Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-09-10

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

All times shown according to UTC.

Time Nick Message
00:00 ageorgop joined #salt
00:04 armguy joined #salt
00:09 DEger joined #salt
00:13 penguinpowernz joined #salt
00:15 onlyanegg so now my orchestrate runner seems to be matching my mgmt box rather than what I've specified for 'tgt'
00:15 onlyanegg https://gist.github.com/onlyanegg/2779479652b7e5d50b5c819f92441a94
00:18 DEger joined #salt
00:19 PerilousApricot joined #salt
00:25 flowstate joined #salt
00:28 DEger joined #salt
00:29 amy joined #salt
00:31 skeezix-hf joined #salt
00:32 raspado joined #salt
00:33 raspado i have a state like zfs.setup and zfs.users, how can I get my state in zfs/init.sls to be dependent on zfs.setup?
00:37 DEger joined #salt
00:37 PerilousApricot joined #salt
00:37 hemebond raspado: Depend on or include?
00:38 flowstate joined #salt
00:38 raspado i guess depend, so I have zfs.setup that needs to run before the init.sls is called in zfs
00:39 hemebond Well, you can depend on entire SLS files, but it'd be better if your states in init.sls depended on the relevant states in zfs.setup.sls
00:39 hemebond /zfs/setup.sls
00:41 raspado doing an include?
00:41 hemebond Doing a require probably.
00:42 hemebond Have you read https://docs.saltstack.com/en/latest/ref/states/requisites.html
00:42 DEger joined #salt
00:42 raspado ill take a read
00:49 skeezix-hf joined #salt
00:56 DEger joined #salt
00:56 barmaley joined #salt
01:02 edrocks joined #salt
01:10 DEger joined #salt
01:24 nicksloan joined #salt
01:33 raspado if i do a include in a state file, does that imply that I dont need to define that state in the top file?
01:33 hemebond Correct.
01:34 raspado so an include is almost like a mini top file within a state file?
01:34 hemebond It's a way of pulling in dependencies so you don't have to manually do it for each state.
01:35 hemebond So yeah, a minitop.
01:35 raspado okay cool, my top file was getting pretty long having to define zfs.users, zfs.setup, zfs.config etc etc etc
01:35 raspado thx hemebond
01:35 hemebond ????
01:37 flowstate joined #salt
01:39 sparticl joined #salt
01:46 XenophonF ugh osfinger is only defined for a handful of operating systems
01:46 hemebond osfinger?
01:46 XenophonF yeah it's a combo of the os and osmajorrelease grains (basically)
01:46 hemebond ah
01:46 malinoff joined #salt
01:46 catpigger joined #salt
01:46 XenophonF looked at the code and it's only defined for RedHat, Ubuntu, Debian, FreeBSD, NetBSD, and OpenBSD
01:47 XenophonF you know, what i really want instead of grains.filter_by is a lisp-like COND statement
01:47 XenophonF i guess i'll just write a big if block into map.jinja
01:48 raspado if i change a schedule on the salt master, do I need to clear the schedule cache on the minion for it to take effect?
