Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2014-01-05

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

All times shown according to UTC.

Time Nick Message
00:02 john_934251 left #salt
00:03 oz_akan_ joined #salt
00:06 pass_by_value joined #salt
00:07 cym3try joined #salt
00:08 pdayton joined #salt
00:20 pass_by_value joined #salt
00:27 cro joined #salt
00:38 pdayton joined #salt
00:40 beach joined #salt
00:42 beach hey gang! Anyone around to help a fool trying to get gitfs rolling? Everytime I enable it in the master config state.highstate can't find '<role>' in env 'base'
00:42 redmin joined #salt
00:43 forresta beach, hi there, I assume you've followed through the doc tutorial?
00:44 beach yes, as best as I can
00:44 beach seemed simple enough...
00:44 pdayton1 joined #salt
00:45 forresta ok, so it's all in your master branch right?
00:45 forresta or for your testing
00:46 beach if you're asking if my .sls files are in my repo... well yeah. But, I was only planning on using git for managed files not states
00:47 beach well... let me restate that. I was starting out getting familar with using gitfs with managed files. I hadn't got to the point of using states there yet
00:47 forresta ok
00:47 forresta so you have no states in your gitfs tree?
00:48 beach yeah they are there, but I didn't build the path to them. I was simply trying to get a minion to pull down a readme.txt file into /tmp to prove things out
00:49 forresta hmm ok, I haven't played with gitfs too much.
00:50 forresta can you try to follow through the tutorial on adding a test sls file?
00:50 forresta could just be a managed file or something
00:52 pdayton joined #salt
00:59 srage joined #salt
00:59 beach well, I've changed the order of backend_filesserver order to roots, git and I'm able to synch states okay now. and in the master log I'm getting a Host key verification failed -- which is good i guess. I just need to figure out why my key is failing. it works fine from the cmd line
01:03 oz_akan_ joined #salt
01:15 srage joined #salt
01:20 xzarth_ joined #salt
01:34 MohShami_ joined #salt
01:37 jfzhu_us I am running into a situation whenever I run state.highstate for my vagrant vm using salt 0.17.4, it only occasionally returns something. Does salt have a timeout where if a minion has not returned anything in a certain time then the master won't output anything to the cli
01:40 kedo39 jfzhu_us: try using --show-timeout
01:41 kedo39 you can also set the timeout with --timeout
01:41 kedo39 default behavior is to not print any error, which i'm not a big fan of :/
01:42 jfzhu_us kedo39: Is it salt --show-timeout or?
01:42 kedo39 yea, yea
01:43 kedo39 whoops*
01:43 kedo39 yea
01:43 toastedpenguin joined #salt
01:43 redmin joined #salt
01:44 jfzhu_us kedo39: Is there any way to force errors to come out because its frustrating not knowing what's going on
01:44 kedo39 oh sorry, i meant any error for timeouts
01:45 kedo39 when it doesn't timeout and there's an error, it shows it fine
01:45 higgs001 joined #salt
01:46 jfzhu_us kedo39: Yea i meant the error for timeout
01:46 jfzhu_us kedo39: Do you know if there's any way for me to force salt to stop being silent about timeouts
01:46 kedo39 i just alias salt to be 'salt --show-timeout'
01:49 jfzhu_us kedo39: That actually works really well. Thanks
01:49 kedo39 np
02:00 blarghmatey joined #salt
02:04 oz_akan_ joined #salt
02:05 taion809 joined #salt
02:29 kiorky joined #salt
02:49 Gifflen joined #salt
02:54 redmin joined #salt
03:05 oz_akan_ joined #salt
03:05 logix812 joined #salt
03:11 darrend joined #salt
03:17 Gifflen joined #salt
03:23 pass_by_value joined #salt
03:28 Gifflen joined #salt
03:30 CheKoLyN joined #salt
03:38 kermit joined #salt
03:41 Gifflen_ joined #salt
03:42 Gifflen__ joined #salt
03:45 cachedout joined #salt
04:02 kadel joined #salt
04:06 oz_akan_ joined #salt
04:06 redmin joined #salt
04:07 Gifflen joined #salt
04:20 linx_0x I'm trying to install salt from the ubuntu ppa: can anyone help... this is what I have but it's not working: http://pastebin.com/SHQhYi38
04:22 kedo39 did you repo "pkgrepo" ?
