Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2015-09-29

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

All times shown according to UTC.

Time Nick Message
00:03 tzero joined #salt
00:15 joe_n joined #salt
00:18 GrueMaster joined #salt
00:24 JDiPierro joined #salt
00:35 eliasp joined #salt
00:38 cornfeedhobo left #salt
00:38 pravka joined #salt
00:47 Garo_ joined #salt
00:48 iggy joined #salt
00:49 cwright joined #salt
00:49 pratikmallya joined #salt
00:52 nlb joined #salt
00:52 rhodgin joined #salt
00:52 gtmanfred joined #salt
00:53 zmalone joined #salt
00:54 tkharju joined #salt
00:55 Gilou joined #salt
00:55 JDiPierro joined #salt
00:56 breakingmatter joined #salt
01:00 larsfronius joined #salt
01:01 joe_n joined #salt
01:03 lesterc joined #salt
01:04 Ryan_Lane basepi: :'( https://github.com/saltstack/salt/pull/27475 <-- failing tests. says __states__ isn't defined
01:05 Ryan_Lane btw. __states__ took a 15-20 minute state run to ~2-3 minutes
01:07 iggy Ryan_Lane: you need to write up a blog post about that
01:08 tzero joined #salt
01:09 jalbretsen joined #salt
01:12 theologian joined #salt
01:15 ronrib joined #salt
01:16 David_B55 joined #salt
01:16 tzero joined #salt
01:17 murrdoc joined #salt
01:18 basepi Hrm, guess I'll have to take a look. But that's awesome!
01:19 zmalone joined #salt
01:21 murrdoc ?
01:22 iggy we could definitely stand to improve our highstate runtime
01:27 JDiPierro joined #salt
01:32 timoguin joined #salt
01:33 pravka joined #salt
01:35 joe_n joined #salt
01:36 aqua^c joined #salt
01:36 agj joined #salt
01:36 timoguin_ joined #salt
01:38 opensource_ninja joined #salt
01:39 sunkist joined #salt
01:43 markm joined #salt
01:46 tzero joined #salt
01:47 otter768 joined #salt
01:50 ageorgop joined #salt
01:53 anotherZero joined #salt
01:56 _JZ_ joined #salt
01:56 ajw0100 joined #salt
01:58 catpigger joined #salt
02:00 tzero joined #salt
02:04 hasues joined #salt
02:04 hasues left #salt
02:13 malinoff joined #salt
02:17 tzero joined #salt
02:27 tzero joined #salt
02:35 baweaver joined #salt
02:37 baweaver_ joined #salt
02:39 tzero joined #salt
02:47 murrdoc joined #salt
02:49 tzero joined #salt
02:54 ageorgop joined #salt
02:57 breakingmatter joined #salt
02:58 zmalone joined #salt
02:59 ITChap joined #salt
03:14 clintberry2 joined #salt
03:27 n8n joined #salt
03:30 anmol joined #salt
03:31 subsignal joined #salt
03:35 tzero joined #salt
03:43 subsignal joined #salt
03:44 ramteid joined #salt
03:45 favadi joined #salt
03:50 armguy joined #salt
03:50 tzero joined #salt
03:58 subsignal joined #salt
04:09 subsignal joined #salt
04:10 ALLmightySPIFF joined #salt
04:20 tzero joined #salt
04:24 pratikmallya joined #salt
04:27 pratikmallya joined #salt
04:29 tzero joined #salt
04:30 bbradley joined #salt
04:37 clintberry2 joined #salt
04:38 tzero joined #salt
04:58 breakingmatter joined #salt
04:59 markm_ joined #salt
05:04 anmolb joined #salt
05:07 tzero joined #salt
05:13 subsignal joined #salt
05:13 aqua^c joined #salt
05:14 tzero joined #salt
05:22 moogyver anyone have a quick benchmark about how quickly it takes them to do, say, 'ls -l /tmp' across 100 hosts?
05:30 zmalone joined #salt
05:32 svinota joined #salt
05:34 evle1 joined #salt
05:34 zmalone joined #salt
05:36 anmolb joined #salt
05:37 Bryson joined #salt
05:43 tzero joined #salt
05:49 otter768 joined #salt
05:50 _JZ_ joined #salt
05:51 malinoff joined #salt
05:52 dopesong joined #salt
06:00 rim-k joined #salt
06:02 markm__ joined #salt
06:02 colttt joined #salt
06:08 tzero joined #salt
06:10 joe_n joined #salt
06:10 cyborglone joined #salt
06:11 zmalone joined #salt
06:15 Eureka703 joined #salt
06:18 sirex joined #salt
06:20 AndreasLutro joined #salt
06:21 tzero joined #salt
06:22 edulix joined #salt
06:25 katyucha joined #salt
06:27 lb joined #salt
06:39 Steven| joined #salt
06:54 elsmo joined #salt
06:58 KermitTheFragger joined #salt
06:58 sarlalian joined #salt
07:00 breakingmatter joined #salt
07:00 eseyman_ joined #salt
07:02 cberndt joined #salt
07:02 larsfronius joined #salt
07:03 subsignal joined #salt
07:22 alexlist joined #salt
07:24 colegatron joined #salt
07:25 rodio_ua joined #salt
07:25 rodio_ua left #salt
07:26 colegatron hello. I am frustrated with saltstack because I am not able to understand why states are not executed in the right order and why some of them are not executed as expected (probably because a misunderstanding of 'requirements').
07:27 colegatron I have a couple of questions: How can I force a restart of a service at the very end the states execution?
07:28 colegatron and : there is a saltstack tutorial, with lots of examples of 'requirements' ?
07:29 AndreasLutro colegatron: you probably don't want the service to restart *every time* you run a state, or?
07:29 ze- colegatron: not sure about a tutorial. But, order: last should help
07:29 ze- https://docs.saltstack.com/en/latest/ref/states/ordering.html
07:30 * ze- should probably try to publish his tool to get a dependency graph around.
07:30 felskrone joined #salt
07:30 colegatron AndreasLutro: maybe, but until I do not completely understand the 'requisites', I need to ensure to restart some services after every highstate
07:30 colegatron ze- reading. tnx
07:31 ze- colegatron: i'm going to work. Hilight me again in 30 minutes, and I can try to hook you up with a small tool that might help see the dependencies.
07:31 AndreasLutro colegatron: requisites consist of a lot of different things - require, watch/listen, onchanges... do you have any more specific questions?
07:32 AndreasLutro restarting a service at the end of every highstate is something I can't help you with
07:32 colegatron ze- thank you very much, probably it will help a lot
07:33 colegatron AndreasLutro: seems 'order: last' could do the trick, I'm going to try
07:33 AndreasLutro doubt it
07:34 colegatron AndreasLutro: me too, but I have no other thing to try ... :(
07:35 AndreasLutro you could explain your problem more
07:35 AndreasLutro why do you want to restart the service?
07:38 colegatron I have to provision 1 nginx and 3 tomcat instances in the same server. on nginx will be 2 sites setup and enabled. every tomcat instance is setup with 1 application (war file) and one server.xml/tomcat-users.xml files.