01:55 pipps joined #salt
02:03 pipps joined #salt
02:07 bastiand1 joined #salt
02:08 mpanetta_ joined #salt
02:18 Nahual joined #salt
02:18 west575 joined #salt
02:22 jeddi joined #salt
02:30 ninjada joined #salt
02:31 PerilousApricot joined #salt
02:37 flowstate joined #salt
02:40 DEger joined #salt
02:54 DEger joined #salt
02:59 akitada joined #salt
02:59 DEger joined #salt
03:00 skeezix-hf joined #salt
03:01 pppingme joined #salt
03:02 chainey joined #salt
03:02 chainey joined #salt
03:05 edrocks joined #salt
03:07 DEger joined #salt
03:13 fgimian joined #salt
03:14 fl33t joined #salt
03:14 Xenophon1 joined #salt
03:21 sparticl joined #salt
03:25 skeezix-hf joined #salt
03:27 amy joined #salt
03:31 tuxx joined #salt
03:39 flowstate joined #salt
03:45 fl33t joined #salt
04:05 ablemann joined #salt
04:12 sparticl joined #salt
04:15 lompik joined #salt
04:16 ablemann joined #salt
04:16 justanotheruser joined #salt
04:27 ninjada joined #salt
04:28 barmaley joined #salt
04:30 skeezix-hf joined #salt
04:37 subsignal joined #salt
04:38 flowstate joined #salt
04:45 fl33t joined #salt
04:46 fl33t left #salt
04:51 mavhq joined #salt
05:07 edrocks joined #salt
05:13 monokrome joined #salt
05:15 skeezix-hf joined #salt
05:16 WKNiGHT joined #salt
05:18 ramblinpeck joined #salt
05:18 antonw joined #salt
05:22 twodayslate joined #salt
05:24 bmcorser joined #salt
05:24 imanc joined #salt
05:24 CaptTofu joined #salt
05:25 mihait joined #salt
05:25 mirko joined #salt
05:25 hax404 joined #salt
05:25 justyns joined #salt
05:25 gladiatr joined #salt
05:25 JPaul joined #salt
05:25 M-liberdiko joined #salt
05:25 Bryson joined #salt
05:25 TOoSmOotH joined #salt
05:25 alrayyes joined #salt
05:25 simonmcc joined #salt
05:25 nebuchadnezzar joined #salt
05:25 MikaT_ joined #salt
05:25 coldbrew- joined #salt
05:25 deei joined #salt
05:25 devdearia joined #salt
05:25 kevc joined #salt
05:25 jholtom joined #salt
05:25 skrobul joined #salt
05:25 Sokel joined #salt
05:25 ajv joined #salt
05:25 cliluw joined #salt
05:25 coldbrewedbrew joined #salt
05:25 Vye joined #salt
05:25 adeschamps joined #salt
05:25 al joined #salt
05:25 zifnab joined #salt
05:25 canci_ joined #salt
05:25 jrklein_ joined #salt
05:25 gtmanfred joined #salt
05:25 patrek_ joined #salt
05:25 mihait joined #salt
05:25 hasues joined #salt
05:25 hasues left #salt
05:27 Awesomecase joined #salt
05:34 whitenoise_ joined #salt
05:37 flowstate joined #salt
05:43 DEger joined #salt
06:14 malinoff joined #salt
06:15 ninjada joined #salt
06:17 Devaraj joined #salt
06:18 Devaraj Hi,Is there any document we deploy openstack using salt on ubuntu VMs
06:19 Devaraj Please share the link incase any doc which explains openstack deployment using salt on ubuntu VMs
06:37 jxm_ joined #salt
06:38 flowstate joined #salt
06:50 DEger joined #salt
07:06 fannet joined #salt
07:09 edrocks joined #salt
07:20 ssplatt joined #salt
07:38 flowstate joined #salt
07:41 DEger joined #salt
07:47 justanotheruser joined #salt
07:55 DEger joined #salt
07:55 cliluw joined #salt
08:04 viq joined #salt
08:04 XenophonF joined #salt
08:04 amy joined #salt
08:04 futuredale joined #salt
08:04 aljosa joined #salt
08:04 phtes joined #salt
08:04 lkannan joined #salt
08:04 gazarsgo joined #salt
08:04 west575 joined #salt
08:04 mpanetta_ joined #salt
08:04 LiamMon joined #salt
08:04 gnord joined #salt
08:04 Aikar joined #salt
08:04 hemebond joined #salt
08:04 krymzon joined #salt
08:04 fxhp joined #salt
08:04 mTeK joined #salt
08:04 ub1quit33 joined #salt
08:04 kuromagi joined #salt
08:04 Laogeodritt joined #salt
08:04 dwfreed joined #salt
08:04 esc\ joined #salt
08:04 Sammichmaker joined #salt
08:04 qman__ joined #salt
08:04 Derailed joined #salt
08:04 jnials joined #salt
08:04 Edgan joined #salt
08:04 Ryan_Lane joined #salt
08:04 agentnoel joined #salt
08:04 twiedenbein joined #salt
08:04 chutzpah joined #salt
08:04 esharpmajor joined #salt
08:04 saltsa joined #salt
