Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2018-01-08

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

All times shown according to UTC.

Time Nick Message
00:00 saltslackbridge joined #salt
00:08 RandyT_ joined #salt
00:08 RandyT_ joined #salt
00:10 hammer065 joined #salt
00:11 Deliant joined #salt
00:11 nledez joined #salt
00:15 Processus42 joined #salt
00:21 masber joined #salt
00:24 APLU joined #salt
00:36 tflo joined #salt
00:36 freelock joined #salt
00:36 rtr63gdh[m] joined #salt
00:36 jerrykan[m] joined #salt
00:36 hackel joined #salt
00:36 sxar joined #salt
00:36 toofoo[m] joined #salt
00:36 benjiale[m] joined #salt
00:36 viq[m] joined #salt
00:36 kbaikov[m] joined #salt
00:36 aboe[m] joined #salt
00:36 gomerus[m] joined #salt
00:36 benasse joined #salt
00:36 ThomasJ|m joined #salt
00:36 theblazehen joined #salt
00:37 fujexo[m] joined #salt
00:37 glock69[m] joined #salt
00:37 gomerus[m]1 joined #salt
00:37 Tenyun[m] joined #salt
01:10 masber joined #salt
01:22 nethershaw joined #salt
01:24 Zachary_DuBois joined #salt
01:49 cyteen joined #salt
01:56 kojiro joined #salt
02:15 jhujhiti "The minion function caused an exception: AttributeError: cffi library '_openssl' has no function, constant or global variable named '__loader__'" what might cause this?
02:15 jhujhiti highstate runs quickly fail with that error. state.sls runs don't
02:16 kojiro joined #salt
02:28 zerocoolback joined #salt
02:36 Larry joined #salt
02:42 Larry left #salt
03:03 XenophonF something's seriously borked
03:03 XenophonF A/S/L?
03:03 XenophonF er, I mean, O/S and output of "salt -V"?
03:03 XenophonF post it to ix.io or gist or sommat
03:04 ilbot3 joined #salt
03:04 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, and https://saltstackcommunity.herokuapp.com (for slack) <+> We are volunteers and may not have immediate answers
03:05 jhujhiti nope, can't reproduce it. very strange. i have a script that does the vast majority of the jail setup so i know my multiple previous tests were identical
03:05 XenophonF gotcha
03:05 XenophonF it's a make.conf setting
03:06 jhujhiti ah, duh, that's not a ports option :p
03:06 saltslackbridge joined #salt
03:06 XenophonF well kinda it's in ports/Mk/bsd.default-versions.mk
03:07 jhujhiti hmm, i did one thing differently: i didn't use the rc script to start salt-minion this time...
03:07 XenophonF the new school way to do it is to set DEFAULT_VERSIONS
03:07 XenophonF I should probably update my config.
03:07 tiwula joined #salt
03:08 XenophonF hm - that shouldn't matter
03:08 jhujhiti i'd certainly hope not
03:08 jhujhiti argh, i'm also having issues with minions hanging and never returning, and then they need to be signal 9ed to die, and this new minion just did it
03:10 jhujhiti [INFO    ] Executing command ['pkg', 'install', '-yf', 'postfix'] in directory '/root' <- last log line before it hung
03:11 XenophonF oh I know what that is
03:11 XenophonF press enter
03:11 XenophonF I really, really hate the maintainer of mail/postfix for this
03:12 XenophonF there's a prompt in the post-install script for postfix that asks you if you want to modify mailer.conf to point at postfix
03:12 jhujhiti ah, yes, this isn't the same hang. so many issues :/
03:12 XenophonF it doesn't always detect when it's being called non-interactively right
03:12 jhujhiti we could patch that :p
03:13 jhujhiti the other hang i keep seeing happens wiht minions that should be idle aside from the occasional mine update
03:16 XenophonF re: postfix https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=213557 and https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=213132 (I filed the latter)
03:17 XenophonF we have some version differences, you and me - http://ix.io/DCk
03:17 XenophonF oh
03:17 XenophonF you're running against Python 3
03:18 jhujhiti thanks for those bugs
03:19 jhujhiti yep, python 3, it must be related somehow to the problems i'm having
03:19 jhujhiti or 11.1
03:19 XenophonF could be that or cffi
03:20 jhujhiti minions have been stable for a long time, and then i update the host from 10.3 to 11.1 and the jails start exhibiting this hanging behavior, even though i haven't touched the jail. and yes i do have compat10x installed
03:20 XenophonF yeah that should have been fine
03:21 jhujhiti the hang happens on 10.3 jails, on 11.1 jails the process seems to actually die but i haven't caught it happening yet
03:24 zerocoolback joined #salt
03:26 jhujhiti found one: "sockstat: struct xtcpcb size mismatch" / "sockstat: struct xinpcb size mismatch" / "[INFO    ] Connection to master telesto-salt.adjectivism.org lost"
03:31 XenophonF have you rebuilt your packages against 11.1?