07:38 Ph-x joined #salt
07:38 colegatron seems everything is in the right place once provisioned/deployed, but I have to manually restart nginx and every tomcat instance services, because they are not automatically restarted by saltstack
07:39 colegatron that's why I want to force a services restart at the very end
07:39 illern joined #salt
07:40 rim-k joined #salt
07:42 AndreasLutro well you need to map out which states' changes should trigger a service reload, then use listen or listen_in
07:42 TyrfingMjolnir joined #salt
07:43 AndreasLutro for your nginx sites it's easy, just add a listen_in: [ service: nginx ] to your vhost files or whatever
07:43 fredvd joined #salt
07:43 AndreasLutro not sure what you need to do with the tomcat stuff. I don't think you need to reload nginx every time the tomcat service changes
07:45 forrest joined #salt
07:46 sleetdrop joined #salt
07:46 colegatron AndreasLutro: just to clarify (maybe some of my mistakes are because this):
07:46 colegatron when you say : listen_in: [ service: nginx ]
07:46 forrest joined #salt
07:46 colegatron 'nginx' is the same of the label of the service that installs nginx or it is the name of the real service 'nginx' ?
07:47 colegatron s/same/name/g
07:47 AndreasLutro colegatron: the state ID, or the state name argument - see https://gist.github.com/anlutro/cd0c16d1d23d55ded19b
07:47 |Trend| joined #salt
07:48 colegatron AndreasLutro: tnx, I was doing well then
07:50 otter768 joined #salt
07:51 colegatron Is the documentation complete? looking at https://docs.saltstack.com/en/latest/ref/states/all/salt.states.service.html you can see an example of service.running using - reload: true, but this argument is not specified in the service.running documentation (at the end of the page)
07:52 colegatron I am missing something?
07:53 AndreasLutro it's complicated
07:53 AndreasLutro basically the reload arg is passed onto mod_watch
07:54 amarashov joined #salt
07:55 impi joined #salt
07:57 babilen colegatron: You should, essentially, figure out which changes to your configuration should trigger a service restart. You then use listen/listen_in (or watch/watch_in) to tie them together so that the service is automatically restarted whenever the other state has changes.
07:58 babilen I really wouldn't hardcode a "restart service at the end of every highstate". This is way too procedural and therefore a bad fit for salt. Describe the state you want to achieve, not how to achieve it.
07:58 kawa2014 joined #salt
07:59 colegatron AndreasLutro: so everything that is not defined in the docs is kwargs that are literally passed to the funcion, is it right?
08:00 AndreasLutro in this case yes, there are other special cases all over the place
08:01 Rumbles joined #salt
08:01 colegatron babilen: tnx for your answer. to install a service chain, lets say tomcat, there are four steps: 10. install tomcat, 20. general tomcat setup, 30. app1 setup, 40. app2 setup.
08:02 colegatron I am not sure in which point (from 10-20, from 20-30, from 30-40 or 40-end) should I write the listen/listen_in to let saltstack know when exactly it has to reload/restart the service
08:03 colegatron and because listen/listen_in behaves different, probably the answer is different for both
08:04 losh joined #salt
08:05 AndreasLutro app1/app2 requires general tomcat setup, tomcat setup requires tomcat package installed
08:05 AndreasLutro app1/app2 has listen_in: [ service: whatever ]
08:05 AndreasLutro probably
08:06 babilen Well, multiple places probably .. (tomcat config changes should trigger a restart/reload as well as app configuration changes)
08:06 babilen https://github.com/saltstack-formulas/tomcat-formula might have a number of solutions to this already
08:06 babilen (not familiar with that formula, we are a delightfully tomcat free shop)
08:08 MadsRC After updating from 2015.5.3 to 2015.5.8 (CentOS 7 - SaltStack's new official repo) my log is getting hammered with this error: http://pastebin.com/97900jzZ - Can't seem to locate the missing module or if it's a bug?
08:09 colegatron thank you for the answers. I prefer to understand before try to cut and paste and then expect an unexpected behavior :)  But I will have a look to that formula anyway :)
08:09 babilen (repasted MadsRC paste at https://www.refheap.com/110057 )
08:09 illern joined #salt
08:11 chiui joined #salt
08:11 babilen colegatron: https://github.com/saltstack-formulas/tomcat-formula/blob/master/tomcat/vhosts.sls and https://github.com/saltstack-formulas/tomcat-formula/blob/master/tomcat/package.sls#L10 is where the "sites" magic happens
08:11 colegatron babilen: thank you
08:11 MadsRC babilen: Thanks man, forgot to syntax highlight it :)
08:13 S_Walker joined #salt
08:14 Xevian joined #salt
08:15 babilen MadsRC: I am also allergic to pastebin.com as it is an ad-laden and ugly website that preys on the great unwashed who simply google for "pastebin". Much prefer http://refheap.com, http://paste.debian.net, https://gist.github.com, http://sprunge.us, http://dpaste.de, …
08:16 babilen MadsRC: Could you paste the output of "salt --versions-report" ?
08:18 babilen colegatron: You essentially figure out which states, if they have changes, should trigger a service restart. This will mostly be a relation between the "service.running" state that ensures that the service is, well, running and *all* file.managed, file.absent, file.* states that, in their entirety, constitute the configuration of the service. You would then add a "listen_in" requisite to each of those states
08:18 babilen .
08:18 babilen If you make changes to one of these files the service will be automagically restarted.
08:19 babilen But this isn't really restricted to file.* at all (just often the case as you manage configuration files), but could be mounts, firewall rules and whatnot
08:20 Grokzen joined #salt
08:20 MadsRC babilen: Just a second - Right now Salt even refuses to work properly... Guess something went wrong with the update...
08:23 aw110f joined #salt
08:23 babilen MadsRC: How did you update?
08:24 colegatron babilen: but I if you have lots of sls files, how saltstack knows which file.manged, file.* is related to a service? only because the requisites or because they should be in the same sls or because other reason?
08:25 MadsRC babilen: I did it by removing every file referenced when doing a python setup.py install --record file.txt and then followed the CentOS guide from repo.saltstack.com
08:25 thalleralexander joined #salt
08:26 babilen colegatron: You express relations between individual states, not SLS files
08:26 babilen MadsRC: Sounds as if you have some stuff left from the earlier, manual, install
08:26 colegatron ok. tnx
08:26 babilen (pyc files maybe)
08:26 MadsRC babilen: Though I suspect that some files where removed and not reinstalled - By doing a complete reinstall again I managed to get rid of the error (Plus a wierd one where it refiesed to run cause "salt.scripts" couldn't be imported)
08:26 babilen Okay, so all is good now?
08:27 GreatSnoopy joined #salt
08:27 MadsRC It seems like it, Haven't seen the error in the master log anymore (On this test system atleast)
08:29 aw110f_ joined #salt
08:31 s_kunk joined #salt
08:40 auzty joined #salt
08:47 X67r joined #salt
08:53 Nebraskka joined #salt
08:55 svinota joined #salt
08:59 PuckD guys, how would you change a minions hostname? I didn't accept the key yet, what would I need to do?
09:00 subsignal joined #salt
09:01 breakingmatter joined #salt
09:01 svinota joined #salt
09:17 bluenemo joined #salt
09:24 babilen PuckD: You can set the minion id in the minion config or /etc/salt/minion_id
09:24 larsfronius joined #salt
09:24 PuckD babilen: but I'd like it to use the hostname of the machine, i'd have to reinstall the minion?
09:25 babilen PuckD: https://docs.saltstack.com/en/latest/ref/configuration/minion.html#std:conf_minion-id (and https://github.com/saltstack-formulas/salt-formula in "salt:minion:id" -- https://github.com/saltstack-formulas/salt-formula/blob/master/salt/files/minion.d/f_defaults.conf#L69 )
09:26 babilen PuckD: What is the ID now and what do you want it to be?
