 juanlittledevil jeddi & UtahDave: yes I don't intend to run salt unattended, but I do intend to use it in production eventually. I will do some testing… alternatively I could just write some kind of watchdog for the minion process. I suppose I COULD use inittab hehe.
01:59 middleman_ joined #salt
02:05 teebes joined #salt
02:05 jacksontj joined #salt
02:08 saurabhs left #salt
02:12 xl1 joined #salt
02:15 chrisgilmerproj joined #salt
02:17 chrisgilmerproj left #salt
02:22 sssslang joined #salt
02:22 middleman_ joined #salt
02:36 middleman_ joined #salt
02:37 jacksontj joined #salt
02:40 oz_akan_ joined #salt
02:43 mgw joined #salt
02:44 mgw joined #salt
02:45 middleman_ joined #salt
02:51 whit joined #salt
02:54 kermit joined #salt
02:59 middleman_ joined #salt
03:07 jaequery joined #salt
03:10 intchanter joined #salt
03:10 Ryan_Lane joined #salt
03:14 malinoff joined #salt
03:14 malinoff Hi all
03:15 jeddi juanlittledevil: systemd ftw :)   but yeah - i think similarly to you that if you've got a repo that's trustworthy, most of your risk is removed.    there haven't been many (that i've heard of) backwards compatibility problems so far.  but still, it's probably considered best practice to not automatically update the thing that you use to automatically update everything else on all your machines :)
03:16 dthom91 joined #salt
03:17 juicer2 joined #salt
03:24 littleidea joined #salt
03:25 middleman_ joined #salt
03:30 aat joined #salt
03:44 mgw joined #salt
03:47 yidhra joined #salt
03:50 dthom91 joined #salt
03:57 Jahkeup joined #salt
04:01 middleman_ joined #salt
04:07 xenoxaos joined #salt
04:18 middleman_ joined #salt
04:26 malinoff joined #salt
04:35 jalbretsen joined #salt
04:37 Anb___ joined #salt
04:38 dthom91 joined #salt
04:41 middleman_ joined #salt
04:45 jacksontj joined #salt
04:46 malinoff joined #salt
04:49 druonysuse joined #salt
04:49 druonysuse joined #salt
04:49 namtab joined #salt
04:50 middleman_ joined #salt
04:55 z3uS joined #salt
04:59 middleman_ joined #salt
05:06 z3uS joined #salt
05:08 rgarcia_ joined #salt
05:08 malinoff joined #salt
05:10 middleman_ joined #salt
05:10 rgarcia_ joined #salt
05:13 oz_akan_ joined #salt
05:19 waverider joined #salt
05:23 middleman_ joined #salt
05:25 agend joined #salt
05:26 druonysuse joined #salt
05:29 malinoff joined #salt
05:41 kleinishere joined #salt
05:42 dthom91 joined #salt
05:51 pdayton joined #salt
05:51 middleman_ joined #salt
05:55 elbaschid joined #salt
05:55 jalbretsen joined #salt
05:57 jalbretsen left #salt
05:58 druonysuse joined #salt
05:59 middleman_ joined #salt
06:03 waverider joined #salt
06:04 linjan_ joined #salt
06:07 middleman_ joined #salt
06:10 MrTango joined #salt
06:13 dthom91 joined #salt
06:15 middleman_ joined #salt
06:19 dthom91 joined #salt
06:19 pdayton joined #salt
06:22 xmj what could possibly go wrong! ? :D
06:28 juanlittledevil joined #salt
06:33 sssslang just finished reading cfengine's doc, then I known salt... sigh
06:37 balboah_ joined #salt
06:45 druonysuse joined #salt
06:52 vaxholm_ joined #salt
06:53 jpcw joined #salt
06:56 TheCodeAssassin joined #salt
06:57 TheCodeAssassin hi everyone
07:00 Nexpro joined #salt
07:00 ml_1 joined #salt
07:02 malinoff Hi
07:07 isomorphic joined #salt
07:14 carlos joined #salt
07:16 TheCodeAssassin hi malinoff
07:17 Lue_4911 joined #salt
07:24 gildegoma joined #salt
07:24 dthom91 joined #salt
07:28 ggoZ joined #salt
07:28 scott_w joined #salt
07:30 grunge joined #salt
07:37 ml_1 joined #salt
07:40 az87c_ joined #salt
07:49 felixhummel joined #salt
07:52 kleinishere joined #salt
07:54 it_dude joined #salt
07:57 ggoZ joined #salt
08:00 ml_11 joined #salt
08:00 unicoletti_ joined #salt
08:02 ricicle joined #salt
08:02 entil joined #salt
08:03 entil hi guys!
08:03 entil is there special magic involved in including a pydsl sls or is it just me? probably just me ;) include: - foo.postgresql_conf_render says Unknown include
08:03 entil it's not available on the salt master in environment(s): base
08:04 entil I've tried that kind of absolute path, and just postgresql_conf_render and pretty much everything
08:08 gildegoma joined #salt
08:15 entil oh whoah, I screwed up the path, there was supposed to be one more element! Now I got a valid python error, so I'll work from here
08:19 LotusBleu joined #salt
08:24 zooz joined #salt
08:31 Xeago joined #salt
08:33 Xeago joined #salt
08:34 Xeago joined #salt
08:36 nielsbusch joined #salt
08:43 xl1 joined #salt
08:45 ricicle joined #salt
09:18 felixhummel joined #salt
09:26 bhosmer joined #salt
09:27 liuyq joined #salt
09:32 chubrub joined #salt
09:34 chubrub Hi all ;] I'm having top.sls file and I'd like to match states in following way: 'prod*|testing*|staging*': \n - some states here. Is it possible?
09:44 chubrub sorry, stupid question :/
09:44 chubrub rtfm ;]
09:50 lemao joined #salt
09:57 kleinishere joined #salt
10:09 whiskybar joined #salt
10:11 Damoun joined #salt
10:14 TJ1980 joined #salt
10:22 giantlock joined #salt
10:24 slow_ joined #salt
10:33 bemehow joined #salt
10:40 ricicle joined #salt
10:51 Spads joined #salt
11:01 aat joined #salt
11:03 david_a joined #salt
11:15 whiskybar joined #salt
11:24 blee joined #salt
11:28 n3storm joined #salt
11:28 n3storm good morning
11:28 n3storm is there a way to add a repository to debian sources.list?
