Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2018-01-11

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

All times shown according to UTC.

Time Nick Message
00:00 saltslackbridge joined #salt
00:10 karlthane joined #salt
00:14 aldevar joined #salt
00:15 pipps joined #salt
00:16 hoonetorg joined #salt
00:34 pipps joined #salt
00:36 pipps99 joined #salt
00:40 schasi joined #salt
00:42 edrocks joined #salt
01:01 chesty joined #salt
01:07 nethershaw joined #salt
01:14 gnomethrower joined #salt
01:14 gnomethrower joined #salt
01:42 jhujhiti how do i use the new reactor syntax to pass positional arguments? the example in "Local Reactions" at https://docs.saltstack.com/en/latest/topics/reactor/index.html uses state.single, which uses only named arguments
02:06 RF_____ joined #salt
02:07 XenophonF whytewolf: hold on, you mean actually contribute code back to an open source project? #madlads
02:11 whytewolf Well yeah. Besides if others patch in good code with good docs it makes my job a lot easier
02:14 RF__ joined #salt
02:23 kettlewell joined #salt
02:25 nomeed joined #salt
02:27 stduolc joined #salt
02:27 jeffspeff joined #salt
02:27 stduolc hi all, I want to write my own module for some minions, which will be filter by grains. Can I do that? Is there any demo or source code? I only found the demo code like this: https://docs.saltstack.com/en/latest/topics/utils/index.html
02:34 nomeed joined #salt
02:39 shiranaihito joined #salt
03:02 ilbot3 joined #salt
03:02 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 masber joined #salt
03:13 jab416171 joined #salt
03:15 stduolc eightyeight: thx, but it's not solved my quiz
03:17 chowmeined joined #salt
03:30 eightyeight MTecknology: thx. I didn't get a chance to look at the issue today
03:30 eightyeight stduolc: ?
03:31 MTecknology eightyeight: I just happened to need that to get proxmox stuff working correctly, figured I should share. :)
03:31 jab416171 joined #salt
03:36 eightyeight MTecknology: I appreciate it. I'll get to it tomorrow. Hopefully. Heh.
03:50 jeffspeff joined #salt
03:58 jeffspeff joined #salt
04:05 swa_work joined #salt
04:20 zerocool_ joined #salt
04:32 Guest73 joined #salt
04:44 Guest73 joined #salt
04:48 Guest73 joined #salt
04:59 whytewolf stduolc: https://docs.saltstack.com/en/latest/ref/modules/ might be better
05:11 NightMonkey joined #salt
05:33 rubenb_ joined #salt
05:46 golodhrim|work joined #salt
05:47 rubenb_ left #salt
05:47 rubenb joined #salt
05:56 onlyanegg joined #salt
06:20 hoonetorg joined #salt
06:24 LocaMocha joined #salt
06:51 NightMonkey joined #salt
06:58 pualj joined #salt
07:16 pualj_ joined #salt
07:19 sergeyt joined #salt
07:33 aerbax joined #salt
07:43 nku i have a debian package that reads an env var during install. setting it doesn't seem to be possible with pkg?
07:49 MTecknology debian packages should not read environment variables... that's typically a no-no
07:51 nku yeah, it's completely brain-dead.. even includes a compiled nginx, postgres, etc. wtf..
07:51 nku in *one* package..
07:53 aerbax joined #salt
08:02 MTecknology yuck
08:02 MTecknology could you just use "FOO=bar cmd"?
08:10 Yoda-BZH joined #salt
08:10 Yoda-BZH joined #salt
08:10 nku nah, i won't use it. fortunately there are other options
08:13 aldevar joined #salt
08:14 Ricardo1000 joined #salt
08:20 aviau joined #salt
08:45 Yoda-BZH` joined #salt
08:51 Yoda-BZH` left #salt
08:51 Yoda-BZH joined #salt
08:52 Yoda-BZH joined #salt
08:52 MTecknology woo! I just finished renaming/rebuilding my salt host. One more vm host to got and I'll be done with this tedious bit of my project!