09:26 PuckD it's CentOS6-7, and I need to be nginx.host
09:26 PuckD *need it to be
09:27 alexhayes joined #salt
09:28 babilen Hmm, curious. It defaults to the FQDN for me.
09:28 babilen (but then I'm using Debian packages)
09:28 PuckD the hostname of the machine was not setup correctly when salt minion was installed
09:29 S_Walker joined #salt
09:30 cliluw joined #salt
09:31 PuckD and if I change the hostname in the minion config, I'll need to revoke the key and try again, right?
09:33 babilen correct
09:39 impi joined #salt
09:43 aqua^c joined #salt
09:47 mage_ hello, I'm trying to setup gitfs with pygit2 provider but I'm still getting a: 2015-09-29 11:44:24,369 [salt.utils.gitfs][ERROR][76533] Exception 'The SSL certificate is invalid' caught while fetching gitfs remote 'https://xxx/git/bbpf/automation.git'. Yes, I have "gitfs_ssl_verify: False" in my "master" conf file ... any idea what could be wrong ?
09:48 mage_ is it a bug ? it looks like this config is not used in Salt: https://dpaste.de/63VE
09:51 otter768 joined #salt
09:52 AndreasLutro mage_: I can't actually see that config value being used anywhere... so it might be!
09:53 mage_ looks like the problem is in init_repo() function of salt/utils.py for the Pygit2 class
09:53 mage_ I'll fill a bug report, and try to provide a patch as soon as I have little time :)
09:55 GrueMaster joined #salt
09:56 AndreasLutro ah maybe it's passed on to pygit2 via kwargs
10:00 mage_ perhaps.. I haven't enough time to check now :( will do that later
10:01 mage_ I've added a bug report #27495
10:03 kbyrne joined #salt
10:09 BogdanR joined #salt
10:14 leev did anything change from 2015.5.3 from 2015.8.0 about how merging of pillar data happens?
10:15 Nebraskka joined #salt
10:17 jagguli` hi guys check out https://bitbucket.org/jagguli/iodine
10:17 jagguli` A salt-api client based on IPython
10:18 jagguli` using/using Ipython ??
10:18 FredFoo joined #salt
10:20 scoates joined #salt
10:20 giantlock joined #salt
10:22 fabiand joined #salt
10:22 fabiand hey
10:22 fabiand is there something like an sls registry?
10:22 fabiand besides
10:22 fabiand https://github.com/saltstack-formulas
10:25 fabiand joined #salt
10:26 jagguli` are you looking for formulas ?
10:27 jagguli` thers the winrepo that has a bunch of sls
10:31 TyrfingMjolnir joined #salt
10:33 fabiand jagguli`, hey
10:33 fabiand yep
10:33 fabiand the winrepo? me searches
10:33 fabiand ah i see
10:40 markm__ joined #salt
10:40 subsignal joined #salt
10:42 TyrfingMjolnir joined #salt
10:50 X67r joined #salt
10:51 XenophonF joined #salt
10:51 XenophonF does anyone have a state formula they'd be willing to share that changes the hostname of an EC2 instance?  i'm about to write one but would much prefer to use someone else's code :)
10:59 Puck` joined #salt
11:02 breakingmatter joined #salt
11:04 illern joined #salt
11:04 bhosmer joined #salt
11:04 mage_ what it the recommended gitfs provider ?
11:04 svinota joined #salt
11:11 XenophonF I think the general consensus is pygit2, but I don't like it and use GitPython instead.
11:11 chiui joined #salt
11:11 XenophonF Unless you have technical reasons for using something different (like I do), go with pygit2, which is the default.
11:11 bluenemo can I write a jinja macro that returns something, so I can use the macro inside jinja to assign it to new objects? I have a bunch of jinja replace filters and only want to write the logic once
11:13 bluenemo ah found what I want
11:13 XenophonF Yes, but return values are tricky to get right.
11:14 XenophonF bluenemo: https://github.com/irtnog/active-directory-formula/blob/master/ad/ds/template.sls and https://github.com/irtnog/active-directory-formula/blob/master/ad/ds/options.jinja show how I did it, using the Jinja call statement.
11:15 mage_ I think pygit2 doesn't work with self-signed certificates
11:15 XenophonF I've seen macros that return strings or YAML dictionaries, but depending the return value's complexity, you'll have to serialize/de-serialize it yourself, which can be lead to weird errors if you miss a step.
11:16 XenophonF mage_: don't know, don't use it
11:17 bluenemo ah just use set. ok cool, thank you
11:17 bluenemo XenophonF,
11:17 bluenemo I actually just started expanding my own _modules for everything jinja annoys me ;)
11:18 XenophonF hah
11:18 XenophonF I've done the same.
11:18 XenophonF hell, that Active Directory state formula has a custom grain, custom execution module, and custom state module
11:19 XenophonF I think at some point I need to learn how to write state formulas in Python and give up on Jinja entirely.  it's got some significant limitations.
11:25 bluenemo XenophonF, I'm currenlty writing a HA openvpn Formula with "everything" (I really like vpn :) I write it in pyobjects, its not as heavy as python, its ok so far. Actually the formula is half pyobjects half jinja|yaml. I also wrote a state module for easyrsa.. or actually I forked easyrsa, then worte a module on it..
11:25 XenophonF ugh, easyrsa
11:26 XenophonF my condolences :)
11:26 XenophonF actually, that sounds really, really cool
11:26 bluenemo plays "easy" with openvpn ;) but yeah.. I read some of the bash code. Its not that bad.
11:26 XenophonF are you going to be able to make it public?
11:26 bluenemo but not for automation. thats why I forked it
11:26 bluenemo anyway, it only executes a couple of openssl commands..
11:26 bluenemo well yeah I plan to, my problem is i'm kinda shy about my code :D
11:27 XenophonF for me it was easier to just run openssl directly
11:27 bluenemo but yeah it rocks so far, let me give you an example:
11:28 bluenemo XenophonF, https://paste.debian.net/hidden/8f0a8355/ sets up a openvpn server with all the ssl stuff required, actually checks if the certs / keys / ca check out against each other, packages client configs so you can move them to /etc/openvpn and then just tar xvzf them.. and have a README with it.. and more stuff..
11:28 bluenemo you can have multiple VPN Servers / subnets, each with its own ca
11:28 breakingmatter joined #salt
11:29 denys joined #salt
11:29 bluenemo and you can then NOT use client-to-client but iptables forward to allow connections.. currenlty working on the template for the iptables script so you can say client A in VPN alpha can ssh to client X in VPN omega, but not the other way around and so on..
11:33 XenophonF very nice!
11:33 bluenemo XenophonF, what I disliked about just running openssl is that it gives cryptic errors.. I mean easyrsa - and so my fork autorsa :D is kinda a big bunch of bash stuff.. but it gives you a hand when you forget to set some variable or sth.. basically the autorsa state module is a bunch of py stuff setting up the easy-rsa bash env, exporting $KEY_DIR, and then actually running th ebash commands, interpreting sutff.
11:33 bluenemo so its kind of a py wrapper around the bash scripts
11:33 XenophonF i love all-singing/all-dancing state formulas
11:34 bluenemo but it looks better than cmd.run and it can handle much more logic ;)
11:34 XenophonF sweet!