11:29 n3storm http://docs.saltstack.com/ref/modules/all/salt.modules.apt.html doesn't show a add_repo method
11:29 n3storm what I want to achieve is add mariadb repositories to a debian 7 setup
11:30 n3storm thanks
11:30 unicoletti n3storm: mod_repo will create it if does not exist
11:30 n3storm oh my!
11:31 n3storm thanks unicoletti
11:31 n3storm starting salt is fast
11:31 n3storm but may be now I have to slow down and read more carefully
11:31 unicoletti it's addictive too ;-)
11:32 n3storm yeah!
11:34 xl1 left #salt
11:39 entil oh whoah, I've never even heard of mod_repo
11:39 entil but I can't get a file to render with pydsl, sheesh. it's as if the template-enabled version of the raw file doesn't get through
11:40 entil did you guys ever see anything like this?
11:40 entil shared_buffers = {{ shared_buffers }}MB                 # min 128kB
11:40 entil then my pydsl has a function that gets put in a dictionary, for context, and I say: pgconf.file.managed(context=CONTEXT)
11:40 entil the function, for the time being, returns the integer 16, but the end-result file has shared_buffers = 24MB which is the default
11:41 entil some hard-coded content does get through ok
11:45 entil is the file under /var/cache/salt/minion/ always supposed to be up to date? because it doesn't have the {{ shared_buffers }} curlies in it, only 24MB
11:48 unicoletti entil: try salt 'minion-id-here' saltutil.refresh_pillar
11:49 unicoletti it's quite strange though, it should not be necessary to refresh pillar, as it should refresh auto
11:49 entil that's what I'd think too
11:50 entil it returned a yellow None
11:50 entil and no change in the cached file, wtf
11:50 anteaya joined #salt
11:50 entil I'll try a state.highstate, but this is kind of frustrating, not knowing what's happening
11:52 entil Comment:   State require.environment-selector.duunitori-db.postgres found in sls environment-selector.duunitori-db.postgresql_conf_render is unavailable
11:52 entil that's so odd
11:52 unicoletti entil: to see what is going on logon in the minion then run salt-call -l debug state.highstate
11:52 entil why would highstate say that all of a sudden? some hard-coded changes did get installed on the first run
11:53 entil unicoletti: I did the similar thing remote, I'll do that now locally with salt-call
11:53 unicoletti locally should be more verbose
11:54 entil it's crazy verbose
11:54 entil but the same error came through
11:55 entil but I'm confused as to why the state would be unavailable, what does that mean?
11:58 krissaxton joined #salt
12:00 Stein joined #salt
12:02 kleinishere joined #salt
12:05 EnTeQuAk joined #salt
12:06 entil unicoletti: zomfg
12:07 entil I'm confused, but now it works
12:07 entil I just took a new ec2 instance and now I got the function return value rendered in the template
12:07 entil sheesh
12:07 entil beat your head to the wall for long enough and the wall will give in
12:08 Spads_ joined #salt
12:08 entil does still complain about the state being unavailable, but on a fresh install it worked, at least this one time
12:10 unicoletti it complains about an 'environment-selector', did you turn on/off env in mid-flight?
12:10 Katafalkas joined #salt
12:10 entil as in mid-sls or -highstate? no
12:10 entil but I'll make an alteration and see if salt-call makes the changes
12:11 Katafalkas joined #salt
12:11 LucasCozy joined #salt
12:11 whiskybar1 joined #salt
12:12 Katafalk_ joined #salt
12:12 oz_akan_ joined #salt
12:13 oz_akan_ joined #salt
12:17 ricicle joined #salt
12:21 bemehow joined #salt
12:21 entil it tries to, can't access grains, so at least there is progress
12:21 entil __grains__['meminfo'] not found, gawd, but it is in the files and now there's a meeting coming up so I'll just wrangle this later
12:23 whiskybar1 joined #salt
12:25 abele joined #salt
12:31 n3storm unicoletti: any chance to refer me an example of mod_repo?
12:31 unicoletti nope, I'm more of a rpm guy
12:34 sgviking joined #salt
12:37 teebes joined #salt
12:37 Psi-Jack unicoletti: Not sure how that's related, But okay. :)
12:38 n3storm unicolett: yum example?
12:38 n3storm the problem is using in state
12:38 n3storm using mod_repo in state
12:38 n3storm looks like it was asked here before:
12:38 Psi-Jack mod_repo?
12:38 unicoletti Psi-Jack: apt is used in debian and derivatives, yum-rpm in redhat and derivatives. I mostly used redhat/centos
12:38 n3storm http://irclog.perlgeek.de/salt/2013-05-29
12:38 Psi-Jack I see modules.yumpkg, modules.apt, etc.. But nothing called mod_repo specifically,.
12:39 Psi-Jack To me, sounds more like a communication error. :)
12:39 n3storm http://docs.saltstack.com/ref/modules/all/salt.modules.apt.html
12:39 n3storm it is
12:39 unicoletti Psi-Jack: ahh, now I see, n3storm was referring to apt.mod_repo
12:39 Psi-Jack Ahhhh
12:39 n3storm yes, I want to add mariadb repo to an state
12:39 Psi-Jack And yeah, I'm a yum/rpm guy myself mostly these days. :)
12:40 n3storm there is mod_repo for yum
12:40 Psi-Jack Heck, with apt, that's easy with even something as simple as file.managed. :p
12:40 TheCodeAssassin n3storm: https://gist.github.com/thecodeassassin/97382c253d92ad6704f2
12:40 n3storm you got an example?