04:22 kedo39 typo*
04:23 linx_0x hah .. yeah thanks
04:31 linx_0x what's the syntax for requiring the salt-minion package "greater than or equal to" version 0.17.4 ?
04:31 linx_0x is this possible ?
04:34 kedo39 linx_0x: look under the usagd examples on http://docs.saltstack.com/ref/states/all/salt.states.pkg.html#salt.states.pkg.installed
04:34 kedo39 usage*
04:47 taion809 joined #salt
04:53 sroegner_ joined #salt
04:53 anuvrat joined #salt
05:00 bhosmer joined #salt
05:06 oz_akan_ joined #salt
05:10 Tekni joined #salt
05:17 rojem joined #salt
05:17 redmin joined #salt
05:26 oz_akan_ joined #salt
05:34 smkelly joined #salt
05:34 lyddonb joined #salt
05:34 jphall joined #salt
05:34 smkelly joined #salt
05:35 melinath joined #salt
05:36 ecdhe joined #salt
05:36 jfzhu_us Can you extend salt pillars?
05:38 ecdhe I'm setting up a shared web hosting site, and I'm using salt to manage my infrastructure.
05:38 forresta jfzhu_us, I don't think so right now: https://github.com/saltstack/salt/issues/3991
05:39 ecdhe I was curious about exposing salt to my customers... is there a good way to let shared hosting users use salt to deploy to the shared server without letting them change the packages that are installed, etc?
05:39 forresta jfzhu_us, your best bet would be to use multiple pillars.
05:40 forresta ecdhe, ehh not really, there aren't 'groups' regarding control as far as I know inside of Salt.
05:40 jfzhu_us forresta: Thanks. I should've asked first before i shot myself in the foot
05:40 forresta jfzhu_us, no worries
05:40 ecdhe I could allow a salt-client to run as the customer's user in non-daemon mode--then linux would keep the salt-client from installing packages, changing network configuration etc...
05:41 forresta ecdhe, your best bet that I can think of would be to write states to install what users need, and then provide them some front end interface, maybe like: http://stackd.io/
05:41 forresta if those slackers ever finish it :p
05:41 forresta oh maybe none of them are in here currently, let's see..
05:41 forresta oh yea abemusic isn't here currently, bummer
05:41 ecdhe Shared hosting is nice because it's cheap, but I think it would be handy to give customers a way to scale from the shared hosting environment to a larger architecture.
05:42 forresta ecdhe, I used to work for a web host, I would never give customers salt access.
05:42 forresta it's too powerful, and customers are too clueless.
05:42 forresta trying to lock it down would be a major pain
05:42 ecdhe Is the linux user model too permissive?
05:43 ecdhe If the shared user only has write access to it's own directory, would that work?
05:43 forresta ecdhe, I honestly don't know, when you install salt you need to make sure that it has access to certain things (even when running as a masterless minion)
05:43 forresta so it has to access pki directories and such
05:43 forresta http://docs.saltstack.com/topics/nonroot.html
05:43 forresta otherwise it will crap out
05:44 ecdhe Even so, pki info will relate to the customer...
05:44 forresta only to that customer?
05:44 anuvrat joined #salt
05:44 forresta in a typical salt setup, that pki directory contains data for all machines
05:44 forresta are you using a single master -> minion setup for users?
05:44 ecdhe Yes--if a customer is running their own master, then their client will need pki info for that master.
05:44 forresta ahh interesting...
05:44 jfzhu_us ecdhe: client_acl might be what you are looking for
05:45 jfzhu_us in the master config file it looks like you can specify a subset of teh commands that can be run
05:45 forresta jfzhu_us, yea that's a fair point to make if each user has their own master
05:45 forresta I don't know how that will work for states though
05:46 forresta because I'll say 'screw you buddy', and write my own state that does all of that stuff if I can
05:46 forresta I'm pretty biased though as I am looking at  it from a web hosting engineer perspective
05:46 forresta so maybe your users are a little more trustworthy
05:48 jfzhu_us forresta: well a non admin user won't have write access to /srv/salt so they shouldn't be able to write a sls file to execute or is there another way to run sls files?
