Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-07-15

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

All times shown according to UTC.

Time Nick Message
00:00 Eugene yum/rpm have oddities when dealing with multilib packages
00:01 debian112 joined #salt
00:05 mort Interesting - even Ubuntu is wonky though.
00:06 rem5 joined #salt
00:10 ajw0100 joined #salt
00:11 emaninpa joined #salt
00:13 debian112 joined #salt
00:13 rem5 joined #salt
00:17 ninjada joined #salt
00:18 brotatochip joined #salt
00:25 DEger joined #salt
00:25 netcho joined #salt
00:28 subsignal joined #salt
00:31 woodtablet left #salt
00:35 alwaysatthenoc joined #salt
00:40 nZac joined #salt
00:41 subsignal joined #salt
00:44 scoates joined #salt
00:52 Hazelesque joined #salt
00:58 flowstate joined #salt
00:59 Nahual joined #salt
01:03 aw110f joined #salt
01:04 subsignal joined #salt
01:09 CTaylor_ joined #salt
01:17 iceyao joined #salt
01:22 iceyao_ joined #salt
01:22 catpigger joined #salt
01:22 shoemonkey joined #salt
01:22 ajw0100 joined #salt
01:25 netcho joined #salt
01:29 brotatochip joined #salt
01:32 iceyao joined #salt
01:35 ivanjaros joined #salt
01:46 alexlist joined #salt
01:46 Sammichmaker joined #salt
01:47 ilbot3 joined #salt
01:47 Topic for #salt is now Welcome to #salt! | Latest Versions: 2015.5.10, 2015.8.10, 2016.3.1 | Support: https://www.saltstack.com/support/ | Logs: http://irclog.perlgeek.de/salt/ | Paste: https://gist.github.com/ (please don't multiline paste into channel) | See also: #salt-devel, #salt-offtopic | Ask with patience as we are volunteers and may not have immediate answers
01:51 debian112 left #salt
01:51 RabidCicada joined #salt
01:54 Lionel_Debroux_ joined #salt
01:57 justanotheruser joined #salt
01:58 flowstate joined #salt
01:58 yomateo joined #salt
02:00 lero joined #salt
02:02 bastiand1 joined #salt
02:06 evle joined #salt
02:06 iceyao joined #salt
02:07 tristianc joined #salt
02:07 shoemonkey joined #salt
02:11 squishypebble joined #salt
02:13 debian112 joined #salt
02:16 nZac joined #salt
02:23 shoemonkey joined #salt
02:27 netcho joined #salt
02:33 fannet joined #salt
02:48 badon_ joined #salt
02:49 ninjada joined #salt
02:57 flowstate joined #salt
03:00 bastiandg joined #salt
03:06 auzty joined #salt
03:07 k_sze[work] joined #salt
03:19 nZac joined #salt
03:21 iceyao_ joined #salt
03:26 shoemonkey joined #salt
03:27 treaki__ joined #salt
03:28 netcho joined #salt
03:34 beardedeagle joined #salt
03:36 manji joined #salt
03:46 rem5 joined #salt
03:57 flowstate joined #salt
04:03 armyriad joined #salt
04:17 DEger joined #salt
04:19 DEger joined #salt
04:29 netcho joined #salt
04:47 rem5 joined #salt
04:56 flowstate joined #salt
05:05 ivanjaros joined #salt
05:10 futuredale_ joined #salt
05:12 yomateo- joined #salt
05:16 ivanjaros joined #salt
05:19 nZac joined #salt
05:26 fracklen joined #salt
05:28 vodik joined #salt
05:30 netcho joined #salt
05:39 kshlm joined #salt
05:41 rdas joined #salt
05:43 impi joined #salt
05:48 rem5 joined #salt
05:51 whitenoise joined #salt
05:51 abolte left #salt
05:52 felskrone joined #salt
05:56 ninjada joined #salt
05:59 Dragon535 joined #salt
06:20 nZac joined #salt
06:22 goldielox joined #salt
06:23 infrmnt joined #salt
06:29 onlyanegg joined #salt
06:30 netcho joined #salt
06:31 bastiandg joined #salt
06:37 Lionel_Debroux_ joined #salt
06:41 irctc875 joined #salt
06:49 rem5 joined #salt
06:50 ivanjaros joined #salt
06:51 jhauser joined #salt
06:57 SpX joined #salt
06:57 flowstate joined #salt
07:03 nagri How do I list minions that are spined up by my salt cloud ?
07:06 fracklen joined #salt
07:07 fracklen joined #salt
07:07 infrmnt joined #salt
07:08 ninjada_ joined #salt
07:10 KingJ joined #salt
07:17 ravenx joined #salt
07:23 ninjada joined #salt
07:28 infrmnt joined #salt
07:28 ax562 joined #salt
07:29 ax562 I have a quick question what is salt?
07:30 harkx ax562 , you could try browsing the website : https://docs.saltstack.com/en/latest/
07:31 netcho joined #salt
07:32 nagri Oh salt experts heed my cry. please have a look at this http://stackoverflow.com/questions/38390216/salt-master-salt-cloud-not-acting-idempotent
07:33 ax562 still don't understand harkx
07:36 harkx ax562, https://docs.saltstack.com/en/latest/topics/tutorials/walkthrough.html
07:36 nagri ax562: I am a beginner too. Salt has a steep learning curve. Get your hands dirty, thats the only way.
07:37 ravenx yup
07:37 alvinstarr joined #salt
07:37 ravenx it pays off in the end though
07:37 ravenx much faster than puppet/chef
07:37 ravenx much more stable between releases
07:37 harkx ax562, and google that question "what is saltstack" and make some coffee :)
07:38 ravenx and the language/dsl or w/e is actually clear.
07:38 dariusjs joined #salt
07:39 babilen nagri: Does it work if you install the minions manually?
07:39 * babilen doesn't trust saltify
07:43 manji joined #salt
07:44 krymzon joined #salt
07:46 irctc059 joined #salt
07:46 mikecmpbll joined #salt
07:47 Dragon535 left #salt
07:47 nagri babilen: Manually everything works
07:47 nagri I am trying to automate minion installation
07:48 nagri thats the pain.
07:48 babilen nagri: You'd still have to add them to saltify-map, don't you?
07:49 nagri yes but thats just one time thing. and I can spin up 10 minions at once. just by specifying IP username and password
07:49 rem5 joined #salt
07:50 babilen We use salt-ssh with the salt-formula for that. Just add minions to the roster and run "salt-ssh '*' state.highstate"
07:50 nagri what do you guys do in your prod enviornment ? to install minions ?