12:40 unicoletti n3storm:  https://github.com/unicolet/mappu-deploy-tooling/blob/master/salt/states/salt/postgres.sls#L11
12:40 TJ1980 joined #salt
12:40 unicoletti too many examples, LOL
12:40 TheCodeAssassin n3storm: worked for me
12:40 n3storm TheCodeAssassin: thank you so much
12:41 Psi-Jack unicoletti: Heh, RPM /and/ POstgreSQL, vewwy nawice. :)
12:42 unicoletti help yourself
12:42 TheCodeAssassin n3storm: your welcome ^^
12:44 n3storm odd:
12:44 n3storm mariadb_repo:
12:44 n3storm pkgrepo:
12:44 n3storm - managed
12:44 n3storm - keyid: '0xcbcb082a1bb943db'
12:44 n3storm - keyserver: keyserver.ubuntu.com
12:44 n3storm - name: deb http://repo.maxindo.net.id/mariadb/repo/5.5/ubuntu precise main
12:44 n3storm
12:44 n3storm mariadb-server:
12:44 n3storm pkg:
12:44 n3storm - installed
12:44 n3storm - require:
12:44 n3storm - pkgrepo: mariadb_repo
12:44 TheCodeAssassin n3storm: use a pastie service such as gist or pastebin
12:44 n3storm Data failed to compile:
12:44 n3storm ----------
12:44 n3storm The require or watch statement in state "mariadb-server" in sls "mysql" needs to be formed as a list
12:44 n3storm I know
12:44 n3storm was just this
12:44 TheCodeAssassin don't paste so many lines, is gonna get you kicked
12:44 n3storm yeah, I know (I'm not irc newbie)
12:45 honestly anyone know if the salt cron module works with step instruction?
12:45 honestly e.g. "minute: */5"
12:46 TheCodeAssassin n3storm: what version are you using?
12:46 TheCodeAssassin salt version?
12:46 n3storm just installed on minion and just updated on master
12:46 n3storm let me check
12:47 n3storm salt-master 0.16.3
12:47 Katafalkas joined #salt
12:47 TheCodeAssassin weird
12:47 TheCodeAssassin maybe this: https://gist.github.com/thecodeassassin/5ee4691370582a502c68 ?
12:48 Sypher joined #salt
12:48 TheCodeAssassin worked for me though
12:48 honestly well crud
12:48 honestly it doesn't
12:49 honestly I have to consider that a bug
12:50 n3storm honestly: step instruction is not unix wide supported is it?
12:51 n3storm TheCodeAssassin: fails to compile even wors
12:51 n3storm worse
12:51 n3storm with python code dump
12:53 TheCodeAssassin hehe
12:54 TheCodeAssassin did you remove the comment n3storm?
12:55 honestly n3storm: I don't know
12:55 ipmb joined #salt
12:56 TheCodeAssassin just tested this one: https://gist.github.com/thecodeassassin/830b8c35df6b1dbefd4e
12:56 aat joined #salt
12:57 juicer2 joined #salt
13:00 TheCodeAssassin might need a little tweaking
13:00 honestly n3storm: according to man 5 crontab step values are supported
13:00 honestly and I think man 5 is posix
13:01 oz_akan__ joined #salt
13:02 kenbolton joined #salt
13:05 titubear joined #salt
13:05 unicoletti honestly: 5 refers to section 5 aka: file formats and is specified to distinguish from man 1 crontab which refers to the crontab command
13:06 TJ1980 joined #salt
13:08 Gifflen joined #salt
13:08 honestly unicoletti: I asked ##posix, if anybody knows, it'll be them
13:09 teebes joined #salt
13:09 bemehow joined #salt
13:14 Spads joined #salt
13:14 whiskybar1 joined #salt
13:17 unicoletti seriously, I like how all your lines start with honestly:
13:17 MTecknology How can I see what version of salt-minion is installed on all teh minions?
13:18 mwillhite joined #salt
13:18 MTecknology honestly: I don't honestly have a clue, dishonestly.
13:18 MTecknology :D
13:20 Kholloway joined #salt
13:20 unicoletti test.version
13:22 MTecknology :) perfect
13:25 honestly unicoletti: (:
13:26 racooper joined #salt
13:27 honestly unicoletti: boo, it appears to be an extension
13:28 chortlehoort joined #salt
13:28 unicoletti don't worry too much about it, it's friday after all
13:28 MTecknology ROFriday!!!!!!
13:28 honestly hah
13:28 unicoletti at least we know we can't trust ##posix /grin
13:28 honestly i do need that feature though
13:28 honestly huh?
13:29 honestly no, that's what ##posix told me
13:29 MTecknology Making changes to anything related to production in any way on ROFriday is strictly disallowed.
13:29 unicoletti ahh, that it is an extension?
13:29 unicoletti anyway, what is it that you want to achieve?
13:30 kenbolton joined #salt
13:31 honestly unicoletti: well, i want a cronjob to be executed every ten minutes
13:33 unicoletti can't you use file.append ?
13:33 toastedpenguin joined #salt
13:33 unicoletti it won't append the same line if it's there already
13:33 unicoletti http://docs.saltstack.com/ref/states/all/salt.states.file.html#salt.states.file.append
13:33 honestly i could, but using the cron state module would be way nicer
13:35 unicoletti why, is cron.present not working? (I might have missed some bits before)
13:35 MTecknology for cron?  http://salt.readthedocs.org/en/latest/ref/modules/all/salt.modules.cron.html
13:36 MTecknology Who shall be nominated to write the state for it?
13:36 DerekRBN Just discovered Packer.io Anyone using this with salt??
13:37 unicoletti there is a state for it already
13:37 * MTecknology isn't seeing it.
13:38 unicoletti https://github.com/saltstack/salt/blob/develop/salt/states/cron.py
13:38 unicoletti DerekRBN: is packer.io like Vagrant?
13:38 mwillhite for some reason my ssh-agent is not starting up anymore…
13:38 MTecknology but... it's not here? http://salt.readthedocs.org/en/latest/py-modindex.html  :(
13:39 mwillhite after I updated my mac last night… I can't connect to github anymore on my salty-vagrant machine
13:39 DerekRBN No packer.io builds images. which you can use to boot vagrant
13:39 mwillhite looking into it its because ssh-add doesn't have any of my public keys and the agent isn't running…but it was all working fine before the reboot
13:39 DerekRBN found the docs I was looking for. http://www.packer.io/docs/provisioners/salt-masterless.html    Pretty much you provision with salt and then build a image to decrease provisioning time
13:40 mwillhite has anyone else run into that?
13:40 LucasCozy joined #salt
13:40 LucasCozy joined #salt
13:40 unicoletti cool
13:40 unicoletti written in?
