Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2017-05-29

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

All times shown according to UTC.

Time Nick Message
00:03 fracklen joined #salt
00:03 fritz09_ joined #salt
00:26 mosen joined #salt
00:48 aneeshusa joined #salt
00:53 edrocks joined #salt
01:07 chowmeined joined #salt
01:32 shred joined #salt
01:41 masber joined #salt
01:48 ilbot3 joined #salt
01:48 Topic for #salt is now Welcome to #salt! <+> Latest Versions: 2016.3.6, 2016.11.5 <+> Support: https://www.saltstack.com/support/ <+> 1st Salt Cloud Working Group meeting June 1st, 2017 https://goo.gl/o2OK49 <+> Logs: http://irclog.perlgeek.de/salt/ <+> Paste: https://gist.github.com/ <+> See also: #salt-devel, #salt-offtopic <+> We are volunteers and may not have immediate answers
01:55 shred joined #salt
01:58 zerocool_ joined #salt
02:40 aneeshusa joined #salt
02:42 mikea joined #salt
02:47 JPT joined #salt
03:10 amcorreia joined #salt
03:16 evle2 joined #salt
03:32 miruoy joined #salt
04:08 om2 joined #salt
04:08 aneeshusa joined #salt
04:16 jas02 joined #salt
04:37 justanotheruser joined #salt
04:45 rgrundstrom joined #salt
04:49 rgrundstrom Good morning
04:49 golodhrim|work joined #salt
04:54 om2 joined #salt
04:54 jas02 joined #salt
04:57 edrocks joined #salt
05:03 shred joined #salt
05:04 masber joined #salt
05:05 jhauser joined #salt
05:26 hemebond Does anyone have an example of a reactor that runs state.apply against multiple minions?
05:26 hemebond state.apply with a state file that is.
05:29 aneeshusa joined #salt
05:31 Rubin joined #salt
05:32 impi joined #salt
05:35 patrek joined #salt
05:41 sergeyt joined #salt
05:44 om2 joined #salt
05:46 iggytest joined #salt
05:56 hemebond NVM, master processes were in a weird state.
06:02 do3meli joined #salt
06:03 do3meli left #salt
06:06 shred_ joined #salt
06:11 felskrone joined #salt
06:11 colttt joined #salt
06:23 aldevar joined #salt
06:26 patrek joined #salt
06:27 jas02 joined #salt
06:28 mikecmpbll joined #salt
06:32 dcpc007 joined #salt
06:33 Ricardo1000 joined #salt
06:35 jas02 joined #salt
06:36 jas02 joined #salt
06:44 rgrundstrom I need to set a ip address in a config file. When looking thru grains on the minion i notice that there are several options under for example ipv4. What would be the best way to get the correct ip adress from grains?
06:45 hemebond Probably network.ip_addrs
06:46 rgrundstrom Ty
06:49 nebuchadnezzar joined #salt
06:57 sergeyt joined #salt
07:02 DanyC joined #salt
07:04 o1e9 joined #salt
07:11 JohnnyRun joined #salt
07:16 rgrundstrom Hmm when i do this salt['mine.get']('*', 'network.ip_addrs') output is udp:['172.27.87.10']:161 ... the ['x.x.x.x'] is unwanted. What am i doing wrong?
07:17 rgrundstrom Wrong....
07:17 rgrundstrom Ignore
07:17 whytewolf |last
07:17 whytewolf |first
07:17 whytewolf [0]
07:17 rgrundstrom udp:{{salt['network.ip_addrs']()}}:161
07:18 sergeyt joined #salt
07:18 rgrundstrom output udp:['172.27.87.10']:161
07:18 whytewolf udp:{{salt['network.ip_addrs']()|first}}:161
07:18 rgrundstrom ['x.x.x.x'] unwanted
07:18 whytewolf udp:{{salt['network.ip_addrs']()|last}}:161
07:18 mbologna joined #salt
07:18 whytewolf udp:{{salt['network.ip_addrs']()[0]}}:161
07:18 whytewolf pick one
07:19 rgrundstrom checking
07:20 dcpc007 hi all !