07:50 babilen The difference is that it actually wotrks
07:50 babilen You would also want to copy and paste your stackoverflow question to https://github.com/saltstack/salt/issues/
07:50 nagri but then with salt-ssh how do you accept minion keys on master ?
07:51 babilen https://github.com/saltstack-formulas/salt-formula
07:51 babilen I run "salt-key -A"
07:51 ax562 left #salt
07:51 babilen You could also auto-accept them or write a reactor that does that.
07:52 babilen I mean, saltify looks nice, but it is obviously not working the way it should.
07:57 nagri babilen: Posted the issue.
07:57 babilen ta
07:58 flowstate joined #salt
07:58 nagri #34687
07:59 impi joined #salt
07:59 babilen Yeah
07:59 Rumbles joined #salt
07:59 nagri So you are saying that right now the best way would be using salt-ssh and then manually accepting keys.
07:59 nagri right ?
08:00 nagri for minion installation
08:00 babilen I'm not saying that that would be the best way, but that works at least
08:00 babilen And I'm sure you have better things to do then sort our saltify bugs
08:00 nagri yeah :)
08:00 nagri Thanks.
08:05 mikecmpbll joined #salt
08:12 yomateo joined #salt
08:13 nZac joined #salt
08:13 kawa2014 joined #salt
08:14 hemebond joined #salt
08:17 yomateo- joined #salt
08:18 keimlink joined #salt
08:19 deniszh joined #salt
08:20 peters-tx joined #salt
08:22 citaret joined #salt
08:32 netcho joined #salt
08:36 mikecmpb_ joined #salt
08:37 kawa2014 joined #salt
08:48 s_kunk joined #salt
08:56 flowstate joined #salt
08:59 tuxick i noticed when running against multiple hosts --output=json doesn't produce valid json
08:59 tuxick it at least needs a header like { "minions" : [ and comma after eacht block
09:00 losh joined #salt
09:02 lero joined #salt
09:03 hemebond Hmm.
09:03 hemebond It might not be valid in its entirety, but I suspect that's because each is its own response.
09:04 hemebond Should at least be contained in a list.
09:05 tuxick yes, so i'm wondering if it'd cause trouble if this gets 'fixed'
09:05 tuxick now i have to first edit the output before i can throw it at any script that parses json
09:06 tuxick learning python by writing some reporting scripts :)
09:06 fracklen The output is not given in its entirety... The output is given as the responses come in - So you would have to wait until all responses are back before rendering
09:06 tuxick i know
09:06 fracklen That would IMHO make for a poor user experience
09:06 hemebond Yeah, probably better to wrap it in [] after collecting it all.
09:06 tuxick but all it takes is an initial line, append comma to each response, and end with ] }
09:07 tuxick simple enough
09:08 hemebond Isn't there a way to get each json on one line?
09:08 hemebond Better to have a list, a fetch each line as it comes in and append it to your list.
09:09 hemebond As a line, instead of inserting commas.
09:09 fracklen If you use --output=raw you get it as python dicts... That would be really easy to process
09:10 hemebond Ah, much easier.
09:10 fracklen And even better... --output=yaml actually produces valid yaml :)
09:11 LostSoul hi
09:11 hemebond Ah yes, YAML would give you, effectively, an entire response.
09:11 LostSoul I've met strange issue, salt doesn't want to update ssh keys once I 've changed file
09:11 hemebond LostSoul: hi
09:11 LostSoul Is it normal?
09:11 hemebond How are you updating SSH keys?
09:11 LostSoul I mean I keep keys in few files (in this case I'm using one)
09:12 LostSoul I've added new keys to this file and after state.highstate salts says it's already there
09:13 LostSoul I'm using ssh_auth.present:
09:13 LostSoul hemebond: This way: http://paste.debian.net/hidden/056dfd94/
09:16 hemebond LostSoul: You seem to be adding multiple - source properties.
09:16 hemebond Is that a correct thing to do?
09:16 hemebond Reading the docs it sounds like there should only be one source entry.
09:18 fracklen LostSoul: Just a note on style... I'd much rather put the public key as pillar data. Then you would use file.manged and use contents_pillar: sshkeys:users:{{ username }}:public_keys
09:20 fracklen Like this pillar: http://paste.debian.net/781598/
09:21 fracklen This would easily allow you to manage multiple keys per user
09:21 LostSoul Damn fracklen, thanks :)
09:21 hemebond source file can contain multiple keys.
09:21 hemebond "The source file for the key(s). Can contain any number of public keys, in standard "authorized_keys" format. If this is set, comment and enc will be ignored."
09:21 LostSoul Thanks hemebond on that note too :)
09:21 VSpike Hi. What is the optimum way to ensure static routes on a debian host with Salt?
09:22 VSpike salt.states.network.routes ?
09:22 LostSoul hemebond: in this case I have only 1 file but it still doesn't work
09:22 LostSoul fracklen: "Then you would use file.manged and use contents_pillar: sshkeys:users:{{ username }}:public_keys
09:22 LostSoul I didn't get this part ^
09:23 hemebond They mean to manually manage the file and its contents instead of using the ssh module.
09:23 hemebond VSpike: Looks like it.
09:24 fracklen LostSoul: using the file.managed state, you can specify a "contents_pillar" and refer directly to the pillar
09:25 fracklen https://docs.saltstack.com/en/latest/ref/states/all/salt.states.file.html#salt.states.file.managed
09:25 LostSoul Aaa
09:25 LostSoul You mean authorized_keys?
09:25 fracklen yeah
09:25 LostSoul I mean to manage it with file.managed module and get data from salt-pillars
09:25 LostSoul I see
09:26 fracklen I prefer to have my state handle everything, so to add a new user, I only have to change pillar data
09:27 armyriad joined #salt
09:28 VSpike hemebond: Yeah, despite the warning about debian on that page I guess
09:29 VSpike Next question ... anyone know how to use it, or do I need to dive into the code?
09:30 hemebond VSpike: There's an example in the docs.
09:30 tuxick fracklen: thanks, --output=raw will save some work :)
09:30 hemebond !states.network
09:30 VSpike Oh yeah, I missed it ... sorry
09:30 hemebond Dang it tuxick, for a split second I thought I had the right command.
09:30 sud0x3 joined #salt
09:32 tuxick ?
09:32 tuxick well my suggestion would be 'better' i think
09:33 netcho joined #salt
09:33 hemebond When I posted !states.network your comment popped up immediately, making me think I'd got the command correct.