13:41 ninkotech__ joined #salt
13:42 DerekRBN unicoletti: what's packer io written in? Not sure looks like a good mix of go and bash? idk what go is
13:44 littleidea joined #salt
13:44 m_george|away joined #salt
13:48 Teknix joined #salt
13:50 mstan joined #salt
13:55 mstan hi
13:56 mstan I'm seeing random publish messages being lost
13:56 mstan when multiple minions call publish.publish at the same time
13:56 mstan lookit at the code I see that master.py sets HWM=1 in the Published zmq connection
13:57 mstan does anybody know why ?
13:57 DerekRBN mstan: do they have unique id's?
13:58 mstan yes
13:58 mstan and setting a higher HWM fixed the problem
13:59 sgviking joined #salt
14:02 cron0 joined #salt
14:03 Kholloway joined #salt
14:06 kleinishere joined #salt
14:08 sgviking joined #salt
14:08 Katafalkas joined #salt
14:09 blee joined #salt
14:10 jacob_s So packer.io is like cobbler?
14:10 iPaulo joined #salt
14:10 iPaulo left #salt
14:13 ipmb joined #salt
14:15 DerekRBN I'm looking at cobbler and im not sure it's alike could you explain the use of cobbler?
14:16 whiskybar1 joined #salt
14:17 m_george left #salt
14:20 littleidea joined #salt
14:21 mwillhite say I have this mysql-formula: https://gist.github.com/miwillhite/6319835
14:21 mwillhite and I  have an app that depends on the database-setup state…how do I reference that in my require:?
14:23 ricicle joined #salt
14:23 ricicle hi all, so with a file source, I sometimes need to specify 'templates' as a path part but sometimes it just works. What am I missing?
14:24 StDiluted joined #salt
14:24 kaptk2 joined #salt
14:26 backjlack joined #salt
14:26 unicoletti left #salt
14:28 jacob_s Yeah, nvm Derek, cobbler is more of an installation automation tool.
14:29 jacob_s specifically os installation.
14:30 mannyt joined #salt
14:40 DerekRBN mwillhite: just use mysql_user:
14:41 mwillhite so…from another sls, do I need to first include the mysql/init.sls ?
14:41 mwillhite then I can reference mysql_user from there?
14:41 DerekRBN yeah
14:41 DerekRBN after an import you can reference anything in that other state file as a require
14:42 * honestly sighs
14:42 honestly manually editing crontab using file.append is too damn ugly
14:43 DerekRBN jacob: packer.io is a build tool for creating machine images. My use case is cutting the 10min it takes to provision a machine for auto scaling
14:45 DerekRBN honestly: why not just use file.managed and put a cron file in /etc/cron.d/
14:48 honestly what are the semantics of cron.d?
14:51 ckao joined #salt
14:53 honestly ah, found it
14:58 n8xja joined #salt
14:59 whit joined #salt
15:02 tyler-baker joined #salt
15:02 tyler-baker joined #salt
15:03 mmilano joined #salt
15:04 mstan ricicle, the source file path needs to be exactly the one in your <file_root> folder
15:04 mstan do you have an example file source where 'it just works' ?
15:05 ricicle checking...
15:06 pdayton joined #salt
15:06 TJ1980 joined #salt
15:06 jacob_s I was wondering about what the status of support symbolic links in Salt is.
15:07 jacob_s I wanted to make some of my salt files link to files in one of my code repository, but it looked like Salt wasn't following the symbolic links.
15:08 jpeach joined #salt
15:09 jaequery joined #salt
15:09 DerekRBN The file.symlink works pretty well. are you trying to reference a symlink in a path for a new file?
15:10 ricicle mstan, it looks like we're using a mix of some modules having a templates directory in their root and others not. Coming from puppet-land I assumed (uh-oh) that templates was special. The modules having templates directory specify templates in the source, nothing more strange than that :)
15:10 jacob_s I was symbolic linking a folder.
15:10 ricicle mstan, thanks for your help
15:11 mstan welcome
15:11 jkleckner joined #salt
15:11 ipmb I have a file that has both a file.managed state and a file.touch state which is resulting in a recursive dependency
15:12 ipmb any suggestions on how to accomplish this? https://dpaste.de/tChYs/
15:13 ipmb maybe I could change it to `unless`
15:13 DerekRBN ipmb you're symlinking a file to itself
15:13 StDiluted joined #salt
15:13 forrest joined #salt
15:13 DerekRBN nvp
15:13 DerekRBN nvm it's a available to enabled
15:13 ipmb yep :)
15:14 kermit joined #salt
15:14 jacksontj joined #salt
15:16 DerekRBN you want to create a symlink and then touch the file?
15:16 ipmb create only happens once
15:17 ipmb touch happens every time after that
15:17 ipmb touch gracefully restarts the process when I deploy
15:17 whiskybar joined #salt
15:18 mwillhite ok thanks DerekRBN
15:18 ml_1 joined #salt
15:19 DerekRBN ipmb are you sure thats where the recusion is at? could it be the line above
15:20 lazyguru joined #salt
15:20 ipmb DerekRBN: the one I flagged is ambiguous now. there are two states that match
15:21 ipmb I think I'm just going to change it to cmd.run
15:24 mastrolinux joined #salt
15:25 mastrolinux hey what is the best way to include simple calc in a sls file? I would have something like 60 % pillar['myparam']
15:25 mastrolinux the module operator
15:26 mastrolinux but * + and other simple math operators are fine too
15:27 mastrolinux I need to do it in a cron state
15:28 dthom91 joined #salt
15:29 chortlehoort joined #salt
15:30 ipmb mastrolinux: {{ (grains.mem_total * 0.25)|int }}MB
15:30 ipmb we use that in a few of our states
15:30 ipmb (or something similar)
15:35 devinus joined #salt
15:35 mastrolinux cool but where is documented? what are other operators availaibable? Or is it jsut about jinja support?
15:36 mastrolinux ipmb: thank you as first ;)
15:36 ipmb mastrolinux: it's just jinja
15:36 ipmb which is just python
15:37 mastrolinux so basically http://jinja.pocoo.org/docs/templates/#math works
15:38 mastrolinux thanks
15:38 mastrolinux I had problems with jinja if condition in salstack in the past
15:38 mastrolinux I do not think saltstack allows everything about jinja in .sls files
15:38 mastrolinux maybe just in jinja templates
15:40 conan_the_destro joined #salt
15:40 mstan everything should be allowd in sls files
15:41 mstan but maybe not top.sls and overstate.sls
15:44 tuxIO joined #salt
15:47 Katafalkas joined #salt
15:48 entil ok, now I've been fighting for an hour with the worst race condition there is
15:48 entil you know how postgres wants kernel.shmmax going?