05:48 forresta oh you're only planning on letting them use modules?
05:48 ecdhe So the host 'web' is a minion in my infrastructure.  The main salt client runs as root and installs and configures packages, users, etc.
05:48 ecdhe The goal was to have another salt-client run as a user that only has write access to the user directory.
05:49 ecdhe So, if you're maintaining example.com as user20, you would only have write access to /home/user20/example.com
05:49 forresta yea I understand what you mean
05:50 ecdhe So salt couldn't install system packages--but it could pull down custom PHP versions, deploy your app, etc.
05:50 forresta but you'd still need to execute states to deploy the app
05:51 ecdhe So could I configure the salt minion to use ~/salt instead of /srv/salt ?
05:52 forresta I don't know if it will accept a path in that format, I've never tried it
05:52 ecdhe I know "~" is a bashism.
05:53 ecdhe But I don't mind having a separate minion.conf for every user...
05:54 ecdhe The biggest functional gain here is that you could maintain a large site on AWS, but use shared hosting at a different datacenter to maintain a status page, or use it as an extremely localized CDN.
05:56 ecdhe Alternately, a consulting company could write salt states for an initial revision of a website--then scale it up to AWS when the demand grows too large.
05:56 forresta why wouldn't you just spin up a cheap VPS in that situation
05:57 ecdhe Perhaps that's a poor use case.
05:57 jfzhu_us ecdhe: I think forresta has a really good point on how complicated it will be to set this up. I am also not so sure how many people would want to use configuration management in such a small case.
05:58 forresta I'm not trying to be a downer here
05:58 forresta I'm just approaching it from someone who has dealt with shared hosting customers
05:58 forresta and imagining how this would scale in that situation, plus teaching people how to use it, etc.
05:58 ecdhe I appreciate the input.
05:59 xmltok joined #salt
06:00 ecdhe I'm a big fan of dreamhost, but they don't offer a few things...
06:00 ecdhe They have weak TLS ciphers.
06:00 jfzhu_us In my experience, security is often a pain to set up and maintain. selinux is a good example of security that can be frustrating to have
06:00 forresta does dreamhost let you install packages on their shared hosting?
06:01 ecdhe Dreamhost has one click installs for things they've already configured.
06:01 ecdhe They have lots of great features.
06:02 ecdhe Dreamhost offers unmetered bandwidth BUT won't let you seed a torrent.
06:02 forresta yea it's meant for hosting cheap sites
06:02 forresta stop abusing the shared hosting :P
06:02 ecdhe :-)
06:02 forresta you're sounding like the people who host 150 episodes of star trek on their shared hosting account.
06:02 forresta IT'S A WEBSITE! THERE IS A DIRECTORY LISTING!
06:03 ecdhe They don't really seem to mind if you copy 80GBs of the latest ubuntu repo.
06:03 forresta *shrug*
06:04 forresta 80GB is nothing in the grand scheem
06:04 forresta *scheme
06:04 forresta it's the useres who use TBs of bandwidth that are a problem
06:04 forresta *users
06:04 ecdhe But seeding a torrent doesn't mean lots of bandwidth...
06:05 ecdhe If 1000 people downloaded the whole file from your server before, now they're only downloading shards.
06:05 taion809 joined #salt
06:05 ecdhe Dreamhost is awesome, they're just not flexible enough.
06:05 forresta right, because you're not paying for flexibility
06:06 forresta it's the same with any large hosting provider
06:06 forresta I mean, it's like 5 bucks a month
06:06 ecdhe Well, actually 8 bucks.
06:06 forresta even at 8 bucks
06:06 ecdhe Digital Ocean is cheaper.
06:06 forresta yea, which is what I use
06:06 forresta but bandwidth is limited there
06:07 forresta granted I'm not running torrents off of something like that
06:07 xmltok joined #salt
06:07 ecdhe But with the right feature set, I think I could get some power users with low-volume sites.
06:08 forresta on a shared hosting provider? Sure I've seen people do it on accounts that support multiple domains
06:08 forresta I personally wouldn't, because the performance isn't good enough for me
06:10 ecdhe Well, I use dreamhost to manage several domains and it's been handy to kind of sandbox a future business site there until it needs more serious infrastructure behind it.