09:33 tuxick since with very little changes it'd be valid json i can use for reporting
09:33 hemebond Apparently !states.network does nothing.
09:34 hemebond saltstackbot: states.network
09:34 armyriad joined #salt
09:39 armyriad joined #salt
09:42 LostSoul fracklen: I try to do same way :)
09:42 LostSoul Thanks for tip ;;)
09:42 peters-tx joined #salt
09:42 fracklen Sure thing :)
09:42 LostSoul fracklen: I just never met this "contents_pillar:"
09:43 sjorge joined #salt
09:47 pfallenop joined #salt
09:51 rem5 joined #salt
09:51 pfallenop joined #salt
09:54 mikecmpbll joined #salt
09:54 cyborg-one joined #salt
09:54 bluenemo joined #salt
09:56 flowstate joined #salt
09:58 hunsu joined #salt
10:00 hunsu I want to manage a dotfile. How to set the owner of the file as the owner of the directory where it will be copied ?
10:01 LostSoul I don't understand it
10:01 hemebond hunsu: modules to manage files let you specify perms.
10:01 LostSoul I mean I have 3-4 files in some cases and keys are copied
10:01 LostSoul But when I edit file (and I even use only 1 file) it doesn't add changed :O
10:01 LostSoul changes *
10:03 quarcu hello
10:03 quarcu is it possible to use jinja in topfile ?
10:03 hemebond quarcu: yes
10:04 atf joined #salt
10:04 impi joined #salt
10:05 iceyao joined #salt
10:05 quarcu so, i.e. this will work ? https://gist.github.com/piotrs-kainos/eb73995f1ae6e350bff5afdbff07f213
10:06 hemebond At a glance, sure.
10:06 hemebond Wait, no.
10:06 hemebond You still have to target.
10:07 hemebond Even if it's just '*'
10:07 hunsu hemebond: I use file.managed but if set the mode but not the user the owner is root
10:08 hemebond hunsu: Sure. The user will be root because Salt created it.
10:08 quarcu ah, right, forgot about that ;) was just an example
10:08 quarcu https://gist.github.com/piotrs-kainos/eb73995f1ae6e350bff5afdbff07f213
10:08 hemebond quarcu: Looks fine to me.
10:08 quarcu kk, thx hemebond
10:08 quarcu will test it
10:08 hemebond :thumbsup:
10:11 quarcu hemebond: do you think this one as well might work: https://gist.github.com/piotrs-kainos/eb73995f1ae6e350bff5afdbff07f213
10:11 hunsu hemebond: is it possible to change it to the owner of the directory?
10:11 hemebond quarcu: That doesn't seem right.
10:12 hemebond hunsu: Yes, you can set the user/owner for a file you manage.
10:12 hemebond I don't know of a way to automatically know the directory owner, but you should know that if you're managing the file.
10:12 quarcu hemebond: what do you mean not right ?
10:12 quarcu not valid in syntax or simply approach ?
10:12 netcho joined #salt
10:13 hemebond quarcu: Both I think.
10:13 quarcu well, it is just a PoC for multienv
10:13 quarcu i.e. each env have a separate set of pillars
10:13 nZac joined #salt
10:14 hemebond Sure. No need to target on the grain if you're already applying a state with the environment name in it.
10:14 quarcu and it is meant for immutable infrastructure, where grain is being set at init of image creation
10:14 hemebond Oh this is a pillar top.sls?
10:14 quarcu yeah
10:14 quarcu i.e. I have dev, preprod, prod env
10:15 quarcu each having same set of pillars (quantity) but different data inside
10:15 quarcu this way I would have a very simple topfile, but not sure if that would work
10:15 hemebond Sure. Then you just need to know the env name and use it as a variable.
10:15 quarcu so, possibly this would work ?
10:15 hemebond Targeting on the grain doesn't seem to make sense here.
10:16 quarcu well, it does in my case, as I set grain once on build instance startup
10:16 quarcu it is built standalone
10:16 hemebond Sure, but what does targeting on the grain get you?
10:17 hemebond The targeting is for the minion ID.
10:17 quarcu I keep type of env I'm building against inside grain
10:17 hemebond Sorry that was wrong...
10:17 quarcu and that separates pillars for me
10:18 hemebond Your target will always match the current minion.
10:18 hemebond Actually it won't because you change the . to _ for env.
10:18 hemebond So env_name no longer matches env
10:19 hemebond So your targeting will always fail.
10:19 joshin joined #salt
10:19 joshin joined #salt
10:21 quarcu well, the replace here is just arbitrary, can be removed
10:21 hemebond Well then it will always match :-)
10:21 quarcu env_name could be i.e dev or dev_smth dev.smth
10:21 hemebond So you're not targeting any better than '*'
10:22 thalleralexander joined #salt
10:22 quarcu well, base is base, my case default for all nodes
10:23 quarcu and target via variable in grain would simplify my topfile
10:23 hemebond But you're not really because your target applies to "this" node.
10:23 hemebond So take any minion and walk through your top file.
10:24 hemebond if you set a variable based on this minion and then use that variable in a target, it will match :-)
10:24 quarcu well, it is sort of different approach I'm aiming
10:24 hemebond I'm referring to "env_name:{{env}}" btw.
10:24 quarcu I set env_name and role as grains
10:24 hemebond {{ env }}.nginx makes, sense.
10:25 hemebond That will apply the nginx pillar for the environment the minion is in.
10:25 quarcu then I build a golden image
10:25 quarcu yes, that is the case
10:25 quarcu however, what i was aiming with variable in target grain
10:25 hemebond But that target doesn't make sense.
10:25 quarcu was that currently I have a section for each env
10:26 hemebond Doesn't make sense. It will always match.
10:26 hemebond So you actually want something like "env_name:prd"
10:26 quarcu and this would simplify each section of env to a single one
10:26 quarcu yes
10:26 quarcu I set env_name : prd
10:26 quarcu then topfile would target pillars against that grain
10:27 hemebond That makes sense. Using {{env}} in the target doesn't.
10:27 quarcu and as I wrote, instead fo having section per env i.e. "env_name:prd": "env_name:dev"
10:27 quarcu would shrink to a single "env_name:{{env}}"
10:28 hemebond Well, no, because it will always match
10:29 quarcu what do you mean it will always match ?
