Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2017-04-13

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

All times shown according to UTC.

Time Nick Message
00:07 woodtablet left #salt
00:18 onlyanegg joined #salt
00:25 mpanetta joined #salt
00:37 Tanta joined #salt
00:38 keltim joined #salt
00:47 akunin joined #salt
00:47 sarcasticadmin joined #salt
00:50 akunin how do i print only the changes from a known jid? salt-run --state-output=changes jobs.list_job <jid> still spits out the whole thing. because the job-cache is saved in mongodb?
00:52 onlyanegg joined #salt
00:52 aneeshusa joined #salt
00:55 masber joined #salt
01:03 onlyanegg joined #salt
01:11 akunin joined #salt
01:17 akunin joined #salt
01:27 Karunamon joined #salt
01:28 ksa joined #salt
01:29 egilh joined #salt
01:31 akunin joined #salt
01:32 benner joined #salt
01:32 ventris joined #salt
01:32 Annihitek joined #salt
01:39 akunin joined #salt
01:46 catpigger joined #salt
01:48 ilbot3 joined #salt
01:48 Topic for #salt is now Welcome to #salt! <+> Latest Versions: 2016.3.6, 2016.11.3 <+> 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 mosen joined #salt
01:54 akunin joined #salt
01:57 censorshipwreck joined #salt
01:59 druonysus joined #salt
02:00 pbandark joined #salt
02:02 druonysus does Salt work with Python3?
02:02 hemebond druonysus: Not yet, no.
02:05 whytewolf hopefully next major release. nitrogen is supposed to be the python3 release.
02:05 druonysus hemebond: Is there anywhere were this is tracked to follow along with the progress?
02:06 hemebond There are issues in Github for it.
02:07 druonysus is it this one? https://github.com/saltstack/salt/issues/11995
02:07 saltstackbot [#11995][MERGED] Python 3 Support | Salt is currently not compatible with python 3. Ubuntu trusty and above default to using python 3, so it would be really nice to be able to run salt in the default python version....
02:08 whytewolf yes, that is one
02:08 druonysus great
02:08 druonysus I will keep my eye on it
02:08 whytewolf I'm sure there are more
02:09 hemebond There are. I don't think there is a single tracking issue for it.
02:09 druonysus whytewolf and hemebond thank you both very much :)
02:11 whytewolf druonysus: 2 more https://github.com/saltstack/salt/issues/40435 https://github.com/saltstack/salt/issues/40434
02:11 saltstackbot [#40434][OPEN] Python 3 - Head of develop - Salt-ssh not working | Salt-ssh with python3 on head of develop fails ...
02:15 mpanetta_ joined #salt
02:26 akunin joined #salt
02:29 akunin joined #salt
02:32 akunin joined #salt
02:35 akunin joined #salt
02:41 akunin joined #salt
02:46 akunin joined #salt
02:48 evle joined #salt
02:48 patrek joined #salt
02:51 akunin joined #salt
02:54 akunin joined #salt
02:57 Bock joined #salt
03:01 fooker joined #salt
03:06 akunin joined #salt
03:08 Bock joined #salt
03:17 onlyanegg joined #salt
03:23 Xenophon1 joined #salt
03:33 stooj joined #salt
03:37 onlyanegg joined #salt
03:39 stugs joined #salt
03:46 akunin joined #salt
03:51 akunin joined #salt
03:54 atree joined #salt
03:59 akunin joined #salt
04:02 akunin joined #salt
04:09 akunin joined #salt
04:12 druonysus left #salt
04:49 onlyanegg joined #salt
04:51 justan0theruser joined #salt
04:58 rdas joined #salt
05:03 mpanetta joined #salt
05:16 bluepaladin joined #salt
05:20 yuhl______ joined #salt
05:23 yuhl______ joined #salt
05:34 preludedrew joined #salt
05:52 do3meli joined #salt
05:52 do3meli left #salt
06:00 golodhrim|work|3 joined #salt
06:23 jas02 joined #salt
06:25 Inveracity joined #salt
06:26 jodok joined #salt
06:27 nick123 joined #salt
06:38 Antiarc joined #salt
06:40 strohi joined #salt
06:56 juntalis joined #salt
07:00 aldevar joined #salt
07:04 Ricardo1000 joined #salt
07:14 Praematura joined #salt
07:26 JohnnyRun joined #salt
07:28 dariusjs joined #salt
07:32 Reverend joined #salt
07:32 Reverend left #salt
07:36 pi3r joined #salt
07:37 pi3r Hi, I can't for the sake of it understand why I have got this error: https://paste.ee/p/5bhwL
07:37 pi3r Any idea ?
