Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2017-10-23

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

All times shown according to UTC.

Time Nick Message
00:02 CampusD joined #salt
00:08 nick123 joined #salt
00:13 CampusD joined #salt
00:16 CampusD joined #salt
00:27 CampusD joined #salt
00:31 CampusD Hi all, anybody would be able to tell me why this module execution doesn't work from within the python shell?
00:31 CampusD https://gist.github.com/anonymous/30ae31ccd423a6a334158ca9604294e3
00:35 CampusD seems to work when executed from the CLI
00:39 whytewolf why are you trying to go to kwargs half way through.
00:40 whytewolf just try cmd='c:\salt\bin\python.exe c:\scripts\my_script.py', trigger_type='Daily'
00:43 CampusD some parameters seem to be part of kwargs
00:43 CampusD https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.win_task.html#salt.modules.win_task.create_task
00:44 whytewolf try it
00:45 CampusD ok, let me try it that way
00:45 whytewolf kwargs = keyword arguments. which lets python take an arbitrary number of arguments
00:48 CampusD so I tried the following: "caller.cmd('task.create_task','TestTask1',user_name='System',force=True, cmd='c:\salt\bin\python.exe c:\scripts\my_script.py', trigger_type='Daily')"
00:48 CampusD keep getting False back
00:53 whytewolf humm, just for kicks and giggles. try changing task.create_task to test.kwarg keep everything else the same
00:53 whytewolf [you might need to change 'TestTask1' to name='TestTask1']
00:54 CampusD TypeError: kwarg() takes exactly 0 arguments (5 given)
00:54 whytewolf wtf
00:55 CampusD I switched that parameter to name='TestTask1', still getting false
00:55 whytewolf i meant with the test.kwarg
00:56 CampusD yeah, that one throws that TypeError
00:56 whytewolf that should defintly NOT throw a type error.
00:57 whytewolf esp one saying it takes 0 arguments
00:59 whytewolf test.kwarg should take any arguments you toss at it and spit out everything in what the function sees as **kwargs
00:59 whytewolf such as
00:59 whytewolf caller.cmd('test.kwarg',cmd='echo hi')
00:59 whytewolf {'cmd': 'echo hi'}
01:02 CampusD is it because the first set of parameters are not part of the **kwargs? create_task(name, location='\\', user_name='System', password=None, force=False, **kwargs)
01:02 whytewolf no
01:02 CampusD test.kwarg seems to work if I take those first set of parameters out
01:02 whytewolf that doesn't matter
01:03 CampusD ok
01:03 whytewolf the ONLY input to test.kwargs is **kwargs
01:03 whytewolf sorry test.kwarg
01:04 CampusD so it would not care for name, location='\\', user_name='System', password=None, force=False ?
01:04 whytewolf they are no different to it
01:04 whytewolf everything is a kwarg to it
01:05 CampusD oh, I see
01:10 whytewolf caller.cmd('test.kwarg',name='TestTask1',user_name='System',force=True, cmd='c:\salt\bin\python.exe c:\scripts\my_script.py', trigger_type='Daily')
01:10 whytewolf {'user_name': 'System', 'cmd': 'c:\\salt\x08in\\python.exe c:\\scripts\\my_script.py', 'force': True, 'name': 'TestTask1', 'trigger_type': 'Daily'}
01:10 whytewolf thats what i get when i try on a linux system
01:10 whytewolf [can't try the create_task command cause. i don't have windows]
01:12 CampusD yeah these are windows servers, I've tried on 2008 and 2016
01:12 whytewolf the os shouldn't matter when it comes to kwarg
01:13 whytewolf check the command i passed in there on your os
01:13 whytewolf don't type copy and paste it
01:16 CampusD I got back the following