10:29 hemebond I mean you can do it, but it's the same as '*'
10:29 hemebond The target "env_name:{{env}}" says "apply these pillars if the minion environment matches the minion environment"
10:30 hemebond If env_name is 'dev' the target will be "env_name:dev". If env_name is "sillybilly" the target will be "env_name:sillybilly". It will always match.
10:31 quarcu well yes, it will match particular env
10:31 ninjada joined #salt
10:31 quarcu and inside this match I would specify i.e {{env_name}}.nginx
10:31 hemebond No, it will match any environment.
10:32 hemebond Whatever the current minion has, it will match, because you have {{env}} set to whatever the minion has.
10:32 quarcu well, shouldn't it match the env I specify in grain ?
10:32 hemebond The grain is on the minion.
10:32 hemebond The grain belongs to that minion.
10:33 quarcu well, I set the grain only once and only for build
10:33 quarcu I dont use salt later on
10:33 hemebond The grain is used every time the minion gets a command.
10:33 quarcu yup, and I only use it once
10:34 hemebond I don't know what you mean.
10:34 quarcu trying to get there ;)
10:34 quarcu i.e. lets say I have dev, stage, prod
10:35 quarcu and each has a separate directory with pillars
10:35 quarcu I start the instance, set the grain, and based on grain I want to target a specific set of pillars for that particular env
10:36 quarcu so, ATM I have 3 sections, each one for each env
10:36 hemebond Ah
10:36 quarcu i.e. 1 for dev, 1 for stage, 1 for prod
10:36 hemebond Have you changed the top.sls file?
10:36 hemebond I'm still looking at one with the env_name grain target.
10:36 quarcu if I could use variable in grain match, that would respectively match each node for particular env based on grain
10:37 quarcu and instead having 1 section per env i.e. 3 sections in total for dev stage and prod
10:37 quarcu I would have 1 universal section which would match pillars based on variable grain set on instance start
10:38 hemebond You need to keep in mind that the top.sls targeting is done against each minion.
10:39 hemebond While you _could_ do targeting the way you currently have that file, it doesn't achieve anything since it will _always_ apply.
10:40 quarcu well, it is build as immutable, so it should always match some particular env and role
10:40 hemebond Make up a minion and grain and walk through your top.sls and see if the target matches.
10:40 hemebond immutable?
10:40 quarcu the image is build only once
10:40 quarcu salt is standalone and used only to build the image
10:40 hemebond Sure, but the top.sls is processed each time.
10:40 quarcu well, it is processed only once
10:40 hemebond Wait, what?
10:41 quarcu at build time of instance
10:41 quarcu after I have the image, I dont run salt anymore
10:41 hemebond (doesn't alter the fact that the target is the same as '*')
10:42 quarcu well, I have a separate directory / env
10:42 quarcu so i need to target it somehow
10:42 quarcu * does not do the job for me in that case
10:42 hemebond You've got that will with pillar name, {{env}}.nginx
10:43 hemebond In what case can you see "env_name:{{env}}" _not_ matching a minion?
10:43 hemebond Since it directly takes the env_name from the minion and then uses it in the target.
10:44 hemebond env=dev, env_name:dev
10:44 colegatron_origi joined #salt
10:44 hemebond env=prd, env_name:prd
10:44 hemebond You see how your variable is guaranteeing a match?
10:44 quarcu yeah, got it
10:44 hemebond Since env=env_name
10:45 quarcu yeah, makes sense
10:45 hemebond :thumbsup:
10:45 quarcu was trying to forcefully separate it on syntax level
10:45 quarcu thx for help :)
10:45 hemebond Your {{env}}.pillarname will do that.
10:45 amcorreia joined #salt
10:46 hemebond Assuming you want the exact same pillars used in every environment.
10:47 quarcu well, that is the part I was missing till now
10:47 quarcu ie I have a list of users, lets say admins and devs
10:47 quarcu for some env like dev or stage
10:47 quarcu both type of accounts will be there
10:47 quarcu but lets say on prod, there should only be admins
10:48 quarcu so, after all it is not always exactly the same
10:50 quarcu but I get your point now :)
10:50 kevinquinnyo1 joined #salt
10:51 hemebond I probably could have explained it better just by writing up an example :-P
10:51 hemebond Wrote a lot of words instead.
10:52 rem5 joined #salt
10:52 ninjada joined #salt
10:53 hemebond I was also going to suggest actual separate Salt environments, but they're not so straight forward.
10:53 quarcu no probs
10:54 quarcu I get the idea
10:54 quarcu TBH I can work a way to get separate list of users per env
10:54 quarcu i.e. make base.users.admins
10:54 quarcu and {{env}}.users
10:54 quarcu which will have a env specific accounts
10:54 quarcu so not that bad after all
10:54 hemebond Yeah, using a directory for each env should work fine.
10:55 quarcu yup, currently doing it that way
10:55 hemebond And works almost the same as Salt environments.
10:56 quarcu can you provide a link to environments you are talking about ?
10:56 quarcu don't recall this functionality
10:57 netcho joined #salt
10:57 flowstate joined #salt
10:57 hemebond https://docs.saltstack.com/en/latest/ref/states/top.html#environments
10:58 hemebond It's a way of compartmenting your states and pillars, but if you're not careful it's easy to botch it and get stuff all over the place.
10:58 hemebond *separating your states...
10:59 iceyao joined #salt
11:00 hemebond Actually, it might work well for you since you are already setting minion settings for environments.
11:01 hemebond Rather than setting a grain you can just set the minion config property.
11:04 catpig joined #salt
11:11 _JZ_ joined #salt
11:12 ninjada joined #salt
11:15 quarcu ah, right, remember that
11:15 quarcu we have the old environment where one of my colleagues tried it that way and totally didn't get it the right way
11:15 ninjada joined #salt
11:15 quarcu we needed to have pillars data merging
11:15 quarcu and with envs it didn't work as we wanted
11:36 TomJepp joined #salt
11:37 ryan8403 joined #salt
11:39 rem5 joined #salt
11:41 LostSoul fracklen: Hi
11:41 LostSoul I've tried your solution, but ssh keys are still not changed
11:47 rem5 joined #salt
11:51 kawa2014 joined #salt
12:11 numkem joined #salt
12:11 ninjada joined #salt
12:14 nZac joined #salt
12:14 tuxick ugh, output of iptables.get_rules is *awful*
12:15 XenophonF let's face it, the output of iptables anything is awful
12:15 * XenophonF isn't a fant
12:15 XenophonF er
12:15 * XenophonF isn't a fan.
12:17 tuxick oh i can deal with iptables output
12:17 tuxick but no newlines = really painful
12:18 XenophonF are other output options available? what if you use --output json or something?