08:04 vaelen joined #salt
08:04 Eugene joined #salt
08:04 CustosLimen joined #salt
08:04 xenoxaos joined #salt
08:04 aalmenar joined #salt
08:04 jav joined #salt
08:04 Guest48747 joined #salt
08:04 kossy_ joined #salt
08:04 kylehuff joined #salt
08:04 aarontc_ joined #salt
08:04 khorben joined #salt
08:04 yesimon joined #salt
08:04 briansteffens joined #salt
08:04 c4rc4s joined #salt
08:04 vodik joined #salt
08:04 Llmiseyhaa joined #salt
08:04 saltstackbot joined #salt
08:04 iter joined #salt
08:04 Sketch joined #salt
08:04 ntropy joined #salt
08:04 KingJ joined #salt
08:04 stickmack joined #salt
08:04 cyraxjoe joined #salt
08:04 eightyeight joined #salt
08:04 g3cko joined #salt
08:04 DaveQB joined #salt
08:04 v0rtex joined #salt
08:04 Hipikat joined #salt
08:04 mariusv joined #salt
08:04 llua joined #salt
08:04 alias joined #salt
08:04 wwalker joined #salt
08:04 aberdine_ joined #salt
08:04 tru_tru joined #salt
08:04 Hydrosine joined #salt
08:04 tuxx_ joined #salt
08:04 rickflare joined #salt
08:04 evilrob joined #salt
08:04 bryguy joined #salt
08:04 carmony joined #salt
08:04 Qlawy joined #salt
08:04 __alex joined #salt
08:04 weylin joined #salt
08:04 Gareth joined #salt
08:04 nikogonzo joined #salt
08:04 iggy joined #salt
08:04 daveleigh joined #salt
08:04 devtea joined #salt
08:04 asyncsrc1 joined #salt
08:04 dober joined #salt
08:04 trent__ joined #salt
08:05 oeuftete joined #salt
08:05 binocvlar joined #salt
08:05 codehotter joined #salt
08:05 georgemarshall joined #salt
08:05 komputes joined #salt
08:05 nshttpd joined #salt
08:05 jesusaur joined #salt
08:05 TyrfingMjolnir joined #salt
08:05 honestly joined #salt
08:05 lazybear joined #salt
08:05 eichiro joined #salt
08:05 leev joined #salt
08:05 afics joined #salt
08:05 stopbyte joined #salt
08:05 chitown joined #salt
08:05 dustywusty joined #salt
08:05 TooLmaN joined #salt
08:05 Vaelatern joined #salt
08:05 baffle joined #salt
08:05 notCalle joined #salt
08:05 kidneb joined #salt
08:05 Qwazerty2 joined #salt
08:05 synapse joined #salt
08:05 tehsu joined #salt
08:05 robawt joined #salt
08:05 lynxman joined #salt
08:05 sinh_ joined #salt
08:05 vadimck joined #salt
08:05 mrMute joined #salt
08:05 Roelt joined #salt
08:05 barmaley joined #salt
08:05 Garo_ joined #salt
08:05 NightMonkey joined #salt
08:05 doublemetres joined #salt
08:05 SpX joined #salt
08:05 armyriad joined #salt
08:05 SubOracle joined #salt
08:05 cebreidian joined #salt
08:05 Satyajit joined #salt
08:05 Hazelesque joined #salt
08:05 bVector joined #salt
08:05 ajolo joined #salt
08:05 voytek joined #salt
08:05 jab416171 joined #salt
08:05 gmoro joined #salt
08:05 psy0rz joined #salt
08:05 flebel joined #salt
08:05 unusedPhD_ joined #salt
08:05 my10c joined #salt
08:05 _W_ joined #salt
08:05 lubyou joined #salt
08:05 dunz0r joined #salt
08:05 Micromus joined #salt
08:05 AvengerMoJo joined #salt
08:05 etw joined #salt
08:05 blue joined #salt
08:05 tperale joined #salt
08:05 SWAT joined #salt
08:05 egil joined #salt
08:05 scooby2 joined #salt
08:05 tyler-baker joined #salt
08:05 DEger joined #salt
08:06 eichiro joined #salt
08:09 Rumbles joined #salt
08:11 SWAT joined #salt
08:13 aljosa joined #salt
08:13 DEger joined #salt
08:14 phtes joined #salt
08:14 SteamWells joined #salt
08:15 djural joined #salt
08:15 Ryan_Lane joined #salt
08:15 kutenai joined #salt
08:16 CaptTofu joined #salt
08:16 johtso joined #salt
08:17 gazarsgo joined #salt
08:17 JamieH joined #salt
08:17 futuredale joined #salt
08:19 Awesomecase joined #salt
08:32 DEger joined #salt
08:37 flowstate joined #salt
08:41 DEger joined #salt
08:50 DEger joined #salt
08:52 losh joined #salt
08:59 DEger joined #salt
09:09 DEger joined #salt
09:12 edrocks joined #salt
09:23 DEger joined #salt
09:31 coredumb morning
09:32 DEger joined #salt
09:32 coredumb What are the guidelines when porting pkg functions to a new package manager for functions not available ?