07:23 ronnix joined #salt
07:25 rgrundstrom whytewolf: Works like a charm. Thank you.
07:26 whytewolf no problem
07:30 fracklen joined #salt
07:31 fracklen joined #salt
07:31 fracklen joined #salt
07:32 q1x joined #salt
07:37 felskrone joined #salt
07:40 inad922 joined #salt
07:43 preludedrew joined #salt
07:44 darioleidi joined #salt
07:44 shred_ joined #salt
07:49 Ricardo1000 joined #salt
07:52 Rubin joined #salt
07:52 impi joined #salt
07:54 sh123124213 joined #salt
07:56 teclator joined #salt
07:57 ronnix joined #salt
07:58 edrocks joined #salt
07:58 kbaikov joined #salt
08:01 nku joined #salt
08:08 zerocool_ joined #salt
08:10 bdrung_work joined #salt
08:15 Mattch joined #salt
08:15 zerocool_ joined #salt
08:19 Rumbles joined #salt
08:22 Sammichmaker joined #salt
08:28 dcpc007 on a debian 7 i got this error while updating : E: impossible de récupérer http://repo.saltstack.com/apt/debian/7/amd64/latest/pool/main/libs/libsodium/libsodium13_1.0.3-1_amd64.deb : Taille incohérente
08:28 dcpc007 size error ???
08:28 zerocool_ joined #salt
08:29 dcpc007 any idea about this please ?
08:42 impi joined #salt
08:42 zerocoolback joined #salt
08:43 ronnix joined #salt
08:44 whytewolf given the error and the fact no one has been in here with that error since the release. i would say you have an issue on your system. could be several things. ran out of space. or network issues truncated the file.
08:44 whytewolf something anlong those lines
08:46 SaltyVagrant joined #salt
08:50 jamesp9 joined #salt
08:57 Bock joined #salt
09:02 Terminus joined #salt
09:07 jas02_ joined #salt
09:09 jas02__ joined #salt
09:17 bdeluca joined #salt
09:17 cyteen joined #salt
09:23 jamesp9 joined #salt
09:26 squig joined #salt
09:28 squig hey, I am trying to debug some one elses salt stuff to deal with salt states that keep on running
09:28 squig I have a cmd.run and I am trying to figure out what require: means on that
09:29 iggy require means that the other state has to run first and must be successful
09:29 whytewolf require means that the other state needs to show as compleate or change before the cmd.run will run.
09:30 squig ok, cool so If I also add onchanges IT will only run when the other state changes?
09:31 iggy and change the cmd.run to cmd.wait
09:31 squig cmd.wait is depricated
09:31 whytewolf cmd.wait is not correct with onchanges
09:32 whytewolf cmd.wait has no function. and onchanges doesn't trigger mod_watch
09:32 iggy oh right
09:32 squig so I should use cmd.run with onchanges and require or only onchanges
09:33 whytewolf i would only use onchanges. I don't remeber if requisites are additive.
09:33 squig ok, I will start with that and test it
09:33 iggy I would write a state module instead of calling cmd.run
09:33 whytewolf ^^^^^
09:35 squig so what does that mean in more detail :)
09:35 squig sorry, want to do this right
09:35 whytewolf https://docs.saltstack.com/en/latest/ref/states/writing.html
09:36 squig I know my client will want me to do what ever is fastest :(
09:37 whytewolf ugh, tell them if they don't have time to do it right, when will they have time to do it over.
09:38 squig I will write and ask them :)
09:41 gmoro joined #salt
09:42 federico3 any hint on unblocking https://github.com/saltstack/salt/pull/41142 please?
09:43 squig it looks like the build system is just broken
09:44 squig salt-master was already running
09:45 whytewolf federico3: it doesn't look blocked to me. just not accepted yet. it is in develop so any chance it has is for the up coming nitrogen release
09:45 federico3 whytewolf: one of the tests (apparently unrelated to the PR contents) is failing so I was wondering if it's a blocker
09:46 squig 'WARNING: Unable to bind socket 0.0.0.0:4506, error: [Errno 98] Address already in use; Is there another salt-master running?', u'The salt master is shutdown. The ports are not available to bind']
09:46 whytewolf if it is unrelated to the PR it shouldn't be an issue.