12:18 XenophonF (assuming you're running that function via salt/salt-call from the command line)
12:27 jschoolcraft joined #salt
12:28 zer0def joined #salt
12:30 squishypebble1 joined #salt
12:36 fredvd joined #salt
12:45 sud0x3 joined #salt
12:47 gh34 joined #salt
12:50 M-liberdiko joined #salt
12:51 mage_ joined #salt
12:54 subsignal joined #salt
12:55 TooLmaN joined #salt
12:55 subsigna_ joined #salt
12:58 tracphil joined #salt
12:58 ronnix joined #salt
13:02 lero__ joined #salt
13:03 mikecmpb_ joined #salt
13:03 shoemonkey joined #salt
13:06 jirobulova joined #salt
13:17 west575 joined #salt
13:18 Andrew joined #salt
13:19 lero joined #salt
13:21 amcorreia joined #salt
13:26 M-MadsRC joined #salt
13:27 mapu joined #salt
13:29 kevinquinnyo1 joined #salt
13:31 shoemonkey joined #salt
13:32 protoz joined #salt
13:38 lero joined #salt
13:40 shoemonkey joined #salt
13:42 nZac joined #salt
13:43 ivanjaros joined #salt
13:47 dyasny joined #salt
13:48 perfectsine joined #salt
13:52 lero joined #salt
13:52 DEger joined #salt
13:56 shoemonkey joined #salt
13:57 tapoxi joined #salt
14:09 flowstate joined #salt
14:16 flowstate joined #salt
14:18 kaptk2 joined #salt
14:25 johnkeates joined #salt
14:25 iceyao joined #salt
14:26 racooper joined #salt
14:28 mikecmpbll joined #salt
14:29 nZac joined #salt
14:30 corichar joined #salt
14:35 danielcb joined #salt
14:36 MadHatter42 joined #salt
14:39 lero joined #salt
14:42 llua joined #salt
14:52 danielcb joined #salt
14:58 whitenoise joined #salt
14:58 impi joined #salt
14:59 lero joined #salt
15:02 MajObviousman so 2 spaces is the default indentation width of yaml, but 4 spaces is the default of a lot of code these days
15:02 MajObviousman this is frustrating me for reasons that I can't comprehend
15:04 iceyao joined #salt
15:05 aphor <MajObviousman: makes programmers editors difficult and indespensable.
15:05 aphor Good morning SaltStack!
15:06 MajObviousman since we're re-rolling our salt states, I have the opportunity to change them all to 4 spaces. I wonder how much yelling that would cause
15:06 aphor Is there a reason why salt.cloud driver functions are all loaded when running salt.cloud execution, states, client commands, etc?
15:07 aphor MajObviousman: provide vim settings in the headers of the files to help ease the pain?
15:08 bluenemo joined #salt
15:08 MajObviousman that's ... not a bad idea
15:08 MajObviousman then I wouldn't have to screw with it
15:10 aphor need to add unit tests to make sure style enforcements/facilities are reliably and consistently used.
15:10 MajObviousman and now would be the good time to start that
15:11 MajObviousman though I don't believe we have many custom needs, TBH
15:23 aphor https://github.com/s0undt3ch <-- Is this guy around on #salt?
15:25 ribx joined #salt
15:25 tapoxi joined #salt
15:28 brent_ joined #salt
15:33 aphor I'm kinda wishing salt cloud loader was using _virtual_ instead of LIBCLOUD_FUNCS_NOT_SUPPORTED
15:33 aphor This is 2013 stuff.
15:35 Salander27 joined #salt
15:44 DEger joined #salt
15:44 onlyanegg joined #salt
15:44 debian112 joined #salt
15:47 teryx510 joined #salt
15:49 aphor Is coffee working yet in MDT TZ?
15:49 teryx5101 joined #salt
15:51 nyx joined #salt
15:55 krymzon joined #salt
15:56 flowstate joined #salt
16:05 alinuxninja joined #salt
16:16 corichar Does anyone know how to trigger an update for the salt development documentation?
16:17 aphor corichar: ask jfindlay directly
16:18 corichar jfindlay: can you trigger a update for the development docs?
16:20 lungaro joined #salt
16:21 aw110f joined #salt
16:31 ponyofdeath hi, is there a way for pkg.installed: - sources: to check if package is installed before downloading again?
16:32 nyx joined #salt
16:35 iggy no
16:35 iggy use aptly... it's easy to setup
16:35 johnkeates check if it is installed as a prerequisite
16:37 dyasny joined #salt
16:37 ponyofdeath ok let me look at what aptly is
16:37 ponyofdeath thanks
16:37 msn joined #salt
16:38 msn I am getting while starting salt-master using the default master file IndexError: string index out of range
16:39 perfectsine_ joined #salt
16:41 iggy msn: check the issue tracker?
16:42 shoemonkey joined #salt
16:43 msn i don't think its an issue
16:43 msn something config
16:43 infrmnt joined #salt
16:43 msn it was not there earlier and suddenly pops up
16:43 coval3nce_ joined #salt
16:44 DanyC joined #salt
16:44 iggy roll your configs back to the previous version
16:45 DanyC hi, anyone has any experience with the salt-cloud vmware provider ?
16:46 writtenoff joined #salt
16:46 armyriad joined #salt
16:47 Jaremy joined #salt
16:49 infrmnt joined #salt
16:50 msn left #salt
16:50 msn joined #salt
16:51 shoemonk_ joined #salt
16:54 edrocks joined #salt
16:56 flowstate joined #salt
16:56 skogg joined #salt
17:01 nyx joined #salt
17:03 ageorgop joined #salt
17:06 johnkeates left #salt
17:06 nZac joined #salt
17:06 debian112 left #salt
17:07 brotatochip joined #salt
17:09 onlyanegg joined #salt
17:16 gantan joined #salt
17:28 KingJ joined #salt
17:30 zero_shane joined #salt
17:34 dtsar joined #salt
17:34 dtsar hi all - can we talk about how to deal with the headache of nodejs version management?
17:35 dtsar would love to hear experiences
17:35 POJO joined #salt
17:35 dtsar i'm not at a state where i deploy new machines with every release; machines are longer-living than i would like
17:36 dtsar but occasionally bring new machines up
17:36 dtsar and node updates so frequently that i'm constantly facing version conflict
17:36 dtsar because older packages aren't accessible via the ubuntu PPA
17:37 dtsar so should i just copy down a .deb file and store on the master and distribute that via file.managed?
