Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-07-03

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

All times shown according to UTC.

Time Nick Message
00:01 Lee- joined #salt
00:51 rem5 joined #salt
01:02 fannet_ joined #salt
01:13 iceyao joined #salt
01:27 kus joined #salt
01:34 dendazen joined #salt
01:35 catpigger joined #salt
02:03 dendazen joined #salt
02:11 om2 joined #salt
02:42 justanotheruser joined #salt
03:02 fannet_ joined #salt
03:18 rem5 joined #salt
03:18 rem5 joined #salt
03:55 amcorreia joined #salt
04:07 rem5 joined #salt
04:18 zenlot joined #salt
04:22 khaije1 joined #salt
04:22 khaije1 I'd like to refresh the fileserver cache on the salt master by running a highstate ... how is this done?
04:23 khaije1 This is part of a quick provision salt-master I'm working on
04:25 khaije1 I think this uses the orchestrate capability, just not clear on how to pull it off :0/
04:28 racooper joined #salt
04:29 khaije1 update: I think I found the info at https://docs.saltstack.com/en/latest/ref/states/all/salt.states.saltmod.html#salt.states.saltmod.runner but feel free to ping me all the same :)
04:51 bfig joined #salt
04:51 bfig__ joined #salt
05:28 sjmh joined #salt
05:56 DarkKnightCZ joined #salt
06:07 ruxu joined #salt
06:17 manji joined #salt
06:21 bantone sup
06:22 cprior joined #salt
06:25 garionphx joined #salt
06:28 garionphx Hi all. I've been playing with getting a salt master on an EC2 instance, and using gitfs for the states and pillars.. In v2016.3.1, it works fine, but with devel, it doesnt.
06:29 DEger joined #salt
06:30 garionphx (pausing while I bring my devel instance up to get the exact errors..)
06:33 linjan joined #salt
06:37 garionphx "Exception caught while initializing gitfs remote 'ssh://<snip>@git-codecommit.us-east-1.amazonaws.com/v1/repos/<snip>': value must be a string
06:39 hemebond garionphx: Can you show the exact entry from your config?
06:39 hemebond (obscured is necessary)
06:41 garionphx hemebond: Sure... http://pastebin.com/JYJ2KsXL.. That config works ok in v2016.3.1, but when I switch to devel, it does odd stuff.
06:42 garionphx If I run the master a second time, it starts, I dont get any of my pillars (not sure about the states.)
06:43 hemebond What happens if you put the key in quotes?
06:43 hemebond e.g., - "ssh://...
06:43 garionphx for the pillar?
06:43 garionphx or both?
06:44 garionphx "Failed to checkout master from git_pillar remote 'master "ssh://APKAJVFJ7LD7VFYL46OQ@git-codecommit.us-east-1.amazonaws.com/v1/repos/saltconfig"': remote ref does not exist"
06:45 garionphx time to change my IAM, damn it.
06:46 deniszh joined #salt
06:49 hemebond Should ext_pillar:git not be a string instead of the list you have now?
06:50 hemebond Oh, accepts list too.\
06:50 hemebond Maybe try a string to test
06:51 garionphx Well... As I said, the config works as is on 2016.1.3..
06:51 hemebond Sure, but someone might have broken git_pillar
06:51 hemebond in develop
06:52 garionphx from what I can tell, looks like my states arent there either.
06:53 garionphx It might just be easier for me to backport the one feature I need in devel to 2016.3.1.... (boto_lambda in a vpc)
06:55 garionphx I take that back.. The states are there.
06:55 garionphx So its just git_pillar that seem broken.
06:56 garionphx Failed to checkout master from git_pillar remote 'master ssh://<snip>@git-codecommit.us-east-1.amazonaws.com/v1/repos/saltconfig': _getconf() takes exactly 2 arguments (1 given)
07:00 babilen garionphx: If all you want is a a module from devel, I'd just copy it to _modules and be done with it.