09:33 coredumb I'm going for returning 'Not available' but wanted to be sure before I send a PR
09:36 flowstate joined #salt
09:38 ws2k3 joined #salt
09:39 hemebond Does it not handle it automatically?
09:41 cyborg-one joined #salt
09:44 Rumbles joined #salt
09:45 CheckYourSix joined #salt
09:46 DEger joined #salt
09:55 DEger joined #salt
10:00 ivanjaros joined #salt
10:04 DEger joined #salt
10:05 babilen coredumb: You basically wouldn't implement them
10:10 coredumb babilen: so it's better having 'pkg.toto' is not available.
10:11 coredumb then clean output ?
10:13 coredumb hemebond: true but it returns an error on the minion as well
10:13 DEger joined #salt
10:14 coredumb when applying states on thousands of servers I'm not sure I'd want to have to review errors when it's a known unsupported function
10:22 DEger joined #salt
10:32 DEger joined #salt
10:37 flowstate joined #salt
10:41 DEger joined #salt
10:44 babilen coredumb: The *state* obviously have to deal with the fact that the function in question is not available
10:46 babilen You just wouldn't implement functions that aren't applicable to the platform in the respective module. See, for example, get_selections in aptpkg and yumpkg
10:51 mariusv joined #salt
10:51 mariusv joined #salt
10:56 babilen coredumb: Could you maybe provide more details? That way we might be able to suggest something sensible :)
11:01 coredumb babilen: nah actually I was thinking of the cases you'd run the same pkg.<func> on all your machines were you have yum/apt/something else where indeed it's supported on some and not on others
11:02 coredumb if it's obiviously not supported shouldn't it just cleanly return that it's not - meaning not an error - instead of giving a minion error that would on the result require investigation
11:03 coredumb though I'm just asking, if the rule is to let minion output error on it I'll gladly not implement fake return from those functions :)
11:09 babilen coredumb: The trick here would be to target only boxes that implement it
11:09 DEger joined #salt
11:10 babilen Packaging systems are quite different and it isn't feasible to maintain stubs for everything in every single module (think of Windows, FreeBSD and Linux differences for example)
11:10 coredumb yep but I think we can safely assure it's going to happen :D
11:10 babilen You are a number of functions that will be available everywhere, but some that are quite specific
11:10 babilen Well, then you get an error
11:10 coredumb sure then ok no fake returns :P
11:10 coredumb works for me
11:12 coredumb I'll comment them out and see if I port some of them in the long run
11:13 babilen What kind of functions are these?
11:14 edrocks joined #salt
11:15 coredumb https://github.com/coredumb/salt/blob/alpine_support/salt/modules/apk.py
11:15 coredumb from line 50
11:16 coredumb a bunch of them actually
11:18 DEger joined #salt
11:20 babilen You essentially want to make sure that the pkg state works well with that module and then just remove the functions that aren't available
11:20 babilen (or implement them if possible)
11:23 coredumb yeah that's what I did so I just took the essential that seemed the most important for me
11:30 malinoff joined #salt
11:31 catpig joined #salt
11:31 DEger joined #salt
11:38 flowstate joined #salt
11:41 DEger joined #salt
11:47 drybjed hello, are there any projects like https://voxpupuli.org/ for Salt?