01:16 CampusD {'user_name': 'System', 'cmd': 'c:\\salt\x08in\\python.exe c:\\scripts\\my_script.py', 'force': True, 'name': 'TestTask1 ', 'trigger_type': 'Daily'}
01:16 whytewolf ok, change only test.kwarg to task.create_task
01:18 CampusD done, still getting False
01:18 whytewolf then i don't know what else is wrong
01:19 whytewolf at this point it takes knowing that module. and i don't know that module
01:34 tiwula joined #salt
01:56 ilbot3 joined #salt
01:56 Topic for #salt is now Welcome to #salt! <+> Latest Versions: 2016.11.8, 2017.7.2 <+> Support: https://www.saltstack.com/support/ <+> 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
01:56 nomeed joined #salt
02:28 CampusD sounds good, thanks for the tips, I'll take a look at the module to see if I can decipher it. https://github.com/saltstack/salt/blob/develop/salt/modules/win_task.py
02:36 k_sze[work] joined #salt
02:41 jagguli joined #salt
02:52 JPT joined #salt
02:52 justanotheruser joined #salt
02:53 justanotheruser joined #salt
03:11 dxiri joined #salt
03:30 Vaelatern joined #salt
03:33 masber joined #salt
03:36 dxiri joined #salt
03:39 aldevar joined #salt
04:19 darix joined #salt
04:30 aldevar left #salt
04:44 Shirkdog joined #salt
04:54 evle joined #salt
05:26 oida joined #salt
05:44 omie888777 joined #salt
05:47 lompik joined #salt
05:54 omie888777 joined #salt
05:55 Ricardo1000 joined #salt
06:01 colttt joined #salt
06:02 omie88877777 joined #salt
06:08 omergerd2019 joined #salt
06:12 omie88877777 joined #salt
06:28 do3meli joined #salt
06:28 hammer065 joined #salt
06:35 pualj joined #salt
06:47 do3meli left #salt
06:48 poige joined #salt
06:48 poige hi! What would be the best way to define salt-minions groups w/o using Salt's groups?
06:49 poige Tinkering with salt-master config to define groups doesn't seem to be any reasonable as to me
07:08 LocaMocha joined #salt
07:14 dynamicudpate joined #salt
07:18 lompik joined #salt
07:20 coredumb poige: multiple ways
07:20 coredumb you can use grains or pillars
07:20 poige pillars would be cache first
07:20 poige then use
07:21 coredumb would that be an actual issue?
07:21 poige well, it differs from what you have with groups in ergh master's config
07:22 coredumb well sure
07:22 coredumb but you don't have to tinker with your master config then
07:22 coredumb :)
07:23 poige well sure (c)
07:23 poige but it was obvious from the very beginning
07:24 coredumb clearly approach is different, depends how you want to integrate this "grouping" thing in your process
07:24 coredumb here I use pillars only from an external pillar
07:24 coredumb so roles are defined from pillars per minion in the external pillar
07:25 poige Well, as to me, groups in master config look completely pointless
07:25 poige No idea why someone decided to put it that way
07:31 robman joined #salt
07:33 hemebond Pillars would work well.
07:33 hemebond Though applying pillars to the minions gives the same issue.
07:37 coredumb it's fairly easy to refresh pillars using a reactor
07:38 usernkey joined #salt
07:47 inad922 joined #salt
07:56 mikecmpbll joined #salt
08:00 _KaszpiR_ joined #salt
08:05 poige doesn't it all look like an issue artificially made up?
08:05 poige -- can we use groups? -- sure, just edit master service config
08:05 poige -- can we avoid it? -- Sure, use pillars... and ergh you also would need a reactor
08:06 pbandark joined #salt
08:09 absolutejam Any level of verbosity that will let me see exactly what my custom module is running?