03:33 jhujhiti yep, this looks like a freebsd problem after all, not a salt one. all calls to the sockstat binary fail like that. it's linked against only libc and there's no libc in compat10x
03:36 jhujhiti although it's probably a syscall that's failing
03:47 zerocoolback joined #salt
03:49 JPT joined #salt
04:10 cgiroua joined #salt
04:17 shiranaihito joined #salt
04:29 zerocoolback joined #salt
04:49 zerocoolback joined #salt
04:54 JulianPalmer joined #salt
04:55 JulianPalmer left #salt
05:02 pcdummy joined #salt
05:02 pcdummy joined #salt
05:05 zerocoolback joined #salt
05:17 zerocoolback joined #salt
05:21 felskrone joined #salt
05:40 zerocoolback joined #salt
05:42 MTecknology iggy: unlike me, you are correct! :)
05:43 tzero joined #salt
06:07 zerocoolback joined #salt
07:12 bildz joined #salt
07:13 NeoXiD joined #salt
07:14 munhitsu__ joined #salt
07:14 dimeshake joined #salt
07:14 benasse joined #salt
07:16 sayyid9000 joined #salt
07:18 nledez joined #salt
07:18 nledez joined #salt
07:18 whyzgeek joined #salt
07:23 JawnAuz__ joined #salt
07:25 JawnAuz joined #salt
07:40 yuhl joined #salt
07:41 RandyT joined #salt
07:41 RandyT joined #salt
07:46 zerocoolback joined #salt
07:47 pbandark joined #salt
07:48 zerocoolback joined #salt
07:48 Tucky joined #salt
07:52 jas02 joined #salt
08:00 manji joined #salt
08:03 kshlm joined #salt
08:03 impi joined #salt
08:05 pbandark joined #salt
08:15 Hybrid joined #salt
08:18 XenophonF joined #salt
08:19 aviau joined #salt
08:24 darioleidi joined #salt
08:25 aldevar joined #salt
08:26 colttt joined #salt
08:26 zerocoolback joined #salt
08:28 jhauser joined #salt
08:28 chowmeined joined #salt
08:29 jrenner joined #salt
08:32 ech0_42 joined #salt
08:43 sh123124213 joined #salt
08:44 zerocoolback joined #salt
08:45 zerocoolback joined #salt
08:46 zerocoolback joined #salt
08:46 jbulger joined #salt
08:46 felskrone left #salt
08:59 mikecmpbll joined #salt
09:00 xet7 joined #salt
09:11 q1x joined #salt
09:23 lionel_ left #salt
09:25 yuhl joined #salt
09:30 Mattch joined #salt
09:35 xet7 joined #salt
09:49 oida joined #salt
10:01 mikecmpbll joined #salt
10:10 scooby2 joined #salt
10:13 Hybrid1 joined #salt
10:13 mikecmpb_ joined #salt
10:34 babilen Hmm, I can't find it, but I remember seeing formula(s) that allow users to pass in options to pkg.installed dynamically from pillars
10:34 babilen Does that ring a bell?
10:35 xet7 joined #salt
10:35 cofeineSunshine joined #salt
10:36 xet7 joined #salt
10:50 beardo joined #salt
10:51 simondodsley joined #salt
10:55 Hybrid joined #salt
10:57 keith4 joined #salt
11:03 hax404_ joined #salt
11:10 xet7 joined #salt
11:57 jdipierro joined #salt
12:01 chesty joined #salt
12:07 zerocoolback joined #salt
12:11 gmoro joined #salt
12:15 pcn dynamically?
12:15 pcn Can you clarify what you mean?
12:16 Whissi joined #salt
12:25 tongpu joined #salt
12:26 yidhra joined #salt
12:27 zerocoolback joined #salt
12:40 XenophonF jhujhiti: that should be something you can fix by recompiling salt (et al)
12:53 XenophonF anyone have a formula for Moodle they'd like to share?
12:54 jdipierro joined #salt
13:16 Nahual joined #salt
13:25 mikecmpbll joined #salt
13:29 jdipierro joined #salt
13:30 babilen I'd like to pass arbitrary options to a state via pillars - Something along the lines of http://paste.debian.net/1004007/
13:34 babilen Meltdown seems to have affected my brain :)
13:34 yidhra joined #salt
13:39 mchlumsky joined #salt
13:43 XenophonF babilen: I drive my AWS config states entirely from Pillar data.
13:44 XenophonF https://github.com/irtnog/salt-states/tree/development/aws
13:44 XenophonF it's kind of over-engineered but maybe it will inspire you?
