Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2015-06-05

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

All times shown according to UTC.

Time Nick Message
00:01 d[^_^]b joined #salt
00:04 d[^_^]b joined #salt
00:05 d[^_^]b joined #salt
00:07 d[^_^]b joined #salt
00:09 d[^_^]b joined #salt
00:10 murrdoc Ahlee:  u are evil
00:10 murrdoc and u know it
00:10 murrdoc so pgrep … whats the salty way to do this
00:11 d[^_^]b joined #salt
00:12 MatthewsFace joined #salt
00:13 d[^_^]b joined #salt
00:15 otter768 joined #salt
00:15 d[^_^]b joined #salt
00:17 d[^_^]b joined #salt
00:18 cberndt joined #salt
00:19 d[^_^]b joined #salt
00:20 murrdoc joined #salt
00:21 Ahlee murrdoc: another meetup should be announced soon
00:21 d[^_^]b joined #salt
00:23 d[^_^]b joined #salt
00:25 d[^_^]b joined #salt
00:26 IanVorn joined #salt
00:27 IanVorn joined #salt
00:27 d[^_^]b joined #salt
00:29 d[^_^]b joined #salt
00:31 monokrome joined #salt
00:31 d[^_^]b joined #salt
00:34 d[^_^]b joined #salt
00:34 alfss joined #salt
00:35 d[^_^]b joined #salt
00:37 d[^_^]b joined #salt
00:39 ajw0100 joined #salt
00:39 d[^_^]b joined #salt
00:41 d[^_^]b joined #salt
00:43 d[^_^]b joined #salt
00:45 d[^_^]b joined #salt
00:47 d[^_^]b joined #salt
00:48 baweaver joined #salt
00:49 d[^_^]b joined #salt
00:51 DrBojingle joined #salt
00:51 d[^_^]b joined #salt
00:52 cberndt joined #salt
00:53 d[^_^]b joined #salt
00:55 d[^_^]b joined #salt
00:57 d[^_^]b joined #salt
00:59 d[^_^]b joined #salt
01:01 d[^_^]b joined #salt
01:03 sgargan joined #salt
01:04 d[^_^]b joined #salt
01:05 d[^_^]b joined #salt
01:07 d[^_^]b joined #salt
01:09 d[^_^]b joined #salt
01:10 pacopablo joined #salt
01:11 d[^_^]b joined #salt
01:12 pacopablo I am trying to modify the openvpn-formula. I'd like to store my keys in the pillar data with gpg.  That I can do with the gpg renderer.
01:12 pacopablo however, teh openvpn-formula uses a jinja template for writing the actual config files.
01:13 pacopablo and uses a map.jinja file in the state file to set some paths.
01:13 pacopablo is there a way to reference the data fro the map file in the template?
01:13 d[^_^]b joined #salt
01:14 Nazca__ joined #salt
01:15 d[^_^]b joined #salt
01:16 fllr joined #salt
01:17 d[^_^]b joined #salt
01:19 iggy murrdoc: nothing that I know of
01:19 d[^_^]b joined #salt
01:20 murrdoc sucks
01:20 iggy pacopablo: yes, it's probably called {{ openvpn_settings }} or {{ openvpn }}
01:21 d[^_^]b joined #salt
01:22 murrdoc iggy … can i install a pip that lives on the salt master
01:22 murrdoc like we can do it with packages
01:23 d[^_^]b joined #salt
01:24 iggy .deb's for everything
01:25 d[^_^]b joined #salt
01:27 d[^_^]b joined #salt
01:27 david_an111 joined #salt
01:29 beauby joined #salt
01:29 d[^_^]b joined #salt
01:30 joshin basepi: can you punt d[^_^]b
01:30 basepi Flapping connection again?
01:31 basepi (I can't see it, I have parts and joins filtered)
01:31 joshin Yeah. About once a minute.
01:31 murrdoc u can add ignore on user joins to your client
01:31 d[^_^]b joined #salt
01:31 d[^_^]b was kicked by basepi: d[^_^]b
01:32 joshin Thanks basepi. I know murrdoc, but it's useful for the channels I moderate.
01:32 basepi It's interesting, he seems to be using irccloud, which I am also using. Dunno why he's flapping.
01:32 basepi Am I?
01:32 joshin Nope.
01:32 basepi Hmmmm
01:32 joshin Are you using IPv4 or 6?
01:33 basepi 4, I think.
01:33 joshin He was using 6.
01:33 joshin ae@2600:3c03::f03c:91ff:fe6e:5ed9)
01:33 basepi Well, if he ever talks, see if you can get him to look into it. =P
01:34 basepi I don't particularly want to ban someone who's doing nothing particularly wrong.
01:34 d[^_^]b joined #salt
01:35 d[^_^]b joined #salt
01:36 ageorgop joined #salt
01:37 d[^_^]b joined #salt
01:38 murrdoc1 joined #salt
01:39 d[^_^]b joined #salt
01:41 d[^_^]b joined #salt
01:45 iggy just ban, if they actually have interest in fixing the problem, they can join with another name
01:45 d[^_^]b joined #salt
01:45 d[^_^]b joined #salt
01:46 ITChap joined #salt
01:47 d[^_^]b joined #salt
01:49 d[^_^]b joined #salt
01:51 d[^_^]b joined #salt
01:53 d[^_^]b joined #salt
01:55 d[^_^]b joined #salt
01:57 d[^_^]b joined #salt
01:59 d[^_^]b joined #salt
02:00 coval3nce joined #salt
02:00 elektrix hey, is there a good way to lint salt states and such on OSX? I tried to get pep8 lint working, but it was a nightmare, and I couldn't get it to happen.
02:01 d[^_^]b joined #salt
02:01 elektrix and lint templates?
02:02 dingo i think d[^_^]b is pretty, gives a sort of quilt effect on my terminal, don't punt him
02:02 dingo elektrix: i've saught into similar things
02:03 elektrix dingo: any luck?
02:03 dingo i have yet to find a good solution, i can share with you the minimals that i have written in the form of "unit tests"
02:03 al joined #salt
02:03 elektrix that'd be cool
02:03 dingo in true python projects, i use a ton of linting, probably 10 tools all together
02:03 dingo but salt states have the before and after-rendered -- so pillars are really all i can afford to validate
02:04 d[^_^]b joined #salt
02:04 elektrix I'll take that...am using pillars
02:04 * dingo is preparing gists
02:04 elektrix w00t!
02:05 d[^_^]b joined #salt
02:06 viq joined #salt
02:06 elektrix I came to Salt from puppet, so I really miss puppet-lint and puppet validate
02:07 d[^_^]b joined #salt
02:07 elektrix erm puppet parser validate
02:08 dingo py.test tests: https://gist.github.com/jquast/038c5400535a1276430d
02:08 dingo the supporting accessories.py module: https://github.eng.cloud.io.com/cloud/cloud-salt/blob/master/undercloud/tests/accessories.py
02:08 dingo elektrix: well i could lecture all day why what we have is better -- partly that salt also verifies your state when it renders it, its declaration vs. code, its very different
02:09 dingo the reason not much linting is done with salt, is salt does a fine job linting itself
02:09 d[^_^]b joined #salt
02:09 julez joined #salt
02:10 beauby joined #salt
02:10 KyleG joined #salt
02:10 KyleG joined #salt
02:11 d[^_^]b joined #salt
02:13 d[^_^]b joined #salt
02:15 elektrix yeah...I'm recognizing that...I just know that when I get a little beyond basic Salt stuff...it works, but I don't know that I'm doing it in the best way, or if my syntax will cause issues as I generate more Salt states and such
02:15 dingo anybody got a formular for doing salt['file.dirname'] using only jinja on older version of salt where it isn't available?
02:15 dingo elektrix: i think some style enforcement could be done with yaml
02:15 dingo just ensuring the yaml renders is all that truely matters
02:15 d[^_^]b joined #salt
02:15 dingo the data types and the way they are used, that cannot be determined without rendering
02:16 dingo i wouldn't mind some no-extra-newlines, 4-space-indent, no-trailing-whitespace style guides on yaml myself
02:16 otter768 joined #salt
02:17 elektrix well I've been using Vagrant environments when building stuff so the things I check in don't have any obvious errors. Just like the little things like when I'm using single quotes vs double quotes (or not use quotes at all), octal permissions, etc
02:17 d[^_^]b joined #salt
02:18 dingo i use vagrant too, and build systems that do vagrant up and provisioners that ensure the exit code of highstate is non-zero with salt-call --retcode-passthru or whatever the --arg is
02:18 elektrix my Vim plugins take care of a lot of those things (strip whitespace, turn all hard tabs into spaces, 4-space-indent, etc)
02:19 d[^_^]b joined #salt
02:20 dingo are you using https://github.com/saltstack/salt-vim ?
02:20 coval3nce is there an aggegator service for salt-api for topologies with multimasters (without failover type)?
02:20 elektrix cool...I need to step up on what I can do w/ Vagrant...right now all I have is a master and a minion
02:20 elektrix um...let me check
02:21 coval3nce elektrix: https://github.com/dkiser/salt-multimaster-pki-example  <- just whipped this up today, trying to learn multimaster setups myself
02:21 d[^_^]b joined #salt
02:21 elektrix yup, am using salt-vim
02:23 elektrix haven't got to the point where we need multi-master...we're a small (less than 50 hosts) environment
02:23 d[^_^]b joined #salt
02:23 elektrix and have mostly static hosts...aren't constantly spinning up/down tons of VMs
02:24 smcquay joined #salt
02:24 coval3nce gotcha, yeah im still just fooling around with lab vm’s and stuff at this point.
02:25 coval3nce trying to figure out the best design pattern to role out, so far got gitfs backedn with gpg encryption for states/pillars
02:25 coval3nce now trying to figure out best way to deal with masters dying
02:25 d[^_^]b joined #salt
02:26 elektrix eh...we're working on upgrading a number of static hosts right now that are on old OSes...constantly going back and forth between writing Salt states that do it the most proper/elegant/scalable way, and just getting everything into Salt so that hosts come up working, and we can use Salt effectively in a DR scenario
02:27 coval3nce Cool, i feel the pain ont he old OS front, got a bunch of cent5 stuffs i need to onboard and/or jut nuke/pake with some new salt goodness
02:27 elektrix ha...is cent5 still getting updates?
02:27 d[^_^]b joined #salt
02:27 coval3nce prolly just security ones
02:28 elektrix we're running some internal (read really firewalled off) hosts running Debian Lenny...which quit getting security updates a while ago
02:29 d[^_^]b joined #salt
02:30 elektrix given all the fun stuff like shellshock, and most recently logjam, getting some of these systems current has become a priority...having poor scores on ssl labs didn't hurt this arguement
02:31 coval3nce totally man - and venom now
02:31 coval3nce sucks for pretty much all hypervizor host admins
02:31 elektrix yeah, we lucked out...no KVM in our env
02:31 murrdoc joined #salt
02:31 coval3nce but at least i dont think there is widely distributed POC or anythign for it yet
02:31 d[^_^]b joined #salt
02:32 elektrix nothing I've seen...plus you have to get root before you even get there...
02:33 coval3nce yup, so only VPS type stuff is really a target
02:33 coval3nce but…everyone runs apps as root hehe
02:33 coval3nce so there is that too
02:33 elektrix yup...I've be nervous if I was a VPS provider and couldn't patch
02:34 d[^_^]b joined #salt
02:34 elektrix ehh...not everyone runs as root, but enough do to make it a good target
02:35 elektrix all the low hanging fruit...I know I can't keep up with the craziest security wonks/admins, but my goal is to never be the low hanging fruit
02:35 d[^_^]b joined #salt
02:36 elektrix esp when you have to consider that whole annoying security vs usability issue
02:37 coval3nce yup, its hard keeping up
02:37 d[^_^]b joined #salt
02:38 john5223 joined #salt
02:38 coval3nce 2015-06-05 02:37:54,177 [salt.minion      ][WARNING ][9] Master ip address changed from 192.168.69.20 to 0.0.0.1
02:38 coval3nce hehe, that seems odd
02:39 d[^_^]b joined #salt
02:41 d[^_^]b joined #salt
02:42 beauby joined #salt
02:42 elektrix dingo: just curious...still working on those gists? or get distracted?
02:42 dergrunepunkt joined #salt
02:42 elektrix if so, that's cool, distractions happen
02:43 d[^_^]b joined #salt
02:43 dergrunepunkt hi, quick question, is the same use "{{ pillar.openstack.cinder_dbpass }}"  or "{{ pillar['openstack']['cinder_dbpass'] }} ?