11:34 XenophonF you should share this, even if only for pedagogical purposes
11:34 bluenemo :) ok cool. that gives me confidence to put it on github when its done in a few weeks ;) I'll ping you :)
11:34 XenophonF definitely
11:34 XenophonF i haven't see anything written using pyobjects, and this looks really useful
11:35 bluenemo hehe. I wrote the first version only using bash scripts and cmd.run. it was ugly as f*** but it was very educational about runnin gbash with salt M)
11:35 XenophonF i think that's the canonical way to start a formula
11:35 bluenemo ?
11:35 XenophonF start with a command or script, then slowly refactor into something more, i dunno, native
11:36 bluenemo hm
11:36 bluenemo well my problem was I needed more than two states (true / false)
11:36 XenophonF yup yup
11:36 bluenemo for example the formula tests if "files are present"
11:36 bluenemo so I have a csr, a crt and a key and maybe some other stuff I need for autorsa.package_client or sth
11:37 bluenemo so I check if those are present, not present or only some and if only some I want to exception
11:37 bluenemo and so on
11:37 bluenemo that sucks in jinja :)
11:37 XenophonF for real
11:37 bluenemo also now checking files, then using openssl to verify key with crt and so on would be 10 states, now its one
11:38 XenophonF so autorsa is your own state module, right?
11:38 bluenemo and I also forked a easy-rsa 3 feature into my 2.2 fork, my autorsa state module can create client keys WITH passwords :)
11:38 bluenemo for laptops :)
11:38 bluenemo yes
11:38 XenophonF nice
11:39 XenophonF man, that's cool
11:39 bluenemo the state also installs /usr/share/autorsa.. the bash scripts I forked (a bit). i mostly extended for stuff like ./verify-ca-crt and ./verify-key-crt and so on
11:39 bluenemo and ./package-client
11:39 bluenemo :)
11:39 bluenemo the reason I'm always shy to put on github is i'm more of an admin, not a coder.. my code isnt always that pretty ;)
11:39 XenophonF eh, so what
11:40 XenophonF that's the only way to get feedback
11:40 bluenemo using salt for about a year now, I do get better with py..
11:40 bluenemo nah I have a team and stuff, I mostly ask them. also this formula is for a customer, they know py almost better than me.
11:40 bluenemo :)
11:40 bluenemo actually wo almost..
11:41 XenophonF that's awsome
11:41 bluenemo but yeah this one i will publish, its my biggest project in salt so far
11:42 bluenemo how do I write in yaml that client originating in SOURCE vpn and connecting to DESTINATION vpn can access 1) all clients 2) the vpn server 3) a list of clients / and all that preferably with "all protocols" as well as sth like tcp 22..
11:43 bluenemo I'm in between wanting to abstract iptables logik and wanting to just put stuff like "-j ACCEPT" directly into the pillars
11:44 XenophonF hm, let me look through my bookmarks - someone posted a very neat way of expressing iptables stuff in pillar
11:46 AndreasLutro was it me?
11:46 AndreasLutro if it was, https://gist.github.com/anlutro/644fc588e228f0df14b9
11:46 JDiPierro joined #salt
11:47 XenophonF it was, AndreasLutro
11:47 bluenemo hm --dports could be sth.. just use a []..
11:47 XenophonF brb gotta walk the dog before it starts raining :)
11:48 babilen (from the dog or the sky?)
11:48 favadi joined #salt
11:52 otter768 joined #salt
11:53 alxbse joined #salt
11:56 bluenemo lol
12:00 Nebraskka joined #salt
12:01 calvinh joined #salt
12:07 N-Mi joined #salt
12:08 TyrfingMjolnir joined #salt
12:09 denys joined #salt
12:13 pm90_ joined #salt
12:14 mapu joined #salt
12:15 anmolb joined #salt
12:20 pm90__ joined #salt
12:28 impi joined #salt
12:29 johtso joined #salt
12:31 favadi joined #salt
12:36 harkx joined #salt
12:40 elfixit joined #salt
12:42 calvinh joined #salt
12:42 dendazen joined #salt
12:47 racooper joined #salt
12:50 alexlist joined #salt
12:51 calvinh_ joined #salt
12:56 ferbla joined #salt
12:56 DammitJim joined #salt
12:57 dopesong joined #salt
12:58 subsignal joined #salt
12:58 blueyed joined #salt
13:01 tyler joined #salt
13:03 larsfron_ joined #salt
13:04 blueyed What is the process of backporting fixes to the stable/relevant branches?  https://github.com/saltstack/salt/pull/25162 is missing in 2015.8.  Should I create a PR for it?
13:05 DammitJim is there an easy way to "undo" a state?
13:06 rhodgin joined #salt
13:08 AndreasLutro blueyed: I'd just make a comment and ask about it, though I'm sure a backport pr would be accepted if you do it right
13:08 AndreasLutro DammitJim: no
13:11 JDiPierro joined #salt
13:11 jdesilet joined #salt
13:12 mpanetta joined #salt
13:13 DammitJim ugh
13:13 DammitJim with lots of power comes a lot of responsibility
13:14 AndreasLutro absolutely ;)
13:15 AndreasLutro DammitJim: are the specific things you want to prevent from screwing up in salt?
13:15 favadi joined #salt
13:15 DammitJim no, it was a general question
13:15 DammitJim but I do wonder about logging
13:15 DammitJim is there a good mechanismo to log all the things that are being done?
13:16 AndreasLutro by using a custom returner I guess
13:16 DammitJim custom returner... interesting
13:16 DammitJim Ideally, the output would go into a database
13:17 AndreasLutro I've never used one but there should be a mysql returner for example, which I think inserts highstate result data into a database
13:18 DammitJim I'm looking to do the following... get states, pillar, etc from a database (to have versioning)
13:18 DammitJim then also a database for where data files come from (compiled war files, properties files for applications)..
13:18 DammitJim and lastly execution results to go to a database so we can go back and audit our work
13:19 AndreasLutro think you're in for some work then
13:19 AndreasLutro I just use the cli so can't really help
13:21 DammitJim I use the cli
13:21 DammitJim as well
13:21 DammitJim All I mean is I want ways to audit what we are doing
13:21 DammitJim ya know?
13:22 favadi joined #salt
13:22 drawsmcgraw DammitJim: Idunno about all that other stuff, but if you just want the auditing, make sure no one can "su to root" (i.e. only allow sudo), then use a returner for your favorite database.
13:23 DammitJim ok, great.. I'll dig that out... custom returner
13:23 drawsmcgraw That first part is important, else cowboys will just "sudo su - ", then run salt commands. Then you'll just have a bunch of "commands run by root"
13:23 drawsmcgraw You might not need a custom one
13:23 drawsmcgraw There's a *bunch* of stock returners that may fit your need.
13:24 svinota joined #salt
13:24 XenophonF back
13:24 DammitJim thanks
13:24 DammitJim good point... I do need to disable sudo su on my servers
13:26 XenophonF DammitJim: would enabling sudoreplay meet your auditing needs?