07:38 pi3r I am using the latest 2016.11.3
07:38 dariusjs joined #salt
07:45 om2 joined #salt
07:50 haam3r pi3r: maybe an issue with your pgjsonb returner config?
07:51 pbandark joined #salt
07:51 om2 joined #salt
07:54 Rumbles joined #salt
07:57 pi3r haam3r: I am sure the returner works though as I do receive the job results in the database
08:02 pi3r I can't believe this. What a loosy way to handle errors ! If I try salt-run jobs.lookup_jid 20 I have got a stacktrace with
08:02 pi3r ProgrammingError: operator does not exist: character varying = integer
08:02 pi3r LINE 1: SELECT load FROM jids WHERE jid = 20;
08:02 pi3r I really don't expect this from a product that has been out for years now
08:10 onlyanegg joined #salt
08:11 nick123 joined #salt
08:20 swa_work joined #salt
08:23 hemebond Product?
08:26 jhauser joined #salt
08:27 Mattch joined #salt
08:31 ronnix joined #salt
08:31 mikecmpbll joined #salt
08:38 pi3r hemebond: let's say I would expect Salt (or any other orchestration tool) to handle basic input validation instead of throwing python stack trace ...
08:38 gmoro_ joined #salt
08:40 candyman88 joined #salt
08:49 s_kunk joined #salt
08:50 gmoro joined #salt
08:50 hemebond What is the pgjsonb returner? Returns a json string to postgresql?
08:51 hemebond Apparently so :-)
08:58 N-Mi joined #salt
08:58 N-Mi joined #salt
09:25 dariusjs joined #salt
09:34 strohi joined #salt
09:37 pi3r hemebond: yes. I am really confused because I have another machine (in staging instead of dev) with the same salt config (except in point to another db to the same postgrestql server)
09:38 hemebond The returner fails every time?
09:38 pi3r No it works on the second machine ...
09:38 hemebond lookup_jid queries the database? That's interesting.
09:38 pi3r Of course it always fails when I enter a wrong jid but this is another issue
09:39 pi3r well to be sure I remove all record from the dev postgresql database ... and did it again hoping the problem would disappear but no
09:39 hemebond I'm confused what the issue is now.
09:42 pi3r I have created this issue https://github.com/saltstack/salt/issues/40673 that describes it better
09:42 saltstackbot [#40673][OPEN] Unhelpful user errors when running `salt-run jobs.lookup_jid` | Description of Issue/Question...
09:45 hemebond But you get no error if the jid exists?
09:45 hemebond Do you get an error if the jid is in another store?
09:46 arif-ali joined #salt
09:47 pi3r My main concern is not really the error that always occurs when the jid is unknown
09:47 pi3r but when the jid is known, in one server I have got an exception and on the other I don't
09:47 hemebond And it fails every time?
09:47 pi3r Note that the error is not the same (as the one you have when you enter an unknown jid)
09:48 pi3r yes on dev it is failing everytime
09:48 hemebond The issue you raised is about an error when the jid doesn't exist.
09:49 hemebond Oh, that part of your issue is unrelated.
09:49 hemebond Are you not able to reduce your issue down to just the core information?
09:49 hemebond You have a lot of stuff and config in there that seems unrelated to the issue.
09:50 pi3r I will try to do that.
09:51 ronnix joined #salt
09:52 hemebond Is your dev and prod(?) running the same version of PostgreSQL?
09:54 pi3r Yes the same server, just different database
09:55 pi3r I have just removed a bunch of config (related to ldap authentication)
09:56 pi3r And the problem persists.
09:56 hemebond So they talk to exactly the same PostgreSQL server but the database name is different?
09:56 pi3r yes
09:56 hemebond Can you not remove that "How to reproduce this error" section? Seems unrelated.
09:57 pi3r the script to generate the salt table is also the same (everything is automated with puppet :-)
09:57 hemebond *spit*
09:57 pi3r yes will do
09:58 rburkholder joined #salt
09:58 hemebond So really the only part of the master config related is the master_job_cache and returner.pgjsonb.*?
09:59 hemebond Other that returner.pgjsonb.db the master config is identical, correct?
09:59 pi3r yes
09:59 hemebond Centos....
10:00 hemebond The version output really is identical?
10:00 hemebond What happens if you provide a port for the returner?
10:01 pi3r both with CentOS Linux release 7.3.1611 (Core)
10:01 pi3r Try that already. The INFO about using the default disappears but the error stays
10:02 hemebond Cool, can you update the issue with just that?