02:45 d[^_^]b joined #salt
02:47 d[^_^]b joined #salt
02:49 d[^_^]b joined #salt
02:51 d[^_^]b joined #salt
02:53 d[^_^]b joined #salt
02:55 d[^_^]b joined #salt
02:56 mohae joined #salt
02:57 coval3nce dergrunepunkt: i think so yes -> http://docs.saltstack.com/en/latest/ref/renderers/all/salt.renderers.jinja.html#calling-salt-functions
02:57 dingo 02:42 < elektrix> dingo: just curious...still working on those gists? or get distracted?
02:57 dingo you missed it
02:57 d[^_^]b joined #salt
02:57 dergrunepunkt coval3nce: thanks
02:58 dingo 02:08 < dingo> py.test tests: https://gist.github.com/jquast/038c5400535a1276430d
02:58 dingo 02:08 < dingo> the supporting accessories.py module: https://github.eng.cloud.io.com/cloud/cloud-salt/blob/master/undercloud/tests/accessories.py
02:58 dingo i was also distracted :)
02:58 coval3nce hmm, anyone ever see multimaster change ip trying to go to some bogus fialover ip like 0.0.0.1?
02:58 Furao joined #salt
02:58 dingo its that pretty d[^_^]b pattern gunked up yer buffer
02:58 elektrix distractions happen :D
02:59 dingo in this case it was a female, you can understand my decision to distract
02:59 elektrix totally
02:59 d[^_^]b joined #salt
03:00 elektrix second link is broken...https://github.eng.cloud.io.com/cloud/cloud-salt/blob/master/undercloud/tests/accessories.py
03:00 dergrunepunkt guys, I can' t understand what' s going on here: http://pastebin.com/ky09xBM2
03:01 d[^_^]b joined #salt
03:01 iggy dergrunepunkt: I'd guess you don't have the access info correct
03:02 coval3nce roles looks like a list, type, which then has dicts and/or lists within that
03:03 dergrunepunkt if you look below the roles list is there
03:04 d[^_^]b joined #salt
03:04 iggy roles should be a dict
03:04 iggy not a list
03:04 iggy from the tiny bit of reading I've done
03:05 d[^_^]b joined #salt
03:06 favadi joined #salt
03:06 elektrix dingo: thanks for the gist...speaking of female distractions...mine is asking why I'm still on the computer when I have to get up for work in 6-ish hours...thanks for the gist...will look it over more closely in the morning
03:06 dergrunepunkt the key on the dict is 'service'
03:07 dergrunepunkt I'm so frustrated :(
03:07 iggy no, that's a list
03:07 d[^_^]b joined #salt
03:08 dergrunepunkt ok
03:09 d[^_^]b joined #salt
03:09 beauby joined #salt
03:10 mohae_ joined #salt
03:10 iggy look at the example in the docs, it doesnt' have -'s before service
03:11 iggy and require is not indented
03:11 iggy or is indented one level less
03:11 d[^_^]b joined #salt
03:13 coval3nce iggy: ever see something like this? “Master ip address changed from 192.168.69.20 to 0.0.0.1”
03:13 iggy coval3nce: bad DNS?
03:13 d[^_^]b joined #salt
03:13 coval3nce hmm, if ip’s are put into the master list, does it still try to use dns?
03:13 coval3nce lemme go see if i can find the code for that
03:14 iggy shouldn't I'd think
03:14 iggy multi-master isn't an area of expertise for me
03:14 mohae joined #salt
03:14 mohae_ joined #salt
03:15 d[^_^]b joined #salt
03:16 coval3nce https://github.com/saltstack/salt/blob/develop/salt/minion.py#L136  <- yeah maybe its dns related
03:16 coval3nce my code is inside of a docker container
03:17 d[^_^]b joined #salt
03:19 hasues joined #salt
03:19 d[^_^]b joined #salt
03:19 hasues left #salt
03:20 cberndt joined #salt
03:21 d[^_^]b joined #salt
03:22 mohae joined #salt
03:23 d[^_^]b joined #salt
03:24 aqua^c joined #salt
03:25 d[^_^]b joined #salt
03:25 dexter91x joined #salt
03:27 d[^_^]b joined #salt
03:29 d[^_^]b joined #salt
03:31 d[^_^]b joined #salt
03:34 d[^_^]b joined #salt
03:35 d[^_^]b joined #salt
03:37 d[^_^]b joined #salt
03:38 big_area joined #salt
03:39 d[^_^]b joined #salt
03:41 d[^_^]b joined #salt
03:42 sunkist1 joined #salt
03:43 d[^_^]b joined #salt
03:43 FarrisG_ joined #salt
03:45 d[^_^]b joined #salt
03:46 big_area joined #salt
03:47 numkem_ joined #salt
03:47 aqua^c joined #salt
03:47 d[^_^]b joined #salt
03:49 d[^_^]b joined #salt
03:50 joshin No, no ban. I wonder if nickserv has an email.
03:51 d[^_^]b joined #salt
03:53 d[^_^]b joined #salt
03:55 d[^_^]b joined #salt
03:57 d[^_^]b joined #salt
03:59 clintberry joined #salt
03:59 d[^_^]b joined #salt
04:01 d[^_^]b joined #salt
04:04 d[^_^]b joined #salt
04:05 d[^_^]b joined #salt
04:07 d[^_^]b joined #salt
04:09 d[^_^]b joined #salt
04:10 aqua^c joined #salt
04:10 murrdoc1 joined #salt
04:10 desposo joined #salt
04:11 d[^_^]b joined #salt
04:13 d[^_^]b joined #salt
04:15 d[^_^]b joined #salt
04:17 otter768 joined #salt
04:17 d[^_^]b joined #salt
04:19 d[^_^]b joined #salt
04:20 nikob joined #salt
04:21 d[^_^]b joined #salt
04:21 sunkist joined #salt
04:23 coval3nce iggy: https://github.com/saltstack/salt/blob/5eae6de6093fb878fc7004c0549f34955ca3c21e/salt/minion.py#L136  <- looks like in multimaster list, it calls dns_check with ‘1’ in my case for some reasons, e.g opts[‘master’] somehow gets set to ‘1’
04:23 d[^_^]b joined #salt
04:24 iggy well, at least you tracked it down
04:24 nikob Is there any reason a lot of formulas tend to use the grain 'os' instead of 'os_family' is this a convention/best practice thing? or should this really be asked of individual repos?
04:24 iggy sounds nasty though
04:25 coval3nce yeah makes no sense hehe
04:25 iggy nikob: it's a per formula thing mostly
04:25 nikob ah k
04:25 d[^_^]b joined #salt
04:26 coval3nce iggy: yeah before the call to salt.utils.dns_check, opts[‘master’] resolves to 1, so something is changing that value originally taken from th config file
04:26 nikob running centos and have had to update about 10 formulas which grain os==RedHat
04:26 sunkist1 joined #salt
04:26 nikob started wondering if I was doing something wrong
04:27 iggy nikob: the problem with os_family is that it can be slightly overbroad
04:27 nikob yeah i dig that
04:27 d[^_^]b joined #salt
04:28 iggy i.e. amazon linux shows as RedHat, but breaks a lot when used that way
04:28 nikob oh really? I kinda just assumed it was a Deb/Ubuntu thing
04:29 d[^_^]b joined #salt
04:30 L2SHO joined #salt
04:30 ramaseshan joined #salt
04:31 stoogenmeyer_ joined #salt
04:31 d[^_^]b joined #salt
04:31 sunkist joined #salt
04:34 d[^_^]b joined #salt
04:35 d[^_^]b joined #salt
04:36 sunkist1 joined #salt
04:37 d[^_^]b joined #salt
04:38 zer0def joined #salt
04:39 d[^_^]b joined #salt
04:40 cztanu hey folks, is there some way to have two pillar files, each with an element which has the same name, and have them both applied to a minion?
04:41 d[^_^]b joined #salt
04:43 jamesog joined #salt
04:43 d[^_^]b joined #salt
04:45 d[^_^]b joined #salt
04:47 d[^_^]b joined #salt
04:49 d[^_^]b joined #salt
04:51 d[^_^]b joined #salt
04:53 d[^_^]b joined #salt
04:55 d[^_^]b joined #salt
04:57 d[^_^]b joined #salt
04:59 d[^_^]b joined #salt
05:00 coval3nce iggy: i think https://github.com/saltstack/salt/blob/5eae6de6093fb878fc7004c0549f34955ca3c21e/salt/minion.py#L334 blows away the notion of the master list
05:01 d[^_^]b joined #salt
05:02 julez joined #salt
05:03 catpigger joined #salt
05:04 d[^_^]b joined #salt
05:05 d[^_^]b joined #salt
05:07 d[^_^]b joined #salt
05:08 iggy cztanu: no
05:08 iggy not reliably anyway
05:09 d[^_^]b joined #salt
05:11 d[^_^]b joined #salt
05:13 d[^_^]b joined #salt
05:15 impi joined #salt
05:15 d[^_^]b joined #salt
05:17 d[^_^]b joined #salt
05:17 mosen joined #salt
05:19 d[^_^]b joined #salt
05:21 d[^_^]b joined #salt
05:23 evle1 joined #salt
05:23 d[^_^]b joined #salt
05:25 d[^_^]b joined #salt
05:27 d[^_^]b joined #salt
05:28 borgstrom joined #salt
05:29 d[^_^]b joined #salt
05:30 TyrfingMjolnir joined #salt
05:31 d[^_^]b joined #salt
05:34 d[^_^]b joined #salt
05:35 d[^_^]b joined #salt
05:36 borgstrom joined #salt
05:37 d[^_^]b joined #salt
05:38 big_area joined #salt
05:39 d[^_^]b joined #salt
05:41 d[^_^]b joined #salt
05:43 d[^_^]b joined #salt
05:45 d[^_^]b joined #salt
05:47 d[^_^]b joined #salt
05:49 stoogenmeyer_ joined #salt
05:49 d[^_^]b joined #salt
05:51 d[^_^]b joined #salt
05:53 d[^_^]b joined #salt
05:53 rdas joined #salt
05:55 d[^_^]b joined #salt
05:57 d[^_^]b joined #salt
05:59 d[^_^]b joined #salt
06:01 AndreasLutro joined #salt
06:01 d[^_^]b joined #salt
06:02 flyboy joined #salt
06:02 colttt joined #salt
06:03 badon_ joined #salt
06:04 d[^_^]b joined #salt
06:05 Furao_ joined #salt
06:05 d[^_^]b joined #salt
06:07 d[^_^]b joined #salt
06:09 d[^_^]b joined #salt
06:10 golodhrim joined #salt
06:11 d[^_^]b joined #salt
06:12 Pulp joined #salt
06:13 d[^_^]b joined #salt
06:15 d[^_^]b joined #salt
06:17 d[^_^]b joined #salt
06:17 fllr joined #salt
06:18 otter768 joined #salt
06:19 d[^_^]b joined #salt
06:21 d[^_^]b joined #salt
06:23 Furao joined #salt
06:23 d[^_^]b joined #salt
06:25 d[^_^]b joined #salt
06:26 epcim_ joined #salt
06:27 d[^_^]b joined #salt
06:29 d[^_^]b joined #salt
06:31 d[^_^]b joined #salt
06:32 Grokzen joined #salt
06:34 d[^_^]b joined #salt
06:35 d[^_^]b joined #salt
06:36 ramaseshan_ joined #salt
06:37 d[^_^]b joined #salt
06:39 d[^_^]b joined #salt
06:39 aqua^c joined #salt
06:41 d[^_^]b joined #salt
06:43 d[^_^]b joined #salt
06:45 Furao joined #salt
06:45 d[^_^]b joined #salt
06:46 coval3nce https://github.com/saltstack/salt/issues/24434 and submitted a PR for it - salt-minion multi master failover to a bogus ip address
06:47 d[^_^]b joined #salt
06:49 ggoZ joined #salt
06:49 d[^_^]b joined #salt
06:51 d[^_^]b joined #salt
06:53 d[^_^]b joined #salt
06:55 d[^_^]b joined #salt
06:55 scuwolf joined #salt
06:57 d[^_^]b joined #salt
06:59 d[^_^]b joined #salt
07:00 julez joined #salt
07:01 d[^_^]b joined #salt
07:01 stephanbuys joined #salt
07:01 bhosmer joined #salt
07:02 gfa joined #salt
07:04 d[^_^]b joined #salt
07:05 d[^_^]b joined #salt
07:06 thalleralexander joined #salt
07:06 MatthewsFace joined #salt
07:07 d[^_^]b joined #salt
07:09 Auroch joined #salt
07:09 d[^_^]b joined #salt
07:11 d[^_^]b joined #salt
07:13 d[^_^]b joined #salt
07:14 favadi joined #salt
07:14 supersheep joined #salt
07:15 d[^_^]b joined #salt
07:16 karlthane joined #salt
07:17 d[^_^]b joined #salt
07:18 DrBojingle joined #salt
07:19 d[^_^]b joined #salt
07:20 saifi joined #salt
07:21 d[^_^]b joined #salt
07:23 d[^_^]b joined #salt
07:24 shantanoo joined #salt
07:25 d[^_^]b joined #salt
07:27 saifi joined #salt
07:27 d[^_^]b joined #salt
07:29 d[^_^]b joined #salt
07:31 d[^_^]b joined #salt
07:34 d[^_^]b joined #salt
07:34 shantanoo hi all
07:34 eseyman joined #salt
07:35 shantanoo i want to make sure that states in the formula are run in sequence provided in the formula
07:35 d[^_^]b joined #salt
07:35 shantanoo what should i use to make sure it works correctly. was trying 'require'. sample code http://pastebin.com/shHsqpQS
07:35 shantanoo is this correct?