07:00 babilen (same for state modules)
07:01 garionphx More ofthat last error: http://pastebin.com/u4pm3jw5
07:02 garionphx If I do that, will it override the existing one, or will I have to rename it? in other words, can I just copy it into the _modules and _states, without renaming/editting?
07:04 fannet_ joined #salt
07:10 manji joined #salt
07:12 ruxu joined #salt
07:16 DarkKnightCZ joined #salt
07:21 Fiber^ joined #salt
07:33 blueelvis joined #salt
07:39 cprior garionphx: Remembering https://docs.saltstack.com/en/latest/ref/file_server/dynamic-modules.html and re-reading this quickly it seems you can sync modules from the master fileserver ("sync" as in overwrite, methinks)
07:44 garionphx yeah, I found what I needed in the docs: "Any custom states which have been synced to a minion, that are named the same as one of Salt's default set of states, will take the place of the default state with the same name"
07:45 garionphx It looks like its working too.. Went back to 2016.3.1, added those files to _modules/_states, and it seems to be working.
07:45 babilen garionphx: Yes, that is why I suggested copying it to _states / _modules
07:47 babilen It's an easy way to introduce customised modules into salt. And this does not only work for states and execution modules either, but many other parts of salt also (grains, beacons, ...)
07:53 ruxu joined #salt
07:56 garionphx Yeah.. I work at a company that has done some magical stuff with salt...I'm just a lowly dev, playing with salt at home for some side stuff.
07:56 cprior joined #salt
07:58 keimlink joined #salt
08:15 CeBe joined #salt
08:39 blueelvis joined #salt
09:05 fannet_ joined #salt
09:10 Ashald joined #salt
09:22 markus__ joined #salt
09:26 markus__ question about matching minions. For example: All minions with pillar "patch:patch_a or patch_patch_c": salt -C 'I@patch:patch_a or I@patch:patch_c
09:27 markus__ All minions, exept those with pillar "patch:patch_a": salt -C '* and not I@patch:patch_a '
09:28 markus__ But i cant find no valid salt compound match for the following: All minions, exept those with pillar "patch:patch_a or patch_patch_c"
09:29 markus__ So how to do this, what is the right syntax?
09:29 markus__ https://docs.saltstack.com/en/latest/topics/targeting/compound.html does not help (or i can't find)
09:33 babilen foo and not (bar or baz)
09:34 babilen So, something like "* and not ("patch:patch_a or patch_patch_c)" or just "not (patch:patch_a or patch_patch_c)" or "not patch:patch_a and not patch:patch_a"
09:35 manji joined #salt
09:38 markus__ <babilen> i tried both before without success. But may a pcre match should work ie. J@(foo|bar) ...trying...
09:39 fannet_ joined #salt
09:39 babilen Should work, you don't need a PCRE (even though you could use it)
09:42 markus__ Problem solved. This ist a working example: salt -C '* and not J@patch:(patch_a|patch_c)$'
09:43 babilen IIRC you can drop the "* and"
09:43 markus__ so pcre still does the job :)
09:44 babilen Mind pasting the output of that and "salt -C "* and not (I@patch:patch_a or I@patch:patch_b)" test.ping" for comparison?
09:44 manji joined #salt
09:44 manji joined #salt
09:45 babilen A PCRE is probably still a cleaner approach, but logically the latter should work
09:46 markus__ ^^ack. but practically it does not work.(here) Hmm...second please....
09:47 babilen It's particularly surprising as "salt -C 'I@patch:patch_a or I@patch:patch_c' test.ping" does the right thing according to you
09:48 markus__ ^^thats exactly why i am here. ;-)
09:48 babilen Sounds like a bug if that is the case, tbh
09:52 markus__ i will investigate further. Maybe just a "layer8" problem.
