Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-10-17

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

All times shown according to UTC.

Time Nick Message
00:12 shakalaka joined #salt
00:29 ZachLanich joined #salt
00:30 cowza joined #salt
00:39 aagbds joined #salt
00:51 cilkay How would I have a daemon, B, that is dependent on another daemon, A, restart only if A is restarted?
00:52 hemebond Umm. Maybe onchanges or something?
00:52 cilkay I can restart A based on a file watch, like above. I suppose I could watch the same file in B.
00:53 hemebond Yes, or watch the service state for changes.
00:54 cilkay That's probably a cleaner way of doing it since it decouples B from the file state that A is watching. If I have to monitor another file state for A, I don't have to repeat myself for B.
00:57 cilkay The use case is a Django channels app. If anything changes in the app, uwsgi should be restarted. If uwsgi is restarted, asgi and the websocket service should also be restarted.
01:02 jas02 joined #salt
01:03 catpigger joined #salt
01:08 Karunamon joined #salt
01:10 DEger joined #salt
01:25 shakalaka joined #salt
01:34 knikolov joined #salt
01:40 sebastian-w joined #salt
01:47 ilbot3 joined #salt
01:47 Topic for #salt is now Welcome to #salt! | Latest Versions: 2015.8.12, 2016.3.3 | 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
01:52 edrocks joined #salt
02:00 netcho_ joined #salt
02:14 cppking joined #salt
02:16 cppking anybody help me about this https://bpaste.net/show/e7eafe56af09
02:16 cppking why cant "awk" work in salt "cmd.run"
02:17 godlike joined #salt
02:17 hemebond Do you have a simpler awk command I can test with?
02:17 hemebond One that is non-destructive?
02:17 cppking hemebond: it's simple enought , just print the pid of process
02:17 jab416171 joined #salt
02:20 hemebond The command doesn't work for me on my local machine.
02:20 hemebond Even without Salt.
02:23 hemebond cppking: ^
02:25 cppking hemebond: what do you mean , it's the problem of "command" or "execution of salt" ?
02:25 hemebond The command
02:25 hemebond The command is supposed to just print a number, right?
02:25 hemebond It doesn't, it just prints out the entire line.
02:25 cppking not a number, a list of number end with "\n"
02:25 hemebond The awk part of the command does nothing.
02:26 hemebond If I remove it I get the same output.
02:26 cppking yes, with or without "awk", it always get same output , why ?
02:26 hemebond Because your awk is wrong somehow.
02:26 hemebond I'm not very familiar with awk.
02:29 cppking here is the full command , I mean to kill all zabbix_agent process on all hosts, https://bpaste.net/show/b896ed8bd33b
02:29 cppking but it doesn't work, because "awk" doesn't work properly
02:29 hemebond Okay the awk part works with single quotes
02:30 hemebond awk '{print $2}'
02:30 hemebond But not the double-quotes.
02:30 cppking yes, I change it already
02:30 hemebond Do you not have access to pgrep?
02:30 cppking salt-minion run as root
02:31 hemebond Yes
02:31 cppking run as root already
02:31 hemebond Yeah
02:31 hemebond (not sure what you're getting at)
02:32 cppking well , these "jst*" host is proxied by a "salt-syndic"
02:32 jas02 joined #salt
02:32 hemebond uhuh
02:33 hemebond I don't understand what that has to do with pgrep.
02:33 cppking you mean I should use "pgrep" not "grep" ?
02:33 hemebond Correct.
02:33 hemebond pgrep does everything you are doing with ps, grep and awk.
02:34 hemebond So you don't need any of those.
02:34 hemebond Also, you have the ps.pkill execution module function in Salt.
02:34 hemebond https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.ps.html#salt.modules.ps.pkill
02:35 cppking I use "pgrep" successfully, but I still don't understand why "grep awk" etc not work
02:35 cppking Is it because used too many "pipe" ?
02:35 hemebond Your awk wasn't working because you had "{print $2}" instead of '{print $2}'
02:35 hemebond The quotes apparently matter.
02:35 hemebond You had " but awk wants '
02:36 cppking hemebond: no, I've changed it to single quote already, in my later post of full command
02:36 hemebond Then it should work.
02:37 hemebond Wait...
02:37 hemebond I don't know what the while is doing.
02:37 hemebond Everything up to and including the awk works for me.
02:38 hemebond Wait, it doesn't work in Salt.
02:38 hemebond One sec.
02:40 hemebond You should just use pkill
02:40 hemebond salt minion ps.pkill processname
02:42 evle2 joined #salt
02:45 hemebond I can't get it to work via Salt. I don't know enough about awk to know what it wants.
02:53 k_sze[work] joined #salt
02:59 shakalaka joined #salt
02:59 k_sze[work] joined #salt
03:21 shakalaka joined #salt
03:22 Micromus joined #salt
03:22 Lionel_Debroux joined #salt
03:24 raspado joined #salt
03:26 cppking hemebond: so now you think it is a problme of "command" or "execution of salt"?
03:33 flowstate joined #salt
03:35 knikolov joined #salt
03:39 hemebond cppking: It looks like it's being affected, at some point, by Salt.
03:39 hemebond It doesn't appear to be going through as-is.
03:46 mikea joined #salt
03:46 mikea any idea why a schedule state would keep reporting a change with +run_on_start:False every state run?
03:49 John_Kang joined #salt
03:50 cppking joined #salt
03:51 mikea Relevant info: http://pastebin.com/nXgmiqLW
03:54 edrocks joined #salt
03:58 hemebond mikea: Something was done with that property on 19 Aug
03:59 hemebond "fix 35420, add run_on_start in build_schedule_item, remove redundancy…"
03:59 mikea ah
03:59 mikea so its a bug
03:59 hemebond Maybe. Or it didn't exist until 19 Aug
03:59 hemebond https://github.com/saltstack/salt/commit/2d3a882cc2b998c7e4e7ba4b336b56dee0384f15
04:00 mikea I'm not running devel branch, I'm running 2016.3.3
04:00 hemebond https://github.com/saltstack/salt/issues/35420
04:00 saltstackbot [#35420][MERGED] run_on_start missing on salt.states.schedule | Description of Issue/Question...
04:01 mikea ah
04:01 mikea well I'm definitely running that code
04:01 mikea because it wouldn't work at all otherwise
04:01 netcho_ joined #salt
04:02 mikea hmm
04:02 mikea instead of a highstate, do you think it'd be better as a reactor on minion start?
04:02 hemebond To run a highstate? I use the start event for that.
04:02 mikea me too
04:03 hemebond I don't currently use any of the minion scheduling stuff.