07:37 babilen shantanoo: A service dead followed by a service running doesn't make sense and SLS files are evaluated top to bottom anyway
07:37 d[^_^]b joined #salt
07:37 babilen (as not both of those states can be true)
07:37 shantanoo babilen, am facing issue with service.restart
07:38 babilen are you?
07:38 markm joined #salt
07:38 babilen Could you elaborate on that? (and please do yourself and the rest of the internet a favour and don't use pastebin.com)
07:39 shantanoo babilen, getting "Comment: State 'service.restart' was not found in SLS 'demo'"
07:39 babilen Why would you want jenkins to be in a "restarting state" ? I'd argue that you want it either to be running or not running
07:39 shantanoo after installing the plugins, i need to restart the jenkins
07:39 d[^_^]b joined #salt
07:40 babilen Then define a watch_in on the service state in the state that installs the plugins. That'll restart the service automatically whenever there are changes to the watching states
07:41 d[^_^]b joined #salt
07:43 babilen (or a listen_in if you don't have to restart it immediately)
07:43 d[^_^]b joined #salt
07:44 babilen http://docs.saltstack.com/en/latest/ref/states/requisites.html discusses requisites between states and their effect
07:45 d[^_^]b joined #salt
07:47 d[^_^]b joined #salt
07:47 fredvd joined #salt
07:49 slav0nic joined #salt
07:49 d[^_^]b joined #salt
07:49 kawa2014 joined #salt
07:51 d[^_^]b joined #salt
07:53 d[^_^]b joined #salt
07:54 arnoldB joined #salt
07:55 d[^_^]b joined #salt
07:56 chiui joined #salt
07:56 impi joined #salt
07:57 d[^_^]b joined #salt
07:59 d[^_^]b joined #salt
08:01 d[^_^]b joined #salt
08:04 d[^_^]b joined #salt
08:05 IanVorn joined #salt
08:05 d[^_^]b joined #salt
08:07 d[^_^]b joined #salt
08:09 d[^_^]b joined #salt
08:09 s_kunk joined #salt
08:10 s_kunk_ joined #salt
08:11 d[^_^]b joined #salt
08:12 shantanoo joined #salt
08:12 shantanoo joined #salt
08:13 shantanoo babilen, checking requisites link
08:13 d[^_^]b joined #salt
08:14 s_kunk joined #salt
08:15 d[^_^]b joined #salt
08:15 shantanoo babilen, i was using 'require'. i suppose that is the correct approach.
08:17 KermitTheFragger joined #salt
08:17 Furao let me know when d[^_^]b don’t flood join the channel, bye
08:17 Furao left #salt
08:17 d[^_^]b joined #salt
08:18 babilen shantanoo: I prefer to not list all requisites with the service state, but to use "watch_in/listen_in: service: foo" in other states that require a restart of the service if they have changes.
08:19 otter768 joined #salt
08:19 d[^_^]b joined #salt
08:21 d[^_^]b joined #salt
08:23 d[^_^]b joined #salt
08:25 d[^_^]b joined #salt
08:26 babilen shantanoo: And the difference between watch* and require is that require simply changes the order in which states are executed, but that watch*/listen* trigger a restart of a service (which is, I gather, what you want)
08:27 d[^_^]b joined #salt
08:29 d[^_^]b joined #salt
08:31 d[^_^]b joined #salt
08:34 d[^_^]b joined #salt
08:34 linjan joined #salt
08:35 d[^_^]b joined #salt
08:35 fl0w0lf left #salt
08:36 arnoldB hm, salt bootstrapping using git is somehow broken now
08:36 arnoldB the init scripts are missing on debian
08:37 d[^_^]b joined #salt
08:39 d[^_^]b joined #salt
08:40 shantanoo joined #salt
08:40 shantanoo joined #salt
08:41 arnoldB the /debian folder is gone
08:41 d[^_^]b joined #salt
08:43 d[^_^]b joined #salt
08:45 d[^_^]b joined #salt
08:47 d[^_^]b joined #salt
08:49 d[^_^]b joined #salt
08:50 ajw0100 joined #salt
08:51 d[^_^]b joined #salt
08:53 d[^_^]b joined #salt
08:55 lloesche joined #salt
08:55 d[^_^]b joined #salt
08:56 babilen arnoldB: Gone from where?
08:56 denys joined #salt
08:57 arnoldB babilen: salt repo. => https://github.com/saltstack/salt/pull/23219
08:57 d[^_^]b joined #salt
08:59 d[^_^]b joined #salt
09:01 d[^_^]b joined #salt
09:02 babilen arnoldB: That hasn't been used for creating packages in a while
09:02 zerthimon joined #salt
09:03 arnoldB babilen: are you sure? https://github.com/saltstack/salt-bootstrap/blob/develop/bootstrap-salt.sh#L2511
09:03 babilen Okay, that is a bug in the bootstrap script then
09:04 d[^_^]b joined #salt
09:05 fllr joined #salt
09:05 babilen arnoldB: Do you want to file it?
09:05 d[^_^]b joined #salt
09:05 arnoldB babilen: already done: https://github.com/saltstack/salt-bootstrap/issues/607
09:06 babilen (fwiw, the init script wouldn't have been in line with the "official" one for quite a while I gather)
09:06 babilen I see that you are a man of many words :)
09:06 arnoldB well, why duplicating information?
09:07 arnoldB do you miss any information in the issue?
09:07 d[^_^]b joined #salt
09:09 sgargan joined #salt
09:09 d[^_^]b joined #salt
09:11 d[^_^]b joined #salt
09:13 d[^_^]b joined #salt
09:15 impi joined #salt
09:15 d[^_^]b joined #salt
09:15 favadi joined #salt
09:16 arnoldB the salt-minion fails to start due to an assertion error in tornado (debian wheezy, salt develop)
09:17 d[^_^]b joined #salt
09:19 d[^_^]b joined #salt
09:21 d[^_^]b joined #salt
09:23 d[^_^]b joined #salt
09:25 MatthewsFace joined #salt
09:25 supersheep joined #salt
09:25 d[^_^]b joined #salt
09:25 pcdummy joined #salt
09:27 d[^_^]b joined #salt
09:29 d[^_^]b joined #salt
09:30 hger joined #salt
09:31 d[^_^]b joined #salt
09:31 hger I have a problem with my master server, trying to run salt 'web*' state.highstate results in TypeError: traverse_dict_and_list() got an unexpected keyword argument 'delimiter'
09:32 arnoldB hger: do you have the full backtrace?
09:32 hger however if I do a salt-call from any client in that range (web*) it works no problem
09:33 hger the python trace is like so:
09:33 hger webdb02.esss.lu.se:     Data failed to compile: ----------     Traceback (most recent call last):   File "/usr/lib/python2.7/site-packages/salt/state.py", line 2874, in call_highstate     return False   File "/usr/lib/python2.7/site-packages/salt/state.py", line 2376, in get_top     return errors   File "/usr/lib/python2.7/site-packages/salt/state.py", line 2253, in get_tops     opts['state_auto_order'])   File "/usr/lib/python2.7/site-p
09:33 hger File "/usr/lib/python2.7/site-packages/salt/state.py", line 2874, in call_highstate     return False
09:33 hger File "/usr/lib/python2.7/site-packages/salt/state.py", line 2376, in get_top     return errors
09:34 hger File "/usr/lib/python2.7/site-packages/salt/state.py", line 2253, in get_tops     opts['state_auto_order'])
09:34 hger File "/usr/lib/python2.7/site-packages/salt/template.py", line 84, in compile_template     input_data = ret
09:34 d[^_^]b joined #salt
09:34 hger File "/usr/lib/python2.7/site-packages/salt/renderers/yaml.py", line 68, in render     if __salt__['config.get']('yaml_utf8', False):
09:34 hger File "/usr/lib/python2.7/site-packages/salt/modules/config.py", line 321, in get     delimiter=delimiter)
09:34 the_lalelu hger: please use a paste service
09:34 hger TypeError: traverse_dict_and_list() got an unexpected keyword argument 'delimiter'
09:35 bluenemo joined #salt
09:35 d[^_^]b joined #salt
09:36 favadi joined #salt
09:37 d[^_^]b joined #salt
09:38 arnoldB issue on GH for the tornado issue: https://github.com/saltstack/salt/issues/24440
09:39 hger Sorry about the trace spam
09:39 hger will use paste service next time
09:39 d[^_^]b joined #salt
09:40 arnoldB the trace is hard to read
09:40 hger Yes will make paste
09:40 arnoldB :) thx
09:41 arnoldB please also show the output of "salt-minion --versions"
09:41 keimlink joined #salt
09:41 d[^_^]b joined #salt
09:43 hger https://gist.github.com/anonymous/51547e9451b52da093bd
09:43 d[^_^]b joined #salt
09:45 hger salt-minion --versions from salt master: https://gist.github.com/hger/8434fd94ac03419bc11d
09:45 d[^_^]b joined #salt
09:46 Kelsar can i delete /var/cache/salt? will it be regenerated?
09:46 arnoldB Kelsar: rm -r /var/cache/salt/*
09:46 arnoldB Kelsar: never had problems with it
09:46 Kelsar arnoldB: yes, i am moving it to different device...
09:47 d[^_^]b joined #salt
09:47 matthew-parlette joined #salt
09:48 Kelsar My / was a "bit" full
09:49 matthew-1arlette joined #salt
09:49 d[^_^]b joined #salt
09:51 arnoldB hger: your give me the output of "grep -C3 'def traverse_dict_and_list' /usr/lib/python2.7/site-packages/salt/utils/__init__.py" ?
09:51 d[^_^]b joined #salt
09:51 arnoldB hger: the issue looks for me to being fixed by https://github.com/saltstack/salt/commit/cc30cd95aba9221a39a70f3bc1ecda21e4d72d04
09:52 hger this line is in the output def traverse_dict_and_list(data, key, default, delimiter=DEFAULT_TARGET_DELIM):
09:53 d[^_^]b joined #salt
09:54 arnoldB hger: weird. could it be true that you have several salt versions installed on your system?
09:55 arnoldB hger: python says it doesn't know the argument "delimiter", but you see it's in the function definition
09:55 hger yes
09:55 d[^_^]b joined #salt
09:56 hger I have these packages installed: salt-2015.5.0-1.el6.noarch, salt-minion-2015.5.0-1.el6.noarch and salt-master-2015.5.0-1.el6.noarch
09:56 hger bad? good?
09:57 hger need to go for lunch will be back in an hour. thanks for your help!
09:57 d[^_^]b joined #salt
09:58 favadi joined #salt
09:59 d[^_^]b joined #salt
10:01 d[^_^]b joined #salt
10:04 d[^_^]b joined #salt
10:05 d[^_^]b joined #salt
10:07 d[^_^]b joined #salt
10:09 CeBe joined #salt
10:09 d[^_^]b joined #salt
10:11 d[^_^]b joined #salt
10:13 d[^_^]b joined #salt
10:15 Sacro can one of you ops ban hi8m
10:15 d[^_^]b joined #salt
10:15 Sacro basepi: do something about d[^_^]b
10:17 d[^_^]b joined #salt
10:18 favadi we should kick d[^_^]b
10:19 Sacro Maybe if we highlight d[^_^]b enough he can fix things
10:19 babilen This channel needs more ops
10:19 d[^_^]b joined #salt
10:19 babilen (in different time zones)
10:20 otter768 joined #salt
10:21 pcdummy joined #salt
10:21 pcdummy joined #salt
10:21 d[^_^]b joined #salt
10:22 keimlink joined #salt
10:22 babilen whiteinge, herlo, SEJeff, thatch45, Utah_Dave, basepi, cro, terminalmage, cachedout, rallytime: d[^_^]b join/part flood. Please run "/mode +b *!ae@2600:3c03::f03c:91ff:fe6e:5ed9$##fix_your_connection" to mitigate the problem
10:23 arnoldB what's the matter? are these users PM-spamming?