09:53 babilen ack
09:58 blueelvis joined #salt
10:20 cprior markus__: I found it quite easy to read the Python code in such situations. For example I was perplexed recently with the bahaviour of salt.states.file.line and the code was an easy-to-grasp if/elif/else thing
10:21 cprior Afterwards I understood the docs! *g* (code and docs are fine, no offence meant)
10:22 ruxu joined #salt
10:31 lero joined #salt
10:33 workthrick joined #salt
10:34 eichiro joined #salt
10:37 lkannan joined #salt
10:37 macheck joined #salt
10:37 bmcorser joined #salt
10:37 smakar joined #salt
10:37 hacks joined #salt
10:37 wiqd joined #salt
10:39 liviudm joined #salt
10:57 reith joined #salt
11:12 markus__ <babilen>: please confirm: https://gist.github.com/anonymous/67a8c3a795462d0b6fb82fd8044cdd86
11:14 markus__ if you think thats a bug, i will open a issue for this.
11:22 darvon joined #salt
11:34 kiorky joined #salt
11:34 dyaln joined #salt
11:38 viq markus__: how about instead of 'not (A or B)' using 'not A and not B' ?
11:39 viq "salt -C 'not I@patch:patch_a and not I@patch:patch_c'"
11:39 markus__ just a minute please, upgrading to 2016.3.1...
11:43 markus__ viq: "salt -C 'not I@tc:patch:patch_a and not I@tc:patch:patch_c' test.ping --output=txt |sort -n |wc -l" returns 44. ie. this is ok
11:44 viq and after upgrade the one with brackets still gives you the wrong count?
11:45 bluenemo joined #salt
11:52 markus__ viq: yes, same behave. after upgrade, i fight with https://github.com/saltstack/salt/issues/31978 so compound matching issues have to wait until i fixed older bugs
11:52 saltstackbot [#31978][OPEN] The tty option is not recognized in salt-cloud, so requiretty is not enforced in my setup | Option "tty:" in salt-cloud (salt/utils/cloud.py) is not recognized in my setup. I do not understand why....
11:56 yomateo joined #salt
12:01 TyrfingMjolnir joined #salt
12:02 linjan joined #salt
12:08 iceyao joined #salt
12:10 blueelvis joined #salt
12:14 linjan joined #salt
12:51 dmaiocchi joined #salt
13:41 oida joined #salt
13:41 fannet_ joined #salt
14:06 Fiber^ joined #salt
14:16 snc joined #salt
14:18 whitenoise joined #salt
14:25 macheck joined #salt
14:27 mohae joined #salt
14:30 west575 joined #salt
14:34 sjmh joined #salt
14:46 rem5 joined #salt
14:56 manji joined #salt
15:14 subsignal joined #salt
15:17 darvon joined #salt
15:23 amcorreia joined #salt
15:36 sjmh joined #salt
15:42 fannet_ joined #salt
15:48 darvon joined #salt
15:51 rem5 joined #salt
16:11 DEger joined #salt
16:12 charli joined #salt
16:17 KingJ joined #salt
16:25 rem5 joined #salt
16:33 POJO joined #salt
16:40 rem5 joined #salt
16:43 alexlist joined #salt
16:50 ruxu from a salt minion (which is on AWS), can I find the instance's external IP address?
16:51 s_kunk joined #salt
16:53 bfig_ joined #salt
16:53 bfig___ joined #salt
16:53 atmosx ruxu: salt '<minion>' cmd.run 'curl ifconfig.me'
16:54 ruxu thanks, but isn't this command ran from the salt master?
16:54 ruxu atmosx: thanks, but isn't this command ran from the salt master?
16:55 atmosx ifconfig.me is down hm
16:55 iggy you're telling the minion to run that command
16:55 iggy so no
16:55 keldwud joined #salt
16:55 iggy here are also some grains in salt-contrib
16:55 keldwud joined #salt
16:56 iggy https://github.com/saltstack/salt-contrib/tree/master/grains
16:56 atmosx ruxu: curl ipecho.net/plain ..