17:37 bastiandg joined #salt
17:39 brotatochip joined #salt
17:41 yomateo joined #salt
17:41 aphor DanyC: I'm not familiar with VMWare yet, but I've tried EC2 and DigitalOcean drivers, and there's a lot of old stuff that wants refactoring if you find bugs.
17:42 gimpy2938 I need a simple way to say "if a certain IP in a certain subnet" in an SLS file, what's the easiest way to do this?  I can do it in Python easily but then what?  Make that an execution module?  How do you use them in an SLS file?
17:43 whytewolf gimpy2938: exacution modules are run through jinja. also there already is a network exacution module
17:43 gimpy2938 whytewolf: got a link?  couldn't find anything in Salt that can do that
17:44 whytewolf gimpy2938: https://docs.saltstack.com/en/2015.8/ref/modules/all/salt.modules.network.html#salt.modules.network.in_subnet
17:44 aphor gimpy2938 you can do python expressions in Jinja.
17:45 whytewolf gimpy2938: https://docs.saltstack.com/en/2015.8/ref/modules/all/salt.modules.network.html#salt.modules.network.ip_in_subnet
17:45 pfallenop joined #salt
17:46 whytewolf {% if salt.network.ip_in_subnet(ip_addr='192.168.1.1',cidr='192.168.1.0/24') %}
17:48 gimpy2938 whytewolf: it would be fantastic if docs included actual real life useful and working examples like that
17:49 gimpy2938 for some reason over 90% of salt docs are just "x does x, to use x call x()" then 10 minutes of Google later I have 7 examples from random blogs that don't work and 1 that does
17:50 whytewolf it gets a bit one sided if they did that for every function
17:50 GreatSnoopy joined #salt
17:50 whytewolf so they teach that exacution modules can be run through jinja like that. then leave it up to the user to remeber it
17:51 gimpy2938 whytewolf: I don't see anything even similar to your example in the main "modules" doc
17:52 whytewolf it is in the jinja docs IIRC
17:52 whytewolf which is where it would be expected to be
17:53 whytewolf https://docs.saltstack.com/en/latest/topics/jinja/index.html#calling-salt-functions
17:53 jfindlay corichar: I will look into that.  They are regenerated daily from what I remember.  Is there docs on a specific branch that you need updated?
17:55 gimpy2938 I see, if I am learning modules I would expect to have the docs be there; if related docs are elsewehere it should should clearly reference and link them
17:55 flowstate joined #salt
17:55 whytewolf gimpy2938: but you are not learning modules in that instance you are learning jinja
17:56 fracklen joined #salt
17:56 corichar jfindlay: The system module's documentation is not up to date with the development branch: https://docs.saltstack.com/en/develop/ref/modules/all/salt.modules.system.html#module-salt.modules.system
17:56 whytewolf salt '*' network.ip_in_subnet 192.168.1.1 192.168.1.0/24
17:56 whytewolf would be the standard module call
17:57 whytewolf but if you feel the documentation can be updated. it is in the github and avalible for anyone to modify
18:01 jfindlay also, unless there has been a change within the past day, the docs should be accurate, so if not, there could be a bug in the doc build system or index page(s)
18:01 yomateo joined #salt
18:02 jfindlay corichar: if you have installed develop, it is possible that your local docs don't match the docs on the website if you haven't updated your install in a while
18:03 infrmnt joined #salt
18:06 teryx510 joined #salt
18:11 mikecmpbll joined #salt
18:12 ALLmightySPIFF joined #salt
18:18 CTaylor joined #salt
18:20 corichar jfindlay: no I'm looking at the online docs and they are not matching the pydocs in the development repo. Although I just noticed a typo with my documentation when I built it locally.
18:20 ageorgop joined #salt
18:20 SpeeR joined #salt
18:22 nZac joined #salt
18:25 ronnix joined #salt
18:33 gimpy2938 left #salt
18:36 saltling joined #salt
18:36 mikecmpbll joined #salt
18:40 pcdummy I started to write a saltstack module for LXD, any interest here? Where to put for review? - https://gist.github.com/pcdummy/67332678fcf7ecb95343d630068a17a8
18:42 iggy pcdummy: interest
18:43 iggy pcdummy: at first glance, to have full end to end, you should add lxd init ability
18:44 pcdummy iggy: god idea, will add that!
18:45 _JZ_ joined #salt
18:45 cableninja joined #salt
18:48 shoemonkey joined #salt
18:48 iggy planning a state module to go with it?
18:48 cableninja joined #salt
18:49 pcdummy iggy: yes
18:56 flowstate joined #salt
18:58 mtw-local_ joined #salt
18:59 aphor Can I use cmd.call from a YAML state.sls file?
19:01 MTecknology OMFG!!! I have missed salt *sosososooooo* much! Feels incredible to be back! :D
19:01 whaity joined #salt
19:02 aphor cmd.call is telling me func: cloud.something  string 'cloud.something' is not callable.
19:02 aphor :(
19:03 MTecknology aphor: did you configure the provider?
19:03 pcdummy iggy: there you go https://gist.github.com/pcdummy/67332678fcf7ecb95343d630068a17a8#file-lxd-py-L65
19:04 aphor MTecknology: yeah, I can run the cloud.action using salt-call or even different syntax with salt-cloud, but not in a cmd.call state.
19:05 MTecknology wait... you're trying to call the command "cloud.something"?
19:05 aphor It's like the cmd.call doesn't LazyLoad the func=whatever argument
19:06 MTecknology aphor: put your state on dpaste.com
19:06 MTecknology or gist
19:06 Ch3LL joined #salt
19:06 aphor Actually, I'm trying to cmd.call name: cloud.action
19:06 MTecknology cloud.action probably isn't a valid command available on your system
19:07 MTecknology perhaps you're looking for '- name: salt-call cloud.action'
19:08 aphor https://gist.github.com/aphor/8488098db3d1b0dbfeaa4bc435b9eb6c
19:10 s_kunk joined #salt
19:11 MTecknology heh... cmd.call isn't what I thought it was
19:11 MTecknology I was thinking cmd.wait
19:11 jfindlay corichar: what are the differences you are seeing?
19:11 MTecknology aphor: "Invoke a pre-defined Python function with arguments specified in the state declaration. This function is mainly used by the salt.renderers.pydsl renderer."
19:11 jhauser joined #salt
19:12 aphor MTecknology: that's not discouraging enough to keep me from trying it...