10:23 arnoldB ah, well. irssi is ignoring joins/parts
10:23 arnoldB :)
10:23 d[^_^]b joined #salt
10:24 babilen No, just part/join flood and some people (like me) don't configure their clients to ignore those. I colour actions like those in a very dark grey so they can be easily ignored (normal text is white)
10:24 babilen I'm op in #debian* so I can't ignore them (nor do I have to as they really don't distract me)
10:25 d[^_^]b joined #salt
10:27 d[^_^]b joined #salt
10:29 d[^_^]b joined #salt
10:31 d[^_^]b joined #salt
10:34 d[^_^]b joined #salt
10:35 d[^_^]b joined #salt
10:36 FarrisG joined #salt
10:37 d[^_^]b joined #salt
10:38 julez joined #salt
10:39 d[^_^]b joined #salt
10:40 hal58th joined #salt
10:41 d[^_^]b joined #salt
10:43 d[^_^]b joined #salt
10:45 d[^_^]b joined #salt
10:46 kruppm joined #salt
10:47 d[^_^]b joined #salt
10:49 d[^_^]b joined #salt
10:51 d[^_^]b joined #salt
10:52 shantanoo joined #salt
10:53 d[^_^]b joined #salt
10:55 d[^_^]b joined #salt
10:57 d[^_^]b joined #salt
10:59 d[^_^]b joined #salt
11:01 Nazca joined #salt
11:01 d[^_^]b joined #salt
11:04 d[^_^]b joined #salt
11:04 kruppm is there a way to tell salt to ignore symlinks when *replacing* files via file.managed? "ignore_symlinks: false" does not seem to have any effect when the symlink is *dead*, eg. the target does not exist
11:05 d[^_^]b joined #salt
11:06 kruppm running version 2015.5 here
11:06 Nazca__ joined #salt
11:07 d[^_^]b joined #salt
11:09 d[^_^]b joined #salt
11:11 d[^_^]b joined #salt
11:13 aberdine Is it possible to have gitfs salt-formula span multiple environments with creating a branch for each?
11:13 Nazca joined #salt
11:13 aberdine or should I just create a branch for each?
11:13 d[^_^]b joined #salt
11:13 giantlock joined #salt
11:15 d[^_^]b joined #salt
11:17 d[^_^]b joined #salt
11:18 bluenemo joined #salt
11:19 d[^_^]b joined #salt
11:21 babilen aberdine: use base?
11:21 scuwolf joined #salt
11:21 d[^_^]b joined #salt
11:22 scuwolf joined #salt
11:23 scuwolf joined #salt
11:23 d[^_^]b joined #salt
11:24 aberdine is that included in all environments?
11:25 aberdine and can I override parts of base in dev, e.g. pillar items?
11:25 d[^_^]b joined #salt
11:26 babilen yes, you'd have additional branches for the actual environments
11:26 babilen "base" is "global"
11:27 babilen (if you put all boxes in that environment that is)
11:27 aberdine can a box exist in two environments?
11:27 babilen In the end it simply boils down to how you target in your top.sls
11:27 babilen sure
11:27 babilen Let me find the example in the docs
11:27 d[^_^]b joined #salt
11:27 aberdine thanks
11:28 babilen http://docs.saltstack.com/en/latest/ref/states/top.html#environments -- if you scroll a bit down to the example that starts with "In this setup we have state trees in three of the four environments, ..."
11:28 babilen States in * will be targeted to every minion and are therefore global
11:29 Ztyx joined #salt
11:29 aberdine right. I think the problem I'm having is related to extra_filerefs then, rather than envs
11:29 d[^_^]b joined #salt
11:29 aberdine whats the salt path to a gitfs mount?
11:29 babilen The idea would then be to have formulas in base and then devide boxes into their actual environments... Please make sure that you use a dedicated repository for your top.sls file if you use GitFS (as they are merged across environments)
11:30 babilen It's somewhere in /var/cache/salt/ ...
11:30 aberdine so base is speclal?
11:31 aberdine uh oh, I branched my question - I'll come back to the salt path thing
11:31 denys joined #salt
11:31 d[^_^]b joined #salt
11:31 babilen Not really ... The only thing that is special about "base" is that the "master" branch is mapped to it in GitFS. The *entire* targeting is handled in top.sls and it really is up to you how distinct you want your environments to be.
11:32 aberdine I'm still not clear if a server in dev sees formula in base
11:32 babilen It only would if you target it in your top.sls
11:33 aberdine so the same server in both envs?
11:33 babilen You might understand it a bit better if you ask yourself the question "What puts a minion into a specific environment?"
11:33 aberdine the top.sls file?
11:34 aberdine and the matching done there?
11:34 d[^_^]b joined #salt
11:34 babilen And if you plan to use "base" (i.e. the "master" branch in git) for states that should be available to all minions you will have to target these states to *all* minions
11:34 aberdine right
11:35 babilen Hence the "base: '*': - state_that_is_available_to_every_minion" idiom in there
11:35 aberdine then adjust pillars accordingly?
11:35 donmichelangelo joined #salt
11:35 babilen indeed
11:35 d[^_^]b joined #salt
11:35 aberdine I've got it, I think. At least enough to try a few things out
11:36 aberdine regarding the salt:// uri - is the path element the full path to the gitfs mount?
11:37 babilen One problem you'll face is that "state that is available" and "state that is being run" is not easy to distinguish in salt. If you target something to every minion it will be applied to every minion. That is fine for formulas that don't do anything unless you explicitly provide applicable pillar data, but you might not want to install everything everywhere.
11:37 hger on the salt master should al of these packages be installed?salt-2015.5.0-1.el6.noarch, salt-minion-2015.5.0-1.el6.noarch and salt-master-2015.5.0-1.el6 ? the minion should be there.
11:37 d[^_^]b joined #salt
11:37 babilen That being said: You can still target more intelligently in your base environment (e.g. "base: '*-database': - database_state")
11:37 aberdine in which case I need a branch in the gitfs repo?
11:38 aberdine if I want to apply a gitfs repo to only one env? or remap base?
11:38 aberdine afaik for a few minutes
11:39 sgargan joined #salt
11:39 d[^_^]b joined #salt
11:39 babilen Branches in a git repo are, first and foremost, just that and are being used for merging changes from one to the other in order to introduce some "testing" method. You could, for example, start making changes to your states in the "dev" branch with the goal of, ultimately, merging those changes to the "prod" branch ...
11:40 babilen Minions to which you target the "dev" environment would then start using that version of the state and would, thereby, constitute your minion group on which you can test things before merging it into "prod" which includes minions that are "in production"
11:41 d[^_^]b joined #salt
11:42 babilen In the end you start with one branch (master / base) and branch of there. Depending on your git development philosophy/methodology you would then introduce multiple other branches for developing and testing changes. GitFS and salt simply gives you the ability to selectively target things to your minions
11:43 d[^_^]b joined #salt
11:43 babilen The problem is that environments in salt and the way GitFS treats them are different and GitFS conflates this by assuming that "branch == environment" which means that all environments start out with an identical set of states
11:44 babilen That might not be what you want and the necessity to needlessly maintain those branches in your formulas is one of the downsides of that scheme
11:45 babilen What *salt* cares about are two things: 1. file_roots (i.e. where are states located that belong to environment FOO) and 2. Which states are being sent to which minion (top.sls)
11:45 d[^_^]b joined #salt
11:45 babilen s/sent/applied
11:45 ggoZ joined #salt
11:47 d[^_^]b joined #salt
11:48 babilen What I would recommend if you like to use GitFS is: One repository for top.sls files, one for the "waterfall/testing environments" with dev, qa and master/base in there and, finally, one distinct repo for states that should be globally available (optionally with a waterfall setup as well)
11:48 babilen That way you don't conflate branches for testing/merging things and targeting
11:49 d[^_^]b joined #salt
11:50 aberdine and the environments in pillar and states are seperate?
11:51 aberdine That is, I can have a minion in prod in states and in dev in pillar?
11:51 d[^_^]b joined #salt
11:51 babilen That entirely boils down to your targeting in the pillar top.sls (once again with distinct repositories)
11:52 aberdine so having different data depending on environment but still the same state branch (e.g. salt formula)
11:52 seblu joined #salt
11:52 aberdine yes, I think I see it now
11:52 aberdine thanks for that babilen, most useful
11:53 d[^_^]b joined #salt
11:53 babilen aberdine: The problem is primarily that "environments" are simply a concept and it doesn't help that GitFS conflates some of the distinct applications for which they could be used with "vanilla salt".
11:54 babilen There is also the entire question of how pillars and states are merged or how one overrides the other and so on
11:54 venom joined #salt
11:55 babilen I'd play a little with it and get a feeling for it. We use a slightly simplified version of the layout I sketched above (no "qa") and I'm happy with it, but people who are new on the team need a bit of "mulling over the documentation" before they arrive at the "that's trivial" stage :)
11:55 vahtangsuperboos joined #salt
11:55 vahtangsuperboos Hi there !
11:55 aberdine yes, that the override issue I'm looking at soon. I'm used to hiera and deep merge of the trees in that, and I understand that salt doesn't so that yet
11:55 vahtangsuperboos is it possible to run calls on salt-master via python and virtual environment  ?
11:55 d[^_^]b joined #salt
11:55 vahtangsuperboos seems like this command running but does not return anything just empty []
11:56 babilen aberdine: pillars are "smartly" merged
11:56 vahtangsuperboos runner = salt.runner.RunnerClient(opts)
11:56 vahtangsuperboos minions_list = runner.cmd('manage.up', [])
11:56 aberdine babilen I'm thinking of just dev & master for now, qa later, or perhaps directly from master
11:56 vahtangsuperboos return minions_list
11:56 babilen http://docs.saltstack.com/en/latest/ref/configuration/master.html#pillar-source-merging-strategy
11:56 vahtangsuperboos it is just a part of python function
11:57 aberdine I kind of like my version control de-coupled from my prod systems
11:57 d[^_^]b joined #salt
11:57 aberdine what about state precedence babilen >
11:57 aberdine ?
11:58 babilen aberdine: I work in git all the time which is why I simply have a "dev" branch/environment in which all my "testboxes" (be it actual minions running in the cloud or vagrant boxes) are in.
11:59 babilen aberdine: You can't target the same state (by id) to the same minion twice (which severely restricts you in the "minion in multiple environments" department)
11:59 yawniek whats the general pattern to get the ipv4 interface of the first interface thats not the loopback?
11:59 yawniek if the interface name is not known
11:59 d[^_^]b joined #salt
11:59 babilen yawniek: You really shouldn't care about interfaces
11:59 aberdine babilen which gets chosen if it exists in both?
12:00 yawniek babilen: ok i need the ip address for a specific network (to be saved in a config file of a server app)
12:00 babilen aberdine: "ERROR! ID CRASH! BAD SYSOP" (i.e. salt refuses to work)
12:00 Ztyx yawniek: Have a look at the "fqdn_ip4" grains. It works for me.
12:00 babilen yawniek: one second
12:01 aberdine babilen sound like I'll learn that one pretty quickly :)
12:01 d[^_^]b joined #salt
12:01 babilen yawniek: https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.network.html#salt.modules.network.ip_addrs is what I'd use. Define a suitable cidr range and find the actual address in the network you care about that way
12:02 babilen yawniek: We actually take it a bit further and have salt mine function aliases using that execution function for all "interesting" networks and are therefore able to retrieve minion ip addresses on every minion
12:03 julez joined #salt
12:04 d[^_^]b joined #salt
12:05 yawniek babilen: cool, thanks
12:05 d[^_^]b joined #salt
12:06 IanVorn joined #salt
12:07 d[^_^]b joined #salt
12:07 babilen yawniek: https://www.refheap.com/102119
12:09 yawniek :y: looks interesting
12:09 d[^_^]b joined #salt
12:10 babilen We target those mine function aliases to every minion.. in that example I defined two mine function aliases (for two private ip4 networks) and use that in a managed configuration file later on. That particular example retrieves the address of the minion itself (hence the "grain['id']" target), but you could target any other id (or all the targeting really)
12:11 babilen http://docs.saltstack.com/en/latest/topics/mine/ has some information on that
12:11 babilen Background to this is that you *really* don't want to start hardcoding interfaces in your setup.