04:03 Ni3mm4nd joined #salt
04:03 mikea I run it at startup and I schedule it for every 30 minutes (we're migrating from puppet, so I had to mimic the puppet behavior to make certain people feel more comfortable
04:03 hemebond lol
04:03 mikea without the run_on_start=False the highstate would run twice on startup
04:04 hemebond Because of your reactor?
04:04 mikea once from the reactor, again because of the scheduled job
04:04 mikea I can probably keep the schedule, remove the run_on_start and the reactor
04:21 jas02 joined #salt
04:32 cppking joined #salt
04:38 flowstate joined #salt
04:42 mikea I filed an issue for it
04:43 mikea just so it gets fixed eventually
04:43 mikea https://github.com/saltstack/salt/issues/37037
04:43 saltstackbot [#37037][OPEN] schedule state always reports changed when running in highstate | Description of Issue/Question...
04:43 raspado joined #salt
04:59 netcho_ joined #salt
05:00 flowstate joined #salt
05:01 DarkKnightCZ joined #salt
05:02 DEger joined #salt
05:05 knikolov joined #salt
05:12 flowstate joined #salt
05:18 samodid joined #salt
05:21 netcho_ joined #salt
05:22 DEger joined #salt
05:29 mohae_ joined #salt
05:30 coldbrewedbrew joined #salt
05:30 coldbrewedbrew joined #salt
05:32 coldbrewedbrew_ joined #salt
05:32 FreeSpencer joined #salt
05:32 FreeSpencer joined #salt
05:35 DEger joined #salt
05:42 mavhq joined #salt
05:48 felskrone joined #salt
05:50 flowstate joined #salt
05:52 jas02 joined #salt
05:53 CeBe1 joined #salt
05:55 flowstate joined #salt
05:55 whytewolf left #salt
05:56 whytewolf joined #salt
05:56 edrocks joined #salt
05:57 rdas joined #salt
05:59 vodik joined #salt
06:04 zulutango joined #salt
06:06 DEger joined #salt
06:13 colttt joined #salt
06:13 haam3r joined #salt
06:19 netcho_ joined #salt
06:20 DEger joined #salt
06:20 CeBe1 joined #salt
06:23 yuhlw_____ joined #salt
06:28 DEger joined #salt
06:31 sh123124213 joined #salt
06:33 yuhlw_____ joined #salt
06:35 barmaley joined #salt
06:37 sgo_ joined #salt
06:41 DEger joined #salt
06:45 theblazehen_ joined #salt
06:46 adb-mz joined #salt
06:46 jas02 joined #salt
06:47 sh123124213 joined #salt
06:50 DEger joined #salt
06:57 m4rx joined #salt
07:03 DEger joined #salt
07:03 dariusjs joined #salt
07:03 Electron^- joined #salt
07:05 sh123124213 joined #salt
07:13 samodid joined #salt
07:14 DEger joined #salt
07:18 netcho_ joined #salt
07:23 DEger joined #salt
07:24 ronnix joined #salt
07:33 infrmnt joined #salt
07:34 DEger joined #salt
07:35 barmaley joined #salt
07:35 JohnnyRun joined #salt
07:38 DEger_ joined #salt
07:46 _KaszpiR_ joined #salt
07:48 felskrone how do i create empty (single asterisk *) passwords with user.present?
07:51 stotch joined #salt
07:53 sh123124213 felskrone : try to escape it \*
07:55 sh123124213 if you mean user without a password and not an asterisk password then try to have ''
07:55 sh123124213 https://docs.saltstack.com/en/latest/ref/states/all/salt.states.user.html
07:56 sh123124213 empty_password=False
07:57 felskrone sh123124213: thx, tried that but \* produces the actual \* in /etc/shadow (not what i want), i specifically want the single * to be in shadow to show that the user never had a password
07:58 felskrone i guess i can live with '!', even though that indicates that the account is locked which it is not
07:58 edrocks joined #salt
07:59 sh123124213 felskrone : try empty_password=True
07:59 sh123124213 check the parameters of the function
07:59 mikecmpbll joined #salt
08:02 felskrone i dont get it, what would password-less login do for me if i want a single * (which indicates that the user shall never be able to login)
08:05 sh123124213 sorry, I was confused of you wanting a passwordless or a single asterisk for password
08:05 DEger joined #salt
08:05 theblazehen_ joined #salt
08:05 felskrone no problem, thanks anyway :-)
08:14 Reverend joined #salt
08:15 Rumbles joined #salt
08:15 sh123124213 joined #salt
08:15 cyteen joined #salt
08:18 jhauser joined #salt
08:23 sh123124213 joined #salt
08:25 _KaszpiR_ joined #salt
08:30 sh123124213 joined #salt
08:33 ronnix_ joined #salt
08:33 sh123124213 joined #salt
08:36 netcho_ joined #salt
08:38 Mattch joined #salt
08:38 sh123124213 joined #salt
08:42 hlub_ is there a way to query pillar of several machines in an orchestrate sls? I need to get credentials of databases to be able to perform backups.
08:44 hlub_ Alternatively I could run db dump commands over SSH.
08:45 sh123124213 joined #salt
08:45 hemebond I didn't think you could use pillars in orchestration.
08:46 hemebond Except what you provided via the command line.
08:47 sybix joined #salt
08:47 JohnnyRun joined #salt
08:49 lero joined #salt
08:52 DarkKnightCZ joined #salt
08:56 sh123124213 joined #salt
08:58 JohnnyRun joined #salt
09:05 theblazehen_ joined #salt
09:06 sh123124213 joined #salt
09:08 jas02 joined #salt
09:12 flowstate joined #salt
09:14 DarkKnightCZ joined #salt
09:14 jas02 joined #salt
09:15 SaltyVagrant joined #salt
09:22 _aeris_ joined #salt
09:27 jp47itc joined #salt
09:33 sh123124213 joined #salt
09:34 krymzon joined #salt
09:34 bigfoot_ joined #salt
09:39 KkLl joined #salt
09:44 bigfoot_ Hi! could anybody suggest how I could add MAILFROM line to crontab? (I use cron.present to apply the highstate)
09:45 KkLl left #salt
09:46 bigfoot_ file.prepend?