13:45 XenophonF unfortunately, I can't think of a way to do what you showed without a for loop that generates the relevant YAML
13:45 yujunz joined #salt
13:47 XenophonF but you could do that pretty easily with something like {%- for settings in salt.pillar.get('foo_pkg_options')|dictsort %}
13:47 babilen I had the hope that a simple conversion to yaml/json/... would work, but could only come up with an iteration
13:47 babilen I guess, I'll do that
13:49 XenophonF yeah there isn't a pretty way to append a Jinja list to a YAML list
13:49 XenophonF there isn't a pretty way to concatenate two YAML lists either
13:50 XenophonF it's kind of a shame that states take lists of arguments instead of dictionaries, although I kind of understand why that is
13:51 babilen Thank you for the confirmation!
13:53 Morrolan joined #salt
13:55 major joined #salt
14:01 cgiroua joined #salt
14:07 kojiro joined #salt
14:30 gh34 joined #salt
14:36 om2 joined #salt
14:39 pbandark1 joined #salt
14:41 yidhra joined #salt
14:46 edrocks joined #salt
14:47 racooper joined #salt
14:54 ouemt joined #salt
14:55 nixjdm joined #salt
15:07 cgiroua joined #salt
15:16 Whissi joined #salt
15:26 WoJ joined #salt
15:32 babilen XenophonF: Ended up using http://paste.debian.net/1004044/
15:36 yuhl joined #salt
15:40 yujunz joined #salt
15:52 jas02 joined #salt
15:53 mianosm joined #salt
15:59 keith4 joined #salt
16:03 AdamSewell joined #salt
16:04 yujunz joined #salt
16:11 AdamSewell I'm trying to remotely remove the salt-minion package on a windows minion, using "pkg.remove 'salt-minion'" but salt is giving me an error saying it can't find the package. if i list the installed packages, salt-minion is not in the list. I noticed that the package has changed in the ng repo as well to py2 or py3 but neither of those work either. Does anyone have an idea of what's going on?
16:11 sh123124213 can I evaluate jinja on the master ? or how do I manage a file that needs the output from a command from the master ?
16:12 DanyC joined #salt
16:13 Neighbour sh123124213: Yes, but only when done from orchestration scripts...It would be a security breach if a random minion could execute states on the master and get its results
16:14 xet7 joined #salt
16:17 yujunz-zte joined #salt
16:22 heaje joined #salt
16:23 overyander joined #salt
16:29 jeffspeff joined #salt
16:41 lordcirth_work joined #salt
16:57 ciastek joined #salt
16:59 ciastek Are events lost, when salt master is restarted? I'd like to move from running highstate manually to more event-based approach, but can't find an answer how and when orchestration can fail and how to prevent it.
17:12 CrummyGummy joined #salt
17:17 yujunz joined #salt
17:17 sjorge joined #salt
17:19 DammitJim joined #salt
17:21 tiwula joined #salt
17:23 gh34 joined #salt
17:24 schemanic joined #salt
17:24 saltslackbridge <chris.ruscio> Trying to run Salt-master from within docker. The minion can connect, salt master accepts the key, see debug messages on both master and minion logs due to ping_interval on the minion ... but nothing happens when the master issues a command (test.ping on the master -> Minion did not return. [No response]) - Any thoughts on what might be missing?
17:25 schemanic joined #salt
17:26 schemanic Hi. on 2017.7.2 master/minion I have a minion returning with Minion did not return. [No response]
17:26 schemanic after calling highstate.
17:27 schemanic Where should I begin troubleshooting this?
17:30 yujunz-zte joined #salt
17:32 schemanic Can anyone tell me where I can find out what the minion thinks happened last?
17:32 schemanic I dont see a /var/log/salt/minion
17:34 whytewolf check if the minion started
17:35 saltslackbridge <chris.ruscio> or manually start the minion with debug logging (https://docs.saltstack.com/en/latest/topics/troubleshooting/minion.html)
17:35 onlyanegg joined #salt
17:41 schemanic whytewolf yes the minion is started
17:42 schemanic whytewolf, I'm able to run some states and get return output but others just get the no response
17:44 pipps joined #salt
17:44 pipps joined #salt
17:48 yujunz joined #salt
17:48 aldevar left #salt
17:57 edrocks joined #salt
17:58 sjorge joined #salt
17:59 saltslackbridge <chris.ruscio> cd
17:59 tflo joined #salt
18:03 Edgan joined #salt
18:08 yujunz joined #salt
18:10 dacoda joined #salt
18:12 dacoda Hello!