11:52 babilen https://github.com/saltstack-formulas
11:53 drybjed thanks
12:02 DEger joined #salt
12:04 nicksloan joined #salt
12:14 barmaley joined #salt
12:20 DEger joined #salt
12:24 PerilousApricot joined #salt
12:28 ivanjaros joined #salt
12:29 DEger joined #salt
12:43 DEger joined #salt
12:47 lompik joined #salt
12:53 DEger joined #salt
12:57 SpX joined #salt
13:01 fredvd joined #salt
13:02 DEger joined #salt
13:05 ninjada joined #salt
13:10 deus_ex joined #salt
13:14 SpX joined #salt
13:16 DEger joined #salt
13:17 edrocks joined #salt
13:20 XenophonF joined #salt
13:25 DEger joined #salt
13:28 PerilousApricot joined #salt
13:34 DEger joined #salt
13:39 PerilousApricot joined #salt
13:42 Lionel_Debroux_ joined #salt
13:43 DEger joined #salt
14:00 PerilousApricot joined #salt
14:08 ekristen joined #salt
14:09 DEger joined #salt
14:16 hasues joined #salt
14:16 hasues left #salt
14:18 DEger joined #salt
14:22 dendazen joined #salt
14:32 DEger joined #salt
14:34 riftman joined #salt
14:39 nidr0x joined #salt
14:41 DEger joined #salt
14:41 malinoff joined #salt
14:50 DEger joined #salt
14:57 malinoff joined #salt
15:00 fxhp joined #salt
15:01 pppingme joined #salt
15:05 riftman joined #salt
15:05 whitenoise_ joined #salt
15:08 DEger joined #salt
15:09 fgimian joined #salt
15:12 fgimian joined #salt
15:15 perfectsine joined #salt
15:17 DEger joined #salt
15:20 perfectsine_ joined #salt
15:26 ekristen joined #salt
15:36 DEger joined #salt
15:48 raspado joined #salt
15:52 Nahual joined #salt
15:55 DEger joined #salt
15:58 pipps joined #salt
16:02 subsignal joined #salt
16:04 subsignal joined #salt
16:20 DEger joined #salt
16:26 nicksloan joined #salt
16:28 ageorgop joined #salt
16:30 subsignal joined #salt
16:31 ronnix joined #salt
16:34 fannet joined #salt
16:37 nebuchadnezzar joined #salt
16:40 flowstate joined #salt
16:40 DEger joined #salt
16:44 raspado joined #salt
16:45 raspado if adding an include with state files defined in my init.sls, does that automatically run through those state files before it starts running whatever is in init.sls?
16:46 raspado example: http://pastebin.com/W0XyyPXL
16:51 babilen raspado: Normally yes, but if you want to enforce that you'd have to define explicit requisites between states
16:51 babilen Salt might reorder states
16:52 raspado in my top.sls, I would just have a zfs state then the init.sls would have the includes
16:52 raspado the includes themselves may be reordered?
16:52 babilen Individual states might get reordered
16:52 raspado ok
16:53 babilen It's top to bottom by default
16:53 babilen But if state A absolutely has to run after state B then you should make that explicit
16:55 ronnix joined #salt
16:57 ekristen joined #salt
16:58 DEger joined #salt
17:00 flowstate joined #salt
17:02 ninjada joined #salt
17:13 ronnix joined #salt
17:16 pipps joined #salt
17:19 pipps joined #salt
17:20 DEger joined #salt
17:21 edrocks joined #salt
17:25 sp0097 joined #salt
17:25 pipps joined #salt
17:28 ronnix joined #salt
17:31 pipps joined #salt
17:32 DEger joined #salt
17:33 iggy coredumb: you're doing the lords work
17:34 iggy raspado: requisites also prevent later states from running if one fails
17:35 raspado iggy: i like that
17:35 pipps joined #salt
17:36 iggy you like states to not run when requisite states fail or you like them to just keep running?
17:36 iggy if A, there's also `failhard`
17:36 eerca joined #salt
17:45 _JZ_ joined #salt
17:51 raspado is there a shorter way of saying this? "{% if grains['clusterType'] is defined and grains['clusterType'] == 'SingleNode' %}"
17:52 mavhq joined #salt
17:53 iggy {% if grains.get('clusterType') == 'SingleNode' %}
17:53 iggy .get() will return None if the key doesn't exist which won't match 'SingleNode'
17:54 raspado oh very nice, im trying to get around the failure " Rendering SLS 'playground:firstrun' failed: Jinja variable 'dict object' has no attribute 'clusterType'" because some of my minions dont have that grain
17:54 raspado thx iggy!
17:55 iggy another more pythonic way than your way would be {% if 'clusterType in grains and grains['clusterType'] == 'SingleNode' %}
17:55 iggy but the .get() way is perfectly acceptable too
17:56 raspado very nice
18:02 babilen You typically only need the "is defined" if you are not sure if the variable exists at all
18:02 babilen (which would be the case if 'grains' were not defined)
18:03 coredumb iggy: you mean supporting Alpine ?