09:54 cyteen joined #salt
09:55 toanju joined #salt
09:56 bluenemo joined #salt
10:00 edrocks joined #salt
10:03 raspado joined #salt
10:12 flowstate joined #salt
10:14 cyteen joined #salt
10:16 _KaszpiR_ joined #salt
10:16 amcorreia joined #salt
10:23 DarkKnightCZ joined #salt
10:27 bigfoot_ well now I did add the lines to wrong file so I have a new question, how to remove lines that are prepended with file.prepend :)
10:28 DEger joined #salt
10:29 babilen bigfoot_: You could replace them with nothing
10:29 bigfoot_ ok
10:29 babilen (file.replace for example)
10:34 bigfoot__ joined #salt
10:35 al llua, yeah, you're right, but that's ok for systemd conf files, so works for me in this case
10:40 DEger joined #salt
10:43 jas02 joined #salt
10:45 mike25de_ joined #salt
10:45 * mike25de_ hi all
10:47 mike25de_ guys is it possible to install a package like? ...  xorg-x11-fonts-*  using pkg.installed? IT seems that i have an issue:   The following packages failed to install/update: xorg-x11-fonts-*   . i use salt 2016.3.3 on centos
10:50 DEger joined #salt
10:50 dariusjs joined #salt
10:50 babilen mike25de_: No, you'd have to list them individually
10:50 babilen (not 100% sure with centos, but it apparently doesn't work)
10:52 mike25de_ babilen: it seems like it. Thanks man.
10:52 mike25de_ i will have to do it manually.
10:54 Edgan joined #salt
10:55 jordir joined #salt
10:58 DEger joined #salt
11:02 abednarik joined #salt
11:03 irctc261 joined #salt
11:03 irctc261 hey guys
11:04 irctc261 I wanted to copy a file from master to minion
11:04 irctc261 but in file.managed want to give a path that is absolute
11:04 irctc261 i.e. outside the salt directory
11:04 sgo_ joined #salt
11:04 irctc261 is it possible ?
11:05 haam3r joined #salt
11:06 keimlink joined #salt
11:10 babilen irctc261: You should add the directory to file_roots so you can address it with salt://
11:11 irctc261 babilen: Thank you.
11:11 DEger joined #salt
11:11 babilen irctc261: You are not restricted to /srv/salt there
11:14 flowstate joined #salt
11:15 jas02_ joined #salt
11:16 impi joined #salt
11:22 DarkKnightCZ joined #salt
11:23 cyteen joined #salt
11:31 sh123124213 joined #salt
11:32 athaller joined #salt
11:34 Bico_Fino joined #salt
11:36 jas02 joined #salt
11:41 ronnix joined #salt
11:41 DEger joined #salt
11:50 Bico_Fino joined #salt
11:51 netcho_ joined #salt
11:54 DEger joined #salt
11:57 _KaszpiR_ joined #salt
11:59 DEger joined #salt
12:02 edrocks joined #salt
12:03 _KaszpiR_ joined #salt
12:06 jas02 joined #salt
12:09 alrayyes joined #salt
12:12 Redcat joined #salt
12:12 flowstate joined #salt
12:13 Redcat Hi guys
12:14 Redcat I'm currently working with saltstack LXC module and I'm adding a feature : adding veth interfaces on the fly whitout the need to restart the containers
12:15 Redcat however this feature use few tricks with linux network namespaces and isn't standard in lxc, so it doesn't use lxc-* commands
12:15 Redcat do you think this would be OK to update + PR on the lxc module ?
12:15 numkem joined #salt
12:17 DEger joined #salt
12:20 rdas joined #salt
12:20 jas02 joined #salt
12:22 LostSoul joined #salt
12:25 Qlawy How to tell to pkg.installed to update package if there is newer one available?
12:26 fxhp joined #salt
12:26 amontalban joined #salt
12:26 amontalban joined #salt
12:28 m4rx joined #salt
12:30 phx Qlawy, are you totally sure you want that?
12:30 phx especially if you are not running a controlled and QAd inhouse repo
12:31 phx but even if you do, just stuff installing "whenever" can cause service interruptions
12:32 Redcat Qlawy: and packages have dependancies, that mean updating libopenssl, your pkg management system may kill MySQL/Apache/Nginx/Whatever service is using this dependancy
12:40 lionel joined #salt
12:41 deus_ex joined #salt
12:44 amontalban joined #salt
12:45 sfxandy joined #salt
12:45 sfxandy hi everyone
12:46 sfxandy quick question - anyone had an experience or any suggestions regarding a reactor not firing for 'minion start' events
12:46 xmj libopenssl?
12:46 xmj what's that!
12:47 Qlawy phx: yes, 100% sure
12:48 Qlawy Redcat: I would at least have optional functionality to do that
12:48 xmj use pkg.latest ?
12:48 Qlawy hmm, seems legit
12:48 jp47itc Qlawy: what if he uses it a home for playing admin
12:48 xmj or latest as version-string, as per
12:48 xmj https://docs.saltstack.com/en/latest/ref/states/all/salt.states.pkg.html
12:48 * xmj points to manual
12:49 Qlawy jp47itc: I would live to have option to update my 50 hosts, where I need to update 2 packages and config file for them ;)
12:52 netcho_ joined #salt
12:52 lionel joined #salt
12:52 ronnix joined #salt
12:52 Redcat xmj: sry for the mispelling, I was talking of libssl of course :)
12:52 KingOfFools Why I could get False when do salt-run cache.clear_grains?
12:53 xmj libressl? :}
12:54 iadept joined #salt
12:55 APLU joined #salt
12:55 Redcat no libssl :)
12:56 * Redcat feed the trool
12:58 guerby joined #salt
12:59 m4rx joined #salt
13:01 ronnix_ joined #salt
13:03 keimlink joined #salt
13:06 Redcat sfxandy: Are you sure that you don't have any parsing error in your reactor ?
13:07 sfxandy positive because it fires the majority of the time.  so for example when bringing 40 minions online as part of an environment build, the reactor will fire perfectly well for say 37 or 38 minions.  but for the remaining two, it simply doesnt fire
13:08 sfxandy Redcat: i've experimented with various values for  randomizing the auth delay so that not all minions hit the master at the same tme with auth requests etc., but to no avail
13:10 jav joined #salt
13:12 flowstate joined #salt
13:13 DarkKnightCZ joined #salt
13:13 oida joined #salt
13:14 aagbds joined #salt
13:16 netcho_ joined #salt
13:16 jp47itc Qlawy: me 2
13:17 sgo_ joined #salt
13:20 DarkKnightCZ joined #salt
13:22 Redcat Qlawy: jp47itc: You should use the "version" parameter for pkg.installed, as xmj said
13:23 Redcat I use this for salt-minion updates since it depends on salt-common (at least on debian)
13:36 jp47itc Redcat: ok
13:37 Mads[m] joined #salt
13:37 edrocks joined #salt
13:37 DEger joined #salt
13:38 mikecmpbll i have a design question: i'm defining some firewall configuration in pillar data, and I have a pillar file e.g. app_server.sls that has firewall config specific for app servers. however, depending on the data centre the app server is at the config has to be slightly different; should i make app_server_dc1.sls and app_server_dc2.sls or can i use jinja in my pillar data to simplify things?