13:27 XenophonF DammitJim: i'm using sudoers-formula to set the `log_output` option in .../etc/sudoers
13:27 XenophonF that should catch people who run `sudo su -` or `sudo -s`
13:27 av_ joined #salt
13:29 DammitJim that's OK XenophonF .. thanks for the input.. probably not necessary at this point, but will need to look at it
13:30 XenophonF iirc you can make states like file.managed keep backup files
13:30 XenophonF dunno about others
13:31 XenophonF it'd be nice if salt kept deltas of everything though, to make reverting easier
13:31 XenophonF tbh i work around this in my workflow
13:31 moski joined #salt
13:32 numkem joined #salt
13:32 XenophonF where i have states and pillars in git, with my existing config in the production branch and the proposed config in the staging branch
13:32 XenophonF and i run staged changes via state.sls ... saltenv=staging
13:33 XenophonF if my user acceptance tests fail, i can always go back to the original config by running a highstate or state.sls ... saltenv=production
13:33 larsfronius joined #salt
13:33 XenophonF it's kind of like github flow only for states, as opposed to code
13:34 XenophonF the production branch of my state repo always reflects what's currently running in production
13:34 DammitJim XenophonF, yes, what you said.... iirc you can make states like file.managed keep backup files
13:35 DammitJim that's a good suggestion
13:35 DammitJim I'd like to get there... but honestly, I'm trying to learn git to do it right
13:35 av_ Hi guys. I am using salt cloud to deploy on AWS. I have specified ssh username and private key(that is .pem file) in `cloud.providers`. I have also specified security group which makes port 22 open. When I create a new instance with salt cloud, I keep getting file permissions for private key are open. How do I fix it?
13:35 DammitJim I've done subversion in the past
13:36 XenophonF same here
13:36 XenophonF although i went svn->hg->git
13:36 babilen My condolences
13:36 XenophonF av_: how are you deploying the private key to the computer on which salt-cloud runs?
13:37 protoz joined #salt
13:37 larsfron_ joined #salt
13:37 XenophonF av_: as long as you set the permissions of the file to 600 (u=rw,go=), you should be set
13:38 av_ XenophonF: as one of state, permissions are 600
13:38 av_ yes, permissions I have set is 600
13:38 XenophonF av_: does that match the file system?
13:38 av_ however it is in one states file. So I am guess the error happens before it starts running states?
13:38 av_ yes, matches the file system
13:38 XenophonF babilen: hah well, i kind of liked hg but the ecosystem around git is tons better
13:39 XenophonF av_: dunno - can you post the actual error message on gist or something?
13:39 av_ sure, I will post the full log, I am running the command again
13:39 Aidin1 joined #salt
13:40 markm__ joined #salt
13:40 TyrfingMjolnir joined #salt
13:44 perfectsine joined #salt
13:44 av_ XenophonF: here you go: http://dpaste.com/3NRMA76
13:44 dopesong_ joined #salt
13:45 larsfronius joined #salt
13:45 AndreasLutro well, it clearly says the pem file is 644
13:46 JDiPierro joined #salt
13:46 av_ AndreasLutro: yes
13:47 av_ but how do I change it or specify it?
13:47 AndreasLutro chmod?
13:47 av_ I have already specified it in one of the states
13:47 av_ wait, you want me to set it locally and try?
13:47 av_ wait let me do that
13:47 XenophonF av_: yes
13:47 AndreasLutro I think the message is coming from the local system, not the remote
13:47 av_ ohh
13:47 av_ oops
13:47 av_ let me try
13:47 AndreasLutro why would the remote node try to ssh somewhere?
13:47 XenophonF that error's coming from the computer on which you're running salt-cloud
13:48 Akhter joined #salt
13:48 av_ yes, got it
13:48 av_ let me fix it
13:48 pdayton joined #salt
13:50 Akhter joined #salt
13:51 timoguin joined #salt
13:52 zmalone joined #salt
13:52 winsalt joined #salt
13:52 otter768 joined #salt
13:53 larsfron_ joined #salt
13:55 I joined #salt
13:56 TheoSLC joined #salt
13:58 pm90_ joined #salt
13:58 favadi joined #salt
13:58 pdayton joined #salt
13:59 JDiPierro I need to change the permissions of a file after it's extracted via archive.extracted, is there a better way to do that than just a chmod through cmd.run?
14:00 babilen JDiPierro: no
14:00 breakingmatter joined #salt
14:00 cpowell joined #salt
14:00 babilen Well, you could fix the archive
14:03 protoz_ joined #salt
14:06 debian112 joined #salt
14:06 ssc joined #salt
14:06 andrew_v joined #salt
14:10 malinoff joined #salt
14:11 TooLmaN joined #salt
14:12 favadi joined #salt
14:13 aqua^c joined #salt
14:16 numkem joined #salt
14:16 fredvd joined #salt
14:16 khaije1 joined #salt
14:17 teebes joined #salt
14:20 XenophonF JDiPierro: cmd.script executing mtree
14:20 XenophonF JDiPierro: or on windows a batch/powershell script that fixes up permissions
14:20 XenophonF JDiPierro: or a file.directory state
14:20 XenophonF unfortunately they're all kind of ugly, especially if you're on windows
14:22 giantlock joined #salt
14:24 calvinh joined #salt
14:25 Rumbles joined #salt
14:25 rhodgin joined #salt
14:26 berserk joined #salt
14:28 Brew joined #salt
14:30 ekristen joined #salt
14:31 calvinh_ joined #salt
14:32 scoates anyone have a project to collect AWS inventory and pipe it into the salt mine?
14:34 ALLmightySPIFF joined #salt
14:34 jalbretsen joined #salt
14:36 TheoSLC joined #salt
14:37 rhodgin joined #salt
14:38 jhauser joined #salt
14:39 froztbyte dammit, I think I'm failing at mines
14:40 losh joined #salt
14:40 froztbyte a mine.update on my targets returns True, the initial stuff I push into pillar shows up in the pillar data on the minions, but mine.get is all empty
14:40 favadi joined #salt
14:40 froztbyte (trying with network.ip_addrs atm, from the docs example)
14:43 winsalt is the win repo exclusively through git now?
14:43 favadi joined #salt
14:43 winsalt for 2015.8?
14:45 fe92 joined #salt
14:49 ImNotARobot joined #salt
14:49 favadi joined #salt
14:49 bmac2 joined #salt
14:51 Rumbles joined #salt
14:51 froztbyte okay seriously. I'm trying to copy the upstream mine example, and it's just not working
14:52 froztbyte trying to use pillar data to drive what functions I want. how do I start debugging this?
14:52 froztbyte running `salt-call -l debug mine.update` on a minion shows me stuff I'd expect
14:53 froztbyte but mine.get shortly thereafter is still blank
14:53 froztbyte (the latter on the master)
14:55 froztbyte actually what. I just got a change. so I'd been testing everything from the master, constantly to no avail. on the first run from the minion, I now seem to have mine data
14:56 froztbyte is there a bug with `salt '*' mine.update` initiated from the master?
14:57 froztbyte looks like 2015.5.0+ds-1~bpo70+1 here still
14:57 fredvd joined #salt
14:58 forrest joined #salt
14:59 Roll joined #salt
14:59 baba_ joined #salt
15:00 dfinn joined #salt
15:01 Roll Just heard about salt. Where's the best place to start?
15:02 khaije1 Hi all, kind of an open ended question about minion naming conventions... when using mostly virtualized minions, (Docker, VMware, et al), that are likely to be cloned and forked, snapshotted and reverted ... is there ever a problem in having a minion show up with duplicate a 'id' ?
15:02 dyasny joined #salt
15:02 khaije1 I'm wondering if I shouldn't address minions according to their MAC address, though this is a _lot_ less friendly.
15:03 pm90__ joined #salt
15:03 irctc878 joined #salt
15:03 pm90___ joined #salt
15:04 khaije1 At present minion names are $(hostname -s), but I wonder if they should include the fqdn ... any organized though, convention and/or concensus on how best to handle this topic?
15:04 khaije1 organized *thought*
15:05 irctc878 good morning all. Is there a difference between the variable names 'id' and 'name' when calling a reactor/sls?