12:11 babilen (unless they *really* are important)
12:11 d[^_^]b joined #salt
12:11 yawniek babilen: for my case i might go and get the network via libcloud or salt-cloud as we run on aws
12:12 yawniek absolutely agree, especially with archlinux where if names change often
12:12 babilen We had one customer who insisted that their public address had to be on eth1 because that hardcoded that all over the place ... that lead to a number of boxes with eth1 but no eth0 (which would be the private one) and other horrible things like that
12:13 babilen yawniek: Sure, in the end all you need is the cidr mask of the network you care about. You can even get that dynamically from grains (if your cloud provider provides that information already)
12:13 d[^_^]b joined #salt
12:14 babilen Once you have that you can use the network.ip_addrs execution function to ask questions such as "give me the addresses of all minions matching TARGET_EXPRESSION in network CIDRMASK
12:15 shantanoo joined #salt
12:15 d[^_^]b joined #salt
12:16 aberdine where can I find docs on salt:// uris?
12:16 aberdine searching it removes the ://
12:17 d[^_^]b joined #salt
12:17 markm_ joined #salt
12:18 sunkist joined #salt
12:18 babilen aberdine: There isn't much magic to them .. what do you want to know?
12:19 aberdine what form the paths are and any restrictions
12:19 d[^_^]b joined #salt
12:19 ponpanderer joined #salt
12:19 aberdine specifically, referencing a file coming in from gitfs
12:20 aberdine do I need to give salt:///var/cache/....
12:20 babilen oh .. oh ... noooo
12:20 giouna joined #salt
12:20 aberdine (it's to work around a bug with maps and salt-ssh)
12:20 babilen GitFS maps the repository "root" as salt:///
12:20 otter768 joined #salt
12:21 aberdine that's not working for me
12:21 aberdine it did when I used roots
12:21 babilen roots?
12:21 aberdine not with gitfs and salt-ssh
12:21 babilen You mean file_roots ?
12:21 aberdine yes
12:21 babilen Is that a salt-ssh specific bug?
12:21 d[^_^]b joined #salt
12:21 keimlink joined #salt
12:21 aberdine yes
12:22 aberdine https://github.com/saltstack/salt/issues/9878
12:22 aberdine I have   extra_filerefs:
12:22 aberdine - salt://salt/formulas.jinja
12:22 aberdine in my Saltfile
12:22 aberdine https://github.com/saltstack-formulas/salt-formula is the gitfs
12:23 aberdine am I missing a /?
12:23 aberdine hmm
12:23 sunkist1 joined #salt
12:23 aberdine no, that didn't help
12:23 d[^_^]b joined #salt
12:23 giouna i am trying to load specific values from a pillar with if statemnets. Inside if statements i get a custom grain (called role) and according to its value it shall load the value or not
12:23 giouna http://jpst.it/zlhZ
12:24 kruppm joined #salt
12:24 giouna i pasted an example in the above link
12:24 pannon joined #salt
12:24 giouna thing is that if the "role" is not the first word inside the if statement its not working
12:25 giouna {% if ("app" or "front" or "back") in salt['grains.get']('role') %}
12:25 d[^_^]b joined #salt
12:26 giouna for example if the minios grain "role" is set to "front" the above will not work
12:26 giouna if it is app , it then works
12:26 giouna is this a bug or am i doing something wrong?
12:27 AndreasLutro aberdine: try clearing the cache both locally and remotely
12:27 d[^_^]b joined #salt
12:28 AndreasLutro I use the extra_filerefs thing and it seems to work fine
12:28 aberdine How do I clear the cache? manual rm it?
12:29 AndreasLutro yeah, I think by default it's in /var/cache/salt locally, and somewhere in /tmp remotely
12:29 AndreasLutro /tmp/salt_root or something
12:29 d[^_^]b joined #salt
12:31 d[^_^]b joined #salt
12:34 d[^_^]b joined #salt
12:35 d[^_^]b joined #salt
12:37 d[^_^]b joined #salt
12:39 d[^_^]b joined #salt
12:41 d[^_^]b joined #salt
12:42 shantanoo_ joined #salt
12:42 babilen giouna: You want "foo in bar" or "baz in bar" or .....
12:43 d[^_^]b joined #salt
12:45 d[^_^]b joined #salt
12:47 d[^_^]b joined #salt
12:47 pfallenop joined #salt
12:48 aberdine AndreasLutro I deleted the cache. For some reason the gitfs cache only contains a partial copy of the full remote repo
12:48 warthog42 joined #salt
12:49 AndreasLutro hm, well, I'm not sure how gitfs works over salt-ssh
12:49 babilen aberdine: Yes, it only checks out what it needs
12:49 d[^_^]b joined #salt
12:49 aberdine ah
12:50 aberdine I think that's the problem
12:50 aberdine it doesn't know about the map
12:50 aberdine as it's refered to via a template
12:51 Steven- joined #salt
12:51 d[^_^]b joined #salt
12:52 jdesilet joined #salt
12:53 aberdine I'll file a bug and see if adding an option to force a full clone is viable
12:53 d[^_^]b joined #salt
12:54 babilen Just clone it somewhere? GitFS is supposed to work like that
12:55 d[^_^]b joined #salt
12:57 d[^_^]b joined #salt
12:57 aberdine yes, I get that, but where a file is reference in a jinja2 template like https://github.com/saltstack-formulas/salt-formula/blob/master/salt/formulas.sls#L4 it's not cloning the referenced file
12:58 wopwop joined #salt
12:59 d[^_^]b joined #salt
13:01 d[^_^]b joined #salt
13:01 aberdine actually, it might be my mistake in my saltfile
13:04 d[^_^]b joined #salt
13:04 hojgaard joined #salt
13:05 d[^_^]b joined #salt
13:06 giantlock joined #salt
13:06 brandonk joined #salt
13:07 aberdine fixed my saltfile, still no joy
13:07 rojem joined #salt
13:07 d[^_^]b joined #salt
13:09 d[^_^]b joined #salt
13:10 rdxc joined #salt
13:11 aberdine It can't be the case that {% from "salt/map.jinja" import salt_settings with context %} doesn't work with gitfs, can it?
13:11 d[^_^]b joined #salt
13:12 aberdine the error I get is http://pastebin.com/g9kJh34h
13:13 sunkist joined #salt
13:13 d[^_^]b joined #salt
13:14 IanVorn joined #salt
13:15 d[^_^]b joined #salt
13:17 racooper joined #salt
13:17 d[^_^]b joined #salt
13:18 sunkist1 joined #salt
13:19 d[^_^]b joined #salt
13:20 bhosmer joined #salt
13:20 supersheep joined #salt
13:21 d[^_^]b joined #salt
13:21 subsignal joined #salt
13:23 sunkist joined #salt
13:23 shantanoo joined #salt
13:23 cpowell joined #salt
13:23 d[^_^]b joined #salt
13:25 _JZ_ joined #salt
13:25 primechuck joined #salt
13:25 d[^_^]b joined #salt
13:26 dyasny joined #salt
13:27 d[^_^]b joined #salt
13:28 sunkist1 joined #salt
13:29 ggoZ joined #salt
13:29 d[^_^]b joined #salt
13:30 perfectsine joined #salt
13:31 aberdine I filled https://github.com/saltstack/salt/issues/24449
13:31 d[^_^]b joined #salt
13:31 Auroch joined #salt
13:32 brandonk Having some issues with saltstack creating raids properly. Specifically I am trying to use the raid.present function in mdadm.py module as detailed here http://docs.saltstack.com/en/latest/ref/states/all/salt.states.mdadm.html
13:33 brandonk it seems, however, that this function simply does not exist?
13:33 AndreasLutro brandonk: it's mdadm.present, not raid.present
13:34 d[^_^]b joined #salt
13:34 amcorreia joined #salt
13:35 AndreasLutro oh nevermind, I didn't read the code examples properly
13:35 brandonk I'll give that a shot but the module contains __virtualname__ = 'raid'
13:35 AndreasLutro yes indeed
13:35 d[^_^]b joined #salt
13:36 AndreasLutro https://github.com/saltstack/salt/blob/develop/salt/states/mdadm.py#L42-L45 as you can see here it has a couple of other prerequisites
13:36 AndreasLutro if any of those fail, it would probably prevent the module from working
13:37 brandonk i've run which mdadm on the system and it resolves a path and it is an AWS Amazon Linux AMI so the kernel is Linux
13:37 d[^_^]b joined #salt
13:37 LotR joined #salt
13:38 brandonk What troubles me is I just don't even see the "present" function anywhere in that python code
13:38 AndreasLutro salt's which function is its own implementation, could be a bug there
13:38 AndreasLutro https://github.com/saltstack/salt/blob/develop/salt/states/mdadm.py#L49 ?
13:39 d[^_^]b joined #salt
13:41 brandonk interesting - that is not present in the version I am looking at having cloned the repo and checked out to HEAD
13:41 d[^_^]b joined #salt
13:41 brandonk within salt/modules/mdadm.py
13:42 julez joined #salt
13:42 AndreasLutro well, you want salt/states/mdadm.py
13:42 mpanetta joined #salt
13:43 sgargan joined #salt
13:43 d[^_^]b joined #salt
13:43 brandonk well thats good to know thanks
13:44 drawsmcgraw joined #salt
13:45 AndreasLutro brandonk: could you maybe share the full error message you get?
13:45 d[^_^]b joined #salt
13:45 brandonk yeah let me try to make another run at it I'll get back to you in a few. Thanks for the help
13:47 d[^_^]b joined #salt
13:47 john5223 joined #salt
13:48 shantanoo_ joined #salt
13:49 d[^_^]b joined #salt
13:51 clintberry joined #salt
13:51 d[^_^]b joined #salt
13:52 shantanoo__ joined #salt
13:52 emaninpa joined #salt
13:53 d[^_^]b joined #salt
13:55 d[^_^]b joined #salt
13:57 d[^_^]b joined #salt
13:59 d[^_^]b joined #salt
14:01 d[^_^]b joined #salt
14:01 timoguin joined #salt
14:03 andrew_v joined #salt
14:04 d[^_^]b joined #salt
14:05 rm_jorge joined #salt
14:05 d[^_^]b joined #salt
14:06 impi joined #salt
14:06 timoguin joined #salt
14:07 anotherZero joined #salt
14:07 d[^_^]b joined #salt
14:08 cpowell_ joined #salt
14:09 d[^_^]b joined #salt
14:10 hger joined #salt
14:11 hger Anyone have problems with salt-minion-2015.5.0-1.el7.noarch or salt-2015.5.0-1.el7.noarch package on CentOS7?
14:11 d[^_^]b joined #salt
14:12 eper joined #salt
14:12 hger if I issue a salt '*' state.highstate it fails on CentOS7 minions and works on CentOs6/ScientificLinux6
14:13 eper hello, is there a method using the python salt client API when doing queries which also returns the disconnected clients, similar to if I run it via CLI with show_timeout ?
14:13 iggy all the same minion version? What master version?
14:13 d[^_^]b joined #salt
14:14 hger on centos6 salt-minion-2015.5.0-1.el6.noarch and salt-2015.5.0-1.el6.noarch
14:14 hger salt-master-2015.5.0-1.el6.noarch on master
14:15 iggy what's the error?
14:15 hger will paste trace
14:15 hger https://gist.github.com/anonymous/51547e9451b52da093bd
14:15 d[^_^]b joined #salt
14:16 hger difference is python version, 2.6.6 on centos6 and 2.7.5 on centos7
14:17 d[^_^]b joined #salt
14:18 hger got this tip: grep -C3 'def traverse_dict_and_list' /usr/lib/python2.7/site-packages/salt/utils/__init__.py
14:18 hger returned this: def traverse_dict_and_list(data, key, default, delimiter=DEFAULT_TARGET_DELIM):
14:19 d[^_^]b joined #salt
14:19 hger on both centos6 (with slight change to python2.6 in path) as well as on centos7
14:20 timoguin joined #salt
14:21 iggy yeah, seems weird, salt.modules.config calls it once with delimiter as a positional arg and then a couple lines later as a kw arg
14:21 d[^_^]b joined #salt
14:21 otter768 joined #salt
14:21 iggy what does the state look like?