09:46 squig I think if you push a blank space the test will run again and probably pass
09:47 squig in this case the tests could run coz salt is broke
09:48 mike25_ joined #salt
09:49 whytewolf just add a note to the PR. let the guys at salt be the judge
09:50 federico3 looks like I can comment "Go Go Jenkins!" to trigger a rebuild
09:50 federico3 thanks
09:51 mede joined #salt
09:54 dcpc007 whytewolf: i find the same bug for more recent release .. that was Salt repo not up to date in jessie. i open a case on github to see if they corrrect for wheezy too
09:54 dcpc007 could be "only" a problem of signature between package avec description file ....
09:55 dcpc007 i try on a jessie and seems not this prob, and same with jessie-backport. But can't modifiy on the wheezy server, it's the prod
10:01 edrocks joined #salt
10:06 whytewolf of coarse. debian didn't make a cloud image until debian 8
10:11 dcpc007 ? don't understand
10:11 dcpc007 i'll try find a wheezy test machine to add this repo
10:12 dcpc007 and see what happens
10:12 whytewolf dcpc007: i use a openstack cloud to test things. and it is 3am. I do not want to build a new cloud image at 3 in the morning to test this
10:12 dcpc007 http://repo.saltstack.com/apt/debian/7/amd64/latest/pool/main/libs/libsodium/ seems ok
10:12 cyborg-one joined #salt
10:12 dcpc007 lol lunch time here :)
10:13 zerocoolback joined #salt
10:15 whytewolf this also reminds me i need to install vagrent and the hasicorp tools on my laptop
10:15 whytewolf soonish
10:16 Ricardo1000 joined #salt
10:18 Whissi How do I target same minions multiple times in one top.sls file? I.e. I have base_settings, nodegroup_settings and individual_minion_settings. When I do '*'\nbase_settings, 'foo-group'\nmatch:nodegroup\nnodegroup_settings,  '*'\nindividual_minion_settings I am getting conflicting IDs. Do I need to create an additional Nodegroup which will target everything to I have a different "ID"?
10:19 whytewolf Whissi: can you put that info in a gist or something. it is losing something in translation to here
10:20 Whissi Yeah, one moment please
10:25 dcpc007 same problem with libsodium13 on another wheezy for the 1st install
10:26 dcpc007 goind to lunch.? good night whytewolf :)
10:26 jas02 joined #salt
10:26 Whissi whytewolf: https://gist.github.com/Whissi/722d2a51f4ba2ef89786e020e0844904
10:26 whytewolf dcpc007: okay, then you might wanna create a ticket on the salt-pack
10:27 whytewolf https://github.com/saltstack/salt-pack
10:28 whytewolf Whissi: move config_include_individual_host_configuration under the first '*'
10:28 dcpc007 i've done : https://github.com/saltstack/salt/issues/41482
10:28 whytewolf you can have more then one item block
10:28 whytewolf dcpc007: yes but this is more a pack thing then a general salt issue. salt-pack is what they use for building the repo
10:28 Whissi whytewolf: I think you missed my point. Then the individual config will be set first, but the node config will overwrite. But the individual config should be the winner...
10:31 whytewolf Whissi: not sure even if you got it to work. if it would merge the way you are thinking anyway. but you can try adding a node group that has everything and see if that does it for you
10:32 pbandark joined #salt
10:33 kbaikov joined #salt
10:43 vlebo joined #salt
10:46 Whissi For some reason I am getting now "[DEBUG   ] Could not LazyLoad pkg.normalize_name: 'pkg.normalize_name' is not available." errors when running highstate. When I run the failing state manually, everything works ;[
10:49 Whissi Mh. It is passing when I change state order
11:10 sh123124213 joined #salt
11:13 mavhq joined #salt
11:13 Whissi For some reasons I had to change an ID name.