19:13 corichar jfindlay: https://docs.saltstack.com/en/develop/ref/modules/all/salt.modules.system.html#module-salt.modules.system differs from https://github.com/saltstack/salt/blob/develop/salt/modules/system.py
19:13 corichar jfindlay: for example the function get_system_time the parameters are now strings but the docs say they are booleans
19:14 MTecknology aphor: have you done much with -watch:? It sounds like that's what you really need for this state, and for it to be a cloud state instead of cmd.
19:15 MTecknology I could be misunderstanding, but it seems like you're trying to abuse a system that doesn't need to be abused. :)
19:15 corichar jfindlay: I did notice a typo in the my docs that I have a PR for so it really doesn't need to be updated until after that PR is merged.
19:18 jfindlay corichar: would you mind opening an issue on github with the differences you are seeing?
19:18 corichar jfindlay: yeah sure
19:20 jfindlay thanks
19:24 shoemonkey joined #salt
19:26 colegatron_origi joined #salt
19:31 lero joined #salt
19:31 aphor ec2 cloud driver does not implement the necessary volume_list function which state.cloud.volume_present uses.
19:32 aphor :( :( :(
19:32 aphor also the cloud loader is a hairball.
19:33 aphor ... and messing with the loader will require a lot of buy in from Salt ppl before a merge request would even be considered.
19:36 deniszh joined #salt
19:38 MTecknology aphor: open a bug report to begin discussion?
19:45 mikecmpbll joined #salt
19:45 mavhq joined #salt
19:45 scoates joined #salt
19:47 iggy pcdummy: <3 that's been on my TODO for a while
19:49 iggy "15:15 < MTecknology> I could be misunderstanding, but it seems like you're trying to abuse a system that doesn't need to be abused. :)" lol, most of salt
19:50 pcdummy iggy: same for me, i have the dream of a Web Panel for containers - i'm not far away of it.
19:51 pcdummy Where you can say i want a Ubuntu-Xenial-Nginx-PHP7.0 container and it creates you that one.
19:52 pcdummy and ofc all the other stuff around (configure a HTTP-router, configure dns and so on.)
19:52 shoemonkey joined #salt
19:55 flowstate joined #salt
19:55 edrocks joined #salt
20:02 aphor MTecknology: need to meditate on the path forward. I have an initial issue ticket, but not sure if/how I would recommend fixing it.
20:05 iggy pcdummy: you're probably going to want a way to set the config options (outside of init)... so when the state module has a different password than the one that's set (or if there wasn't one initially set, etc)
20:05 krymzon joined #salt
20:06 debian112 joined #salt
20:06 lero joined #salt
20:07 fredvd joined #salt
20:09 pcdummy iggy: already done.
20:10 pcdummy https://gist.github.com/pcdummy/67332678fcf7ecb95343d630068a17a8#file-lxd-py-L149
20:10 iggy lol, dirty minds
20:10 iggy I mean great minds..
20:11 CimmerianX joined #salt
20:12 CimmerianX ANyone here successfully use the windows minion command for system.join_domain?
20:14 iggy pcdummy: would some of this be better achieved with python-pylxd ?
20:15 nZac joined #salt
20:18 dyasny joined #salt
20:21 theblazehen joined #salt
20:28 CimmerianX THe Windos minion is returning False when I run the command with a netbios domain name, and returns a 1355: can't find domain when I give it the DNS entry for the domain.     I verified that the domain name is resolvable (I can join domain manually).
20:28 theblazehen Can someone explain why formulas see the best way to create third party "plugins" if that's the correct word? All the built in states take their arguments directly, whereas with formulas you have to pass the data in pillars. For example you don't define the packages that pkg.install installs in your pillar data, but by passing data to it directly
20:28 theblazehen CimmerianX: Which command?
20:28 CimmerianX from master I use the following:       sa
20:29 tapoxi joined #salt
20:29 pcdummy iggy: no that cant do this as pylxd doesn't have "lxc config set <server-key> <server-value>"
20:29 aphor theblazehen: can you do it with _modules in formulae?
20:29 pcdummy iggy: all what is doable with python-pylxd will be done with it.
20:29 tapoxi anyone on centos/rhel and know a good formula for configuring ldap?
20:29 dyasny joined #salt
20:29 theblazehen aphor: Yes, I can
20:29 iggy pcdummy: rgr, just wanted to make sure you weren't going crazy with the cmd.run's when there was a cleaner way
20:30 theblazehen I just think it would be nice to have a way for the community to share the _modules etc like they can do with the saltstack-formulas repo on github
20:30 CimmerianX salt minion-name system.join_domain domain='domainname', username='ldapadmin, password='hispassword', account_ou='cn=Computers,dc=thedomain,dc=local', account_exists=False
20:30 iggy theblazehen: salt-contrib
20:31 pcdummy iggy: thanks for the input.
20:31 iggy theblazehen: and formulas are generally preferred when there's a lot of variablity in the setup of something
20:31 shoemonkey joined #salt
20:32 theblazehen iggy: Yes, I saw that. I feel that there should be a single repository per contributed module. That way I would be able to also add it to file_roots with gitfs
20:33 iggy theblazehen: nobody is arguing (I would actually like it because salt-contrib hasn't had a merge in ages and I've got PR's just sitting there doing nothing)
20:34 theblazehen I would like it if I could create a repo, and share a whole state+module that has as much functionality as the built in states does
20:34 theblazehen Yeah. I wonder if the developers would be opposed if I created an organization on github for salt modules?
20:35 theblazehen Not sure what the best term for them would be. Modules? Plugins?
20:35 iggy https://github.com/saltstack-contrib
20:35 theblazehen iggy: Hmm. There's nothing there
20:35 iggy that's because I just made it
20:36 theblazehen Ah. cool
20:36 theblazehen https://github.com/tony/salt-states-configs/tree/master/salt/roots/salt would be nice to add some of these in there.
20:37 CimmerianX @theblazehen
20:38 theblazehen iggy: How should repo naming be done? For example, even if a salt forumula isn't in the saltstack-formulas organization you will recognize it by the foobar-formula name
20:41 babilen iggy: What's the plan with salt-contrib ?
20:43 babilen Community maintained https://github.com/saltstack/salt-contrib ?
20:44 subsignal joined #salt
20:45 adelcast joined #salt
20:45 edrocks joined #salt
20:45 pcdummy iggy: thanks for the input! current version of salt-lxd is online: https://gist.github.com/pcdummy/67332678fcf7ecb95343d630068a17a8
20:49 whytewolf iggy: first contribution? https://github.com/whytewolf/salt-debug
20:49 adelcast joined #salt
20:54 subsigna_ joined #salt
20:55 theblazehen pcdummy: Great!  Was going to do it myself, didn't get around to it yet
20:55 flowstate joined #salt
20:56 pcdummy Does "salt * my.cmd" parse input data like json dicts ?