10:03 hemebond I don't know this returner so I don't know what might be noise and what might be related to the issue.
10:03 hemebond That's not to say I'll be able to help you with the issue. Just trying to get the issue trimmed down to only important info,.
10:04 pi3r The only difference I could find which seems unrelated is the fact that '/var/cache/salt/master/jobs' is empty in one machine and on the other it contains all folders
10:04 hemebond Have you confirmed that the database schemas are identical?
10:04 hemebond Oh, interesting.
10:05 hemebond What if you change the master_job_cache off, restart to let it create the dirs, then change it back?
10:05 hemebond The failing master is the one without the dirs, correct?
10:06 pi3r Yes
10:06 pi3r but on the other ones, these dirs haven't been updated for years so I thought that it would not be used anymore
10:07 pi3r when using an external master job cache, I would assume the folder would not be used anymore ?
10:08 hemebond That should be the case.
10:08 hemebond But perhaps there's an assumption being made somewhere in the code.
10:11 onlyanegg joined #salt
10:11 pi3r hum I have just put the master_job_cache to off and I have got a huge stack trace when I do  salt '*' --async test.ping
10:11 hemebond Off?
10:11 hemebond You didn't just comment it out?
10:12 pi3r so the error is [CRITICAL] Could not deserialize msgpack message.This often happens when trying to read a file not in binary mode
10:12 pi3r I will now comment this off instead of writing off ;-)
10:13 pi3r Ok better now
10:14 pi3r so at least when I comment 'master_job_cache: pgjsonb' salt-run jobs.lookup_jid works
10:15 hemebond Were the directories created?
10:15 candyman89 joined #salt
10:16 pi3r Yes. I will now uncomment and see it works
10:18 pi3r hum no the error is back when I unable `pgjsonb` again ...
10:20 pi3r The only other difference between these environments is the following: I have started to upgrade minions to Carbon in dev but not in staging
10:20 hemebond Oh.
10:20 pi3r I now that Carbon have some issues with RHEL6 minion
10:21 pi3r So the plan is to only upgrade to Carbon for RHEL 7 in staging
10:21 hemebond As long as the masters are the same version they should behave the same.
10:21 pi3r both masters are using Carbon
10:21 hemebond So does it work correctly now that the master has the directories?
10:21 pi3r no
10:22 hemebond Okay.
10:22 pi3r it is still failing in dev as soon as I put 'master_job_cache: pgjsonb' back
10:23 hemebond Can I recommend you update your issue to have less of the master config and update the error to take out the port stuff?
10:23 pi3r I guess I could try a job that only targets RHEL 7 minions
10:23 pi3r yes sure I will
10:23 hemebond That would be a good test.
10:29 dendazen joined #salt
10:30 Ricardo1000 joined #salt
10:31 pi3r no luck I have just try salt 'xxx.dev.srv.lan' --async test.ping (where xxx is CentOS 7 with a Carbon minion), take the jid, use it with 'salt-run jobs.lookup_jid' but it is still failing
10:33 pi3r whatever is the command I use with '--async' it fails with the same error (so it is not related to a specific job)
10:34 hemebond Does the error occur with jobs that aren't run with --async?
10:38 ronnix joined #salt
10:41 pi3r no it doesn't
10:41 hemebond Oh really? That needs to go into the issue too. That's very odd.
10:42 Kruge Anyone using salt-cloud to deploy onto vsphere? I'm having a hard time customising a template to put it into the right network
10:42 pi3r I might have misunderstood your question ;-) How do I get the jid from such job ?
10:43 hemebond --verbose
10:43 Kruge I have a conf set up which works in all respects apart from the network adapter.  If I customise the definition I'm using to put Network Adapter 1 onto a particular net, I get two nics in the VM instead of it reassigning nic1 to the intended network.
10:45 pi3r well sorry it does fail too in that case
10:45 hemebond Okay.
10:55 SarsTW joined #salt
10:58 Ricardo1000 joined #salt
11:09 Praematura joined #salt
11:15 amcorreia joined #salt
11:27 onlyanegg joined #salt
11:37 Rumbles joined #salt
11:39 Tanta joined #salt
11:42 ronnix joined #salt
11:44 lstor joined #salt
11:44 philpep joined #salt
11:45 philpep hi, can we schedule jobs (using "schedule" option in minion) that run in local mode, eg. that run even if master is not reachable
11:45 philpep ?