06:10 ecdhe Dreamhost is cheap and it works.
06:11 ecdhe But what I can't do is deploy a site there using CM so that I can transition it to other infrastructure easily.
06:11 forresta right, part of that is to avoid people botnetting
06:11 ecdhe ??
06:11 forresta and obviously, reduced permissions = reduced risk
06:12 forresta if you could use config management, you'd have enough permissions on the box to install software, and do nasty things
06:12 ecdhe But I have ssh access, meaning I could use ansible...
06:12 ecdhe Just cause I can use CM doesn't grant me root...
06:12 forresta right but you can't install packages
06:12 ecdhe Yeah.
06:14 ksalman joined #salt
06:15 ecdhe It would just be nice to commoditize that transition from shared hosting to more performant setups.
06:16 forresta yea I don't disagree
06:16 forresta but why not just put the user on a 5 dollar droplet?
06:16 forresta you're almost assured to get better performance, especially if you're managing the sie.
06:16 forresta *site
06:16 forresta unless you're trying to put a bunch of customers on a single 8 dollar plan
06:21 ecdhe Say you've got a brochure site serving a small municipal area... 10-100 hits/day.  The droplet is overkill and nobody is maintaining your security.  Shared hosting is awesome for that.
06:21 ecdhe Many business owners aspire to growing...
06:22 forresta oh I thought you were saying you were administrating the siets
06:22 forresta *sites
06:22 ecdhe Well not me personally.
06:24 ecdhe So you don't start out serving the tri-state area--but your site may need scaling.
06:25 taion809 joined #salt
06:26 ecdhe That was my premise.  Nobody else seems to offer CM on shared hosting, certainly for security purposes, but also because of low demand.
06:26 forresta yea not many wordpress users asking for config management, and the headache it would be to include it would be major
06:28 ecdhe But I think offering salt capabilities could be good for customers, and gives me the chance to evangelize salt.
06:29 ecdhe forresta, you're awesome, your irc participation has helped me a lot the last few weeks.
06:29 redmin joined #salt
06:29 forresta ecdhe, yea np. It's a fun community
06:30 ecdhe Do you read HN?
06:30 forresta no not really
06:31 ecdhe There was a link to a writeup on Jesse Willms, a Canadian kid who grossed $360M from those "one weird trick" ads online.
06:32 ecdhe His business tactics were slimy, but he did great advertising.
06:32 forresta heh
06:32 forresta yea I'm not a fan of slimy
06:33 ecdhe He would rebrand a chinese acai berry weight loss supplement as "SuperAcai", "AcaiLife", "AcaiBurn" etc.
06:33 forresta wow that is bad
06:34 forresta kedo39, are you around?
06:34 kedo39 always :3
06:34 forresta ok, are you planning on pushing the openssh formula back at some point
06:34 forresta you're stealing my easy commits here
06:35 forresta :P
06:35 kedo39 i have a pull request open
06:35 forresta did you merge in nterupt's changes?
06:36 forresta he also has a pull request open
06:36 kedo39 ah, i saw that. wasn't too sure what to do, since his hanges weren't pulled yet
06:37 forresta yea
06:37 forresta his has some odd if statements in there, I don't know what whiteinge will want to do in that scenario, we'll have to ask him monday
06:38 kedo39 i feel like his key management stuff should be split off into another state
06:38 kedo39 because it's rather specific
06:39 forresta I agree
06:39 forresta that's why I'm not really sure what to do about it
06:41 forresta ecdhe, like I said before, I'm not trying to be a downer or give you grief, I'm just thinking about it from a different perspective, where I'd quickly shut that down if I saw it on shared hosting for security/control purposes.
06:42 sroegner_ joined #salt
06:56 fllr joined #salt
06:59 cym3try joined #salt
07:08 xmltok joined #salt
07:26 oz_akan_ joined #salt
07:27 matanya joined #salt
07:40 redmin joined #salt
07:42 chuyskywalker joined #salt
07:45 chuyskywalker Hey folks, just getting my feet wet and I'm wondering if an empty salt run should take a smidge over a minute. Here's a log and some setup details for my ENV (https://gist.github.com/chuyskywalker/8265547)  -- should Salt be taking this long, or have I got something that needs fixing on my end? Thanks!