13:39 mikecmpbll i have minion name -> data centre stuff set up in other pillar data, if i could access that maybe?
13:39 impi joined #salt
13:40 gtmanfred you can use some jinja in your pillar files, you should be able to get the minion name as long as the grains for that minion are synced to the master
13:42 xmj and abstract things such that what stays the same - stays the same
13:42 Redcat mikecmpbll: I don't know the best practice, but I use the pillar merging feature and minion name-based datacenter selection
13:42 edrocks joined #salt
13:43 Redcat https://gist.github.com/theredcat/388536370b19fe5fbe3ddb230d7e3119
13:43 gtmanfred IMO, it is less complicated to just have app_server/dc1.sls app_server/dc2.sls, and use the top file to assign the correct one, but i also hate using jinja
13:43 Redcat you should be able to use the same technique using grains
13:44 abednarik joined #salt
13:45 mikecmpbll ah, good food for thought, hadn't considered the minion naming angle.
13:46 Redcat mikecmpbll: My whole infrastructure is name-based :)
13:47 mikecmpbll i've not even changed the minion names from default, which was in the back of my mind as a 'to-do'. is it just a case of changing the minion config?
13:50 Redcat mikecmpbll: My infra is based on lxc containers managed by salt, so i just provision containers in a way that the "hostname" command, return the name i want for the minion
13:50 mikecmpbll Redcat : 👍🏼 gotcha
13:50 Redcat That way i'm consistent in salt, dns, lxc names, etc....
13:51 ekristen joined #salt
14:00 ivanjaros joined #salt
14:05 abednarik joined #salt
14:05 mapu joined #salt
14:09 racooper joined #salt
14:10 GordonTX joined #salt
14:10 DammitJim joined #salt
14:11 flowstate joined #salt
14:23 DEger joined #salt
14:24 inad922 joined #salt
14:27 ALLmightySPIFF joined #salt
14:29 DEger joined #salt
14:29 hasues joined #salt
14:29 hasues left #salt
14:30 DEger joined #salt
14:32 fusionx86 joined #salt
14:34 DarkKnightCZ joined #salt
14:35 cscf joined #salt
14:36 rherna joined #salt
14:42 Tanta joined #salt
14:42 heaje joined #salt
14:47 DammitJim joined #salt
14:48 Ni3mm4nd joined #salt
14:49 fusionx86 joined #salt
14:49 lompik joined #salt
14:53 markizano Having issues authenticating against netapi - I have acl options set in /etc/salt/master, and I see it authenticating - and passing authentication. However, it's having issues getting a token from salt.auth.Resolver.mk_token() - I'm trying to dig deeper into the code to figure out what I'm missing, but does anyone have any pointers to help save me ?
14:55 keimlink joined #salt
14:55 markizano http://pastebin.com/eQdLkLim /etc/salt/master config snippet.
14:57 q1x joined #salt
14:59 CeBe1 joined #salt
15:00 llua does someone have a write up about implementing profiles in salt? i have a 'base' profile right now but it is pretty much one big include statement
15:00 llua and i see that being a problem when i want to change ordering of those states
15:01 markizano llua: I think you want to look into matching grains and minion_id's with top.sls.
15:01 flowstate joined #salt
15:12 goudale joined #salt
15:12 goudale hi
15:12 flowstate joined #salt
15:12 goudale is there any way to run command in a template on the master before sending the data to the minion ?
15:13 jenastar joined #salt
15:13 goudale specifically; I want the master to fetch some data (which require authentication, though should append in the master), then insert it in an sls, then let the minion do the normal job (ie, compiling the template and running the state)
15:14 flowstate joined #salt
15:14 goudale I think I can do it with the salt mine, but maybe there is a better road
15:15 markizano goudale: what kind of data ?
15:15 TomJepp joined #salt
15:16 cscf goudale, you could use a python template
15:16 cscf But there might be a better way
15:16 goudale cscf: python is interpreted on the master before sending the template to the minion ?
15:16 cscf goudale, just like any other template, yeah
15:16 goudale markizano: aws data. specifically, some kind of dict mapping security group names to ids
15:17 dariusjs joined #salt
15:18 goudale markizano: in order to retrive this dict, one need some AWS credentials. As my master already have some, it is easier to get the dict on it, instead of a minion which does not have credentials (and it will be a pain to manage specific access key per host)
15:18 markizano goudale: Could you put the aws client on your minions and have them execute the request ? Alternatively, yeah, the salt-mine might be an option, or some kind of persistent store that would make that data available.
15:18 jas02 joined #salt
15:19 goudale markizano: I precisely want to avoid running the request on the minion
15:19 markizano sorry - missed messages, didn't get the chance to edit before I saw your second response.
15:20 cscf a python template could get the data and insert it
15:21 cscf Might be complex though
15:23 markizano goudale: I'm being creative here, but you could have a module that runs on the master that performs the aws interaction, and saves the data structures you want as yaml files in a place the salt master can find the data files. Store them as .sls files, and write up some index file that includes them. Reference the index.sls in your top.sls on the minion you want to have access to the  aws data structures. :)
15:24 goudale that's kind of a reengineered mine, isn't it ?
15:24 goudale I already wrote the module to fetch data, so I'm halfway I guess
15:24 beowuff joined #salt
15:24 Redcat gtmanfred: do you think implementing this function in the lxc module would be a nice idea, or is it too off-scope since it desn't use lxc native commands ? => https://gist.github.com/theredcat/604994e612df7e77f323bb502809bc49
15:26 markizano goudale: possibly, but don't take my word for it, lol, I'm just a customer like yourself.
15:26 markizano goudale: https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.boto_ec2.html <- hrmm.. Interesting, no ?
15:27 goudale markizano: the problem stay; the module executes on the minion, so they need credentials to connect to aws, which I'm not willing to give to all of them.
15:28 goudale I'll try to store data in pillars, which are compiled on the master
15:32 whytewolf humm. cscf not all rendering happens on the master before being transfered to the minion. state files are rendered completly on the minion. only pillars, reactors and orchestration are rendered on the master
15:32 MeltedLux joined #salt
15:33 cscf whytewolf, oops, my bad
15:36 theredcat joined #salt
15:40 XenophonF holy crap! /var/cache/salt/master/pillar_gitfs is world readable?!