08:10 absolutejam I'm having issues with one of my kwargs not being rendered properly
08:10 absolutejam I'm trying to wrap state.apply with a custom module that accepts **kwargs for the pillar kwarg, because writing json at the cli is stupid
08:10 coredumb poige: you'd need a reactor only - if I'm correct - you're using external pillars
08:11 coredumb AFAIK there's no cache latency issue with local pillars
08:13 toanju joined #salt
08:30 Mattch joined #salt
08:31 usernkey joined #salt
08:41 stooj joined #salt
08:50 aldevar joined #salt
09:02 aldevar left #salt
09:16 AvengerMoJo joined #salt
09:17 darioleidi joined #salt
09:44 W4RL0RD joined #salt
09:46 W4RL0RD__ joined #salt
09:49 W4RL0RD__ joined #salt
09:51 ExtraCrispy joined #salt
09:56 chowmein__ joined #salt
10:25 Shirkdog joined #salt
10:40 Naresh joined #salt
11:03 DanyC joined #salt
11:15 pbandark is it possible to store response of salt-cloud command anywhere ?
11:16 stooj joined #salt
11:20 XenophonF pbandark: like with a returner or something?
11:22 pbandark XenophonF: yeah anything would be ok. i am creating instance using salt-cloud. i want to forward the response to other java application in yaml/json format. similarly, if i try to get details of instance/disk etc. i want to forward the response to java application
11:23 XenophonF have you looked at the cloud states?
11:24 XenophonF maybe that plus orchestration or custom returners would be "better"
11:24 XenophonF brb time to make the donuts
11:24 pbandark XenophonF: with salt-cloud i am able to featch required details, but i am unable to find option to forward/save the response
11:26 pbandark XenophonF: for ex. from java application user will try to featch details of existing instance --> which would trigger `salt-cloud` command to featch the details --> salt-cloud should be able to forward the response back to java application.
11:39 robman joined #salt
11:42 hemebond left #salt
11:51 flughafen joined #salt
12:02 flughafen how can i pass the ignore_repo_failure to the command line of "salt-call pkg.install foobar"  ?
12:09 babilen ignore_repo_failure=True
12:09 XenophonF pbandark: you could always pick a different outputter
12:10 XenophonF it outputs YAML by default but "--out json" or whatever will change it
12:10 pbandark XenophonF: with salt-cloud i dont find "--returner" option. can you please explain how i can capture the "salt-cloud"  command output
12:10 XenophonF java application as in CLI/GUI app or as in servlet?
12:11 XenophonF I mean bog-standard Unix pipes
12:11 pbandark cli/gui
12:11 XenophonF ok so you're shelling out to run this command
12:12 XenophonF I don't know how you'd do this in Java, but it C you'd use popen()
12:12 XenophonF s/it/in/
12:12 flughafen babilen: ha. that was simple enough, thanks
12:12 babilen flughafen: enjoy :)
12:13 babilen (it works the same for all options)
12:13 flughafen ok. good to know.
12:14 XenophonF pbandark: you probably have to do something like Process process = Runtime.getRuntime().exec("salt-cloud ... --output json");
12:14 XenophonF and then read the data using Process streams
12:14 XenophonF but it's been years and years and years since I last wrote serious Java programs
12:15 XenophonF oh there's a Popen object too, I think
12:15 XenophonF or maybe that's Python
12:15 Nahual joined #salt
12:15 pbandark XenophonF: from java application, there will be post request to get/create instance.from which reactor will be triggered to create/get instance using "salt-cloud"
12:15 XenophonF Wouldn't it be easier to use the Salt API?
12:16 pbandark XenophonF: yes. i will be using salt-api to get request
12:16 XenophonF why not use the Salt API to launch the instances?
12:16 XenophonF and get data about them?
12:17 sybix Hi, anybody use the zabbix state/module ?
12:17 babilen flughafen: The only tricky bit is that you essentially pass Python data(structures), so if you want to pass a list you'd use foo="['bar', 'baz', ...]" and dictionaries as foo="{'foo': 'bar'}"
12:17 babilen sybix: Be the first!
12:18 sybix I'm trying :)
12:19 sybix The doc say : ' optdepends: zabbix server' this mean i have to install zabbix_server on the overlord ?