11:46 Tanta yes, there is a built-in scheduler called 'cron'
11:46 Tanta just use that
11:46 backslash joined #salt
11:46 philpep Tanta: thanks, I'll test this
11:47 Tanta https://docs.saltstack.com/en/latest/ref/states/all/salt.states.cron.html
11:47 Tanta you might find this useful
11:47 Tanta you can create crons with Salt, for the full circle
11:48 Tanta crond is part of any standard unix or linux system, it is not a Salt component
11:48 hemebond The cron state module just write cron entries.
11:49 hemebond It doesn't use Salts scheduler.
11:49 philpep so, you where not talking about https://docs.saltstack.com/en/latest/topics/jobs/#cron-like-schedule ?
11:50 Tanta oh that too
11:50 Tanta I'm sure Salt has poorly implemented cron somehow
11:50 Tanta lots of software does that, but it's always worse than the native scheduler
11:51 backslash just because that's true for systemd doesn't mean it's true for salt, too.
11:52 philpep using the schedule option can avoid create a new process and run into salt initialisation code, that't why I'm looking at it over the unix cron
11:52 swa joined #salt
11:53 Tanta I'm sure my decade of experience pales in comparison to your 5 minutes of googling
11:53 swa hi, is it me or format of inotify beacon events changed in 2016.11.3?
11:58 hemebond philpep: You want it to execute a Salt command, yeah?
11:58 it_dude joined #salt
11:58 philpep hemebond: yeah
11:59 hemebond The salt scheduler is a good choice then.
12:04 hemebond philpep: What do you mean by "local mode"?
12:09 philpep hemebond: same to salt-call --local (don't attempt to connect to master and don't send jobs events)
12:09 numkem joined #salt
12:10 dendazen joined #salt
12:14 DammitJim joined #salt
12:16 _JZ_ joined #salt
12:17 muxdaemon joined #salt
12:19 hemebond Oh. You can only use the scheduler if you have salt-minion service running.
12:19 hemebond I have no idea if you can have salt-minion running without a master; I've never tried.
12:25 Xenophon1 you can use salt-minion without a master
12:25 Xenophon1 search the docs for "masterless
12:25 Xenophon1 ""
12:25 XenophonF joined #salt
12:26 hemebond philpep: ^^ there you go
12:28 philpep hemebond: I've have salt-minion running, I don't run masterless, but connection with master is not reliable (it can be down for hours) and for this particular job I don't care of the connection with master
12:28 XenophonF iirc the job scheduler works the same but RTFM b/c I don't use masterless
12:28 XenophonF won't the job run regardless of the connection to the master?
12:28 philpep with my setup, if salt master is down, the job don't run (block on SaltReqTimeoutError retries)
12:28 squishypebble joined #salt
12:29 XenophonF ah
12:29 philpep that's why I want to run my "schedule" job in local mode
12:29 XenophonF what does the job do?
12:29 philpep collecting metrics
12:29 XenophonF for mine or grains or something?
12:30 hemebond You can tell scheduler to not return the job to the master. That's what I use for some monitoring stuff.
12:30 hemebond But I don't know if that would allow it to work without the master.
12:30 evle joined #salt
12:30 XenophonF i mean if you're running a job to collect stuff and send it to the master, that kind of assumes the master is up
12:30 XenophonF but if you're just collecting data locally or something, wouldn't a cron/windows task scheduler job be a better solution?
12:31 hemebond (mine return to an icinga2 server)
12:32 philpep hemebond: to not return the job you mean the "jid_include" option (I'll try this) ?
12:33 hemebond return_job: False
12:35 philpep hemebond: thx
12:48 concerti__ joined #salt
12:51 nick123 joined #salt
12:53 hemebond left #salt
12:58 bastiandg joined #salt
13:02 bastiandg joined #salt
13:08 bastiandg joined #salt
13:22 racooper joined #salt
13:23 nick123 joined #salt
13:24 Rumbles joined #salt
13:28 onlyanegg joined #salt
13:33 squishypebble joined #salt
13:35 CeBe joined #salt
13:37 ronnix joined #salt
13:47 Cottser joined #salt
13:48 EvaSDK joined #salt
13:48 EvaSDK hello there
13:51 Tanta hi EvaSDK
13:53 EvaSDK is there someone with write access to formulas, I have some pending PR that I would like to clear up (https://github.com/saltstack-formulas/salt-formula/pull/305, and https://github.com/saltstack-formulas/sensu-formula/pull/38, 39 and 40)
13:53 saltstackbot [#38][OPEN] Fix Debian defaults | Fix rendering error in `repos_map.jinja` and add missing installation of `apt-transport-https` to deal with the new default from #37.