15:42 lero joined #salt
15:42 esckroh joined #salt
15:43 martoss joined #salt
15:43 cscf That does seem unwise
15:43 cscf What exactly does it contain?
15:43 impi joined #salt
15:44 XenophonF well it contains my pillar data, which shouldn't be accessible to unprivileged user accounts
15:44 esckroh_ joined #salt
15:44 cscf Yeah, that's strange then.
15:45 whytewolf XenophonF: what about the files within?
15:45 XenophonF all 0644
15:46 XenophonF everything under /var/cache/salt/master appears to be world-readable
15:46 whytewolf not everythin
15:46 XenophonF i take that back - .root_key isn't, either is files
15:46 theredcat XenophonF: on my setup /var/cache/salt/master/files/base/ is 600
15:47 VMF joined #salt
15:47 whytewolf on my setup /var/cache./salt/master/minion/<minion name>/data.p is 600
15:47 whytewolf but everything up to that point is 755
15:48 prg3_ joined #salt
15:48 jas02 joined #salt
15:49 VMF hello there
15:49 VMF pretty new to salt and CM in general
15:49 VMF have few questions
15:50 VMF how are you all doing?
15:50 whytewolf pretty good.
15:51 VMF quick question, when i run cmd.run, i get an output into stdout
15:51 VMF which is lets say 1 or 0
15:51 VMF now i need to build the logic based on this output
15:51 VMF how can i incorporate it?
15:52 VMF as in "if 1 do this, if 0 do that"
15:54 whytewolf that question has so many answers and non answers that I'm not sure i could form it into a reconizable response
15:54 jenastar joined #salt
15:55 cscf Yeah I think you're going about it wrong, XY problem.  What do you actually want to do?
15:56 VMF ok, i have 3 states, i execute first one and based on the stdout output i execute either state 2 or state 3
15:56 whytewolf is state 1 actuqlly changing anything?
15:57 VMF no, it executes a shell script
15:57 cscf VMF, and what does this script do?
15:57 whytewolf okay, then most likely it shouldn't be a state. it should be an unless or an onlyif
15:57 VMF that checks if deployment exists
15:57 whytewolf https://docs.saltstack.com/en/latest/ref/states/requisites.html
15:57 VMF in gcloud
15:58 SaltyVagrant VMF: Do you have control of the script, that is, can you change what appears on STDOUT? Can you control it’s exit status?
15:58 whytewolf https://docs.saltstack.com/en/latest/ref/states/requisites.html#unless
15:58 VMF i dont think salt has native functionality to check if kubernetes has a deployment
15:58 VMF yes, and yes
15:58 VMF let me show you the script, its a one liner
15:58 VMF kubectl get deployments |grep jbiam| wc -l
15:59 cscf VMF, did you read the link?
15:59 VMF first or second?
15:59 whytewolf both
15:59 cscf Make your script return 0 or 1, then use unless and onlyif
16:00 whytewolf first is over acing about state modifiers, second is in that page about unless
16:00 VMF it does return either 1 or 0
16:00 whytewolf exit code 1 or 0 not text output of 1 or 0
16:00 cscf VMF, return code, or stdout?
16:01 VMF https://gist.github.com/vital4ik/fa842a3b6a5ffdf82c7ef8e8a2b7662a
16:02 VMF I am pretty new
16:02 BlackMaria left #salt
16:02 VMF here is the output from the state executing the script
16:03 VMF reading the requisite
16:04 flowstate joined #salt
16:04 VMF so i need to mod the script to return 0 is there is 1 line and return 1 if there is 0 lines?
16:05 debian112 joined #salt
16:05 VMF ok I think im grasping it, so unless will monitor the execution of the script? script is a managed file
16:05 bicchi joined #salt
16:06 DammitJim joined #salt
16:06 noraatepernos joined #salt
16:10 bicchi Is it possible to setup the salt master so that it has multiple as slaves and the slaves act as the masters to the minions. Basically looking to spread the load of the master onto multiple slaves.
16:10 VMF syndic master
16:10 bicchi perfect, didn't know what that would be called. thanks
16:19 lovecraftian joined #salt
16:19 lovecraftian joined #salt
16:21 CampusD joined #salt
16:22 sarlalian joined #salt
16:24 _aeris_ joined #salt
16:26 VMF Gents, so for unless to work, do i need to give exit code to stgerr?
16:26 VMF i have the following script to test it
16:26 VMF https://gist.github.com/vital4ik/3d72ebc347df83a73072e9f3ee7a4298
16:27 mikecmpbll is there a way to look at a particular grain value of all minions from salt master? like salt '*' grains.items but only one grain? :/
16:27 mikecmpbll fuck nvm
16:27 * mikecmpbll is a moron, it's right there in front of me.
16:29 whytewolf VMF: 0 and 1 [0 = true the command ran fine >=1 is the program didn't run fine aka false]] so using 1 and 2 is basicly telling salt the same thing
16:29 VR-Jack2 joined #salt
16:30 sgo_ joined #salt
16:31 whytewolf least i think thats right... I have drowning in so many meetings right now i can hadley think right
16:31 VMF i feel ya, most of my day is spent in meetings as well
16:31 VMF so lets say first exit is just exit, not 1
16:32 VMF but does the unless need to read from stdout or stderr?
16:32 VMF do i need to 1>&2?
16:32 whytewolf unless reads the retcode
16:32 whytewolf it doens't care about stderr or stdout
16:32 VMF i c
16:32 pipps joined #salt
16:32 VMF let me give it a try
16:32 VMF thanks!
16:35 VMF learned a lot today already:)
16:36 whytewolf learning is good :)
16:36 nidr0x joined #salt
16:37 KingOfFools Guys, how to run saltutil.refresh_pillar with orchestrate file?
16:37 whytewolf salt.function
16:38 jenastar joined #salt
16:39 whytewolf KingOfFools: https://gist.github.com/whytewolf/c72e17bd1e88c461cb50e3e29aeadee6
16:39 whytewolf like that?
16:40 samodid joined #salt
16:40 benny_ joined #salt
16:41 KingOfFools whytewolf: i'm getting error 'No changes made for refresh_pillar'
16:42 jcockhren left #salt
16:44 whytewolf can you post your state?
16:44 KingOfFools whytewolf: oops, i'm retarded, forget about tgt_type
16:44 KingOfFools whytewolf: thanks (:
16:45 whytewolf hehe no problem, even if i did post a wrong state [i forgot the  -]
16:50 CampusD Hi guys, quick question, if I use the local_cache returner, any idea where the data is store and it's format? I am looking at https://docs.saltstack.com/en/latest/ref/returners/all/salt.returners.local_cache.html#module-salt.returners.local_cache
16:50 onlyanegg joined #salt
16:52 nicksloan joined #salt
16:53 whytewolf most likely somewhere in /var/cache/salt/* and msgpack format
16:57 benny_ Hey guys, is anyone running salt using raet?