14:22 iggy it seems like maybe you have something exacerbating the issue
14:22 hger so i setup a new setup with a master and two minions and installed old versions of salt packages. worked. after upgrade of packages no longer working
14:23 iggy otherwise there would be widespread reports of this issue and I've never seen anything like it
14:23 codehotter how do I remove multiple files in one go? (for ex, with a glob?)
14:23 d[^_^]b joined #salt
14:23 hger Ok I will investigate futher, thanks for your help!
14:24 iggy hger: can you paste the state that's failing?
14:24 bluenemo joined #salt
14:24 bluenemo joined #salt
14:25 hger it fails immediatily but it works if I do a salt-call state.highstate from the client
14:25 hger and on the second system there are perhaps three separate states. In production I have approx 50
14:25 d[^_^]b joined #salt
14:26 Pulp joined #salt
14:27 hger anyway, need to go, thanks for your help
14:27 iggy well shit, that was a crappy ending
14:27 d[^_^]b joined #salt
14:28 ek6 feeling unsatisfied?
14:29 d[^_^]b joined #salt
14:31 d[^_^]b joined #salt
14:32 bhosmer joined #salt
14:34 d[^_^]b joined #salt
14:35 illern joined #salt
14:35 d[^_^]b joined #salt
14:37 d[^_^]b joined #salt
14:39 d[^_^]b joined #salt
14:40 MTecknology d[^_^]b: having some connection issues?
14:40 MTecknology for a lot of days...
14:41 sandah joined #salt
14:41 d[^_^]b joined #salt
14:41 dayid joined #salt
14:43 d[^_^]b joined #salt
14:44 coval3nce_ joined #salt
14:44 DammitJim joined #salt
14:45 d[^_^]b joined #salt
14:46 kawa2014 joined #salt
14:46 malinoff joined #salt
14:46 asoc I have written a custom grain that I have in a file_root for a specific node group. I assumed that only the machines in that nodegroup would get that grain but when I sync all the grains it adds that grain to all machines.
14:46 asoc Is this normal behavior?
14:47 Heartsbane {% set HOSTNAME=salt['grains.get']('fqdn') %}
14:47 Heartsbane sorry wrong window :(
14:47 d[^_^]b joined #salt
14:47 * Heartsbane hasn't finished his coffee yet.
14:49 TheoSLC joined #salt
14:49 d[^_^]b joined #salt
14:50 TheoSLC Greetings.  I'm messing with the new Slack modules.  The only thing not working for me is the slack returner.  Does anybody know how to get that to work?
14:51 ferbla joined #salt
14:51 d[^_^]b joined #salt
14:52 Brew joined #salt
14:52 Heartsbane But can I match a wildcard against a group of domains like this: {% if 'demoserver20?.docker.organization.com' in HOSTNAME %}
14:53 Heartsbane ?
14:53 d[^_^]b joined #salt
14:55 jalbretsen joined #salt
14:55 d[^_^]b joined #salt
14:56 debian112 joined #salt
14:57 d[^_^]b joined #salt
14:58 kawa2014 joined #salt
14:59 d[^_^]b joined #salt
15:01 d[^_^]b joined #salt
15:04 d[^_^]b joined #salt
15:05 d[^_^]b joined #salt
15:07 Heartsbane Guess I missed a curly bracket, please ignore
15:07 d[^_^]b joined #salt
15:08 * Heartsbane needs stronger coffee, something with a extra shot of expresso.
15:09 d[^_^]b joined #salt
15:10 malinoff Heartsbane, https://en.wikipedia.org/wiki/Espresso
15:10 viq joined #salt
15:11 bhosmer joined #salt
15:11 d[^_^]b joined #salt
15:12 asoc Ok this is getting even weirder. I can run a highstate that calls the state in the file root but if I run a state.apply with the same state it tells me that there are no matching sls found in env 'base'
15:13 asoc do I have to specify the env that it should look at when running the state.apply?
15:13 asoc I thought that it should do that automagically
15:13 d[^_^]b joined #salt
15:13 coval3nce_ joined #salt
15:14 DammitJim I'm looking at the remote execution tutorial and I see tomcat as a function; however, in ubuntu, I think it is called tomcat7
15:15 DammitJim what do I need to do to update those functions? or do I need to create brand new ones? like something as simple as tomcat7.status
15:15 rojem joined #salt
15:15 d[^_^]b joined #salt
15:17 d[^_^]b joined #salt
15:19 d[^_^]b joined #salt
15:21 d[^_^]b joined #salt
15:21 giantlock joined #salt
15:23 d[^_^]b joined #salt
15:25 d[^_^]b joined #salt
15:27 riftman joined #salt
15:27 d[^_^]b joined #salt
15:29 d[^_^]b joined #salt
15:30 aparsons joined #salt
15:31 d[^_^]b joined #salt
15:32 spiette joined #salt
15:34 d[^_^]b joined #salt
15:35 d[^_^]b joined #salt
15:36 thedodd joined #salt
15:37 aparsons joined #salt
15:37 d[^_^]b joined #salt
15:39 d[^_^]b joined #salt
15:40 programmerq joined #salt
15:41 dfgd joined #salt
15:41 tmclaugh[work] joined #salt
15:41 d[^_^]b joined #salt
15:43 mordonez joined #salt
15:43 d[^_^]b joined #salt
15:45 d[^_^]b joined #salt
15:46 jdesilet joined #salt
15:47 d[^_^]b joined #salt
15:48 ajw0100 joined #salt
15:49 dalexander joined #salt
15:49 pgoetz joined #salt
15:49 d[^_^]b joined #salt
15:49 DammitJim I'm looking at the remote execution tutorial and I see tomcat as a function; however, in ubuntu, I think it is called tomcat7
15:49 DammitJim what do I need to do to update those functions? or do I need to create brand new ones? like something as simple as tomcat7.status
15:50 pgoetz Anyone know how to report typos in the documentation?  I'm reading through the States tutorial and have found quite a lot of them.
15:51 d[^_^]b joined #salt
15:52 pgoetz joined #salt
15:52 linuxdaemon joined #salt
15:52 tkharju joined #salt
15:53 d[^_^]b joined #salt
15:55 d[^_^]b joined #salt
15:55 kawa2014 joined #salt
15:57 mohae joined #salt
15:57 pgoetz_ joined #salt
15:57 tmclaugh[work] joined #salt
15:57 d[^_^]b joined #salt
15:57 riftman joined #salt
15:58 7GHAAEAI0 joined #salt
15:59 d[^_^]b joined #salt
16:01 d[^_^]b joined #salt
16:02 brandonk having some serious issues using orchestration with mdadm (raid.present)
16:03 bhosmer joined #salt
16:04 d[^_^]b joined #salt
16:05 d[^_^]b joined #salt
16:06 desposo joined #salt
16:07 d[^_^]b joined #salt
16:08 hasues joined #salt
16:08 hasues left #salt
16:08 rojem joined #salt
16:09 peters-tx joined #salt
16:09 flebel joined #salt
16:09 d[^_^]b joined #salt
16:11 tfield joined #salt
16:11 d[^_^]b joined #salt
16:12 smcquay joined #salt
16:12 basepi babilen: is that command a timed ban? That's what I'd like to do is just do a temporary ban, not a permanent ban...
16:12 smcquay joined #salt
16:13 d[^_^]b joined #salt
16:14 MTecknology not without a bot to manage it
16:15 MTecknology but there is *!*@2600:3c03::f03c:91ff:fe6e:5ed9$##FIX_YOUR_CONNECTION
16:15 d[^_^]b joined #salt
16:16 clintberry joined #salt
16:17 d[^_^]b joined #salt
16:18 clintber_ joined #salt
16:19 d[^_^]b joined #salt
16:21 d[^_^]b joined #salt
16:22 otter768 joined #salt
16:23 spookah joined #salt
16:23 d[^_^]b joined #salt
16:25 perfectsine joined #salt
16:25 d[^_^]b joined #salt
16:27 fllr joined #salt
16:27 d[^_^]b joined #salt
16:29 bhosmer joined #salt
16:29 d[^_^]b joined #salt
16:31 d[^_^]b joined #salt
16:32 mohae joined #salt
16:34 d[^_^]b joined #salt
16:35 d[^_^]b joined #salt
16:37 cpowell joined #salt
16:37 d[^_^]b joined #salt
16:37 cpowell_ joined #salt
16:38 wendall911 joined #salt
16:39 d[^_^]b joined #salt
16:41 d[^_^]b joined #salt
16:43 d[^_^]b joined #salt
16:45 KyleG joined #salt
16:45 KyleG joined #salt
16:45 tmclaugh[work] Hi, I’d like to have a salt state fail if certain pillar values are set
16:45 d[^_^]b joined #salt
16:46 tmclaugh[work] I’m managing AWS resoruces with salt an minion IDs are <instance_id>.<service>.<environment>
16:46 khris joined #salt
16:47 tmclaugh[work] I want the state to thropw a failure if the environment isn’t valid
16:47 d[^_^]b joined #salt
16:49 d[^_^]b joined #salt
16:51 timoguin tmclaugh[work]: try the test.fail_without_changes state: http://docs.saltstack.com/en/latest/ref/states/all/salt.states.test.html
16:51 timoguin wrap a conditional around it checking for the environment
16:51 d[^_^]b joined #salt
16:52 tmclaugh[work] thank you  this is what I was looking for!
16:52 L2SHO_ joined #salt
16:53 tmclaugh[work] The conditional being a jinja expression right?
16:53 timoguin yea
16:53 tmclaugh[work] Sorry, new to salt and want to make sure I’m on the right path
16:53 tmclaugh[work] cool
16:53 d[^_^]b joined #salt
16:55 d[^_^]b joined #salt
16:56 murrdoc joined #salt
16:57 tedski can someone identify this?  seen in the saltstack innovation award video on youtube: http://i.imgur.com/6SKSNCR.png
16:57 d[^_^]b joined #salt
16:57 writtenoff joined #salt
16:58 IanVorn joined #salt
16:59 L2SHO is it possible to chain top.sls files together to make something like sub-environments?
16:59 d[^_^]b joined #salt
17:00 tedski L2SHO: you can have multiple top files that are merged together
17:00 tedski but not sub environments
17:00 tedski L2SHO: see http://docs.saltstack.com/en/latest/ref/states/top.html#how-top-files-are-compiled
17:01 L2SHO ok, thanks
17:01 MatthewsFace joined #salt
17:01 L2SHO maybe I can just get a little creative with my targeting
17:01 d[^_^]b joined #salt
17:02 L2SHO can an environment name have a / in it?
17:04 d[^_^]b joined #salt
17:05 murrdoc joined #salt
17:05 d[^_^]b joined #salt
17:05 john5223 joined #salt
17:07 d[^_^]b joined #salt
17:09 sgargan joined #salt
17:09 forrest joined #salt
17:09 d[^_^]b joined #salt
17:11 d[^_^]b joined #salt
17:13 L2SHO Can confirm that using a / in an environment name breaks things
17:13 d[^_^]b joined #salt
17:13 RedundancyD joined #salt
17:15 d[^_^]b joined #salt
17:16 ageorgop joined #salt
17:17 d[^_^]b joined #salt
17:19 d[^_^]b joined #salt
17:21 d[^_^]b joined #salt
17:22 sono joined #salt
17:23 d[^_^]b joined #salt
17:23 rgarcia_ joined #salt
17:25 murrdoc try .
17:25 d[^_^]b joined #salt
17:25 L2SHO I went with _ and it worked fine for me
17:27 d[^_^]b joined #salt
17:29 sono Hi - thought I would pop in and ask if anyone is managing a YAML config file with salt and see how they are doing it.