11:15 Aayush_ joined #salt
11:16 N-Mi joined #salt
11:16 N-Mi joined #salt
11:16 Aayush_ left #salt
11:58 numkem joined #salt
12:03 edrocks joined #salt
12:05 sh123124213 joined #salt
12:14 dcpc007 back :)
12:15 dcpc007 whytewolf: maybe only one part of the block not updated, i never create deb package, but i see comment in other bug speaking about new package => new siganature, and maybe they not update the description file with the new signature
12:19 o1e9 joined #salt
12:21 ronnix joined #salt
12:28 numkem joined #salt
12:37 Rumbles joined #salt
12:39 sh123124213 joined #salt
12:41 seanz joined #salt
12:42 xet7 joined #salt
12:42 seanz Greetings. How does salt choose the environment in a multi git-pillar situation?
12:44 seanz Do the branches correspond directly to the branches, and that's the only option?
12:44 seanz Ah, never mind. I found what I need:
12:44 seanz To remap a specific branch to a specific Pillar environment, use the format <branch>:<env>:
12:44 babilen :)
12:47 seanz I might have a wacky situation. I am using two separate git repositories for staging / prod, both with the same master branch. Can I still use that remapping?
12:47 seanz I basically have something like this:
12:47 seanz - master:testing git@repo
12:47 seanz - master:production git@other-repo
12:48 seanz Is this supported?
12:48 seanz I am thinking I might have to set the environment config value explicitly.
12:52 rgrundstrom Im having 2 problems that i would relly like some help with. Please see https://gist.github.com/anonymous/78c3d4b3bc44eb80c7d9b0e80794d810#file-gistfile1-txt for problem description and code.
12:53 jhauser joined #salt
12:56 SaltyVagrant rgrundstrom: It look slike you need pillar stack (https://docs.saltstack.com/en/latest/ref/pillar/all/salt.pillar.stack.html)
12:56 _KaszpiR_ joined #salt
13:12 babilen seanz: That should™ work
13:12 rgrundstrom Good afternoon babilen :)
13:12 seanz babilen: It did not.
13:12 seanz I get an error that the subfolder in the wrong environment is not available.
13:13 seanz At this point, I'm going to recombine my pillar data into a single repo and target through grains.
13:13 seanz That's the recommended way, anyway.
13:23 amcorreia joined #salt
13:24 fxhp joined #salt
13:26 poliva joined #salt
13:27 _KaszpiR_ joined #salt
13:31 seanz Does anyone know why I'd get this error: http://dpaste.com/3A1DJ4Y ?
13:32 seanz From the logs, it looks like salt is still trying to use an old git ext_pillar, but I've changed it. How would I clear the cache?
13:33 amcorreia joined #salt
13:34 seanz Going with a good old-fashioned server reboot in lieu of something better.
13:36 seanz That seemed to work.
13:36 seanz Hm…maybe not.
13:36 seanz Nope, we're good. :)
13:47 zerocoo__ joined #salt
13:50 User__ joined #salt
13:50 benix joined #salt
13:54 relidy joined #salt
13:58 dxiri joined #salt
14:01 matt11_ joined #salt
14:01 matt11_ left #salt
14:02 matt11_ joined #salt
14:04 matt11_ is pillar data supposed to be ordered?
14:07 nku left #salt
14:12 SaltyVagrant matt11_
14:13 SaltyVagrant matt11_: Rather depends what you mean by ordered
14:13 matt11_ ordered dict
14:14 matt11_ appears to be this bug:https://github.com/saltstack/salt/issues/12161
14:18 SaltyVagrant matt11_: I don’t think you can rely on it being ordered. I wouldn’t :)
14:19 matt11_ @SaltyVagrant yeah, that pretty much sucks
14:20 SaltyVagrant matt11_: That said, it looks like it does use OrderedDict https://github.com/saltstack/salt/issues/11599
14:20 matt11_ @SaltyVagrant it says that it should, but then you read the last comments and it still doesnt....