16:57 CampusD whytewolf: thanks, I'll check there. I am assuming I can enable it on the minion config with return: local_cache
16:57 whytewolf most likely default location would maybe be something like /var/cache/salt/minion/jobs/<jid>
16:57 whytewolf CampusD: in thoery yes. i havn't messed with the returners in a while
16:58 impi joined #salt
16:59 CampusD is there a default time that the cached data stays there and does it store any command/state run on the minion? Looks like there is a way to clear it with clean_old_jobs()
17:00 * whytewolf shrugs
17:01 Nahual joined #salt
17:03 sh123124213 joined #salt
17:03 iggy benny_: I'm no dev, but I'd assume raet is basically abandoned
17:04 benny_ Oh, I thought it was going to be the way to go for salt
17:04 benny_ Thanks iggy!!
17:05 mikecmpbll joined #salt
17:06 Bico_Fino joined #salt
17:10 mavhq joined #salt
17:10 whytewolf everytime i say it is most likely abanded an op pops up that there is still work happening on it. but it is not a priority
17:14 wm-bot4 joined #salt
17:15 sh123124213 joined #salt
17:15 KajiMaster joined #salt
17:17 benny_ whytewolf: Thanks, there is no mention of raet being dropped any where. It was supposed to be better than 0mq so I thought I'd give it a try
17:17 benny_ It's also been 2 years since it was announced.
17:18 benny_ I thought they would be done implementing by now.
17:18 Trauma joined #salt
17:21 whytewolf they ran into some snaggs with the fact udp is not exactly the most reliable of network protocols
17:28 jas02 joined #salt
17:30 flowstate joined #salt
17:33 KingOfFools Can anyone suggest how do i run orchestration from from orchestration file?
17:34 whytewolf i would think https://docs.saltstack.com/en/latest/ref/states/all/salt.states.saltmod.html#salt.states.saltmod.runner
17:35 flowstate joined #salt
17:37 KingOfFools whytewolf: yea, but i dont get what variables should i pass. I'm trying to pass ' - name: salt.orchestrate or salt.sls' and '- sls: slsfile' but getting errors
17:38 whytewolf when you run it from the cli you use state.orchestrate as the name
17:38 benny_ whytewolf: that's interesting because there are quite a few products that transfer files using udp and do their own transaction checking.
17:42 cyborg-one joined #salt
17:43 whytewolf KingOfFools: I would think something like this https://gist.github.com/whytewolf/61f06029166fe2b547df604d26504db1
17:43 ramblinpeck joined #salt
17:45 KingOfFools whytewolf: nah, 'KeyError: 'salt.orch''
17:45 KingOfFools whytewolf: same with 'orchestrate'
17:45 whytewolf state not salt
17:46 whytewolf salt.orch = nothing
17:46 Bico_Fino_ joined #salt
17:47 KingOfFools whytewolf: yea yea i got it, but getting The following keyword arguments are not valid: kwargs=OrderedDict([('mods', 'http_frontend.nginx_conf_generate')])
17:48 whytewolf humm, maybe it needs to be - mods:
17:49 KingOfFools whytewolf: nope (:, i think there's no difference between 'mods' and '- mods' in that situation
17:49 whytewolf there is
17:49 whytewolf mods: is a dict - mods is a list
17:50 KingOfFools whytewolf: hm, ok, but getting error anyway
17:55 M-liberdiko joined #salt
17:55 whytewolf man I must be really tired. ...
17:56 whytewolf KingOfFools: I just updated the gist
17:57 whytewolf I'm so tired i forgot that kwargs is literally just passing anything extra in
17:58 dnull[m] joined #salt
17:58 Mads[m] joined #salt
17:58 freelock[m] joined #salt
17:58 M-MadsRC joined #salt
17:58 saintaquinas[m] joined #salt
17:58 KingOfFools whytewolf: daaayuuuum, now it works :D Thanks!
18:01 hellosalt joined #salt
18:03 hellosalt Has anyone run into an issue with Salt and Docker 1.12?
18:03 hellosalt https://gist.github.com/anonymous/a38d7521ca3bb42ec47d72293bb85df6
18:03 hellosalt "error: starting container with HostConfig was deprecated since v1.10 and removed in v1.12"
18:03 nicksloan is anyone using the nginx.ng formula, and if so, have you come up a strategy that doesn't force you to rewrite the entire configuration just to override (for example) the server_name on a different host?
18:03 nicksloan *different minion
18:11 ivanjaros joined #salt
18:12 sh123124213 joined #salt
18:15 edrocks joined #salt
18:29 jason` so i have a string in a map file that's getting the unicode designation and choking the SLS parses - {'devcpu-data': u'http://blahblah.com/devcpu-data-1.7.txz'}
18:30 jason` no doubt because i'm trying to grab a grain.. but not seeing any work around - ' ~ grains.get('osmajorrelease', '') ~ '
18:30 iggy jason`: can you gist a more complete example?
18:34 jason` iggy: http://pastebin.com/UBEPTHr3
18:35 jason` bah sorry on a flight with terrible connectivity.. the grain comes out properly in the final string, it's just the 'u' throwing SLS off it appears
18:37 dps joined #salt
18:42 sgo_ joined #salt
18:43 sh123124213 joined #salt
18:45 sh123124_ joined #salt
18:45 pipps joined #salt
18:49 sh123124213 joined #salt
18:50 abednarik joined #salt
18:54 UtahDave joined #salt
19:11 haam3r joined #salt
19:12 GreatSnoopy joined #salt
19:13 pipps joined #salt
19:17 joe joined #salt
19:22 roxtrongo joined #salt
19:24 DanyC joined #salt
19:24 pipps joined #salt
19:25 DEger joined #salt
19:26 snc joined #salt
19:28 iggy jason`: maybe use the |string jinja filter
19:33 MeltedLux joined #salt
19:34 pipps joined #salt
19:35 notnotpeter joined #salt
19:36 m4rx joined #salt
19:42 DanyC left #salt
19:45 pipps joined #salt
19:48 Bico_Fino joined #salt
19:48 gentoo_ joined #salt
19:50 edrocks joined #salt
19:54 MeltedLux joined #salt
19:55 sh123124213 joined #salt
19:56 pipps joined #salt
19:58 DEger joined #salt
19:58 sh123124213 joined #salt
20:00 DEger joined #salt
20:00 sh123124213 joined #salt
20:01 frew joined #salt
20:02 sh123124_ joined #salt
20:04 pjs joined #salt
20:07 flowstate joined #salt
20:07 lero joined #salt
20:09 preludedrew joined #salt
20:10 sh123124213 joined #salt
20:11 flowstate joined #salt
20:17 racooper howdy.  can the salt beacon on a minion tell master to run a script on another minion? (say, if a file is changed on a primary node, then pull that file to other nodes)
20:17 roxtrongo joined #salt
20:19 roxtrongo joined #salt
20:19 hemebond racooper: minion1 -> beacon -> reactor -> state.apply -> tgt: minion2
20:20 racooper ah. I figured it had to something simple, I'll read the doc on that more closely.  it's been a long week today.