08:54 Tucky joined #salt
09:02 pbandark joined #salt
09:05 ws2k3 joined #salt
09:05 ws2k3 joined #salt
09:06 ws2k3 joined #salt
09:06 ws2k3 joined #salt
09:07 ws2k3 joined #salt
09:08 ws2k3 joined #salt
09:09 Hybrid joined #salt
09:15 jhauser joined #salt
09:26 sh123124213 joined #salt
09:31 BarBQ joined #salt
09:32 Mattch joined #salt
09:56 ws2k3 joined #salt
10:04 indistylo joined #salt
10:06 schasi joined #salt
10:09 cyteen joined #salt
10:25 Larry joined #salt
10:33 sergeyt joined #salt
10:37 bdrung_work joined #salt
10:42 schasi joined #salt
10:44 pualj joined #salt
10:45 xet7 joined #salt
10:57 RandyT joined #salt
10:57 RandyT joined #salt
11:15 xet7 joined #salt
11:24 pualj joined #salt
11:28 yujunz joined #salt
11:36 yujunz joined #salt
11:42 impi joined #salt
11:47 flagword joined #salt
11:51 stooj joined #salt
11:52 hammer065 joined #salt
12:00 zerocool_ joined #salt
12:05 zerocool_ joined #salt
12:06 oida joined #salt
12:11 zerocool_ joined #salt
12:22 chowmein__ joined #salt
12:23 nebuchadnezzar joined #salt
12:24 mechleg1 joined #salt
12:24 jacksontj joined #salt
12:26 mTeK joined #salt
12:27 Tyrant joined #salt
12:27 yidhra joined #salt
12:28 baikal joined #salt
12:31 justanotheruser joined #salt
12:31 kettlewell joined #salt
12:31 jacksontj joined #salt
12:33 sjorge joined #salt
12:48 _JZ_ joined #salt
12:53 ponyofdeath joined #salt
13:00 sergeyt joined #salt
13:13 Deliant joined #salt
13:24 pualj joined #salt
13:30 Deliant joined #salt
13:46 Nahual joined #salt
13:46 edrocks joined #salt
13:53 edrocks joined #salt
13:55 jdipierro joined #salt
13:57 sjorge joined #salt
14:07 s0undt3ch joined #salt
14:14 RF_ joined #salt
14:33 dlloyd joined #salt
14:35 DammitJim joined #salt
14:37 racooper joined #salt
14:42 mchlumsky joined #salt
14:43 kojiro joined #salt
14:45 mchlumsky joined #salt
14:50 _val_ joined #salt
14:51 _val_ Hi everyone. Does salt have any built-in sort of ... 'sort' ranther than piping the salt command to | sort ?
14:51 pualj joined #salt
14:51 _val_ e.g  salt 'somehost[1-9][a-c]*'  cmd.run 'hostname'  | sort    I'd like to avoid the pipeline
14:52 _val_ So I read: https://github.com/saltstack/salt/issues/21797 but doesn't seem to have an answer there
14:54 DammitJim joined #salt
14:55 mchlumsky joined #salt
14:56 gh34 joined #salt
14:56 edrocks joined #salt
15:03 Ricardo1000 joined #salt
15:04 flagword joined #salt
15:10 cgiroua joined #salt
15:20 nonades joined #salt
15:22 cyteen_ joined #salt
15:25 onlyanegg joined #salt
15:31 tobiasBora Hello,
15:31 kojiro hello!
15:31 huddy joined #salt
15:32 schasi joined #salt
15:37 hax404 joined #salt
15:43 tiwula joined #salt
15:44 Kelsar joined #salt
15:46 DammitJim joined #salt
15:49 mk-fg joined #salt
15:49 mk-fg joined #salt
15:52 tobiasBora In some formula, I can read in the source things like "server.support_email"
15:53 sergeyt joined #salt
15:54 tobiasBora hum, maybe "server.mail_from" is better
15:54 tobiasBora while in the yaml, I set up something like:
15:54 tobiasBora server:
15:54 tobiasBora mail:
15:54 tobiasBora from: "me@hi.com"
15:55 tobiasBora why isn't it then "server.support.email" ?
15:55 tobiasBora [my source: https://github.com/salt-formulas/salt-formula-gitlab]
15:56 tobiasBora are the "_" added automatically in place of "." in yaml files? Does that mean that I can't use any "_" in the variable names?