18:03 iggy coredumb: yes
18:04 coredumb iggy: \o/
18:05 babilen I find alpine to be a rather old mail client :)
18:05 babilen But then .. I'm using mutt, so ..
18:05 coredumb babilen: well it is ... it's also a distribution :D
18:05 \ask left #salt
18:05 \ask joined #salt
18:05 \ask left #salt
18:05 babilen coredumb: Curse you ambiguity!
18:06 iggy I'm talking about the distro
18:06 coredumb iggy: I'll PR the work and patch the community package I've reworked until the PR goes upstream
18:07 babilen Some of us really should get together and write some "summarise state run" outputters. A concise summary at the end of all minions would really be nice.
18:08 babilen It isn't even that hard to implement and I'm quite surprised that I haven't found it or that it hasn't been done yet
18:09 iggy isn't that what the elasticsearch outputter is for?
18:10 coredumb I don't know an outputter is yet
18:10 coredumb :D
18:11 Nahual joined #salt
18:11 iggy I mean once you get past `state_output: changes` presumably you're already at the level that you aren't treating things like pets anymore, and cattle don't need one-on-one attention
18:13 babilen Sure .. but I'd like a "SUCCESS: 112 - ERROR: 1 (minion-foo)" or somesuch
18:13 babilen Maybe with a way to execute a runner to get a complete list of all failed states
18:14 DEger joined #salt
18:14 babilen But you are right. In a way you should use something like the elasticsearch runner, but then there is the bit in between "three boxes that fit on one screen" and "five thousand boxes that you never actually touch manually"
18:15 babilen Well, "manually"
18:16 viq iggy: for for that you'd need all minions to reach ES directly, right? A widely reachable ES sounds like a bad idea...
18:16 iggy as much as widely reachable minions
18:16 viq hm?
18:18 viq joined #salt
18:18 iggy if your minions aren't on a "private" network that can reach ES, presumably the minions are also as wide open to the internet as your ES
18:19 viq ES already had remote arbitrary command execution type of bugs - ssh or apache for example are not as "risky" to expose
18:19 iggy whatever means you use to get your minions to talk to the master, use the same one for ES
18:20 viq And I'm reluctant to leave ES open even to a private network, to more than anything that it's cluster peers
18:22 viq I consider even internal networks as potentially hostile, and ES is more sensitive to that than some of other services.
18:23 viq Yeah, I know, Shield, but $$.
18:26 iggy in any case... I don't think outputting on the command line is scalable past about 50 hosts either way... however you solve the problem at that scale is left as an exercise to the reader
18:28 iggy there are now over 30 returners that can be utilized (and custom ones aren't that hard either)
18:29 aphor babilen: how about the Philosophy of Unix approach where you assume nobody wants default output to tell them what is OK and worked..
18:30 aphor so filter out successful and unchanged states first, then report only the root failures of requisites, with a summary of dependent states for a second level of filtering.
18:31 aphor Then a third level of filtering where the "modes" of failures are reported once with a summary list of minions affected?
18:31 iggy that's what `state_output: changes` does
18:32 iggy well, the first part of what you said anyway
18:32 babilen A number of people I speak to miss Ansible's summaries in SaltStack and I agreed with them ...
18:32 babilen People like to know that their medium size thing worked well, without having to scroll
18:33 babilen That's essentially where my comment stems from. I agree that it doesn't scale beyond a certain number of hosts and states, but many people never reach those numbers
18:33 babilen aphor: I like that, but then we should ship those as part of salt.
18:33 aphor I'd also supply a fourth level of filtering that just displays sums of minions state runs and the total of each root requisite.
18:34 babilen And you did that with the help of json outputter, jq and grep?
18:34 aphor I'd write them as a series of coroutines.
18:35 babilen Ah, this is hypethetical?
18:35 babilen +speling
18:35 aphor yes hypothetical
18:35 flowstate joined #salt
18:36 aphor I wouldn't take this on without some consensus here, and a great chance of getting a pull merged into salt.
18:42 aphor wow. I wish I had jails in OSX. That would be the best way to have hundreds of minions to test.
18:42 aphor I'll need to set up a FreeBSD salt test host.