20:22 pipps joined #salt
20:25 frew left #salt
20:25 frew joined #salt
20:25 honestly hate those days
20:26 hemebond To be honest the reactor stuff can be quite difficult to parse.
20:26 UtahDave hemebond +1
20:26 UtahDave left #salt
20:26 hemebond drive-by plus-oned
20:27 sjorge joined #salt
20:27 sjorge joined #salt
20:27 mapu joined #salt
20:29 VMF hey guys, as previsoulsy suggested, I am trying to use unless requisite while using the exit code from the bash scrip
20:30 VMF but i think i did not do something right
20:31 pipps joined #salt
20:33 edrocks joined #salt
20:33 cscf VMF, well, what happens?
20:36 Bico_Fino joined #salt
20:39 dps joined #salt
20:41 beakerman Is the preferred location to set something like a global variable in the pillars example being something like %{if 'AVR' in grains['host']%} environtment: r2d2
20:42 hemebond I literally use a globals.yaml file
20:42 iggy I prefer to target in the top file, but that's just me
20:42 hemebond And import that into each pillar.
20:42 beakerman Find myself reusing this stuff alot and seems like there is a better more "global" way of setting it so I can use it in my states
20:42 hemebond Oh they're just wanting targeting.
20:43 VMF sorry
20:43 VMF some1 came over
20:43 toanju joined #salt
20:43 VMF here is the gist
20:44 VMF https://gist.github.com/vital4ik/ccd5146b841bf14e19bf4385688dec91
20:44 VMF THIS IS THE SHELL SCRIP
20:44 cscf VMF, why  exit 1 >&2  ?
20:45 VMF it was me trying, honestly it doesnt work without anyway
20:45 cscf I mean it doesn't matter, but it does nothing
20:45 VMF https://gist.github.com/vital4ik/6f171c3571f0b9f8ac8b8723cc4aa4c0
20:45 beakerman maybe I am thinking of it wrong the actual variables are global but they are different depending on the environment. I suppose a global.yaml could work just don't want to have to update it a billion times
20:46 VMF here are the states
20:46 iggy protip: gist supports putting multiple files in a single paste
20:46 VMF hmm good to know
20:46 VMF im very new to this
20:46 VMF thanks
20:46 cscf VMF, the script works fine for me
20:47 VMF so when i the state, looks like the retcode is always 1
20:48 VMF so unless always fires
20:48 VMF *run
20:48 VMF end of day, slipping
20:48 nicksloan joined #salt
20:48 iggy jason`: it looks like you might have some funky single quotes in that paste... hard to tell with all the ads on pastebin.com
20:49 VMF where?
20:49 VMF in the script or in the sate?
20:49 iggy I was talking to someone else
20:49 cscf VMF, he's not talking to you,  I think
20:49 iggy VMF: take the cmd.run out of the unless
20:50 VMF ah
20:50 VMF ok lemme try
20:50 iggy - unless:\n  - /tmp/gdcheck.sh
20:50 cscf iggy, ah, so it's trying to run 'cmd.run' in bash and getting an error?
20:51 iggy yeah
20:51 cscf Makes sense
20:51 netcho_ joined #salt
20:51 VMF worked like a charm
20:52 VMF thank you!
20:53 cscf nice
20:53 lionel joined #salt
21:05 Bico_Fino joined #salt
21:07 DEger_ joined #salt
21:08 VMF Gents, question, I am trying to execute a state from jenkins, but i want to add a jenkins var as a parameter into the state
21:08 VMF is it possible?
21:10 ecdhe VMF, that is possible, I'm doing a little searching to see if I can find that for you
21:10 ecdhe VMF, are you calling state.sls?
21:10 DEger joined #salt
21:11 VMF yep
21:12 VMF its salt 'bla*' state.sls state
21:12 VMF i tried searching but my poor english skills are not allowing me to comprehend a lot in the salt docs
21:13 pipps joined #salt
21:13 ecdhe salt '*' state.sls some.state pillar='{"settings":{"foo":"a"}}'
21:13 lero joined #salt
21:13 VMF json?
21:14 VMF so what is foo and a in this case
21:14 VMF and settings
21:14 ecdhe VMF, are you using pillar yet?
21:14 VMF lets say i want to add the build_number var into the state
21:14 VMF yeah, i do, i use remote pillars
21:14 ecdhe cool
21:16 _KaszpiR_ joined #salt
21:17 ecdhe Okay, so pillar='{<json object}' is going to merge with your pillar data, so the values you pass in will be available the same way your pillar data is.
21:17 amontalban joined #salt
21:17 amontalban joined #salt
21:20 VMF gotcha, thank you, let me give it a try
21:20 ecdhe I assume you could, for instance, override or extend the pillar for the users-formula by passing {"users":["usera", "userb"]}
21:20 ecdhe VMF, good luck!
21:22 sh123124213 joined #salt
21:23 VMF and i need to pass 2 separate values?
21:24 VMF *if
21:25 VMF sudo salt "$JBNAME2*" state.sls gdeploy pillar={"jbname":["$JBNAME2"]}
21:25 MTecknology Is it possible, from within an sls file, require that saltutil.sync_states ran and the state is available for the sls?
21:25 * MTecknology hasn't ever needed to worry about this before..
21:26 VMF where JBNAME2 is a parameter that is being passed from another job
21:26 iggy MTecknology: you're doing it wrong
21:26 MTecknology iggy: probably...
21:26 MTecknology iggy: any hints for better? :)
21:27 iggy use highstate (it sync's at the start)
21:27 MTecknology highstate is unreasonably scary in this environment
21:28 nicksloan joined #salt
21:28 nicksloan I know I brought it up recently, but does anyone have any thoughts on this issue on the nginx formula: https://github.com/saltstack-formulas/nginx-formula/issues/132? I feel like I'm missing something obvious, though I have thought of using grains to define some of the specific values, though I don't love the idea
21:28 saltstackbot [#132][OPEN] Overriding values in nginx.ng is pretty tricky. | The biggest problem that I have with this formula is in overriding values. Different minions have different server_name directives, for example. Given the list structure for most directives, I can't rely on dictionary merging to allow me to supply additional bits of config in other files....