15:05 irctc878 left #salt
15:05 irctc878 joined #salt
15:06 froztbyte rol[tab]
15:06 irctc878 left #salt
15:06 hasues joined #salt
15:06 hasues left #salt
15:07 gadams999 joined #salt
15:07 JDiPierro irctc878: I'm not sure about reactor states, but for regulare states in SLS files that mostly depends on the state function being used. Most of them will clone the ID into the name if name isn't present but I believe some of them have a difference. It's also good practice to use a friendly name instead of a file path as the ID of something like file.managed so you can move things without changing requirements everywh
15:07 JDiPierro ere.
15:07 * gadams999 has rejoined with proper nick
15:08 llb42 left #salt
15:09 gadams999 JDiPierro: thanks. I'm going through some examples that use data['name'] in jinja (reactor called sls file) that barfs that dict object doesn't exist. data['id'] is there (2015.3.0 master/minion).
15:10 gadams999 just trying to crawl by calling highstate at the moment. Working with id for the minion tgt, but curious why examples call out 'name' instead.
15:13 winsalt gadams, are you using data['data']['name'] in the reactor?  for example in my reactor i have {% set data = data['data'] %}
15:16 X67r joined #salt
15:17 gadams999 winsalt: the reactor just has a call to an .sls. Ultimately under a cmd.state.highstate I have a "tgt: {{ data['id] }}" line which invokes correctly
15:17 froztbyte grrr. followed the same command steps as https://github.com/saltstack/salt/issues/21871. and that made me get data.
15:18 froztbyte gah
15:18 froztbyte heisenbugs
15:22 winsalt gadams999, have you run salt-run state.event pretty=True on the master, it lets you see what the dictionary returned from a beacon looks like
15:24 kaptk2 joined #salt
15:25 ageorgop joined #salt
15:25 pm90_ joined #salt
15:26 gadams999 winsalt: gotcha. I was focusing on why the reactor was failing but I see the beacon data from the minion restart on my debug. Looks like _stamp, pretag, cmd, tag, data, and id are all part of the data dict sent over.
15:27 pm90__ joined #salt
15:27 gadams999 winsalt: reactor firing, now to figure out while file.recurse with a "clean" parameter is failing. :)
15:31 Guest47337 joined #salt
15:34 anotherZero joined #salt
15:39 DammitJim joined #salt
15:40 alemeno22 joined #salt
15:42 winsalt_ joined #salt
15:48 evle joined #salt
15:48 jfred I've been running into an issue with a machine that I'm working with, where highstating takes upwards of 10 minutes. Its sls files aren't any more complex than the other machines managed by salt (less, actually) and it doesn't have anything in pillar that the other machines don't have. stracing the process and its threads (-fp), I see output like this: https://gist.github.com/jfrederickson/72853a3ba476c7681468
15:49 zwi joined #salt
15:49 jfred running the minion with '-l debug', it takes about a second between each line printed to stdout, and the strace seems to hang on the sendto() syscalls (e.g. line 33)
15:49 X67r joined #salt
15:50 jfred has anyone seen anything like this before?
15:52 DammitJim so, you don't think this has to do with network connectivity as you try to download stuff to the minion?
15:53 jfred these machines are on a relatively fast private network within the same datacenter, and as far as I can tell the connection between my master and that minion is fine
15:53 otter768 joined #salt
15:54 jfred no packet loss, etc
15:54 jfred sub-1ms latency
15:54 David_B55 joined #salt
15:55 DammitJim no, I mean, getting stuff from the outside
15:55 Kelsar joined #salt
15:56 debian112 can you pull from another repo via top.sls?
15:56 debian112 or do you have to do it via a state file with an include:
15:58 jfred DammitJim: oh, I see - well that machine's packages are currently up to date, so the only thing it would be contacting the outside would be to check for package updates, afaik
15:58 jfred the majority of what it's doing is updating config files that were changed on the master
15:59 jfred (this is a system that's already managed through salt and has been highstated before)
16:03 dopesong joined #salt
16:03 ALLmightySPIFF joined #salt
16:09 tzero joined #salt
16:10 bastion1704 joined #salt
16:12 snarfy joined #salt
16:13 X67r joined #salt
16:15 aparsons joined #salt
16:17 katyucha joined #salt
16:18 clintber_ joined #salt
16:18 dthom91 joined #salt
16:20 Akhter joined #salt
16:21 rhodgin joined #salt
16:23 breakingmatter joined #salt
16:27 Bryson joined #salt
16:28 Rumbles joined #salt
16:29 winsalt_ anyone here using 2015.8 on windows?  winrepo is bugging out and salt cant find any packages anymore.
16:30 tzero joined #salt
16:30 druonysuse joined #salt
16:30 druonysuse joined #salt
16:34 X67r joined #salt
16:35 sbogg joined #salt
16:36 dfinn1 joined #salt
16:37 forrest joined #salt
16:38 ponpanderer joined #salt
16:40 ponpanderer anyone aware of any unofficial custom states implementing either nova or swift execution modules?
16:41 KyleG joined #salt
16:41 KyleG joined #salt
16:41 tzero joined #salt
16:42 sunkist joined #salt
16:43 jodv joined #salt
16:44 writtenoff joined #salt
16:44 jmreicha joined #salt
16:44 bastion1704 joined #salt
16:46 druonysuse joined #salt
16:46 druonysuse joined #salt
16:47 pgoetz joined #salt
16:48 pgoetz joined #salt
16:49 sunkist joined #salt
16:49 druonysus joined #salt
16:50 sunkist1 joined #salt
16:50 druonysus joined #salt
16:50 druonysus joined #salt
16:57 tzero joined #salt
16:58 DaveQB joined #salt
16:59 Ryan_Lane iggy: our change was because many of the boto state modules we wrote call to other state modules
16:59 ALLmightySPIFF joined #salt
16:59 Ryan_Lane like boto_asg calling out to boto_cloudwatch to make alarms for the asg
16:59 Ryan_Lane state.single, when called from another state, causes salt to reload its modules
17:00 Ryan_Lane some of our resources would add like 2-3 alarms for themselves, so we were calling state.single a lot of time
17:00 Ryan_Lane times*
17:00 Ryan_Lane each time we call that, it takes ~20-30s for the modules to reload
17:00 Ryan_Lane it quickly adds up to a lot of wasted time
17:01 Ryan_Lane so we got core to add __states__ so that we don't need to reload modules
17:01 Ryan_Lane there's only a couple other state modules that do this, so it's not likely worth a blog post :)
17:04 druonysus joined #salt
17:08 Fiber^ joined #salt
17:08 pcdummy joined #salt
17:11 dthom911 joined #salt
17:13 pm90_ joined #salt
17:17 Stealtherald joined #salt
17:17 debian112 can you pull from another repo via top.sls, or do you have to do it via a state file with an include:?
17:17 rhodgin joined #salt
17:18 Stealtherald Where can I submit for changes to 2015.8.0 docs?
17:18 dthom91 joined #salt
17:19 zmalone You can make a github pull request
17:19 dopesong joined #salt
17:20 zmalone https://github.com/saltstack/salt/tree/2015.8/doc is the repo, as I recall
17:20 zmalone (for Stealtherald)
17:28 JDiPierro joined #salt
17:29 winsalt joined #salt
17:30 elsmo Hey all,
17:30 elsmo Am getting a rather unhelpfull error: "State blockdev.formatted found in sls mesos.master is unavailable"
17:31 elsmo from this state: http://pastebin.com/vDgiiGLj
17:31 elsmo any clues?