18:12 dacoda Sometimes when I'm running a state.apply on a minion, it fails to return the job status back to the master
18:12 dacoda And it says so in the systemd logs for salt
18:12 dacoda I was wondering if there's anyway to either retry sending the job results back to the master
18:12 dacoda Or if there is a way to view the job result locally on the minion
18:13 dacoda Any help you guys might be able to offer would be greatly appreciated :)
18:14 tflo joined #salt
18:14 om2 joined #salt
18:21 schemanic dacoda - I'm having a similar issue. I don't really know where to start aside from debug mode
18:22 dacoda Yes, I thought there might be something in /var/cache/salt/proc/ or somewhere around there but I'm having trouble finding things
18:23 dacoda Are you going over a weak connection? I believe on our end we're having problems due to a weak cellular signal
18:32 saltslackbridge <chris.ruscio> fwiw I had issues where some fraction of random minions would fail with every run ... after adding "ping_interval: 2" to the minion config everything got better. Seems minions would get disconnected and (eventually) reconnect
18:41 yujunz joined #salt
18:43 schemanic_ joined #salt
18:45 schemanic_ joined #salt
18:53 edrocks joined #salt
18:54 schemanic joined #salt
18:58 fatal_exception joined #salt
19:03 pipps joined #salt
19:09 edrocks joined #salt
19:19 oida joined #salt
19:20 yujunz joined #salt
19:21 edrocks joined #salt
19:32 chesty joined #salt
19:48 mcqueenorama joined #salt
19:50 pipps joined #salt
19:50 aldevar joined #salt
19:50 pipps joined #salt
19:51 fatal_exception joined #salt
20:03 pipps joined #salt
20:06 schemanic joined #salt
20:07 Hybrid joined #salt
20:11 pipps joined #salt
20:16 edrocks joined #salt
20:26 Hybrid joined #salt
20:29 Kax joined #salt
20:34 gh34 joined #salt
20:40 yuhl joined #salt
20:44 jpsharp joined #salt
20:46 jpsharp Hi folks.  I'm getting started in setting up Salt, but the first issue I'm having is that Salt is detecting my Gentoo boxes as Ubuntu boxes.  I'm not enough of a Python guy to chase it through the code yet.
20:56 pipps joined #salt
20:57 saltslackbridge <chris.ruscio> Is it possible to run Salt-Master listening on 4505/4506 and map that to 4515/4516 externally (e.g. with IPTABLES)
21:00 pipps99 joined #salt
21:15 lordcirth_work jpsharp, do you have a file /etc/lsb_release?
21:28 yujunz-zte joined #salt
21:29 lordcirth_work jpsharp, if not, install the package 'lsb-release' and try again
21:29 nonsenso joined #salt
21:56 pipps joined #salt
21:58 FuzzyVeg joined #salt
22:01 pipps joined #salt
22:04 yujunz joined #salt
22:04 onlyanegg joined #salt
22:13 onlyanegg joined #salt
22:19 jpsharp lordcirth_work: I do have that file and it has Ubuntu in it rather than Gentoo
22:21 jpsharp So I gotta figure out what kind of Gentoo stupid that's from.  Thanks.
22:30 cyteen joined #salt
22:40 pipps joined #salt
22:40 mavhq joined #salt
22:46 pipps joined #salt
22:46 yujunz joined #salt
22:47 FuzzyVeg left #salt
22:51 jpsharp left #salt
22:55 sh123124213 joined #salt
22:56 stooj joined #salt
22:58 mavhq joined #salt
23:06 stooj Hi. Is there a nicer way of organising my pillar top file so each host is associated with users.$hostname? Currently I'm going to end up with a big list. https://dpaste.de/7rjt
23:07 saltslackbridge <gtmanfred> you might check out pillar stack
23:07 saltslackbridge <gtmanfred> https://docs.saltstack.com/en/latest/ref/pillar/all/salt.pillar.stack.html
23:13 sh123124213 joined #salt
23:14 pipps joined #salt
23:14 whytewolf there is always this since the grains id was fixed to always be the minion id. https://gist.github.com/whytewolf/455d97a6de4991fc930543045b2e1714
23:15 onlyanegg joined #salt
23:15 saltslackbridge <gtmanfred> does that work in pillars? i guess it could if top.sls is rendered once for every minion
23:15 saltslackbridge <gtmanfred> which makes sesne
23:15 saltslackbridge <gtmanfred> sense*
23:16 whytewolf yeap, it totaly works. as long as your id doesn't have dots. but that is a different problem
23:18 pipps joined #salt
23:20 saltslackbridge <gtmanfred> neat
23:25 yujunz joined #salt
23:28 pipps joined #salt
23:32 sh123124213 joined #salt
23:33 * MTecknology just uses .replace('.', '_')
23:38 yujunz joined #salt
23:39 pipps joined #salt
23:40 onlyanegg joined #salt
23:40 mikecmpbll joined #salt
23:44 systemexit joined #salt
23:48 Renich_ joined #salt

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