13:54 Brew joined #salt
13:56 Hybrid joined #salt
13:57 jdipierro joined #salt
13:58 tapoxi joined #salt
14:05 mpanetta joined #salt
14:19 codeape joined #salt
14:26 XenophonF anyone else getting `ValueError: SID conversion failed` when using the file module on Windows?
14:27 XenophonF I'm getting this error on a user.present state
14:27 cyborg-one joined #salt
14:29 jas02 joined #salt
14:31 toastedpenguin joined #salt
14:33 codeape joined #salt
14:38 elektrix joined #salt
14:38 jas02 joined #salt
14:51 PatrolDoom joined #salt
14:51 PatrolDoom joined #salt
14:55 seanz joined #salt
14:55 seanz robawt: Greetings!
15:01 onlyanegg joined #salt
15:03 Ryan_Lane_ joined #salt
15:03 dragon788_ joined #salt
15:03 doriftoshoes__ joined #salt
15:03 dh___ joined #salt
15:03 hillna_ joined #salt
15:03 djural_ joined #salt
15:04 phtes_ joined #salt
15:04 SteamWells_ joined #salt
15:04 WKNiGHT_ joined #salt
15:04 zifnab_ joined #salt
15:05 hacks_ joined #salt
15:05 Nightcinder_ joined #salt
15:05 descrepes_ joined #salt
15:05 darix- joined #salt
15:05 Corey_ joined #salt
15:05 dev_tea_ joined #salt
15:06 wybczu_ joined #salt
15:06 bergei_ joined #salt
15:06 Ludo_ joined #salt
15:06 hlub_ joined #salt
15:07 n1x0n joined #salt
15:07 arapaho_ joined #salt
15:07 doglike joined #salt
15:07 doglike joined #salt
15:07 funabashi joined #salt
15:07 jholtom joined #salt
15:09 stupidni` joined #salt
15:09 dankolbr- joined #salt
15:09 ajv joined #salt
15:09 setkeh joined #salt
15:09 setkeh joined #salt
15:09 aerbax_ joined #salt
15:09 eprice_ joined #salt
15:09 jijiki joined #salt
15:09 tehsu joined #salt
15:10 KingJ_ joined #salt
15:10 elliot1 joined #salt
15:10 NeoXiD_ joined #salt
15:11 al joined #salt
15:11 Guest98080 joined #salt
15:11 bigjazzsound joined #salt
15:11 McNinja joined #salt
15:11 rschneberger joined #salt
15:12 riftman joined #salt
15:13 sjorge joined #salt
15:13 tbrb joined #salt
15:13 skrobul joined #salt
15:14 _0xm68 joined #salt
15:34 dezertol joined #salt
15:40 Satyajit joined #salt
15:40 Praematura joined #salt
15:41 tom29739 joined #salt
15:48 tkojames joined #salt
15:49 XenophonF I'm trying to use an identity file with git.latest, but I'm getting `KeyError: 'ssh.key_is_encrypted'`
15:49 XenophonF it really is not encrypted - I promise
15:56 XenophonF oh
15:56 XenophonF that means the ssh module isn't loaded
15:57 XenophonF so how do I convince git.latest to bypass that check...?
15:58 XenophonF oh, i can run the state using salt-call
16:06 hillna joined #salt
16:10 woodtablet joined #salt
16:11 aldevar left #salt
16:16 djgerm joined #salt
16:18 robawt oh man!  seanz hello!
16:24 tiwula joined #salt
16:24 jdipierro joined #salt
16:26 sarcasticadmin joined #salt
16:32 tkojames_ joined #salt
16:32 lasseknudsen2 joined #salt
16:32 monjwf joined #salt
16:32 jimklo joined #salt
16:32 tapoxi joined #salt
16:32 euidzero_ joined #salt
16:32 dragon788_ joined #salt
16:32 phtes_ joined #salt
16:32 Psy0rz_ joined #salt
16:32 vexati0n joined #salt
16:32 ashmckenzie joined #salt
16:32 Yoda-BZH joined #salt
16:32 Yoda-BZH joined #salt
16:32 pfallenop joined #salt
16:32 Karunamon joined #salt
16:33 swa_work joined #salt
16:33 inetpro joined #salt
16:33 NightMonkey joined #salt
16:33 patrek joined #salt
16:34 Hybrid joined #salt
16:35 wendall911 joined #salt
16:35 doriftoshoes_ joined #salt
16:38 Ryan_Lane joined #salt
16:41 wattwood joined #salt
16:42 wattwood Hello, hello.   I have a pillar that is a list of items.  e.g.:  apache_vhosts: \n -www.server1.com \n -www.server2.com.    php_modules: \n -php-curl \n -php_md5.    If I set these are pillars, what process do I use to loop through them and install them in a state/formula?