07:47 forresta how many states are running through there chuyskywalker
07:48 forresta nothing it looks like from this output, hmm
07:48 forresta can you add a simple state, just like a cmd.run
07:48 forresta to echo test or something
07:48 forresta and is this a local vagrant machine?
07:49 chuyskywalker yes. local vagrant. I took out all the states because I wasn't sure if I had a terrible state or something causing it
07:50 chuyskywalker hence the error at the very end, but sure I can try it with a silly state that does, like, and echo or something.
07:50 forresta yea
07:50 forresta usually the first state run takes a second to compile all the data, but no way should it be taking a minute.
07:50 forresta what version of salt are you on anyways?
07:51 bhosmer joined #salt
07:51 chuyskywalker salt-call 0.17.4
07:51 forresta ok
07:52 forresta what are you doing here so late bhosmer?
07:55 rmt joined #salt
08:00 zooz joined #salt
08:01 chuyskywalker forresta: here's the simple "does something" version (little bit less log output) https://gist.github.com/chuyskywalker/8265649
08:02 forresta was that the first run?
08:03 chuyskywalker yes, I can try it again
08:04 chuyskywalker The log file isn't time stamped, but the waiting happens up until it seems to start actually applying states
08:05 chuyskywalker The log file isn't time stamped, but the waiting happens up until it seems to start actually applying states
08:05 chuyskywalker (ack, sorry)
08:05 forresta np
08:05 forresta I am just interested to see what a second run looks like after the initial 'crunch' of data
08:05 chuyskywalker real    1m1.873s
08:06 jfzhu_us I am running into a situation where my requisite_in is not causing my groups to be executed before my user adds. This is the code I have but it doesn't seem to be working http://pastebin.com/KqnxPE2C and I would appreciate a second set of eyes on it.
08:06 chuyskywalker and a third is.... real    1m1.856s
08:07 forresta that's a pretty crappy run time
08:08 chuyskywalker yeah, and it's clearly just waiting on something silly, since the user/sys timer are < 1 sec
08:08 chuyskywalker I just haven't got the foggiest what it would be
08:08 forresta take a ook at this gist
08:08 forresta *look
08:08 forresta https://gist.github.com/UtahDave/3785738
08:08 Furao joined #salt
08:08 forresta chuyskywalker, I'd say the first step would be to eliminate vagrant as an issue, aka spinning something up on a cloud provider, and running the same config through
08:08 forresta to see if you can dup it in two spots
08:12 chuyskywalker Alright, I'll give that a shot
08:12 chuyskywalker might try a different box first though
08:12 chuyskywalker could be something quirky with this centos install
08:13 forresta ok, I'm gonna head out for the evening since it's getting pretty late here, but I should be around tomorrow evening so I'd be curious to know what your results are
08:14 chuyskywalker Thanks for the help
08:14 forresta yea np, I'll be interested to see what you find out!
08:18 fllr joined #salt
08:23 ekarlso joined #salt
08:27 oz_akan_ joined #salt
08:31 sroegner_ joined #salt
08:37 bhosmer joined #salt
08:53 redmin joined #salt
09:09 BenCoinanke joined #salt
09:14 oleg joined #salt
09:14 oleg Hi guys
09:14 oleg https://github.com/saltstack-formulas/resolver-formula/pulls
09:15 oleg Can anobody accept my pull request and provide feedback what I should fix?
09:15 oleg *s/and/or/g
09:18 fllr joined #salt
09:21 ckao joined #salt
09:26 techdragon joined #salt
09:27 oz_akan_ joined #salt
09:30 techdragon_ joined #salt
09:33 gasbakid joined #salt
09:34 techdrag_ joined #salt
09:35 anuvrat joined #salt
09:38 Furao joined #salt
10:00 techdragon joined #salt
10:04 redmin joined #salt
10:07 Furao joined #salt
10:13 techdragon_ joined #salt
10:17 techdragon joined #salt
10:18 fllr joined #salt
10:19 sroegner_ joined #salt
10:21 viraptor joined #salt
10:24 xmltok joined #salt
10:25 bhosmer joined #salt
10:26 anuvrat joined #salt
10:28 oz_akan_ joined #salt
10:34 redmin joined #salt
10:36 oleg Anybody?