18:45 RandyT hey weekend warriors
18:45 RandyT attempting to deploy ssh private deploy key using example here: https://docs.saltstack.com/en/latest/ref/states/all/salt.states.file.html
18:45 RandyT ending up with zero length file
18:46 RandyT can see the contents in the pillar using the key as described in my state
18:46 RandyT have attempted to place the file several paces on disk including /tmp in case it is a perms issue.
18:46 RandyT file still being created in all places, just zero length.
18:46 RandyT what am I missing
18:47 RandyT state is exactly as written in the contents_pillar: example
18:47 krymzon joined #salt
18:52 Lionel_Debroux joined #salt
18:55 iggy you called your pillar `mypillar`?
18:55 iggy or userdata?
18:56 RandyT nooooo :-)
18:56 RandyT in my case, path to my value in pillar file is ssh_deploy:key
18:56 iggy so then it's not exactly like the example
18:56 RandyT and I can use a pillar.item to see that
18:56 RandyT :-) you are always right
18:57 RandyT I suspect you would like to see it....
18:57 RandyT standby
18:57 iggy you the best
18:57 aphor can anyone sanitize and pickle an ugly ret for me to use as test data for a summary outputter?
19:00 RandyT iggy: https://gist.github.com/rterbush/dd90b14b2bb4c16b2f476873158459a1https://gist.github.com/rterbush/dd90b14b2bb4c16b2f476873158459a1
19:00 RandyT https://gist.github.com/rterbush/dd90b14b2bb4c16b2f476873158459a1
19:01 iggy protip: gist's can have more than one file... no need to put bits of it in the comment
19:01 RandyT iggy: thanks :-)
19:02 iggy your indentation looks off on the pillar
19:02 RandyT just added comment to show return of pillar.item
19:02 RandyT it actually isn't.  4 spaces after |
19:03 ageorgop joined #salt
19:03 iggy huh, I didn't know pillar.item supported nested syntax
19:03 iggy can you do `salt-call pillar.get ssh_deploy:key`
19:04 RandyT iggy: added that as file instead of comment to see proper indent
19:04 RandyT affirmative on doing salt-call from the minion
19:05 RandyT I can see the key
19:05 ivanjaros joined #salt
19:06 iggy okay, so it's not a difference between .item()/.get() kind of issue
19:07 iggy have you tried without quotes around the ssh_deploy:key (on the contents_pillar line?)
19:09 RandyT iggy: will try without quotes, actually getting an INFO on the minion about no source, contents contents_pillar....
19:09 * RandyT smack
19:10 RandyT iggy: ok, so it was not exactly like the example... nor the local file I was editing and pasting into gist
19:10 RandyT I had flipped contents_pillar to be pillar_contents
19:11 RandyT so I have a file with contents now, thank you.
19:11 iggy so what you pasted wasn't what you had in the file?
19:12 iggy because your gist looks right
19:12 RandyT yes, gist was pasted from my local emacs... a few edits ago on the server I must have flipped it...
19:12 iggy and this is why we want pastes of the actual code being used ;)
19:13 RandyT you are so right
19:14 iggy off to enjoy Vancouver... good luck everyone
19:14 RandyT ah, enjoy. A lot to enjoy there
19:24 edrocks joined #salt
19:24 aphor I need to go back to Vancouver soon.
19:24 aphor Have fun!
19:26 aphor does anyone know why salt:// vfs chokes on huge files?
19:31 babilen define huge
19:35 aphor >= 2GB
19:35 flowstate joined #salt
19:36 aphor experience tells me there's an offset index somewhere that uses a 32 bit int or similarly limited data structure.
19:42 subsignal joined #salt
19:43 iggy I think it loads it all into memory
19:43 subsigna_ joined #salt
19:54 flowstate joined #salt
19:55 systo joined #salt
19:57 pipps joined #salt
20:01 west575_ joined #salt
20:20 PerilousApricot joined #salt
20:22 Nahual joined #salt
20:26 sp00ky joined #salt
20:31 hasues joined #salt
20:31 hasues left #salt
20:32 pipps joined #salt
20:34 raspado joined #salt
20:35 flowstate joined #salt
20:36 deus_ex joined #salt
20:37 pipps99 joined #salt
20:38 mavhq joined #salt
20:41 flowstate joined #salt
20:41 cyborg-one joined #salt
20:42 west575 joined #salt
20:43 pipps joined #salt
20:47 west575_ joined #salt
20:48 aphor iggy: if it loads the whole file into memory.. I'm going to look at fixing that now.