16:43 pipps joined #salt
16:44 SaucyElf joined #salt
16:45 wattwood disregard, found an example
16:48 brakkisath joined #salt
16:51 bltmiller joined #salt
16:52 bltmiller hey I need a way to list node groups defined in Salt. any API call that can get me that?
16:53 Kelsar joined #salt
16:54 bltmiller assuming the only way to do it is by exposing my master's config with `pillar_opts = True`
16:57 czchen joined #salt
16:58 Trauma joined #salt
16:59 st8less joined #salt
17:00 copelco joined #salt
17:00 jdipierro joined #salt
17:00 ToeSnacks joined #salt
17:01 munhitsu_ joined #salt
17:03 brakkisa_ joined #salt
17:07 brakkisath joined #salt
17:08 pipps joined #salt
17:12 aldevar joined #salt
17:15 XenophonF wattwood: don't loop
17:16 XenophonF use the pkgs kwarg to pkg.installed plus the |yaml filter
17:16 ping0ra joined #salt
17:16 toastedpenguin trying to setup a state that deletes a directory the first time a minion is deployed so AWS s3 sync starts out consistent, the state creates a custom grain that is used to evaluate if the directory needs to be removed the next time the state is run: https://pastebin.com/TDscy59f
17:16 toastedpenguin however it seems the directory is deleted everytime the state is run even with the grain set
17:16 XenophonF wattwood: https://github.com/irtnog/salt-states/blob/production/accounting/init.sls#L5
17:17 Xenophon1 joined #salt
17:19 Praematura joined #salt
17:21 Xenophon1 joined #salt
17:21 wattwood Thank you XenophonF
17:23 Xenophon1 hemebond, philpep: I suddenly have a use for both schedule.present and return_job: False. coincidence?
17:23 * Xenophon1 queues the X-Files theme.
17:30 ChubYann joined #salt
17:33 lkannan joined #salt
17:49 censorshipwreck joined #salt
18:00 cscf So I have a PXE server using pxelinux, where you can symlink 01-{{ MAC }} to a pxe boot config.  Now, I could for-loop file.manage for each host I want to reinstall, but what's a good way to remove old links?  I suppose I could always delete them all every run, then replace, but that's messy.
18:00 s_kunk joined #salt
18:03 artanicus joined #salt
18:04 artanicus whois
18:04 cyborg-one joined #salt
18:04 artanicus oops / key is rusty
18:04 * gtmanfred gives artanicus some WD-40
18:05 artanicus Thanks man, I'm gonna need it for my brain if I'm ever to get through this gpg stuff :D
18:06 gtmanfred :)
18:06 gtmanfred good luck
18:06 brakkisa_ joined #salt
18:07 artanicus gtmanfred: any pointers on salt.utils.flopen peculiarities? it seems to be trying to IO on a closed handle
18:08 ekkelett gtmanfred: are most functions in a module used from a state recommended to be public and thus used with a loader? :)
18:08 gtmanfred ekkelett: yeah
18:08 cyborg-one joined #salt
18:08 gtmanfred if you are using it in the state, make it public and use it through the loader
18:09 pipps joined #salt
18:12 Joe630 joined #salt
18:12 cmarzullo cscf: ha! dealing with the same thing. Sadly just leaving it messy for now.
18:12 gtmanfred artanicus: i have never used flopen, only fopen?  No real pointers, i just know that I am supposed to use it
18:13 gtmanfred artanicus: s0undt3ch might be able to help.
18:13 cmarzullo coworker has an orchestartion piece setup though. So webhooks trigger those pxe menu files to get created and removed.
18:13 gtmanfred artanicus: but he is only sometimes on
18:14 Joe630 I have a question I hope isn't too dumb - I suplly a salt runner with a hostname and I need its IP later to do some DNS works - what's the right way to grab the IP inside the runner?  I can use dnsutil.A to grab it, I'm just not quite sure how get from salt ns1 dnsutil.A foo to "ip = 10.0.0.0"
18:14 Joe630 *supply
18:15 artanicus gtmanfred: cool, I'll bug him if I can't figure it out. Feeling dumb when I'm doing exactly the same thing as the old code but now it fails in salt internals.