15:48 entil I have procps as its own state, and a file.managed for sysctl, which copies over a fixed version of the file (require: pkg: procps)
15:49 entil I have a cmd.wait called "sysctl reload" which watches the file and requires pkg: procps and file: sysctl
15:49 entil in postgres I have require file: sysctl and pkg: procps
15:50 entil then I do the configuartion and have a "postgres restart" cmd.wait which requires pkg: postgres, the file: config thing, and file: sysctl
15:50 entil on every test postgres tries to restart before the sysctl is done
15:50 entil and sysctl reload happens just after postrges
15:52 mstan shouldn't "postgres restart" have a watch on "cmd: sysctl reload" ?
15:53 bashcoder entil: I've had something similar with setting ulimits before nginx starts up.
15:53 entil mstan: I tried that but it didn't work, then I figured maybe it doesn't matter
15:54 entil I'll try to have the configuration depend on sysctl
15:54 entil maybe even sysctl restart
15:54 entil it's in pydsl so let's see if I get the syntax straight
15:55 bashcoder I use a cmd.wait under the service for nginx, watching file /etc/sysctl.conf
15:55 entil ytf doesn't it work for me then...
15:55 bashcoder And the command that runs is sysctl -p
15:56 entil yes
15:56 entil I'll try to require cmd: sysctl reload first, then screw around with the pydsl
15:56 mstan you could also try adding "onlyif: sysctl kernel.shmmax | grep kernel.shmmax = ..."
15:57 kleinishere joined #salt
15:57 mstan but it could fail the first run
15:57 entil I haven't done that before, I think I'll try the more familiar things first
15:57 StDiluted it would be so awesome if the minion config file was /etc/salt/minion-sample or something
15:57 entil there's enough of a challenge with pydsl
15:57 StDiluted the one included in the package, that is
15:57 entil StDiluted: I'm way past that :D
15:57 entil oh wait
15:57 entil that's another thing
15:57 entil yes
15:58 StDiluted so it didnt stomp your minion config every damn time you update
15:58 StDiluted requires manual intervention to the upgrade
15:58 StDiluted to tell it no
15:58 mstan what package manager ?
15:58 StDiluted apt
15:59 StDiluted i would love if i could just do an apt-get upgrade and not have to interact with it
15:59 StDiluted it's poor packaging behavior to begin with
16:01 mstan we just force a silent install
16:01 mstan with Dpkg::Options::='--force-confdef' -o Dpkg::Options::='--force-confold and stuff like that
16:01 mgw joined #salt
16:01 StDiluted yeah. I know I could do that but that's a lot of typing, and I use the unattended upgrades script
16:01 mstan but maybe somebody else has a real explanation
16:01 StDiluted or manually update
16:02 StDiluted it seems that if you have a config file in your package, you would want it not to stomp on the customer's configured config file
16:02 StDiluted and I get that they want you to have your minions just talk to 'salt'
16:02 racooper .rpmnew :) (oh wait, you said .deb based)
16:02 StDiluted yes
16:03 StDiluted rpm doesnt have the same issue
16:03 racooper oh, it can if the packager doesn't tag things correctly.
16:10 KyleG joined #salt
16:10 KyleG joined #salt
16:25 jacksontj joined #salt
16:26 mwillhite joined #salt
16:28 robertkeizer joined #salt
16:29 anteaya joined #salt
16:31 krissaxton joined #salt
16:32 jimallman joined #salt
16:32 linjan_ joined #salt
16:34 cedwards I'm getting an error trying to apply this state: http://dpaste.com/1354370/
16:34 cedwards I know that syntax used to work. Has it changed?
16:35 cedwards primarily the template and context sections
16:36 chrisgilmerproj joined #salt
16:38 krissaxton joined #salt
16:39 carmony joined #salt
16:40 Xeago joined #salt
16:40 KennethWilke joined #salt
16:45 vaxholm_ joined #salt
16:45 geopet joined #salt
16:46 chrisgilmerproj joined #salt
16:47 whit joined #salt
16:49 whit joined #salt
16:49 mastrolinux just read: As of Salt 0.16.0, it is possible to require an entire sls file. Do this by first including the sls file and then setting a state to require the included sls file.
16:49 mastrolinux I was looking for it sincethe first release, awesome!
16:51 littleidea joined #salt
16:51 chrisgilmerproj joined #salt
16:56 Spads_ joined #salt
16:56 chrisgilmerproj left #salt
16:57 Spads_ joined #salt
17:01 Gifflen joined #salt
17:04 oc joined #salt
17:18 devinus joined #salt
17:19 Kholloway joined #salt
17:20 aurigus joined #salt
17:27 Xeago joined #salt
17:27 KyleG1 joined #salt
17:27 geopet Hi, SaltStackers.
17:28 geopet I have a question about state ordering.
17:28 geopet I have a script that I'd like to run after a state has been completed.
17:28 geopet I'm trying to wrap my head around the syntax, but I'm not getting it.
17:29 geopet Basically I want to install a ruby package using the rbenv state method and then add pieces to my .profile and source it after that's happened.
17:29 geopet How can I test to make sure that the ruby version has been installed before I run that script to update and source the .profile?
17:30 Ryan_Lane joined #salt
17:31 aurigus Anyone know how to change uid for a user? In state.user I'm using uid but getting: These values could not be changed: {'uid': 551}
17:31 Gifflen joined #salt
17:31 aurigus This also fails: salt 'server' user.chuid www 551
17:36 aurigus geopet: have you tried the require: directive?
17:37 aurigus I'm a salt noob but that is what I would try first.
17:37 whiskybar joined #salt
17:37 geopet Right, aurigus! Like this? http://docs.saltstack.com/ref/states/ordering.html#the-require-requisite
17:38 bashcoder geopet: also this - http://docs.saltstack.com/ref/states/requisites.html
17:39 aurigus yup
17:39 geopet Oh! Perfect, bashcoder. I can require a full sls file.
17:39 geopet Awesome.
17:39 geopet Thank you, both! I'll check this out.