17:32 dendazen joined #salt
17:33 whytewolf elsmo: is_windows?
17:33 svinota joined #salt
17:33 elsmo ubuntu :)
17:33 aparsons joined #salt
17:34 ageorgop joined #salt
17:34 whytewolf I ask cause the virtual only checks this if salt.utils.is_windows() and if that is true returns false
17:37 tzero joined #salt
17:37 whytewolf odd, what does sys.platform return on your minion
17:37 Guest47337 joined #salt
17:38 dfinn1 joined #salt
17:39 druonysus joined #salt
17:42 tannermkerr joined #salt
17:42 larsfronius joined #salt
17:42 elsmo hmmm, sys.platform is also not available :?
17:43 whytewolf python -c 'import sys; print sys.platform'
17:43 whytewolf should be linux2
17:43 whytewolf [or something close to it]
17:45 elsmo ahhh
17:45 elsmo it is indeed linux2 :)
17:45 iggy Ryan_Lane: You'd be surprised... I'd probably make it required reading around here if only for "it can be done" reasons
17:47 whytewolf elsmo: then the next step is to use salt-call -l debug on your minion to run the state and see if blockdev is throwing an error
17:48 elsmo whytewolf: i did indeed, but thats all i get :(
17:48 dewdrop joined #salt
17:49 aqua^c joined #salt
17:50 dthom91 joined #salt
17:51 aparsons_ joined #salt
17:51 whytewolf elsmo: at this point i have no idea. what version are we working with here?
17:53 elsmo 2013.5.3  - salt-call 0.17.5
17:53 whytewolf ...
17:53 genediazjr joined #salt
17:53 whytewolf your minion doens't have it
17:53 elsmo sorry
17:53 elsmo 2015.5.3
17:54 whytewolf that the master or the minion?
17:54 elsmo both :)
17:55 otter768 joined #salt
17:55 whytewolf then where did the 0.17.5 you just posted come from?
17:55 tzero joined #salt
17:55 elsmo the salt-call --version :)
17:55 perfectsine joined #salt
17:56 elsmo hmmmmm
17:56 SneakyPhil joined #salt
17:56 whytewolf ... soooo. sounds like the minion is running 0.17.5
17:56 viq joined #salt
17:56 baweaver joined #salt
17:56 elsmo yeah, i think your right :(
17:56 elsmo i added the pps but it chose the ubuntu packaged one anyway :(
17:56 elsmo good spot!
17:57 whytewolf which is why that would be failing [blockdev was put in around 2014.7
17:57 whytewolf I really wish ubuntu would either update the default repo or remove salt from their repo.
17:58 elsmo :)
17:59 elsmo yeah bit odd how they both have a 'score' of 500 from apt after installing the ppa
17:59 wendall911 joined #salt
18:02 zerthimon joined #salt
18:02 tzero joined #salt
18:02 baweaver joined #salt
18:03 oeuftete joined #salt
18:06 Guest47337 joined #salt
18:06 jmreicha joined #salt
18:07 ajw0100 joined #salt
18:08 rhodgin joined #salt
18:08 oeuftete joined #salt
18:09 DammitJim what is the proper way for me to copy a folder?
18:10 JDiPierro DammitJim: File.recurse can copy a folder that the master knows about to the minion.
18:11 DammitJim I meant more a folder copied that already exists in the minion
18:11 DammitJim actually, what I have is a master that has a zip file
18:11 DammitJim I am archive.extracted it
18:11 DammitJim then renaming the folder that it extracts to
18:12 DammitJim but when my for loop in my state goes and tries to do this again, it doesn't decompress the zip file again
18:13 DammitJim can I use file.copy for this?
18:13 JDiPierro Do you have keep: True on the archive.extracted state? If not I think the zip file is discarded.
18:13 DammitJim JDiPierro, I don't have keep: True
18:13 DammitJim but I don't care about the zip file
18:13 Cliph joined #salt
18:13 DammitJim I want the decompressed folder
18:13 JDiPierro Ok, then yeah I'd try the file.copy state
18:14 oeuftete joined #salt
18:14 dopesong joined #salt
18:15 baweaver joined #salt
18:16 dfinn left #salt
18:17 SheetiS1 joined #salt
18:17 _JZ_ joined #salt
18:24 DammitJim file.copy doesn't seem to work
18:25 tzero joined #salt
18:27 aw110f joined #salt
18:29 deus_ex joined #salt
18:32 baweaver joined #salt
18:38 dfinn joined #salt
18:39 rocket joined #salt
18:40 tzero joined #salt
18:40 dopesong_ joined #salt
18:44 msx joined #salt
18:44 larsfronius joined #salt
18:45 GreatSnoopy joined #salt
18:46 blueyed Can the `postgres_database.present` state be made to wait for the DB server to be up?  It fails with 'psql: server closed the connection unexpectedly', probably because of the warmup of the postgres Docker container.
18:47 clintberry2 joined #salt
18:47 blueyed Is there some generic state to wait for a port to be reachable (which might not be enough), or to retry it (https://github.com/saltstack/salt/issues/26350)?
18:48 ajw0100 joined #salt
18:50 tzero joined #salt
18:51 denys joined #salt
18:52 ALLmightySPIFF joined #salt
18:55 snarfy do people use salt-mine as like a system of record or in some kind of svc discovery mechanism
18:57 danlsgiga joined #salt
18:58 tanta_g joined #salt
18:58 danlsgiga hey... I'm trying to have a default reactor for some orchestrator states I have and I'm trying to pass data['data'] | yaml(False) through pillar but the formatted yaml is indentless
18:59 danlsgiga Am I able to do this? Pass my entire event data to my orchestrator state?
19:00 danlsgiga I just don't want to have 1 reactor for each action in my orchestration
19:03 tzero joined #salt
19:03 pdayton1 joined #salt
19:04 _JZ_ joined #salt
19:04 timoguin joined #salt
19:05 berserk joined #salt
19:12 polishdub joined #salt
19:18 tzero joined #salt
19:22 baweaver joined #salt
19:23 Dev0n joined #salt
19:28 fxhp joined #salt
19:29 giantlock joined #salt
19:29 PeterO_ joined #salt
19:31 tzero joined #salt
19:31 diegows joined #salt
19:32 zerthimon joined #salt
19:34 dthom91 joined #salt
19:35 teebes joined #salt
19:35 tomspur joined #salt
19:35 pdayton joined #salt
19:36 pdayton joined #salt
19:39 pdayton joined #salt
19:39 baweaver joined #salt
19:42 elsmo joined #salt
19:45 stefan_ joined #salt
19:45 pdayton joined #salt
19:46 pdayton joined #salt
19:47 tzero joined #salt
19:48 TheoSLC joined #salt
19:55 otter768 joined #salt
19:56 ALLmightySPIFF joined #salt
19:57 bhosmer joined #salt
20:03 dthom91 joined #salt
20:07 tzero joined #salt
20:08 _JZ_ joined #salt
20:09 raygunsix joined #salt
20:11 ajw0100 joined #salt
20:18 jalbretsen joined #salt
20:22 diegows joined #salt
20:27 pmcnabb joined #salt
20:29 stickmack joined #salt
20:30 superseb joined #salt
20:30 cberndt joined #salt
20:31 giantlock joined #salt
20:31 viq joined #salt
20:31 Qwazerty joined #salt
20:31 elsmo joined #salt
20:32 masterkorp joined #salt
20:33 joehh1 joined #salt
20:38 |Fiber^| joined #salt
20:38 racooper joined #salt
20:41 druonysuse joined #salt
20:43 whidbeywalker joined #salt
20:44 Rumbles joined #salt
20:48 dfinn1 joined #salt
21:05 dopesong joined #salt
21:06 writtenoff joined #salt
21:08 msx joined #salt
21:10 BlackFX joined #salt
21:11 BlackFX {% for host in salt['mine.get']('G@environment:'+environment, 'role', expr_form='compound')|groupby(role) %}
21:11 BlackFX is this possible? I cant seem to get it to work :)
21:14 joe_n joined #salt
21:17 dfinn joined #salt
21:22 druonysuse joined #salt
21:25 dthom91 joined #salt
21:33 sjorge joined #salt
21:33 sjorge joined #salt
21:34 opensource_ninja joined #salt
21:35 aparsons joined #salt
21:43 breakingmatter joined #salt
21:43 raygunsix_ joined #salt
21:44 joe_n joined #salt
21:45 baweaver joined #salt
21:48 BlackFX anyone using salt-mine ?