12:19 XenophonF pbandark: https://docs.saltstack.com/en/latest/ref/netapi/all/salt.netapi.rest_cherrypy.html
12:20 XenophonF pbandark: https://github.com/SUSE/salt-netapi-client
12:21 pbandark checking
12:25 Neighbour sybix: I've written my own zabbix state and module to replace the ones that come with salt:)
12:26 mariusv joined #salt
12:26 mariusv joined #salt
12:26 pbandark XenophonF: currently i have configured it as https://paste.fedoraproject.org/paste/-CVq-N2xiilv7I3Bkb4gnw
12:28 babilen Neighbour: How come? And would it make sense to replace Salt's states with yours?
12:28 sybix Neighbour, You did it because you can't make it with those coming with salt ?
12:30 sybix It's look like zabbix server and frontend need to be installed on the same server than salt
12:32 evle1 joined #salt
12:34 ange joined #salt
12:36 ange hello hello
12:37 ange question about the py renderer : is it possible to split py states so that one can share common boilerplate code between them ?
12:38 Neighbour babilen: because I was missing a lot of functionality
12:38 Neighbour I am planning to upstream my version, however it will require some time to clean things up and make things more conforming salt's code standards
12:39 babilen aye
12:39 NightMonkey joined #salt
12:39 babilen I'm sure sybix would be thrilled if you could give us a "teaser" ;)
12:40 Neighbour sybix: Yep, I missed a whole lot. The only thing that's still missing from my implementation are IT services and autodiscovery-related things...all other objects can be manipulated in salt
12:41 Neighbour babilen: what would you consider a 'teaser'? the current codebase? (which consists of 3 files: custom state, custom exectution module and included .py-file with API definitions of zabbix 3.2)
12:45 babilen I guess that would be it, yeah
12:48 sybix I'm searching to add host with salt, this one look's good : salt.states.zabbix_host.present from https://docs.saltstack.com/en/2016.11/ref/states/all/salt.states.zabbix_host.html
12:49 ahrs joined #salt
13:06 gh34 joined #salt
13:07 stooj joined #salt
13:13 edrocks joined #salt
13:14 pualj joined #salt
13:20 alvinstarr joined #salt
13:20 racooper joined #salt
13:21 ExtraCrispy joined #salt
13:21 mrueg joined #salt
13:21 cgiroua joined #salt
13:25 mrueg joined #salt
13:29 numkem joined #salt
13:34 ashepelev joined #salt
13:39 ashepelev joined #salt
13:39 ashepelev Hello! Can you please advise? Im using Salt mine to gather some grains/pillar staff from minions. If I shut down the minion, am I able to receive salt mine cache about this host further (even if it will equal the last UP(healthy) status)?
13:39 tkharju joined #salt
13:39 Neighbour sybix: Are you missing anything from the current zabbix-modules and -states salt provides?
13:44 Brew joined #salt
13:45 spx joined #salt
13:46 spx left #salt
13:49 dxiri joined #salt
13:49 sybix Neighbour, Theoricaly nop, but it send me anything but an error : Reason: 'zabbix_host' __virtual__ returned False
13:50 N-Mi joined #salt
13:50 N-Mi joined #salt
13:50 sybix I assume i'm missing something, but dunno what
13:51 ouemt joined #salt
13:52 Neighbour sybix: yea, that's because the distributed zabbix module expects the zabbix-server to be installed on the minion
13:52 XenophonF pbandark: that paste didn't work. wrong URL?
13:53 Neighbour You can "fix" this by editing /usr/lib/python2.7/site-packages/salt/modules/zabbix.py and changing the __virtual__-function to always return __virtualname__ (i.e. remove the 'if' there)
13:53 pbandark XenophonF: https://paste.fedoraproject.org/paste/NQGN1e6Lb5shVngIhmFAbQ
13:54 pbandark XenophonF: forgot to mention salt-api. https://paste.fedoraproject.org/paste/Y5HLVjdgNg3sh3KgIW22FQ
13:56 XenophonF Two, two pull requests, ah ah ah!