17:41 mattbarto joined #salt
17:45 mattbarto joined #salt
17:49 ricicle joined #salt
17:49 whit joined #salt
17:49 vaxholm_ joined #salt
17:51 UtahDave joined #salt
17:53 whit joined #salt
17:54 dthom91 joined #salt
17:56 tuxIO joined #salt
17:57 KyleG joined #salt
17:57 KyleG joined #salt
17:58 TheCodeAssassin joined #salt
17:58 TheCodeAssassin hey guys, does the minion id default to hostname or hostname --fqdn ?
17:59 juanlittledevil joined #salt
17:59 jeffasinger joined #salt
18:01 giantlock joined #salt
18:02 TheCodeAssassin nvm
18:05 terminalmage TheCodeAssassin: http://docs.saltstack.com/topics/tutorials/walkthrough.html#setting-up-a-salt-minion
18:05 terminalmage recently added that to the docs
18:05 terminalmage well, the part that begins "When the minion is started, it will generate an id value"
18:05 terminalmage that's what you want to look at
18:10 TheCodeAssassin terminalmage: already found it :)
18:10 TheCodeAssassin i do cat /etc/hostname to get it
18:10 TheCodeAssassin i think it does it the same way
18:11 sgviking joined #salt
18:12 isomorphic joined #salt
18:13 n8xja joined #salt
18:14 druonysus joined #salt
18:14 druonysus joined #salt
18:17 chortlehoort joined #salt
18:20 robertkeizer joined #salt
18:21 nliadm Is it impolite to open a pull request without testing your code?
18:21 UtahDave :)  nliadm, if it's something dead simple, it might not be a problem.
18:22 nliadm it's hold/unhold support for apt
18:22 nliadm I know the shell commands work
18:25 forrest will it not be tested by travis CI nliadm?
18:25 mwillhite I don't really understand where salt.modules fit in. I'm using vagrant with salt…and I refere to salt.states in my sls files…how can I use the modules?
18:26 UtahDave mwillhite: salt \* disk.usage    (disk is a Salt execution module)
18:27 mwillhite so they are what is used to run states?
18:27 mwillhite how would I use modules in a masterless setup?
18:28 mwillhite such as salty-vagrant :)
18:29 akoumjian mwillhite: Modules are just (imperative) commands you send to a minion. In a masterless setup, you would run them with salt-call instead of salt, and the only minion you're targetting is, of course, the standalone minion
18:29 mwillhite ok…so they have no place in an .sls file
18:29 akoumjian mwillhite: Comand modules are often used inside of state modules (ie: the file.managed state module uses a series of file command modules)
18:30 mwillhite hm
18:30 mwillhite ok
18:30 akoumjian mwillhite: They could be used in an sls file, but very infrequently
18:30 mwillhite for example, I'm looking at salt.states.git vs salt.modules.git
18:30 mwillhite modules lets me do everything…state has hardly nothing
18:31 akoumjian Typically you use a command module in managing large fleets of servers, ie: "update apache on my 1000 webservers".
18:31 akoumjian mwillhite: The state module is meant to represent idempodent states you want your minion to be in. command modules are more action driven
18:32 mwillhite ok I see
18:32 akoumjian So with the git example, typically what you want in your config management (sls files) is to have the latest copy of the repo and check it out to a certain commit.
18:33 akoumjian Adn this git.latest is really a combination of a bunch fo imperative operations like git.clone and git.checkout
18:33 akoumjian which come from the command module
18:33 mwillhite i see
18:33 mwillhite that makes a lot of sense
18:33 akoumjian Cool, glad I'm making some sense :-)
18:34 mwillhite certainly this is very helpful
18:34 mwillhite :)
18:36 mwillhite my mind is just spinning with the possibilities here…by using this framework
18:37 whit joined #salt
18:42 oz_akan_ joined #salt
18:50 aat joined #salt
18:55 jacksontj joined #salt
18:57 auser joined #salt
18:57 jkleckner joined #salt
18:59 blee joined #salt
19:00 Lue_4911 joined #salt
19:10 dthom91 joined #salt
19:14 Xeago joined #salt
19:23 terminalmage TheCodeAssassin: That behavior, with /etc/hostname, is not in an official release
19:23 terminalmage I personally added it within the last week
19:23 terminalmage right now it's using socket.getfqdn()
19:23 terminalmage unless you're running from git
19:27 aat joined #salt
19:28 tuxIO joined #salt
19:31 dthom91 joined #salt
19:35 JesseC joined #salt
19:40 whit joined #salt
19:44 mattbarto joined #salt
19:49 DerekRBN So just to follow up with what i brought up earlier. I got around to trying out packer.io. With a 18 line json file for packer.io that pointed to my salt master directory. It launched a EC2 instance provisioned it with salt and baked a AMI for me. \
19:50 UtahDave nice!
19:51 zooz joined #salt
19:51 danielbachhuber joined #salt
19:54 Nexpro joined #salt
19:55 Chocobo joined #salt
19:57 robawt hey cool kids
19:57 robawt Tom's FLOSS weekly episode was fantastic by the way
19:57 robawt although now I have a few million questions
19:58 teebes joined #salt
19:58 Ahlee neat.  Test.ping returns None now
20:01 Ahlee When using ext_job_cache, is it possible you have to specify additional information to salt-run jobs.lookup_jid <jid> ?
20:03 UtahDave Ahlee: I'm pretty sure everything should work just the same
20:03 UtahDave robawt: highfive!
20:03 Ahlee hrm, lookup_jid takes jid and ext_source, and then uses __opts__['ext_job_cache'] or ext_source. Hrm.
20:04 Ahlee UtahDave: adding some debug logging now to see, as no ext_job_cache jobs/<jid> return True for minions, defining an ext_job_cache returns None for same minions (after restarting minions/master, of course)
20:04 whit joined #salt
20:05 Boohbah joined #salt
20:06 UtahDave are you setting the ext_job_cache on the minions?  That's a master setting
20:06 jimallman joined #salt
20:06 UtahDave Also, you have to make sure the minions all have proper credentials to use that returner
20:07 Ahlee UtahDave: I did, yes.  I thought logic was master->minion, minion->returner
20:07 Ahlee not minion->master->returner
20:08 UtahDave You're correct.  When you set an external job cache, that automatically sets the default returner on the minions
20:08 Ahlee minions have ext_job_cache: mongo, and same mongo instance defined as master, I see the return in mongo so I know somehow they're winding up
20:09 UtahDave If I remember correctly, ext_job_cache only gets set on the master
20:10 Ahlee so then minions only need to know how to reach it
20:10 Ahlee ok, i'll test in a second
20:10 UtahDave also you set the mongo config info in the master config and the minions inherit that
20:10 UtahDave yep
20:10 Ahlee I seriously wonder how i fucked this up so bad.