14:21 matt11_ @SaltyVagrant currently I'm getting the same bug where it's not ordered, when it seems like it's expected to be
14:21 Inveracity joined #salt
14:23 SaltyVagrant matt11_: Personally, I treat it as unordered. If I really need some part ordering I either explicitly use a array/list or extract keys, sort those and iterate the keys.
14:26 ronnix joined #salt
14:37 moicmike joined #salt
14:38 cro joined #salt
14:42 sergeyt joined #salt
14:42 moicmike joined #salt
14:43 nicksloan joined #salt
14:47 Artik joined #salt
14:48 Guest_Salt_Teste joined #salt
14:51 fracklen joined #salt
14:52 moicmike joined #salt
15:04 edrocks joined #salt
15:11 sergeyt joined #salt
15:23 jas02 joined #salt
15:23 sh123124213 joined #salt
15:27 Praematura joined #salt
15:37 daxroc Evening all
15:40 daxroc I've run into a *bug*, on Amazon Linux 2016.03, Where it seems it tries to install python27-futures from repo and it fails due to that package not being available. Has anyone encountered this  before - any suggestions?
15:40 daxroc *salt-bootstrap script , stable
15:42 sh123124213 joined #salt
15:46 miruoy joined #salt
15:51 edgr joined #salt
16:23 relidy joined #salt
16:30 impi joined #salt
16:31 patrek joined #salt
16:33 Terminus joined #salt
16:33 cmichel joined #salt
17:07 edrocks joined #salt
17:07 penguinpowernz joined #salt
17:18 cyteen joined #salt
17:24 matt11_ joined #salt
17:27 k_sze[work] joined #salt
17:29 Praematura joined #salt
17:36 johnkeates joined #salt
17:42 cyborg-one joined #salt
17:52 fracklen joined #salt
17:53 shred joined #salt
17:55 Inveracity joined #salt
17:59 hexa- does the pillar top file allow `match: pillar` by any chance?
18:00 lordcirth_work hexa-, I don't think so, normal pillar files don't allow depending on other pillar data, to avoid recursion
18:00 hexa- I'm looking for a scheme to put a host in a group and assign pillars to the group
18:01 hexa- currently rolling $weirdstructure + host/ns1/init.sls for example
18:01 lordcirth_work hexa-, what's wrong with match: nodegroup?
18:02 hexa- that they're maintained in the master configuration
18:04 lordcirth_work hexa-, and you want them defined by an ext_pillar?
18:05 hexa- I believe so
18:06 lordcirth_work hexa-, it might be simpler for whatever is generating this group in ext_pillar to just put the relevant data in the minion's pillar?
18:07 hexa- likely
18:17 dyasny joined #salt
18:26 ChubYann joined #salt
18:27 justanotheruser joined #salt
18:30 nicksloan joined #salt
18:35 johnkeates joined #salt
18:40 dxiri hello everyone! quick question, does salt apply states automatically?
18:40 dxiri without me needing to run state.apply?
18:40 dxiri say a file changed at some point and I would like salt to automatically pick it up and replace it without any user intervention
18:42 xet7 joined #salt
18:43 justan0theruser joined #salt
18:50 iggy nope
18:52 iggy salt has a scheduler and beacons (that could react to file changes)
18:52 iggy but neither of those are enabled by default
18:53 Kelsar hmm bacon...
18:54 nafg joined #salt
18:54 nafg_ joined #salt
19:01 aldevar joined #salt
19:06 nicksloan joined #salt
19:17 shred joined #salt
19:18 dxiri ah ok :) thanks for the help!
19:23 Trauma joined #salt
19:26 nicksloan joined #salt
19:29 shanathon joined #salt
19:31 Praematura joined #salt
19:41 cainc joined #salt
19:45 Praematura joined #salt
19:46 cainc left #salt
19:46 cainc joined #salt
19:46 cainc left #salt
19:46 cain_ joined #salt
19:49 nicksloan joined #salt
19:49 cain_ Hey, could someone please recommend how I solve this? I want to use salt to initialize a cluster of new machines, but this involves generating TLS certs for each minion with the correct (using our own CA cert) - at the moment i'm thinking a bash script that iterates over the minions with a particular role and runs a script to generate the certificate for a minion (thinking this might have to be interactive for the CA cert password entry) - what is the
19:49 cain_ idiomatic approach?