15:56 valkyr2e joined #salt
15:57 tobiasBora and what's the difference between identity.engine and identity.get("host")
15:59 sergeyt joined #salt
16:02 RF___ joined #salt
16:02 tobiasBora Hum, it looks like both exists: server.mail.from and server.mail_from
16:02 tobiasBora so maybe server.mail_from refers to "server: mail_from: ..."
16:25 sjorge joined #salt
16:25 schasi joined #salt
16:27 tobiasBora Problem solved
16:27 Naresh joined #salt
16:28 tobiasBora Another question: what is the best way to install and configure postgresql?
16:28 tobiasBora I found states to configure the users/database... but not install the server itself.
16:28 saltslackbridge <gtmanfred> worth mentioning that salt-formulas is not saltstack backed, it is run by the old tcpcloud/mirantis people
16:28 saltslackbridge <gtmanfred> saltstack manages saltstack-formulas
16:29 saltslackbridge <gtmanfred> there is a plan to try and join the two groups, but it has not happened yet
16:36 aldevar left #salt
16:48 lordcirth_work When running a module from the command line, eg "salt '*' status.cpuinfo" is there a way to filter to a dictionary key of the results? like status.cpuinfo()['vendor_id'] in Python
16:49 babilen Output it as JSON and use jq
16:53 arif-ali joined #salt
16:53 raceboyer_ joined #salt
16:59 zerocoolback joined #salt
17:00 lordcirth_work babilen, makes sense.  Reading man jq now
17:01 edrocks joined #salt
17:05 babilen It's a useful tool, you won't regret knowing how to use it
17:13 Hybrid joined #salt
17:14 Heartsbane joined #salt
17:14 Heartsbane joined #salt
17:16 Trauma joined #salt
17:18 onlyanegg joined #salt
17:32 Hybrid joined #salt
17:33 nonades joined #salt
17:37 Aleks3Y joined #salt
17:41 jdipierro joined #salt
17:47 indistylo joined #salt
17:51 pipps joined #salt
17:53 cgiroua joined #salt
17:53 pipps joined #salt
17:59 indistylo joined #salt
18:12 pipps joined #salt
18:17 lordcirth_work babilen, it looks great, but I'm not sure how to do what I want.  Namely, print just minion_id: vendor_id .
18:18 lordcirth_work I got it to work for one minion, jq '."hostname"."vendor_id"'
18:19 cgiroua joined #salt
18:21 lordcirth_work Oh, it's .[]
18:22 lordcirth_work Doesn't actually print the minion id, though
18:28 dstensnes joined #salt
18:36 Edgan joined #salt
18:40 jdipierro joined #salt
18:50 debian112 joined #salt
18:57 debian112 left #salt
19:00 arif-ali joined #salt
19:07 lordcirth_work Got it!  cat ~/out.json | jq 'select( .[].vendor_id | contains("Intel") ) | keys'
19:07 pipps joined #salt
19:07 saltslackbridge <gtmanfred> `jq 'select( .[].vendor_id | contains("Intel") ) | keys' < out.json`
19:08 lordcirth_work Yes, yes, I was sticking as close to piping from salt as possible :P
19:08 saltslackbridge <gtmanfred> :P
19:09 saltslackbridge <gtmanfred> well, you can also do `jq 'select( .[].vendor_id | contains("Intel") ) | keys' <(salt....)` might be `< <(salt...)
19:13 pipps joined #salt
19:14 pipps joined #salt
19:14 huddy joined #salt
19:15 indistylo joined #salt
19:36 J0hnSteel joined #salt
19:39 tobiasBora So no state exists to manage postgresql install and configuration ?
19:40 Lionel_Debroux_ joined #salt
19:43 Trauma joined #salt
19:52 lordcirth_work tobiasBora, well, installation would be pkg.installed.  There's also several postgres states here: https://docs.saltstack.com/en/latest/ref/states/all/
20:01 ymasson joined #salt
20:06 sh123124213 joined #salt
20:11 jeffspeff joined #salt
20:11 edrocks joined #salt
20:33 mk-fg joined #salt
20:33 mk-fg joined #salt
20:46 freelock hmm has anybody seen any major performance regressions in salt events, after applying the kernel updates for meltdown?
20:46 freelock I have a bot that is calling "salt-call event.fire_master ..." on the same server as the salt master
20:48 freelock this gets triggered by a git update on a central git repository... on ubuntu 16.10, with kernel 4.4.0-108-generic, the "python salt-call" process sits there in PS for minutes at a time... even when there isn't a reactor listening for this particular event...