20:10 Ahlee oh, so minions don't need it at all
20:11 Ahlee they just need pymongo or redis or whatever
20:11 Ahlee depending on what ext_job_cache defined. Interesting.
20:11 a1j question how do i get host IP address from grains?
20:12 a1j i'm trying perhaps querying different interfaces but i fail to do so - grains.item ip_interfaces.eth0  gives nothing, and grains.items prints "---------" on top of interface list (what is that?)
20:13 KyleG1 joined #salt
20:13 dthom91 joined #salt
20:13 Ahlee Oh.  If minions inherit from the master, that means they're gona pick up whatever the master had when the master started, when they started.
20:13 Ahlee Oh crap.  I haven't been bouncing minions when changing master settings.  I wager they're quite confused on the lay of the land.
20:13 UtahDave Ahlee: yeah
20:15 Ahlee pretty sure i just did a pretty impressive reproduction of the painting the scream on that realization.
20:15 StDiluted a1j: what about grains.item ipv4?
20:18 jacksontj joined #salt
20:20 a1j StDiluted: that might work IF I'm pretty sure that 127.0.0.1 is always the first item (which i'm not)
20:20 a1j StDiluted: perhaps  salt['network.interfaces']()['eth0']['inet'][0]['address']   should work
20:21 a1j StDiluted: I would not know how to write in jinja to skip address if it is 127* and use next one
20:22 nliadm UtahDave: I seem to have found/fixed my issue
20:23 nliadm I updated the ticket, also
20:23 nliadm problem seems to have been gitfs
20:23 UtahDave nliadm: cool.  what ticket was that?
20:24 nliadm 6787
20:25 nliadm disabling gitfs made highstates go from usually never happening and taking > 90 seconds when they did to something like 5-8 seconds
20:25 Boohbah Thomas uses all the same software as I do, kde, chromium, and clementine
20:26 dthom91 joined #salt
20:27 UtahDave nliadm: wow
20:27 UtahDave basepi: have you read up on nliadm's issue ^^?
20:28 basepi UtahDave: yep, i think i replied to it a few minutes ago
20:28 UtahDave ah, ok
20:28 basepi haven't had a chance to look at it at all yet, though
20:28 basepi probably won't, today.  still have 14 pages of cherry-picking to do
20:38 oz_akan_ joined #salt
20:42 a1j sigh... {% if grains['ipv4'][0] is matching @/^127\\.0\\.0/ -%}  gives "render error: unexpected char u'@' at 78;"
20:43 a1j what is wrong?
20:43 isomorphic joined #salt
20:44 n8xja joined #salt
20:44 StDiluted doesnt like that @ sign in the string
20:45 StDiluted can I do
20:45 StDiluted require:
20:45 StDiluted - rbenv: ruby-1.9.2-p448
20:45 StDiluted does it act like a pkg?
20:46 UtahDave what do you mean?
20:46 a1j StDiluted: but you need @ there for jinja.
20:47 StDiluted a1j: I'm just saying what the error means, I don't know enough jinja yet to know what it should look like there
20:47 StDiluted UtahDave: meaning, i have a ruby state that says
20:47 StDiluted ruby-1.9.3-p448:
20:47 StDiluted rbenv.installed:
20:47 StDiluted yadda yadda
20:47 StDiluted and I want a script to fire AFTER that
20:48 StDiluted so in my later declaration of
20:48 StDiluted thingie:
20:48 a1j thats jinja 1.2 valid syntax
20:48 StDiluted cmd.script:
20:48 StDiluted can i say
20:48 StDiluted - require:
20:48 StDiluted - rbenv: ruby-1.9.3.-p448
20:49 StDiluted that's how it would look if ruby was a pkg
20:49 StDiluted - require:
20:49 StDiluted - pkg: ruby-1.9.3-p448
20:50 TJ1980 joined #salt
20:51 cedwards StDiluted: I believe you would just require: pkg: ruby-1.9.3-p448
20:51 * robawt highfives UtahDave
20:51 StDiluted ruby isn't a package. It's installed using rbenv
20:52 StDiluted rbenv.installed
20:52 cedwards StDiluted: oh, you're right. require: rbenv: ruby-1.9.3-p448 makes sense then
20:52 UtahDave StDiluted: yeah, then you should be able to just require rbenv: ruby-********
20:52 StDiluted ok, that's what I thought. thanks!
20:53 UtahDave you can require ANY state that way
20:54 StDiluted ok. That was my question. That's not really clear anywhere I'm reading from, so I just wanted to check.
20:55 gildegoma joined #salt
20:56 juanlittledevil joined #salt
20:58 mattbarto joined #salt
21:03 JasonSwindle joined #salt
21:04 tuxIO joined #salt
21:10 jslusher joined #salt
21:10 jslusher anyone had success using the iptables salt module?
21:10 jslusher I'm getting the following:
21:10 jslusher omment:   State iptables.append found in "" not available
21:11 TJ1980 joined #salt
21:12 jslusher https://salt.readthedocs.org/en/latest/ref/states/all/salt.states.iptables.html
21:12 mattbarto joined #salt
21:13 jalbretsen joined #salt
21:15 jslusher http://pastebin.com/SyqF1gYG
21:16 dthom91 joined #salt
21:17 a1j ok {{ salt['network.ip_addrs']()[0] }} seems like gives host IP reliably.
21:18 Ahlee well, using redis for ext_job_cache I can successfully return results, whereas mongo doesn't appear, digging deeper
21:18 titubear_ joined #salt
21:18 a1j and salt.states.host is compltely broken in 0.16 (says name already present in any case).
21:19 mwillhite joined #salt
21:19 bemehow joined #salt
21:20 Sypher joined #salt
21:21 keith4 joined #salt
21:23 pdayton joined #salt
21:25 gildegoma joined #salt
21:30 tuxIO joined #salt
21:31 whit joined #salt
21:31 racooper so for which release is the iptables state module to be included?