16:56 iggy there are a few ec2 grains in there that have various levels of functionality (and various levels of dependencies)
16:56 iggy icanhazip.com
16:56 atmosx that's it: https://github.com/saltstack/salt-contrib/blob/master/grains/external_ip.py
16:56 iggy there are literally dozens of them
16:57 atmosx later all
16:57 ruxu thanks :)
17:02 fannet_ joined #salt
17:12 DarkKnightCZ joined #salt
17:26 saltnpeppa joined #salt
17:27 saltnpeppa when running salt-cloud to build an ec2 instance, I get this error:bash: sudo /tmp/.saltcloud-a50cf57c-4012-436c-9ec0-707a5864ab80/deploy.sh -c '/tmp/.saltcloud-a50cf57c-4012-436c-9ec0-707a5864ab80': No such file or directory
17:28 saltnpeppa salt-cloud does ssh to the host
17:28 saltnpeppa anyone seen this before?
17:30 saltnpeppa these paths do exist on the ec2 instance it built too
17:31 saltnpeppa -rwx--x--x. 1 centos centos 204069 Jul  3 17:09 /tmp/.saltcloud-a50cf57c-4012-436c-9ec0-707a5864ab80/deploy.sh
17:32 pranav joined #salt
17:34 saltnpeppa hmmm it might not actually be that it doesnt exist
17:40 saltnpeppa ok i think i see why, this epel rpm does not exist, i get a 404 error: skipping http://download.fedoraproject.org/pub/epel/7/x86_64/e/epel-release-7-5.noarch.rpm - transfer failed
17:42 saltnpeppa i see, the deploy.sh script is specifically calling         rpm -Uvh --force "http://download.fedoraproject.org/pub/epel/7/${EPEL_ARCH}/e/epel-release-7-5.noarch.rpm" || return 1
17:43 saltnpeppa do we need to update the deploy script or something?
17:45 saltnpeppa I see... i think what ill need to do is basically create an ec2 instance with the epel already installed
17:55 iggy saltnpeppa: salt-cloud -u (maybe)
18:03 Lionel_Debroux joined #salt
18:13 systeem left #salt
18:16 darvon joined #salt
18:33 skinnejo joined #salt
19:21 west575 joined #salt
19:29 kus joined #salt
19:32 darvon joined #salt
19:32 eichiro joined #salt
19:44 ruxu joined #salt
19:54 toastedpenguin joined #salt
20:00 punkoivan joined #salt
20:01 punkoivan left #salt
20:07 A||SySt3msG0 joined #salt
20:14 manifold joined #salt
20:44 Lee- joined #salt
20:46 subsignal joined #salt
20:51 ruxu joined #salt
20:54 yomateo joined #salt
21:14 rburkholder joined #salt
21:36 asoc_ joined #salt
22:01 ov joined #salt
22:04 liskl joined #salt
22:15 punkoivan1 joined #salt
22:16 punkoivan1 left #salt
22:34 zenlot6 joined #salt
22:37 yomateo- joined #salt
22:49 liskl joined #salt
22:58 saltnpeppa hi all, salt is creating a repo file in /etc/yum/repos.d/saltstack-salt-epel-7.repo and inside the file its html headers with a 302 redirect
22:58 saltnpeppa anyway to get around this?
23:07 fxhp joined #salt
23:15 xfrost joined #salt
23:15 xfrost left #salt
23:16 johnkeates joined #salt
23:25 snath joined #salt
23:28 zenimal joined #salt
23:38 zenimal joined #salt
23:41 zenimal joined #salt
23:42 fannet_ joined #salt
23:43 zenimal joined #salt
23:44 zenimal joined #salt
23:54 saltnpeppa is there a way to set the hostname once salt-cloud bootstraps a host?
23:55 hoonetorg joined #salt
23:58 badon_ joined #salt

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