20:59 wryfi_ joined #salt
21:04 ninjada joined #salt
21:23 perfectsine joined #salt
21:26 edrocks joined #salt
21:26 krymzon joined #salt
21:35 flowstate joined #salt
21:36 Laogeodritt joined #salt
21:41 hasues joined #salt
21:41 hasues left #salt
21:44 PerilousApricot joined #salt
21:46 Perilous_ joined #salt
21:47 raspado joined #salt
22:02 coredumb iggy: https://github.com/saltstack/salt/pull/36207
22:02 saltstackbot [#36207][OPEN] Feature: Add Alpine Linux support | What does this PR do?...
22:03 coredumb iggy: and if you care about testing the aport: https://github.com/alpinelinux/aports/pull/282
22:03 saltstackbot [#282][OPEN] community/salt: Update and refactoring based on official SRPM |
22:21 badon joined #salt
22:21 ninjada joined #salt
22:24 systo joined #salt
22:26 gazarsgo joined #salt
22:29 johtso joined #salt
22:29 wendall911 left #salt
22:31 futuredale joined #salt
22:31 whitenoise joined #salt
22:31 whitenoise joined #salt
22:31 whitenoise joined #salt
22:32 netzvieh joined #salt
22:35 flowstate joined #salt
22:37 kiorky joined #salt
22:41 whitenoise joined #salt
22:48 Jarus joined #salt
22:50 yomilk joined #salt
22:50 aphor I think we need to wedge something along these lines into salt.fileclient.RemoteClient()
22:50 aphor https://gist.github.com/bdarnell/5bb1bd04a443c4e06ccd
22:53 aphor data = self.channel.send(load, raw=True) doesn't appear to support streaming, eventually just calling tornado.concurrent.Future to get a dict of results.
22:55 aphor the 'data' key of that result should be a generator, and then salt:// would be able to support arbitrarily large files
22:56 yomilk joined #salt
23:25 schemanic_ joined #salt
23:25 schemanic_ hi
23:25 PerilousApricot joined #salt
23:26 schemanic_ anyone here ever set up a jailed sftp install?
23:26 schemanic_ I cant get mine working using the users and openssh formulas
23:26 schemanic_ and I cant figure out why
23:27 krymzon joined #salt
23:28 yomilk joined #salt
23:28 subsignal joined #salt
23:28 edrocks joined #salt
23:31 lompik joined #salt
23:35 yomilk joined #salt
23:36 flowstate joined #salt
23:37 foodo joined #salt
23:38 Shirkdog joined #salt
23:42 foodo i have a problem with a reator file, if i want to set target as {{ data['id'] }} ,i get rendering error "Rendering exception occurred: Jinja variable 'str object' has no attribute 'id' "
23:43 hemebond foodo: data is a string :-)
23:43 SaltySailor joined #salt
23:44 foodo hemebond: i don't get it
23:44 hemebond data is a string. you're trying to use it as if it's a dict.
23:46 foodo hemebond: ok, and who can i change this?
23:46 hemebond I have no idea. You haven't provided any other information. I don't know what "data" is.
23:48 foodo hemebond: it's the minion id of a system who send a beacon event
23:49 hemebond Then you would just use {{ data }}, no?
23:49 hemebond I'm reading the reactor docs now. Is this your first time using a reactor?
23:52 foodo hemebond: yes my first time
23:54 foodo hemebond: if i use tgt; hemebond {{ data }} i get a other error: [ERROR   ] Failed matching available minions with glob pattern: None
23:54 hemebond So data is empty.
23:54 hemebond How are you testing the reactor?
23:55 hemebond Like this? https://docs.saltstack.com/en/latest/topics/event/events.html#firing-events
23:56 hemebond Have you confirmed that your events are being sent and are correct?
23:57 sp00ky is there a configuration flag to disable encryption/decryption of data being sent back and forth between the master and minions?
23:58 foodo hemebond: here is my config https://gist.github.com/anonymous/301c9d9c200463acb61e7f077b6917e6
23:58 hemebond Disable the encryption?
23:59 foodo hemebondm, sp00ky: the event is fiered from a beacon. where can i disable encryption?

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