17:29 d[^_^]b joined #salt
17:30 aparsons joined #salt
17:30 spookah joined #salt
17:31 d[^_^]b joined #salt
17:31 bluenemo_ joined #salt
17:34 d[^_^]b joined #salt
17:35 d[^_^]b joined #salt
17:37 d[^_^]b joined #salt
17:39 d[^_^]b joined #salt
17:40 bhosmer joined #salt
17:41 d[^_^]b joined #salt
17:43 d[^_^]b joined #salt
17:43 Setsuna666 joined #salt
17:45 d[^_^]b joined #salt
17:46 spookah joined #salt
17:47 d[^_^]b joined #salt
17:48 iggy basepi: just ban... they've said 4 lines in the last 9 months I've been in this channel... and it was a question easily answer by the docs had they bothered to look
17:49 d[^_^]b joined #salt
17:51 joeto joined #salt
17:51 impi joined #salt
17:51 d[^_^]b joined #salt
17:51 IanVorn joined #salt
17:53 IanVorn joined #salt
17:53 d[^_^]b joined #salt
17:54 supersheep joined #salt
17:55 d[^_^]b joined #salt
17:56 sgargan joined #salt
17:57 d[^_^]b joined #salt
17:58 basepi Done
17:58 coval3nce_ iggy: btw think i fugired out my multimaster problems from last night
17:59 d[^_^]b joined #salt
18:01 d[^_^]b joined #salt
18:02 hal58th sono, have you tried? Do you have a problem?
18:04 rap424 joined #salt
18:04 d[^_^]b joined #salt
18:05 Matthews_ joined #salt
18:05 denys joined #salt
18:05 d[^_^]b joined #salt
18:06 baweaver joined #salt
18:06 iggy coval3nce_: oh?
18:07 Matthews_ joined #salt
18:07 d[^_^]b joined #salt
18:09 d[^_^]b joined #salt
18:11 d[^_^]b joined #salt
18:13 d[^_^]b joined #salt
18:13 coval3nce_ iggy: https://github.com/saltstack/salt/issues/24434
18:15 DammitJim I'm looking at the remote execution tutorial and I see tomcat as a function; however, in ubuntu, I think it is called tomcat7
18:15 DammitJim what do I need to do to update those functions? or do I need to create brand new ones? like something as simple as tomcat7.status
18:15 d[^_^]b joined #salt
18:16 iggy coval3nce_: what are all those jenkins failures about?
18:16 coval3nce_ iggy: heck if i know, haven’t looked at them yet
18:17 ajw0100 joined #salt
18:17 kevpie joined #salt
18:17 iggy DammitJim: the salt execution modules is called tomcat
18:17 d[^_^]b joined #salt
18:18 coval3nce_ iggy: we are supposed to rebase to develop branch before PR’s right?
18:18 coval3nce_ seems like lots of failures due to missing doc strings etc
18:18 coval3nce_ https://jenkins.saltstack.com/job/salt-pr-linode-cent6-n/114/#showFailuresLink
18:18 iggy coval3nce_: depends on the problem, generally apply it to the oldest branch that has the bug, then they get forward ported
18:19 coval3nce_ iggy: ahh, well probably all my fault then as i made the change off of develop last night
18:19 d[^_^]b joined #salt
18:19 rojem joined #salt
18:21 d[^_^]b joined #salt
18:22 sono hal58th: Right now we are using the file.replace module, doing a regex and replacing the line. But this is prone to error :-/
18:23 otter768 joined #salt
18:23 d[^_^]b joined #salt
18:25 rdxc joined #salt
18:25 IanVorn joined #salt
18:25 d[^_^]b joined #salt
18:26 IanVorn joined #salt
18:27 ajw0100 joined #salt
18:27 d[^_^]b joined #salt
18:28 baweaver joined #salt
18:29 d[^_^]b joined #salt
18:30 cansis joined #salt
18:31 d[^_^]b joined #salt
18:34 d[^_^]b joined #salt
18:35 kevpie hello #salt, i'm looking for help with a way to skip managing a file when the source salt://path points to a non existent file. https://gist.github.com/kevpie/ea8884eef51af7f05160
18:35 d[^_^]b joined #salt
18:37 d[^_^]b joined #salt
18:38 ecerulm joined #salt
18:39 d[^_^]b joined #salt
18:40 viderbit joined #salt
18:42 colonD joined #salt
18:44 forrest kevpie: I don't really know of a good way to do that without some kind of check. You'd be better off looping through with something like the second answer here: https://stackoverflow.com/questions/22582479/how-can-i-check-for-file-existence-in-salt-file-server, or you could pull those into two different states, and not include the state for systems which don't have a log file.
18:45 DammitJim iggy, does that mean I should edit those execution modules or create my own?
18:45 iggy DammitJim: I suspect neither, but having never used them, I have no clue really
18:46 forrest kevpie: If you're really against the multiple sls files, you could have some kind of check based on system that looks for pillar or a custom grain to see whether it should try to drop in the log.sh script.
18:46 drawsmcgraw joined #salt
18:48 timoguin_ joined #salt
18:51 babilen basepi: No, that command isn't a timed ban (that's not possible unless you get eir or another bot in here to manage that). It does, however, redirect that user to the ##fix_your_connection channel (so that they know why they have been banned)
18:52 babilen (eir is a freenode service that tracks bans and quiets)
18:52 basepi Yep, thanks.
18:53 dingo hehe where'd my friend d[^_^]b go?!
18:53 dingo gunna miss him
18:55 iggy for the first time in over a month, I have no github notifications
18:58 kevpie forrest: Thanks for your suggestions. I am building many micro service docker images using runit to handle the supporting services for the container.  In some images there are 5 or so states that follow the same pattern.  I thought it would be nice if I could have this as an include and control it by the presence of the configs alone without the need to restate them in  a pillar or grainsl
18:59 forrest kevpie: Yeah as far as I'm aware there isn't a great way to do that since you're relying on checking the master for a file, maybe if it was masterless you could do some kind of file.exists (http://docs.saltstack.com/en/latest/ref/states/all/salt.states.file.html#salt.states.file.exists)? So do a file.exists on the file (if masterless) and then do a require for that?
19:01 belak I've been messing with salt lately but I've been having issues where jinja is creating a ton of extra lines... so I tried to enable jinja_trim_blocks and jinja_lstrip_blocks but those don't appear to have changed anything - http://pastie.org/private/qj9escjoznfbysuzdd4aq any idea why this might be happening?
19:02 ajw0100 joined #salt
19:02 mpanetta belak: HAve you tried using {%- or -%} instead of just {% and %}?
19:02 mpanetta They remove whitespace before or after the block.
19:02 belak mpanetta: yes... and those work, but I shouldn't those settings do the same thing?
19:03 mpanetta That I don't know
19:03 iggy does the white space bother you that much?
19:03 belak iggy: yes. heh.
19:03 iggy I mean I get it if you're doing a one line loop, but for the most part, yaml (and you) should just ignore the white space
19:04 iggy if you're OCD you're not going to get along with Salt well I don't think
19:05 belak I guess I"ll just change the file around... move the set after the header
19:05 bin_005 joined #salt
19:06 evilrob joined #salt
19:14 murrdoc joined #salt
19:16 notnotpeter joined #salt
19:17 murrdoc joined #salt
19:20 baweaver joined #salt
19:20 jp__ joined #salt
19:20 lokk joined #salt
19:22 lokk hi everyone
19:23 lokk i am working with LDAP and Active Directory authentification using salt stack
19:23 lokk i am stuck at a point and am not able to move further
19:23 lokk can anyone help me through it?
19:24 KyleG I doubt anyone has time to hand hold you through this journey, but we may be able to answer specific questions are you run into problems
19:24 KyleG as you run into problems*
19:25 KyleG SaltStack also offers professional training courses if that's something that interests you.
19:26 lokk yes i am verymuch interested in the course
19:26 KyleG Details are here: http://saltstack.com/training/
19:26 lokk i have posted my issue in git
19:26 lokk https://github.com/saltstack/salt/issues/24443
19:27 KyleG Did you check your LDAP log to see that salt at least made the connection?
19:27 KyleG Your slapd log should tell you if it really was invalid credentials
19:27 KyleG and will also tell you what credentials it received and denied (uid, not password)
19:30 adelcast left #salt
19:30 lokk actually we have a active directory setup
19:30 murrdoc joined #salt
19:30 lokk we use directory serice of AWS
19:31 lokk i have set auth.ldap.activedirectory: True
19:34 Slimmons joined #salt
19:34 adelcast joined #salt
19:35 Slimmons I'm having an issue with virtualenv.managed.  I have a state that runs just fine.  If I add virtualenv.managed to any of the .sls files, when I run state.highstate, salt returns just "minionName", in like 2 seconds.
19:35 Slimmons The only thing that fixes the state is removing virtualenv.managed
19:35 jp__ joined #salt
19:35 kevpie forrest: What I thought I wanted is starting to look a feature request.  I've settled on using '- onlyif: test -f /srv/salt/{{ sls }}/log.sh'.
19:35 bin_005 joined #salt
19:36 forrest kevpie: Yeah that's a solution as well. Still problematic when you have a master though.
19:36 MTecknology iggy: Welp... salt master is now on 2015.5.0. Minions are still on 2014.1.13. I'm getting ready to roll out the updates to all systems and hoping my connectivity issues are resolved as they were for you.
19:37 forrest kevpie: Someone might have already asked for it, so I'd look through the existing issues ;)
19:37 iggy MTecknology: good luck
19:37 jp__ Hello, salt newbie here. Does anybody know if I can store a list of apt-get package names in a file or in a Pillar to be used in a pkg.installed state?
19:38 iggy how would putting them in a pillar be any different than just putting them in a state?
19:38 linjan joined #salt
19:39 jp__ So I could use them in different states
19:39 jp__ or environments
19:40 baweaver joined #salt
19:41 iggy use pkgs and {{ pillar['listofpkgs']|indent(8) }} or whatever
19:45 rojem joined #salt
19:46 murrdoc joined #salt
19:46 tomh- joined #salt
19:48 thehaven joined #salt
19:49 bezaban joined #salt
19:52 smkelly joined #salt
19:54 ajw0100 joined #salt
19:59 ahammond I want to install an RPM with conflicting files. So, I'd like to pass along the --replacefiles paremeter to rpm when it's called from pkg.latest. How can I do that?
20:02 IanVorn joined #salt
20:02 forrest joined #salt
20:02 breakingmatter joined #salt
20:03 breakingmatter Hey, anyone around that's able to help me with a custom state issue?
20:04 ahammond breakingmatter http://sol.gfxile.net/dontask.html
20:05 ahammond I guess until I have a better solution, I'll be doing a cmd.run for the rpm on this. Yuck!
20:06 iggy ahammond: yes, cmd.run it
20:07 Ch3LL_ joined #salt
20:07 breakingmatter Duly noted. I'm using the states from https://github.com/cloudbase/salt-openstack to install OpenStack using Salt. I'm running into an issue with the custom modules and states that are in the hierarchy. Everytime I run the highstate, I get an error that states "State 'keystone.user_present' was not found in SLS 'keystone.openstack_users'". I'm thinking it's because there is a folder in the "file_root" that is named "keystone" and a c
20:07 breakingmatter ustom module/state that's also called "keystone". Is this a known issue, or am I doing something wrong?
20:08 breakingmatter I've already ran "saltutil.sync_all". And I've tested the state itself with "salt '<minion>' keystone.user_present" and I get a response.
20:09 breakingmatter So, it's "loaded" and "available", but it's not running it in the .sls file.
20:09 s_kunk joined #salt
20:09 Ch3LL_ joined #salt
20:09 breakingmatter Apologies for the wall of text.
20:09 kruppm joined #salt
20:11 iggy breakingmatter: states (sls files) and modules (execution and state modules, py code) are handled completely separately
20:11 ahammond breakingmatter sounds like you didn't install those states correctly (specifically, it sounds like you should have /srv/salt/keystone if you cloned it down, which is probably the only way to do it since the cloudbase guys clearly haven't read about salt's gitfs stuff... otherwise they'd have formatted their repo correctly)
20:11 thedodd joined #salt
20:12 ahammond breakingmatter oh wow, it gets even better
20:12 breakingmatter I have all their states (sls) and modules in a separate file root called "os_file_root"
20:12 iggy breakingmatter: the minions all need the keystoneclient python module installed to use those states/modules
20:12 ahammond they've simply snapshotted their entire fileroot
20:12 ahammond yuck.
20:12 breakingmatter Ahh. See, I've never worked with custom modules.
20:14 s_kunk joined #salt
20:15 breakingmatter So as far as those custom modules/states go, I have to install them on the minions?
20:15 breakingmatter All the SLS files seem to "work". The states that fail are the ones that are dependent on the custom .py files.
20:15 Ztyx joined #salt
20:16 breakingmatter OpenStack gets installed, but no users get added because of the errors. So, it's basically worthless without that. heh.
20:19 iggy well, salt has modules for keystone, etc... so not sure if that's something old or what
20:19 iggy oh, they are in the proper _modules/_states dirs
20:20 iggy that's fine, those are special dirs that salt knows to distribute to the minions and load as custom modules
20:21 breakingmatter iggy: I've ran the "saltutil.sync_all" command, and it returned that it loaded the modules.