21:35 tru_tru joined #salt
21:35 jslusher based on the repo it looks like it might have been pushed to 0.17
21:35 whit joined #salt
21:39 cedwards does anyone have an example of using publish.publish to gather IP information from hosts?
21:42 blee_ joined #salt
21:43 canci joined #salt
21:43 dave_den cedwards: do you wnat only one minion to be able to gather the info?
21:43 JesseC left #salt
21:45 dave_den to allow all minions to get the info, in your /etc/salt/master file, set 'peer:\n  .*:\n    - network.ip_addrs'
21:45 dave_den then you can do: salt-call publish.publish '*' network.ip_addrs
21:45 dave_den on any minion
21:47 geopet joined #salt
21:49 cedwards dave_den: primarily i need my haproxy host(s) to be able to gather the IPs of the backend nodes
21:49 druonysus joined #salt
21:51 dave_den the '.*' part from above is a glob, so if your haproxy minions have a structured id, you can restrict it like 'haproxy*:'
21:54 cedwards thanks
21:55 StDiluted how do i file.append onto a file with a line that has quotation marks in it?
21:55 UtahDave cedwards: https://gist.github.com/UtahDave/5217462
21:57 dave_den cedwards: the 10:00 minute mark is exactly what you're asking about, even using haproxy as an example. https://www.youtube.com/watch?v=jJJ8cfDjcTc
21:57 whit joined #salt
21:58 bemehow joined #salt
21:58 StDiluted i.e: - arg:
21:58 StDiluted - /etc/check_mk/conf.d/wato/vpc_west_stack/hosts.tst
21:58 StDiluted - "all_hosts += ["{{ data{'id' }}|cmk-agent|prod|site:west-vpc|lan|tcp|wato|/" + FOLDER_PATH + "/","
21:59 StDiluted that is unhappy and crashes the salt master's event processing
22:00 aat joined #salt
22:00 tuxIO joined #salt
22:00 UtahDave StDiluted: I wonder if you need to use a raw string there
22:01 StDiluted I tried {% raw %}
22:01 StDiluted but I'm trying to also put a jinja templated id in there
22:02 jesusaurus <3 salt-cloud
22:02 jesusaurus (just sayin)
22:04 UtahDave jesusaurus: me too!
22:05 UtahDave StDiluted: Hm.  what's the error you're getting?
22:05 StDiluted the trigger doesnt work, and then the events interface hangs and wont work until i restart salt-master
22:05 StDiluted i dont get an error
22:06 UtahDave StDiluted: can you do a file.managed on the file instead?
22:06 StDiluted no because other things will be changing it
22:08 StDiluted but it seems like anything that confuses the event reactor makes it stop accepting input
22:08 Spads joined #salt
22:09 StDiluted I guess I will have to write a wrapper app that takes an argument from a cmd.cmd.run and it will do the append
22:09 dave_den StDiluted: have you tried using - >"all_hosts += ["{{ data{'id' }}|cmk-agent|prod|site:west-vpc|lan|tcp|wato|/" + FOLDER_PATH + "/","
22:10 StDiluted that's what i used
22:10 robawt what should I read into to learn more about minions being aware of one another?
22:10 dave_den with the > ?
22:10 StDiluted oh
22:10 StDiluted no
22:11 StDiluted let me try
22:11 StDiluted what's that do
22:11 dave_den it tell YAML that what follows is literal
22:11 StDiluted nope
22:11 StDiluted that didnt work either
22:12 StDiluted 2013-08-23 22:10:23,371 [salt.master      ][INFO    ] AES payload received with command _minion_event
22:12 StDiluted 2013-08-23 22:10:23,372 [salt.utils.event ][DEBUG   ] Gathering reactors for tag wato
22:12 StDiluted 2013-08-23 22:10:23,372 [salt.utils.event ][DEBUG   ] Compiling reactions for tag wato
22:12 StDiluted and then nothing
22:12 dave_den sec
22:12 StDiluted any further attempts at firing an event fail as well
22:12 StDiluted even if it is valid
22:13 StDiluted it's ok, I'll write a wrapper
22:14 jpcw joined #salt
22:15 littleidea joined #salt
22:18 rrauenza_ joined #salt
22:19 dave_den StDiluted: also, typo:  {{ data{'id' }}
22:19 rrauenza_ Is there a way to sync a git repo (initial clone) without pulling the latest rev every time it runs?  I guess I could use unless with a state/test call to see if the dir already exists.
22:19 dave_den should be {{ data['id'] }}
22:20 robawt rrauenza_: can't you pull a specific tag or SHA?
22:20 StDiluted got that fixed
22:20 StDiluted lemme see
22:21 StDiluted better
22:21 StDiluted but hangs after rendering
22:21 rrauenza_ I don't want a specific tag ... was going to keep syncing up as a manual task outside of running salt.  I just want salt to initialize the local repo for me.
22:22 rrauenza_ So maybe I just need to look at <path>/.git as an unless option.
22:22 rrauenza_ Unfortunately git.present *seems* like what I want -- but it isn't.
22:25 dave_den StDiluted: https://gist.github.com/dlanderson/c9e1e1abd642a04ab61a
22:26 dave_den gotta run. good luck
22:26 StDiluted thatnks dave
22:30 tuxIO joined #salt
22:31 StDiluted hm
22:35 StDiluted can someone explain to me what's happening when you call cmd.file.append from a reactor?
22:35 StDiluted does it cause a salt.modules.file.append to be run on the minion?
22:43 Linz joined #salt
22:43 oz_akan_ joined #salt
22:50 UtahDave StDiluted: yes, on the tgt minion
22:51 alekibango joined #salt
22:52 aat joined #salt
23:02 kermit joined #salt
23:02 tuxIO joined #salt
23:09 Linz joined #salt
23:39 TJ1980 joined #salt
23:41 KyleG joined #salt
23:41 KyleG joined #salt
23:41 eculver joined #salt
23:41 eculver joined #salt
23:42 TJ1980 joined #salt
23:47 whit joined #salt
23:47 oz_akan_ joined #salt
23:49 rgarcia_ joined #salt
23:52 druonysus joined #salt
23:54 Ryan_Lane1 joined #salt

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