13:56 sybix Neighbour, That's what i fear :)
13:57 Neighbour sybix: The module doesn't *really* need zabbix_server to be present though, it uses HTTP(s) calls to the API, so there is no need for the zabbix-server to actually run on the minion....however, it does check for it in the __virtual__-function
13:57 Neighbour which is why I think you can safely remove that check :)
13:58 k_sze[work] joined #salt
13:58 sybix It's do the check on the overlord or the minion ?
13:59 Neighbour I think on the minion, since that's where the module gets loaded
13:59 sybix That's what i saw in the code. And it's import only 'common' stuff on the module
13:59 Kira joined #salt
14:00 XenophonF pbandark: dude, enable SSL
14:00 XenophonF omg
14:00 XenophonF you should never ever _not_ enable SSL
14:00 pbandark XenophonF: yeah :) i am on test setup will enable it when i will be in production
14:01 XenophonF :)
14:01 sybix Neighbour, inception powa, let's make a state to patch salt :)
14:01 XenophonF I'm an ass when it comes to demanding that dev/test setups be held to the same rigorous security standards as production.
14:01 XenophonF sorry!
14:02 XenophonF so your config looks like it might even work!  how goes testing?
14:03 pbandark XenophonF: at "create_network" / "deploy_gcp_jump_server" i want to capture output and further i want to send it to java application
14:03 XenophonF hm
14:04 XenophonF hm hm hm
14:04 XenophonF well, I wouldn'
14:04 XenophonF maybe calling the cloud.whatever it is state plus a custom returner would be better, but I don't know
14:05 pbandark ok
14:05 pbandark let me understand how to write runners
14:05 pbandark s/runner/returner
14:06 XenophonF what's the point of the java app? a pretty wrapper around salt-cloud?
14:07 pbandark yes. you can say its wrapper for some salt* operations(create instance,subnet, network using salt state and salt-cloud)
14:07 XenophonF that's cool
14:10 csmule joined #salt
14:16 Brew joined #salt
14:17 cgiroua joined #salt
14:24 pbandark XenophonF: is it possible to create runner for salt-cloud ?
14:24 pbandark s/runner/retruner
14:28 acsir joined #salt
14:33 Miouge joined #salt
14:36 tom[] joined #salt
14:43 XenophonF There's already a cloud runner.
14:43 XenophonF https://docs.saltstack.com/en/latest/ref/runners/all/salt.runners.cloud.html
14:47 cgiroua joined #salt
14:50 lompik joined #salt
14:52 pbandark checking
14:57 toanju joined #salt
15:05 keltim joined #salt
15:06 sarcasticadmin joined #salt
15:15 thefynx left #salt
15:22 relidy joined #salt
15:23 ooboyle Edgan: RE the question from Friday 16:00 EST: I had a chance to think about this a bit more over the weekend. Specifically, I have a list of subnets that map to location codes (1 subnet per code). I need to catch minions when they appear in one of the subnets on the list. Once caught, I need to assign them a permanent location code that doesn't disappear or reset, even if the minion ends up appearing on a different subnet at a later d
15:24 ooboyle Edgan: This jinja is what I can use to catch them on a listed subnet: {% if salt['network.in_subnet']('192.168.37.0/24') == True %}
15:26 ooboyle Edgan: If I use a state to run the jinja and assign the location code as a value of the grain, then the location code/grain value will change each time the minion appears on a different subnet. That's not what I want. I want it to be detected the first time the minion appears on a listed subnet so that I can set a grain value that will never change again.
15:27 _JZ_ joined #salt
15:28 ooboyle Edgan: I think I need a two stage process where: Stage 1: first time appearance on a listed subnet generates a grain value. Stage 2: That grain value is transferred to a permanent grain for that minion once. It can not be overwritten with this same solution.
15:29 nick123_ joined #salt
15:31 ooboyle If this is possible, then my original question is still relevant: Is there a way to have the jinja reference a list of subnets that exists in a file that resides on the master (I don't want to have to copy the file to each minion)?