19:55 aarontc joined #salt
20:04 rewbycraft cain_: Maybe take a look at: https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.x509.html
20:05 Trauma joined #salt
20:08 cain_ ahh thanks
20:23 VR-Jack2-H joined #salt
20:29 fracklen joined #salt
20:31 austin_ joined #salt
20:32 Bryson joined #salt
20:40 amcorreia joined #salt
20:42 nicksloan joined #salt
20:45 MTecknology cain_: for sharing the same cert in a cluster, I've considered picking one minion to request the cert and sticking it into the salt mine encrypted w/ the master's gpg key and then having a pillar w/ the gpg renderer share that to other systems. Never actually tried it, though. All of my shared stuff is pre-generated and stuck encrypted into pillar.
20:46 cain_ ah interesting, ill have a think about that
20:47 aldevar left #salt
20:47 MTecknology cain_: oh!!
20:47 MTecknology cain_: SDB!
20:48 MTecknology ext_pillar + sdb
20:50 MTecknology a reactor could be used to populate sdb from the one minion and ext_pillar could be used to read it
20:50 MTecknology cain_: I'm curious what you end up doing! :)
20:50 dxiri joined #salt
20:51 cain_ i'll look into that too, thanks :)
20:55 johnkeates joined #salt
21:09 schemanic joined #salt
21:09 schemanic Hi. I just read the statement 'Salt can't merge pillars' is that true?
21:10 MTecknology what does "merge pillar" mean?
21:12 schemanic I'm not sure, but I think it means where two dicts have the same key but different sub-keys
21:12 schemanic I read it here: https://github.com/saltstack-formulas/iptables-formula#using-iptablesservice
21:12 hemebond That's not true.
21:13 schemanic This person seems to want you to make separate .sls files for each service you want to add to iptables, but says you have to do something special to the state file they reference to make it work
21:13 schemanic I feel like I've used pillar merging before though
21:13 hemebond I also use pillar merging.
21:16 schemanic i bootstrap my master by cloning salt-fomula to the host, which has a minimal pillar which is a subset of the ext_pillar. They don't collide, they merge i believe
21:20 lordcirth_work pillar list merging is great
21:32 armyriad joined #salt
21:40 MTecknology schemanic: whoever wrote that didn't know what they were talking about. Merging that stuff has been possible for as long as I can remember. You just can't redefine the same key at the same level in the same sls file
21:41 MTecknology and if you need to do that, you should probably rethink what you're doing in the first place :)
21:43 MTecknology holy crap, that iptables formula makes my brain bleed.
21:49 nicksloan joined #salt
21:54 dxiri joined #salt
22:01 schemanic MTecknology, thanks for the look-over
22:02 jas02 joined #salt
22:10 edrocks joined #salt
22:13 schemanic This formula is kindof wierd
22:17 yoppa joined #salt
22:22 Corey joined #salt
22:25 Corey_ joined #salt
22:25 relidy joined #salt
22:26 Corey joined #salt
22:28 Corey joined #salt
22:30 Corey joined #salt
22:35 nafg__ joined #salt
22:45 Phanes joined #salt
22:47 sh123124213 joined #salt
22:55 Praematura joined #salt
23:09 MTecknology schemanic: https://gist.github.com/MTecknology/33b9e5a56fd5503484b5c4e77ae358e5
23:11 MTecknology also, yes, that can be written better. I welcome outside help! :D
23:17 nicksloan joined #salt
23:18 MTecknology I need to turn ferm into a formula...
23:18 MTecknology schemanic: you wanna test a formula for me?
23:32 dxiri joined #salt
23:38 nicksloan joined #salt
23:46 shred joined #salt
23:50 ssplatt joined #salt
23:55 Praematura joined #salt

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