21:35 iggy MTecknology: {% set junk = salt['saltutil.sync_states'] %} is shitty
21:36 MTecknology heh, neat; crappy, but neat
21:37 MTecknology I'm trying to figure out how I can keep a custom state out of this, but not sure it's possible
21:39 iggy you are just asking for trouble
21:41 MTecknology hmmmm
21:41 MTecknology Is it possible to just make a state that flags an error?
21:42 iggy !salt state.test
21:42 saltstackbot https://docs.saltstack.com/en/latest/ref/states/all/salt.states.test.html
21:43 MTecknology that might work! I'll just raise warnings if the test fails rather than actually make changes; because this shouldn't ever trigger in the first place
21:48 noraatepernos joined #salt
21:48 netcho_ joined #salt
21:49 zulutango joined #salt
21:49 VMF hmmm i get weird results
21:50 VMF salt 'jbiam*' state.sls gdeploy pillar='{"jbname":{"jbiam"}}'
21:50 VMF is the command
21:50 VMF this is what i get
21:51 iggy that's not valid
21:51 VMF why?
21:51 iggy dict's need key's and values
21:51 iggy if it's just values, use a list (or just nothing at all
21:52 iggy {"jbiam"}  <--- that bit specifically
21:52 VMF thats the value
21:52 VMF and i assumed jbname is the key
21:52 whytewolf {} = dict
21:52 iggy that's the top level dict
21:52 iggy you have {} inside {}
21:52 catpig joined #salt
21:53 VMF got it
21:53 iggy maybe you just want: pillar='{"jbname":"jbiam"}'
21:53 VMF yep that just worked
21:53 VMF guys you are awesome
21:53 VMF ive been struggling with the salt docs for 2 monnth
21:53 VMF now in few hours i made so much progress!
21:54 VMF and now if i want to have 2 vars passed into the state?
21:54 VMF is it possible?
21:55 whytewolf pillar = '{'var1':'value1','var2':'value2'}'
21:55 Salander27 joined #salt
21:55 whytewolf err. quoting isues
21:55 whytewolf pillar = '{"var1":"value1","var2":"value2"}'
21:56 iggy or pillar='{"jbname":["jbiam", "baz]}'
21:56 iggy if it's a list of multiple values per key that you want
21:56 whytewolf ^ for a nested pair
21:56 VMF so elegant
21:56 VMF thank you
21:56 whytewolf you can get crazy with the cheeze whiz with how you align your extra pillars
21:59 VMF nah, I am too much of a noob at the moment
22:04 edrocks joined #salt
22:04 beowuff I've been poking around, but haven't found any thing... Can salt pxe boot a server? Or do I need something like cobbler that would then hand off to salt?
22:05 whytewolf you need something like cobbler. salt was working on salt-genisis a long time ago but that was an abandoned product as far as i can tell
22:06 whytewolf you can use cobble for pxe booting then use saltify afterwords
22:06 hemebond I just used a regular HTTP+PXE server to build servers. Not sure what Cobbler is.
22:06 whytewolf cobbler is basicly a easy interface for that
22:06 hemebond And Salt managed the PXE server of course.
22:07 mike25de left #salt
22:07 hemebond Ah
22:08 whytewolf lets you manage iso's, kernels and install scripts, for pxe installs. and also allows for configuration of systems based on mac address ect.
22:08 iggy we use FAI with some custom code that does the configuring iDRAC to pxe boot
22:08 whytewolf also has repo mirroring and other features
22:09 noraatepernos joined #salt
22:13 KingOfFools joined #salt
22:13 noraatepernos_ joined #salt
22:14 pipps joined #salt
22:14 whitenoise joined #salt
22:17 conan_the_destro joined #salt
22:21 MTecknology https://github.com/saltstack/salt/blob/develop/salt/states/bigip.py#L2823  <-- Is it safe to assume r''' is a typo, or does it make sense there?
22:22 dendazen joined #salt
22:22 hemebond Seems unnecessary.
22:22 nsidhu joined #salt
22:22 iggy backticks
22:23 hemebond ah
22:23 iggy and probably backslash too
22:24 gentoo_ joined #salt
22:24 hemebond I get identical output in a console so must be something to do with the documentation compilation or something.
22:24 GreatSnoopy joined #salt
22:24 gentoo_ joined #salt
22:25 jas02_ joined #salt
22:25 gentoo_ joined #salt
22:25 gentoo_ joined #salt
22:26 gentoo_ joined #salt
22:26 iggy "in a console"?
22:26 hemebond ipython
22:26 iggy you import that?
22:27 hemebond Nope, just typed it out :-P
22:27 hemebond So yeah, probably not very useful.
22:27 hemebond But I was expecting some sort of error.
22:27 iggy multiline comments have special meaning at the start of modules and functions
22:27 * MTecknology apparently needs to learn more about python strings
22:28 iggy (and they aren't technically multi-line comments...)
22:30 hemebond Removing the r in a test file didn't seem to make a difference to the help() output. Should it?
22:38 pipps99 joined #salt
22:42 ALLmightySPIFF joined #salt
22:44 goudale joined #salt
22:49 justanotheruser joined #salt
22:51 pipps99 joined #salt
22:58 Rumbles joined #salt
23:03 Mattch joined #salt
23:04 karlthane joined #salt
23:13 edrocks joined #salt
23:14 sjmh joined #salt
23:16 cyteen joined #salt
23:19 jenastar joined #salt
23:20 pipps joined #salt
23:25 myraft joined #salt
23:25 pipps joined #salt
23:27 XenophonF r'' is a raw string literal
23:27 XenophonF it changes the rules for escaping within the string
23:27 hemebond Yeah, I think it's to do with the Sphinx compilation.
23:28 hemebond (maybe not. ignore me)
23:28 XenophonF triple-quoting strings also changes how the escaping does
23:28 XenophonF docstrings aren't a sphinx thing
23:29 XenophonF but how you markup text within the docstring can be
23:29 hemebond Yes. I was just reading the Sphinx docs on escaping.
23:29 jas02_ joined #salt
23:29 XenophonF https://www.python.org/dev/peps/pep-0257/
23:29 hemebond Normally you have to escape backslashes.
23:29 pipps joined #salt
23:33 pipps joined #salt
23:42 _JZ_ joined #salt
23:44 zulutango joined #salt

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