20:57 pcdummy I want -- salt-call lxd.create_container mycontainer "['default', 'autostart']" "{'type': 'image', 'alias': 'xenial/amd64'}" 2 false
20:57 pcdummy this would create mycontainer with from the image xenial/amd64
20:58 pcdummy ahh and its container_create
21:01 onlyanegg joined #salt
21:06 ajw0100 joined #salt
21:06 theblazehen pcdummy: Or use the REST API directly?
21:07 theblazehen Wait, nevermind
21:07 brotatochip joined #salt
21:07 theblazehen Read your sentence wrong, didn't check the code yet
21:08 pcdummy theblazehen: basicaly i just want "salt-call" or / "salt" to parse the input from json.
21:08 pcdummy didn't test it yet, so don't know if that works.
21:10 pcdummy https://github.com/lxc/pylxd/blob/master/integration/testing.py#L42 <-- this is the stuff i want to give per commandline
21:17 onlyanegg joined #salt
21:18 shoemonkey joined #salt
21:29 onlyanegg joined #salt
21:32 iggy babilen: yeah, since nobody merges shit on salt-contrib
21:34 iggy pcdummy: I'm pretty sure it does (without needing to specifically tell it that it needs to be parsed)
21:35 pcdummy iggy: just tested, does not or idk howto.
21:36 iggy damn :/
21:37 iggy maybe see how the code looks for salt.modules.state.highstate in regards to pillar handling (it supports that kind of arg passing)
21:37 pcdummy ta!
21:40 deniszh joined #salt
21:43 |_[O_O]_| clear
21:48 keimlink joined #salt
21:51 pcdummy state.highstate is special, not sure howto make that happen....
21:51 pcdummy https://gist.github.com/pcdummy/67332678fcf7ecb95343d630068a17a8#file-lxd-py-L245 <-- this my untested implementation of container_create
21:51 pcdummy https://gist.github.com/pcdummy/67332678fcf7ecb95343d630068a17a8#file-lxd-py-L285 <-- this
21:53 ponyofdeath hi, what is a good way to provision a salt-cloud server into aws or digital ocean? seems the steps i am finding on google are doing it in a manual way. using the salt-formula maybe?
21:54 iggy ponyofdeath: I use the cloud map functionality
21:55 ponyofdeath so basically you call the cloud map after setting up the digital ocean provider
21:55 colegatron_origi joined #salt
21:55 ponyofdeath and one of the instances is the salt-cloud master server?
21:56 ponyofdeath so where is the seed point?
21:56 pcdummy root@saltmaster:~# salt 'srv02*' lxd.container_create test source='{"type:" "image", "alias": "xenial/amd64"}'     srv02.pcdummy.lan:    ERROR executing 'lxd.container_create': 'source' must be formatted as dictionary.
21:57 iggy ponyofdeath: there isn't a bootstrap_all_the_things button in salt... there's is a minimum amount of work to get things up and going
21:57 flowstate joined #salt
21:57 iggy ponyofdeath: such functionality wouldn't exactly make sense in every case (think salt-ssh, masterless, etc)
21:58 ponyofdeath iggy: yeah i understand, what i am asking is for the most acceptible way
21:58 ponyofdeath so you are saying that salt-master server is seeded manually on a cloud provider first then?
22:01 iggy I actually started off with salt-cloud on my laptop, brought up the master from there and then off I went
22:02 iggy at my last job, we had a minimal bootstrap script that talked directly to the GCE api and ran the master bootstrap steps via cloud-init
22:02 onlyanegg joined #salt
22:06 aw110f joined #salt
22:07 mohae joined #salt
22:10 Edgan I use salt-ssh to use the same code I use to maintain a salt master to setup a new salt master.
22:10 debian112 joined #salt
22:11 ponyofdeath cool thanks
22:15 mohae joined #salt
22:16 amcorreia joined #salt
22:17 subsignal joined #salt
22:21 pcdummy it works now: https://gist.github.com/pcdummy/67332678fcf7ecb95343d630068a17a8#file-lxd-py-L288
22:21 pcdummy but still no auto parsing of input values :/
22:38 UForgotten joined #salt
22:39 shoemonkey joined #salt
22:45 ewd84 joined #salt
22:45 ewd84 Hello, a quick question for resident gurus: Any potential issues with running balance and scrub at same time?
22:54 deniszh joined #salt
22:56 flowstate joined #salt
22:58 _JZ_ joined #salt
23:03 keimlink_ joined #salt
23:04 iggy ewd84: misfire?
23:04 scoates joined #salt
23:09 ageorgop joined #salt
23:18 UForgotten joined #salt
23:22 fivehole joined #salt
23:24 mrBen2k2k2k joined #salt
23:25 fivehole Does anybody know if you can get a pillar value based on a variable?  like {% set a = "val1" %}  {{ pillar[a] }}
23:29 pcdummy fivehole: should be easy, "pillar.get(a, <default-value>)"
23:31 badon_ joined #salt
23:33 fivehole @pcdummy, you are right.  I notice now while I was simplifying my question I removed the part I didn't know the answer to.
23:34 ageorgop joined #salt
23:37 fivehole so say you have a nested pillar value:  val1:  -name: home -address 25 main.   doesn't seem like pillar[a]['address'] works
23:38 fivehole assuming {% set a = "val1" %}
23:40 mrBen2k2k2k joined #salt
23:41 DEger joined #salt
23:46 iggy [key] is dict key lookup... you have some lists in there
23:47 fivehole so lose the -
23:48 iggy not necessarily, but your lookups have to match your data structure
23:53 amcorreia joined #salt
23:54 childc2 joined #salt
23:54 fivehole And that did the trick.  Guess I need to meet it half way keep my data structures in sync
23:55 ajw0100 joined #salt
23:56 flowstate joined #salt
23:59 childc2 Hi everyone, I've discovered an issue with the virtual grain after updating to salt 2016.3.1 (Boron). On the two VMware machines that I've restarted the minion post upgrade they are reporting that they are physical hosts. I've been stepping through /salt/grains/core.py from the github page and it looks like it should be detecting my platform with lspci. I don't have systemd-detect-virt, virt-what, or dmidecode installed so lspci should be

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