21:53 nofxrok joined #salt
21:56 otter768 joined #salt
22:02 dthom911 joined #salt
22:04 dustywusty joined #salt
22:04 joe_n joined #salt
22:07 dopesong joined #salt
22:07 dthom91 joined #salt
22:08 danlsgiga joined #salt
22:08 babilen BlackFX: Many people
22:08 danlsgiga how can I call a state.orchestrate inside a orchestrate state?
22:09 danlsgiga I want to call another orchestrate file from inside an orchestrate file... is that possible?
22:09 danlsgiga or I must send another event from the orchestrate and filter it in the reactor?
22:10 dthom911 joined #salt
22:13 dthom91 joined #salt
22:13 dthom91 joined #salt
22:19 dthom91 joined #salt
22:20 xMopxShell joined #salt
22:22 dthom911 joined #salt
22:31 dthom91 joined #salt
22:38 conan_the_destro joined #salt
22:42 Ryan_Lane basepi: https://github.com/saltstack/salt/pull/27485 <-- I don't understand this test failure
22:42 Ryan_Lane I understand the lint ones, but the unit tests are telling me that config.option isn't a keyword in __salt__
22:43 protoz joined #salt
22:44 BlackFX I have a list in Jinja and i am appending items
22:44 BlackFX I want to only append unique items
22:44 BlackFX is there a way to check if an item is already in the list
22:45 Ryan_Lane BlackFX: can you use a set?
22:46 larsfronius joined #salt
22:46 BlackFX I dont see why not - how would I do that ?
22:46 Ryan_Lane if not, you can do 'if value in list', just like in python
22:46 rim-k joined #salt
22:46 Ryan_Lane hm. maybe sets aren't available
22:47 opensource_ninja joined #salt
22:48 sjorge joined #salt
22:48 sjorge joined #salt
22:48 Ryan_Lane BlackFX: anyway, you can do {% if value not in list %} {% list.append(value) %}
22:48 BlackFX thanks
22:48 Ryan_Lane yw
22:52 dthom91 joined #salt
22:53 drawsmcgraw left #salt
22:58 dthom911 joined #salt
22:59 orion joined #salt
22:59 orion Hi. How do I install 2015.8.0 on Ubuntu 14.04?
23:00 orion (As opposed to 2015.5.3, which is the only one available at this time)
23:00 ahammond orion have you checked repo.saltstack.com?
23:01 whytewolf http://repo.saltstack.com/#ubuntu
23:01 orion Oh cool. I wasn't aware that was available.
23:01 ahammond orion, best thing since sliced bread
23:01 sjorge joined #salt
23:01 sjorge joined #salt
23:02 orion http://repo.saltstack.com/ubuntu/ <-- 404 not found. :(
23:02 whytewolf http://repo.saltstack.com/#ubuntu < notice the #
23:02 orion (I clicked the Browse Repo button)
23:02 dramagods joined #salt
23:02 whytewolf oh, that link is wrong
23:03 whytewolf http://repo.saltstack.com/apt/ubuntu/
23:03 orion Ah, cool.
23:04 orion So, if I install 2015.8.0 on the salt-master, when salt-cloud bootstraps servers, will it configure the repo automatically?
23:04 whytewolf I don't think the bootstrap script has been updated yet for ubuntu
23:05 orion So, I need to bake the repo in to the AMI.
23:06 ageorgop joined #salt
23:07 CheKoLyN joined #salt
23:07 pdayton joined #salt
23:08 orion whytewolf: Oh, it looks like the PPA is very much diverged from the repo you pointed me to.
23:09 baweaver joined #salt
23:09 orion That's strange to me because the PPA is considered official.
23:09 whytewolf yes, the ppa was kind of third party. and the guy that was running it has let it go into disrepair.
23:09 orion Oh man. :/
23:10 fivehole joined #salt
23:10 Aidin joined #salt
23:10 whytewolf thats why it is still on 2015.5.3 even though the 2015.5 release is 2015.5.5
23:10 whytewolf and has been for months
23:11 orion What a shame. He should turn over the PPA to someone else.
23:11 orion Who cares.
23:11 pdayton1 joined #salt
23:12 jalaziz joined #salt
23:12 whytewolf orrr, salt should put together a trully offical repo that they host on their own ... oh wait thats what they did.
23:13 whytewolf which is actually better because it puts together the base packages that were not being covered through the other sources anyway. like tornado.
23:15 conan_the_destro joined #salt
23:15 msx joined #salt
23:18 fivehole Does anybody know how to debug “Package <package name> not found in the repository” in the windows repo?  This there a way to figure out where it is searching?
23:20 fivehole I have tried puting it in debug and trace but there isn’t any extra info.  Just that the package can’t be found in the repo.
23:23 denys joined #salt
23:24 Aidin left #salt
23:30 breakingmatter joined #salt
23:30 rickflare2 joined #salt
23:30 Eureka70_ joined #salt
23:31 leev_ joined #salt
23:31 KyleG1 joined #salt
23:31 jalaziz joined #salt
23:32 omenlabs joined #salt
23:32 druonysus joined #salt
23:32 druonysus joined #salt
23:32 nledez joined #salt
23:32 nledez joined #salt
23:32 honestly_ joined #salt
23:33 Hipikat joined #salt
23:33 Lionel_Debroux_ joined #salt
23:33 amatas joined #salt
23:33 Number6 joined #salt
23:33 carmony joined #salt
23:33 nihe joined #salt
23:33 dyasny joined #salt
23:34 kevinquinnyo joined #salt
23:34 Ssquidly joined #salt
23:34 JPaul joined #salt
23:34 eliasp joined #salt
23:34 Guest25336 joined #salt
23:34 jerrcs joined #salt
23:34 jorb joined #salt
23:34 MadsRC joined #salt
23:34 onovy joined #salt
23:34 \ask joined #salt
23:34 twobitsprite joined #salt
23:34 PI-Lloyd joined #salt
23:34 vlcn joined #salt
23:34 katyucha joined #salt
23:34 kiper joined #salt
23:34 fxhp joined #salt
23:34 keekz joined #salt
23:35 orion left #salt
23:35 pppingme joined #salt
23:35 ALLmightySPIFF joined #salt
23:37 dramagods left #salt
23:39 kaictl joined #salt
23:44 breakingmatter joined #salt
23:45 micko joined #salt
23:50 bfoxwell joined #salt
23:57 basepi Ryan_Lane: I commented
23:57 otter768 joined #salt

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