20:49 freelock these stack up on our server with a bunch of commits to a pool repository, and soon the master is hitting a load of 35, and getting very unresponsive
20:50 freelock I've rebooted, tried a bunch of other things, but it seems like the "event.fire_master" has a serious performance regression now...
20:51 freelock going back to kernel 4.4.0-104-generic, it's totally fine, finishes quickly and disappears from ps in a few seconds or less
20:51 whytewolf freelock: https://www.theregister.co.uk/2018/01/09/meltdown_spectre_slowdown/
20:51 freelock what might explain this?
20:52 gtmanfred freelock: we have seen some errors in our test suite because of that
20:53 saltslackbridge joined #salt
20:55 freelock interesting...
20:56 freelock it just seems strange that the biggest issue we're seeing so far after upgrading ~30 servers/workstations (aside from one server that failed to boot) is this salt-call that is essentially a no-op
20:59 DammitJim joined #salt
21:02 sh123124213 joined #salt
21:09 sayyid9000 joined #salt
21:10 masber joined #salt
21:23 onlyanegg joined #salt
21:35 sjorge joined #salt
21:41 it_dude joined #salt
21:43 pipps joined #salt
21:59 edrocks joined #salt
22:08 Guest73 joined #salt
22:08 sjorge joined #salt
22:09 onlyanegg joined #salt
22:09 cgiroua joined #salt
22:26 Edgan freelock: I have seen some weird responsiveness issues since I installed the meltdown kernel update, and spectre bios update(microcode). I suspect there is a serious kernel regression. It reminds me of the issue where your Linux X session would freeze on heavy disk I/O like compiling a Linux kernel. Which I think was caused by scheduler issues.
22:28 irated joined #salt
22:36 sh123124213 joined #salt
22:38 sh123124213 joined #salt
22:40 jdipierro joined #salt
22:40 schasi joined #salt
22:40 packeteer joined #salt
22:44 nonades joined #salt
22:52 ahrs joined #salt
22:54 sayyid9000 joined #salt
22:54 marcinkuzminski joined #salt
22:57 RandyT joined #salt
22:57 RandyT joined #salt
22:59 tom][ joined #salt
23:09 Trauma joined #salt
23:12 relidy I'm encountering some inconsistent behavior with the new module.run format. I've added the appropriate configuration on the minion, and using salt-call locally everything's fine. However, if I try to call the appropriate state from the master, I get an error that looks like wha tI'd get if the minion didn't know about the new format. Configuration, state, outputs, and versions: https://gist.github.com/rhoths/abb14cbc6c93b5e1d65dfedfa9bd3561
23:13 relidy Have I missed some configuration somewhere, maybe on the master, that simply isn't documented (or stupid me skipped)?
23:17 saltslackbridge <gtmanfred> did you restart the salt-minion after you set use_superseed
23:18 relidy gtmanfred: I have, yes.
23:18 saltslackbridge <gtmanfred> hrm, weird
23:18 relidy gtmanfred: You know, let me check. I may not have. I've been messing with several machines.
23:20 relidy gtmanfred: Well, that was apparently the problem. I assumed (you know what they say) that the fact that the minion picked up the new configuration meant everything would/should be fine.
23:20 relidy Thanks again gtmanfred.
23:20 saltslackbridge <gtmanfred> np, salt-call starts a whole new salt-minion process, and doesn’t communicate with the running salt-minion
23:37 ry_ joined #salt
23:38 ry_ left #salt
23:45 nonades joined #salt
23:48 wwalker https://gist.github.com/wwalker/aeac99836c9f8a03db2658f73fec60e0 - gitfs_remotes is apparently wrong... point out my stupidity please
23:51 cgiroua joined #salt
23:56 relidy wwalker: Does 'git@github.threatbuild.com:threatgrid/tg-salt.git' == 'git@github.com:tg/tg-salt.git' in your ext_pillar? Is that the problem (I don't know much about gitfs)?
23:57 relidy ^ Specifically the "tg/" vs "threatgrid/"
23:58 wwalker incomplete redaction
23:59 wwalker all 4 URLs are really the same.
23:59 * relidy nods

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