20:21 breakingmatter ----------
20:21 breakingmatter grains:
20:21 breakingmatter modules:
20:21 breakingmatter - modules.glance
20:21 breakingmatter - modules.ini_manage
20:21 breakingmatter - modules.keystone
20:21 breakingmatter - modules.neutron
20:21 breakingmatter outputters:
20:21 breakingmatter renderers:
20:21 breakingmatter returners:
20:21 breakingmatter states:
20:21 iggy my eyes!!!!!!
20:21 breakingmatter - states.glance
20:21 breakingmatter - states.ini_manage
20:21 breakingmatter - states.keystone
20:21 breakingmatter - states.neutron
20:21 breakingmatter utils:
20:21 breakingmatter That paste did not work the way I had hoped.
20:22 iggy the minions automatically do that (sync_all) before a highstate (just fyi)
20:22 drawsmcgraw left #salt
20:24 otter768 joined #salt
20:26 sgargan is it possible to use a security_group name instead of id in the boto modules?
20:27 breakingmatter iggy: I still just don't understand why the minions aren't seeing the custom module.
20:27 spookah joined #salt
20:29 DammitJim does one normally do the configuration of a piece of software separate than the install?
20:29 DammitJim like, should I have a state for the install
20:30 DammitJim and a state for copying the config files?
20:30 Nazca__ joined #salt
20:31 mikepea joined #salt
20:32 basepi breakingmatter: do those modules have __virtual__ functions? It's possible the minions do see them, but they aren't loading them because of a missing dep or a __virtual__ function returning False.
20:32 breakingmatter basepi: Yes, they have __virtual__ functions.
20:34 goki joined #salt
20:35 moderation joined #salt
20:36 supermike__ joined #salt
20:36 wiqd joined #salt
20:36 OliverMT joined #salt
20:38 DammitJim when one specifies a file in a service: watch: does the service get automatically restarted because the minion noticed someone updated that file
20:38 DammitJim or because we pushed a file from the master; therefore, the service gets restarted?
20:39 Ryan_Lane sgargan: yes
20:39 Ryan_Lane it will automatically translate the name of the group into its id
20:39 bbhoss joined #salt
20:39 Ryan_Lane if you're in vpc
20:40 Ryan_Lane if you're in classic it'll just use the group name
20:40 breakingmatter iggy: basepi: this error makes me think there is a naming conflict. "State 'keystone.tenant_present' was not found in SLS 'keystone.openstack_tenants'"
20:40 sgargan great! it seemed like it didn’t but I most probably goofed
20:40 Ryan_Lane you have to specify the vpc info or it'll use classic! :)
20:40 Ryan_Lane just a warning
20:40 breakingmatter 'keystone.tenant_present' is a function defined in that custom module.
20:40 sgargan I’ll check again now that I know it should be supported
20:40 akitada joined #salt
20:41 Ryan_Lane I wrote most of that, so I know it works :)
20:41 Ryan_Lane I'm using that extensively in prod
20:41 basepi breakingmatter: but if for some reason the keystone state module weren't loaded, that's the error you would get
20:41 Ryan_Lane let me know if you need help
20:41 basepi breakingmatter: still could be a __virtual__ problem
20:41 sgargan thanks Ryan_Lane I’ll try again and get back to you
20:41 sgargan cheers
20:41 Ryan_Lane cool
20:43 breakingmatter basepi: Here's the __virtual__ function: https://gist.github.com/anonymous/7f19fb6301c63164b4f2
20:43 breakingmatter Does that look "right"?
20:43 gyre007 joined #salt
20:43 abele joined #salt
20:44 basepi Sure, seems right. Can you run `keystone.auth` on that minion? Have you ensured that it is in fact in __salt__?
20:44 simonmcc joined #salt
20:45 breakingmatter basepi: Just tried it and it runs fine.
20:45 L2SHO is there a way to insert grain data into a managed file?
20:45 breakingmatter with 'salt-call keystone.auth'
20:46 gazarsgo joined #salt
20:49 CaptTofu joined #salt
20:49 codekobe joined #salt
20:49 basepi Hmmm, then I'm at a loss....
20:50 basepi For kicks and giggles, try putting `__virtualname__ = 'keystone'` above your virtual function
20:51 doriftoshoes joined #salt
20:52 LyndsySimon joined #salt
20:52 colonD joined #salt
20:55 EWDurbin joined #salt
20:57 breakingmatter basepi: Added it, and getting the same error.
20:57 breakingmatter Hmm.
20:57 breakingmatter Also, I /did/ resync the module just in case.
20:59 imanc_ joined #salt
20:59 sporkd2 joined #salt
20:59 sgargan Ryan_Lane: I tried again with my load balancer config and various combinations of security_groups: my_balancer_security but no joy
21:00 basepi breakingmatter: alright, next try changing `keystone` to `keystone2` or something. Salt ships with a keystone state module, it's possible that it's conflicting or something. Also double check the spelling of the function name in both the module and the sls file
21:00 epcim_ joined #salt
21:00 breakingmatter basepi: You mean, change the `keystone` module name?
21:01 sgargan running with 2015.5.0
21:01 jay_d joined #salt
21:04 andrew_v I am having an exceedingly strange issue with mine_functions. Some functions return data for all minions, some only return it for the master. I can't figure out why.
21:05 andrew_v Example here: https://gist.github.com/andrew-vant/de678f00bb1cd61e46ff
21:05 OGuilherme joined #salt
21:11 andrew_v Never mind. Found it.
21:12 bin_005 joined #salt
21:13 tcolvin joined #salt
21:14 sandah joined #salt
21:16 toddnni joined #salt
21:17 ashb joined #salt
21:22 SEJeff joined #salt
21:23 basepi breakingmatter: the state module, yes. Oh, but before you do that you should change your virtual function to just `return True` just to make triple sure there are no shenanigans there.
21:29 breakingmatter basepi: Interesting, now I get this mixed error.
21:29 breakingmatter https://gist.github.com/anonymous/ced96fc2bd154c6b468f
21:30 breakingmatter Hm.
21:30 breakingmatter There's a "state" module and a "module".
21:30 breakingmatter I'll rename both of them and see what happens.
21:30 iggy I'm going to guess that "formula" isn't in the best shape for direct consumption
21:31 breakingmatter I dont' think it's anywhere close to a "formula"
21:32 hasues joined #salt
21:32 iggy either way... it's not working out of the box, you should file bugs against that repo
21:32 breakingmatter I was planning on it after I ruled out my own ignorance.
21:33 breakingmatter Plus it's used to deploy an outdated version of OpenStack anyways.
21:33 hasues left #salt
21:34 Ryan_Lane sgargan: can you gist your state, minus any private info?
21:34 Ryan_Lane you sure that security group name exists in your vpc?
21:34 Ryan_Lane also, does it have a name tag?
21:35 Ryan_Lane the salt modules basically operate on the idea that they own everything
21:35 Ryan_Lane they use the name tags for finding things
21:35 froztbyte joined #salt
21:36 kevpie forrest: Thanks for the help. I'll look the issues to see if I can find anything related.
21:36 forrest kevpie: Sounds good!
21:36 breakingmatter basepi: Same problem.
21:37 basepi le sigh
21:38 basepi I don't have any more ideas at the moment. I guess file an issue...
21:39 breakingmatter So, is the issue with Salt or the "formula"? ;)
21:43 sgargan Ryan_Lane: here’s the gist https://gist.github.com/sgargan/c88b838845fd51a41c9c
21:43 sgargan looking at the logs on the minion I see its looking via the id and expecting sg-
21:45 HeftySmurf joined #salt
21:50 SEJeff left #salt
21:52 smcquay joined #salt
21:52 fiery joined #salt
21:54 ndrei joined #salt
21:55 spookah joined #salt
21:58 Ztyx joined #salt
22:00 Ztyx joined #salt
22:03 forrest damn you iggy
22:03 iggy :)
22:04 forrest This issue annoys me
22:04 forrest well, the issue of .d dirs in general annoys me
22:04 forrest the formulas are just templates :|
22:05 dhscholb joined #salt
22:06 Ryan_Lane sgargan: this is for classiv?
22:06 Ryan_Lane *classic
22:06 Ryan_Lane for scheme you want 'internal', not 'internal-facing'
22:06 sgargan no the loadbalancer would be for a vpc
22:07 iggy forrest: you should say what you have to say on the bug where I called you out by name
22:07 iggy ;)
22:07 forrest I'm already typing
22:07 Ryan_Lane sgargan: then you need to put the vpc info in it
22:08 sgargan it think its related to the subnets, if they’re not specified the security_group names are not converted to ids
22:08 Ryan_Lane sgargan: you need this arg set:     - vpc_id: vpc-myid
22:08 Ryan_Lane if you don't specify the vpc id, it won't be able to find the group
22:09 forrest iggy: Discussing it makes me zzzzz
22:09 andrew_v iggy: I am going to laugh fairly hard if I just lost a day trying to figure out what was going on and it turns out to be my own change that created the problem.
22:10 forrest andrew_v: I'm pretty sure that is the case.
22:11 Ryan_Lane sgargan: whoops
22:11 forrest since you were the one who did the .d directory change in the first place I believe ;)
22:11 forrest I've responded though with a proposed solution.
22:11 Ryan_Lane sgargan: yeah, it's subnets that you need
22:11 sgargan 'vpc_id' is an invalid keyword argument for 'boto_elb.present'.
22:11 sgargan that’s what i was thinking
22:11 Ryan_Lane you don't want availability zones
22:12 Ryan_Lane oh. this is an external ELB
22:12 Ryan_Lane you can leave out the scheme
22:12 murrdoc joined #salt
22:12 Ryan_Lane internet-facing is the default
22:12 aw110f joined #salt
22:12 Ryan_Lane I guess it's nice to leave in to make it clearer it faces the outside world :)
22:12 sgargan yes i guess its redundant
22:13 Ryan_Lane doesn't hurt
22:13 sgargan i was just ruling stuff out
22:13 murrdoc ryan have you run into a case , where you need to reboot a minion mid highstate
22:15 Ryan_Lane murrdoc: nope
22:15 Ryan_Lane I never reboot nodes
22:15 Ryan_Lane if it needs a reboot it gets terminated
22:15 murrdoc u lucky lucky guy
22:15 Ryan_Lane and a new node comes in
22:16 jimmybee88 joined #salt
22:17 jimmybee88 left #salt
22:17 aw110f hi, i’m using gitpython in salt and i sometimes i don’t get updates from files when running state.sls.  In the master log i see “unable to update local ref”
22:18 aw110f I think it’s because the repo needs to prune old branches before it can pull updates from the repo
22:18 aw110f git remote prune origin
22:19 murrdoc joined #salt
22:19 aw110f the fix right now is to delete /var/cache/salt/master and let salt re-clone the git_remotes repos
22:19 sgargan Ryan_Lane: I’ve added in one of my subnets, it complains
22:19 sgargan Subnets subnet-8d3e64cb do not map to a valid vpc id.
22:20 ageorgop joined #salt
22:24 spookah joined #salt
22:25 otter768 joined #salt
22:25 smcquay joined #salt
22:25 jerematic joined #salt
22:26 smcquay joined #salt
22:26 P0bailey joined #salt
22:26 P0bailey joined #salt
22:30 sgargan Ryan_Lane: that was a permissions error, the vpc is being resolved correctly now and the sec group too. Nice work on the plugin and thanks for your help
22:35 bin_005 joined #salt
22:40 ajw0100 joined #salt
22:46 Auroch joined #salt
22:47 bfoxwell joined #salt
22:52 cberndt joined #salt
23:02 giantlock joined #salt
23:09 TheoSLC joined #salt
23:09 premera joined #salt
23:20 mummer joined #salt
23:24 Ryan_Lane sgargan: ah. cool :)
23:24 lexter joined #salt
23:24 Ryan_Lane glad it's working for you
23:26 Ryan_Lane sgargan: if you use cloudwatch alarms, just wait till you see the magical stuff you can do with ELBs, cloudwatch alarms and default pillars :)
23:27 Ryan_Lane (http://ryandlane.com/blog/2015/01/06/saltstack-automated-cloudwatch-alarm-management-for-aws-resources/)
23:38 kruppm joined #salt
23:39 hasues joined #salt
23:40 hasues left #salt
23:43 baweaver joined #salt
23:47 badon joined #salt
23:50 tkharju joined #salt
23:51 nebuchadnezzar joined #salt

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