18:15 gtmanfred don't feel dumb, it happens, there is a lot of magic
18:15 artanicus hah, as with any sufficiently large codebase
18:16 gtmanfred i felt super dumb yesterday when I was trying to get salt-master and salt-minion installed on windows to do some testing... that was rough
18:17 jdipierro joined #salt
18:19 artanicus gtmanfred: I noticed this morning that modules/file is only for posix and there's a separate one for windows.. haven't dared to look at it yet. But I guess it would be fair to give them pgp signatures as well
18:25 cscf cmarzullo, anything useful you could share?
18:26 cscf cmarzullo, I've just gotten the PXE setup all into Salt (afaik) but the constant machine-toggling is not.
18:28 debian1121 joined #salt
18:29 jas02 joined #salt
18:30 rebbdohr joined #salt
18:31 Joe630 man I feel dumb right now
18:31 Joe630 i thoguht I knew python and salt, but i do not.
18:32 pcn Joe630: in the runner you should be able to query the minion's grains for that IP
18:32 Joe630 the minion isn't up - this is part of a removal process
18:33 Joe630 also, it may not even be a minion'd host - this is just using salt to delete some DNS records
18:33 gtmanfred artanicus: i do not know.  idk how gpg/pgp works on windowas
18:34 Joe630 but I need a hostname and an IP do kill the PTR records, too
18:34 gtmanfred if you can figure out how to do it in pure python i would say yes?
18:34 gtmanfred Joe630: has the minion connected to the master before?
18:34 pcn Oh, then if you have a dns name for it, then could you  install dnspython to resolve the name (or you could just use socket.gethostbyname I think )
18:34 gtmanfred oh, nevermind
18:35 Joe630 dnspython is looking like the right way to go, it just complicates thing a little
18:35 Joe630 i have salt.modules.dnsutl.A, I just don't know how to incorporate it's out put into my script
18:37 pcn How are you trying to call it?
18:38 Joe630 i'm not sure how
18:38 s_kunk joined #salt
18:38 Joe630 i think thats my question
18:38 Joe630 how do i call a module and assign it's output to a variable inside a runner
18:38 Joe630 and soeery for my its it's
18:38 mikecmpbll joined #salt
18:38 Joe630 and my abhorrent typing, slack corrections have killed my immutable lines
18:40 jas02 joined #salt
18:41 pcn Joe630:
18:41 artanicus gtmanfred: I'm actually a bit surprised, python-gnupg supports Windows. So sounds like Windows peeps will get it as well.
18:41 pcn In [5]: l.cmd('saltmaster', 'dnsutil.A', ['google.com'])
18:41 pcn Out[5]: {'saltmaster': ['172.217.7.238']}
18:42 pcn Joe630: let me see if I can get that into a gist quickly
18:46 pcn Joe630: https://gist.github.com/pcn/01662767119f195d65aa2026ae23bbd5
18:46 Joe630 thank you, that was the direction I was heading
18:46 Joe630 that really helps
18:47 Joe630 it felt wrong when I wrote it, but your doesn't, thanks
18:47 pcn No problem.  It took me forever to realize that the arguments had to be a list even when the arity of the invoked module seemed to be 1
18:48 bltmiller joined #salt
18:48 pcn I honestly think that the python API would be much better if the argument and return types were documented for cases like this. E.g. modules, returners.
18:48 pcn The conventions are opaque/confusing when you just start out.  I'm still starting out and feel that way a lot.
18:51 jas02 joined #salt
18:52 pcn The payoff for the difficulty is definitely worth it
18:58 cmarzullo cscf: sadly not too much to share that isn't hacky.
18:58 cmarzullo We have our formula build our a pxe file for all classes of a machine. Then there's manual intervention to do the symlink from the mac add.
18:59 cmarzullo We are working on an external pillar to hit our inventory and build out all the files so if the box is in 'provision' mode it'll get the installer. And if not it only gets the boot from local disk option
18:59 cmarzullo BUt that's still in development.
18:59 cmarzullo We are trying to make it work like foreman.