15:33 ooboyle Edgan: It would like something like this: {% if salt['network.in_subnet']('<reference the file that contains a list of subnets to see if there's a key match>') == True %}
15:34 ooboyle Edgan: If there is a key match, assign the value for that key to a grain
15:44 keltim joined #salt
15:49 nixjdm joined #salt
15:49 nomeed joined #salt
15:56 dezertol joined #salt
16:00 csmule joined #salt
16:09 DanyC joined #salt
16:21 Deliant joined #salt
16:22 edrocks joined #salt
16:26 edrocks joined #salt
16:35 jmedinar joined #salt
16:36 jmedinar Question on Modules: How can I set a default value in case a pillar is not set?
16:36 jmedinar cpu_th = __pillar__['cloudant']['cpu_th', 80]
16:37 jmedinar I want the default to be 80 in this case
16:40 jmedinar The only way is to change the format to...
16:41 jmedinar cpu_th = __pillar__.get('cloudant.cpu_th', 80)    ??
16:41 whytewolf well, .get is how defaults are handled.
16:41 whytewolf so yeah
16:42 whytewolf cpu_th = __pillar__['cloudant'].get('cpu_th', 80) could also work
16:42 jmedinar ok I see
16:43 jmedinar thanks whytewolf as usual !
16:43 jmedinar :)
16:43 wavded joined #salt
16:45 wavded joined #salt
16:46 tiwula joined #salt
16:57 teratoma /go phpc
16:57 whytewolf go go gadget helicopter
17:01 fatal_exception joined #salt
17:02 nixjdm joined #salt
17:05 dxiri joined #salt
17:10 DanyC joined #salt
17:11 wavded joined #salt
17:17 edrocks joined #salt
17:20 jhauser joined #salt
17:27 jmedinar How can I avoid the message " __virtual__ returned False: Exception raised when processing __virtual__ function for. Module will not be loaded" when the __virtual__ results false???
17:28 jmedinar Is looking for a pillar value that might not be there... and if the value is True it continues...
17:29 DanyC_ joined #salt
17:31 ecdhe joined #salt
17:32 whytewolf find ut what the exception is.
17:33 wavded joined #salt
17:34 astronouth7303 is there a way to specify a file to `pkg.installed`? http://docs.graylog.org/en/latest/pages/collector_sidecar.html doesn't publish any repos, just package files (although it does nicely have rpms, debs, an exe, and tarballs)
17:34 astronouth7303 i don't see a way documented, i wanted to make sure i didn't miss anything
17:35 whytewolf astronouth7303: pkg.install has the sources option. but you still have to supply a name to it
17:35 whytewolf so it knows if the package installed or not
17:35 Neighbour and it has to match the name of the package
17:36 astronouth7303 oh! ok, cool
17:36 astronouth7303 that makes life easier
17:37 inad922 joined #salt
17:37 notemerson joined #salt
17:43 mikecmpbll joined #salt
17:49 jmedinar well I debugged the problem and is doing something funny
17:50 jmedinar https://pastebin.com/mdreqq7V
17:56 mayk joined #salt
18:03 nixjdm joined #salt
18:06 impi joined #salt
18:06 XenophonF Anyone here using an SCL version of Python with the pip state module?
18:06 felskrone joined #salt
18:08 XenophonF usually the Red Hat way to invoke pip would be something like `scl enable rh-python36 -- pip ...`
18:08 XenophonF and the scl command alters PATH and LD_LIBRARY_PATH and friends to make that version of pip work
18:08 XenophonF but it isn't clear how to do that with a pip.installed state
18:09 DanyC joined #salt
18:09 wavded joined #salt
18:10 _KaszpiR_ joined #salt
18:11 keltim joined #salt
18:20 sarcasticadmin joined #salt
18:24 pbandark how to use "returner" with sate file? i want to use "mongo_return" to redirect the output of state execution to mongo db
18:28 pjs joined #salt
18:28 keltim joined #salt
18:39 nick123 joined #salt
18:44 pbandark also, how i can make use of "returner" with salt-runner ?
18:44 DanyC joined #salt
18:44 pbandark with execution module, i am able to use "returner" without any issue. `salt 'salt-master.novalocal' test.ping --return mongo_return` <==
18:46 edrocks joined #salt
18:47 csmule joined #salt
18:58 nick123 joined #salt
19:03 nixjdm joined #salt
19:06 tkharju joined #salt
19:10 ChubYann joined #salt
19:13 toanju joined #salt
19:14 edrocks joined #salt
19:18 swa_work joined #salt
19:21 grains joined #salt
19:26 linovia mmm, is it possible to perform something such as: 1. set a temporary configuration file, 2. reload a server 3. perform some action 4. set the definitive configuration file 5. reload the server ?
19:27 linovia I'm having a hard time since salt does: 1. 4. 2/5 and then 3 which fails because the configuration used is the one from step 4
19:30 swa_work joined #salt
19:36 CampusD joined #salt
19:36 mike_verhulst joined #salt
19:38 swa_work joined #salt
19:39 Ahlee require: to specify an order
19:39 Ahlee make step 5 depend on 4, 4 on 3, etc
19:40 linovia I think I gave it try. Let's see
19:42 Hybrid joined #salt
19:49 linovia nop, doesn't work
19:49 wavded joined #salt
19:51 jhauser joined #salt
19:51 cyborg-one joined #salt
19:52 linovia my bad, it's require, not requires
19:52 linovia thanks Ahlee
19:55 linovia mmm, need to figure out how to get rid of the recursive requisite now :(
20:02 nixjdm joined #salt
20:08 mavhq joined #salt
20:08 sjorge joined #salt
20:13 ult joined #salt
20:16 Hybrid joined #salt
20:24 sarcasticadmin joined #salt
20:31 jfelchner joined #salt
20:42 pipps joined #salt
20:44 bluenemo joined #salt
20:57 lkolstad joined #salt
21:00 pbandark joined #salt
21:00 wavded joined #salt
21:01 swa_work joined #salt
21:01 pipps joined #salt
21:02 nixjdm joined #salt
21:04 wavded joined #salt
21:06 swa_work joined #salt
21:08 pipps joined #salt
21:11 pipps99 joined #salt
21:13 whytewolf oh, i kind of said this yesterday but no one is really here on sundays. I need some testers for a new function on my debug module. a yamllint function that takes into account any rendering that needs to happen first. https://github.com/whytewolf/salt-debug/tree/devel/lint
21:13 swa_work joined #salt
21:13 pipps joined #salt
21:14 spiette joined #salt
21:16 jhauser joined #salt
21:18 sarcasticadmin joined #salt
21:20 swa_work joined #salt
21:23 viq ooh, neat
21:24 pipps joined #salt
21:26 pipps joined #salt
21:37 mayk joined #salt
21:41 pipps joined #salt
21:44 grains left #salt
21:59 cgiroua joined #salt
21:59 cyborg-one joined #salt
22:00 ouemt joined #salt
22:02 nixjdm joined #salt
22:03 pipps joined #salt
22:27 swa_work joined #salt
22:53 swa_work joined #salt
23:13 Edgan ooboyle: If you want it more static it sounds more like a pillar, and less like a grain. You could use a custom grain, and then have the code use that value to once write a different /etc/salt/grains.yml grain
23:14 Edgan ooboyle: If grain foo(yml grain) doesn't exist, set it to the value of bar(custom py)
23:16 Edgan ooboyle: or a custom grain that checks a central database, and if not adds an entry to the database. If there is already an entry, it returns it.
23:18 Edgan ooboyle: Depends on if you want it to persist across system recreation. The yml grain method would recalculate if the system got created
23:18 Edgan ooboyle: Where as with the database method it would be independent of the system
23:47 mikecmpbll joined #salt

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