19:05 llua joined #salt
19:06 ping0ra joined #salt
19:07 cscf cmarzullo, yeah, we are hoping for future inventory integration as well
19:07 GMAzrael_ joined #salt
19:12 pipps joined #salt
19:14 pipps99 joined #salt
19:16 nikdatrix joined #salt
19:19 mpanetta joined #salt
19:21 llua joined #salt
19:21 Praematura joined #salt
19:24 DammitJim joined #salt
19:27 sjorge joined #salt
19:30 ry joined #salt
19:33 lane_ joined #salt
19:36 RabidCicada joined #salt
19:39 mpanetta joined #salt
19:42 jimklo joined #salt
19:43 DammitJim joined #salt
19:50 ekkelett gtmanfred: ugh, reproduced the bug. :(
19:53 llua joined #salt
19:57 daxroc joined #salt
20:01 cyteen joined #salt
20:02 pipps joined #salt
20:05 rich_ joined #salt
20:11 eduardo joined #salt
20:11 bltmiller joined #salt
20:13 pipps joined #salt
20:13 Xenophon1 joined #salt
20:15 rem5 joined #salt
20:36 jas02 joined #salt
20:37 cscf joined #salt
20:45 jas02 joined #salt
20:48 st8less joined #salt
20:48 mpanetta joined #salt
20:50 phtes joined #salt
20:51 speedlight joined #salt
20:52 theblazehen joined #salt
20:53 sysadminboy1 joined #salt
20:58 st8less joined #salt
20:59 tongpu joined #salt
21:05 pbandark joined #salt
21:15 pipps joined #salt
21:17 nikdatrix joined #salt
21:20 druonysus joined #salt
21:21 adelcast joined #salt
21:23 pi3r joined #salt
21:25 pi3r I have a minion that misbehaves somehow. It keeps asking me for accepting the key. If I do, I won't be able to ping or do anything to know what this minion is.
21:25 pi3r I keep receiving "Authentication accepted from TPLW2K8R2" for that minion every sec in the log.
21:25 pi3r "
21:26 pipps joined #salt
21:26 pi3r As you guess it is a windows machine. How can I get rid of it ?
21:27 pi3r I know that installing the minion on that machine was an error. But is it possible to get the IP of TPLW2K8R2 ?
21:27 pi3r Somehow we can't find out what this machine is ;-)
21:28 pi3r So my question is: can I blacklist that minion_id or can I at least get the IP of it ?
21:32 pipps joined #salt
21:35 eduardo joined #salt
21:41 eduardo does anyone have a experience with azure deploy?
21:42 eduardo i followed these steps: https://docs.saltstack.com/en/latest/topics/cloud/azure.html but got error
21:43 eduardo actuallly multiple errors : <Error xmlns="http://schemas.microsoft.com/windowsazure" xmlns:i="http://www.w3.org/2001/XMLSchema-instance"><Code>ResourceNotFound</Code><Message>Role BackupManagerWorkerRole was not found in the currently deployed service.</Message></Error>
21:44 eduardo and this too
21:45 eduardo The Virtual Machine could not be created. If you are using an already existing Cloud Service, make sure you set up the `port` variable corresponding to the SSH port exists and that the port number is not already in use. The following exception was thrown when trying to run the initial deployment:  Not Found <Error xmlns="http://schemas.microsoft.com/windowsazure" xmlns:i="http://www.w3.org/2001/XMLSchema-instance"><Code>ResourceNotFou
21:45 slupers joined #salt
21:45 eduardo connection to azure is available, i can get list of images and etc.
21:46 eduardo Salt Version:             Salt: 2016.11.3   Dependency Versions:  Apache Libcloud: Not Installed             cffi: 1.10.0         cherrypy: Not Installed         dateutil: 2.6.0            gitdb: 0.5.4        gitpython: 0.3.2 RC1            ioflo: Not Installed           Jinja2: 2.7.3          libgit2: Not Installed          libnacl: Not Installed         M2Crypto: Not Installed             Mako: Not Installed     msgpack-pure: Not In
21:47 eduardo azure-cli (2.0.2)
21:51 pipps joined #salt
21:53 Praematura joined #salt
21:58 eduardo anyone?
22:02 pipps99 joined #salt
22:07 nick123 joined #salt
22:17 chitown joined #salt
22:18 DEger joined #salt
22:26 Xenophon1 eduardo: you'll get more help if you post your configs plus detailed error messages (minus any sensitive data, of course)
22:26 Xenophon1 use gist.github.com or something
22:26 pipps joined #salt
22:27 Tanta that's what professional support contracts are for
22:28 Tanta and training classes
22:28 XenophonF yeah well this is a community support forum, so we'll help too
22:39 pipps joined #salt
22:48 jas02 joined #salt
22:57 jimklo joined #salt
22:59 jimklo joined #salt
23:03 antpa joined #salt
23:08 nick123 joined #salt
23:18 nikdatrix joined #salt
23:30 zulutango joined #salt
23:48 Rubin joined #salt
23:59 jdipierro joined #salt

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