10:48 psyl0n joined #salt
10:49 techdragon yes?
10:55 oleg https://github.com/saltstack-formulas/resolver-formula/pulls
10:56 oleg Can anobody accepts my pull request or provide feedback what I should fix here?
10:59 oleg ping? :)
11:00 techdragon oleg: nice, I don't have rights to accept it myself. But that just got pulled into my own private salt repo's right now. more documentation is always a + in my book
11:00 oleg Very nice!
11:00 oleg Can we discuss another another thing?
11:00 techdragon sure.
11:02 oleg One thing, which very frustrated me:
11:03 oleg Does not work as expected: http://pastebin.com/FVzz5ki6
11:03 oleg Work as expected: http://pastebin.com/5FTxULBh
11:03 oleg Reason of problem: http://pastebin.com/v0AXvXsx
11:03 oleg My proposal is following: salt should verify arguments of functions and return warning/error about unknown arguments for function calls.
11:03 oleg Otherwise we have problem like I found (and killed 2 days for understand the reason).
11:03 oleg Small warning about "unknown 'searchpath', 'nameservers' arguments" can safe a lot of time.
11:03 oleg I can write patch and create merge proposal, if this problem frustrate not only me.
11:05 techdragon its a well known YAML quirk
11:05 oleg Not to me
11:05 oleg Not to new guys
11:05 oleg And better if salt would verify for mistake and provide warnings or errors according to settings
11:06 oleg Otherwise you can miss some serious problem here, right?
11:06 techdragon http://docs.saltstack.com/topics/troubleshooting/yaml_idiosyncrasies.html#nested-dicts-key-value
11:06 oleg Yes, yes, it described, I found this later
11:06 oleg But my proposal little different: in ADDITION to available documentation provide warnings/errors about unknown arguments for function calls
11:07 oleg Warning by default, error if you want in config
11:07 oleg it would find all problems from starting point
11:07 oleg What do you think about that?
11:08 techdragon Sort of a linter or parser to check the state before running it and throw warnings on this sort of thing?
11:08 oleg Yes. In this example you should receive two warnings: about unknown argument "searchpath" for function call "file.managed" and about unknown argument "nameservers" for same function
11:09 oleg Right now it silently skip incorrect arguments
11:09 oleg On "all" log level as well
11:09 techdragon Theres already an issue registered for this one. :) https://github.com/saltstack/salt/issues/5378
11:09 techdragon Your not alone in wanting this.
11:09 oleg I was forced to read source code and inject extra logging for dig this problem
11:09 techdragon I sure want it to.
11:09 oleg Very nice! Thanks for reference!
11:09 techdragon Wow. thats way more work than you should have needed.
11:09 techdragon I can see why you were frustrated.
11:11 oleg Can you please  elaborate what did you mean in "thats way more work than you should have needed"?
11:12 oleg Did you mean - issue 5378 required more work than enough for prevent problems like my? :)
11:13 techdragon Well, if you had to inject extra logging in the source code to work out what was going wrong, before you found that YAML idiosyncrasy documentation. Then that is more work than you should have needed to do.
11:13 oleg Oh no, incorrect
11:13 oleg Let me explain...
11:14 oleg When I try to call some fuction in python with incorrect positional arguments count or with extra named arguments, python raise exception, right?
11:14 techdragon Of course.
11:14 oleg My proposal - modify function call in salt in way, where we try to directly call function with all provided arguments and catch exceptions about incorrect arguments - and log about found problems.
11:15 oleg Actual behavior silently ignore invalid extra arguments
11:16 oleg Let me show....
11:16 linjan joined #salt
11:16 oleg File /usr/lib/python2.7/dist-packages/salt/states/file.py
11:16 oleg def managed(name,
11:16 oleg source=None,
11:16 oleg source_hash='',
11:16 oleg user=None,
11:16 oleg group=None,
11:16 oleg mode=None,
11:16 oleg template=None,
11:16 oleg makedirs=False,
11:16 oleg dir_mode=None,
11:16 oleg context=None,
11:16 oleg replace=True,
11:16 oleg defaults=None,
11:16 oleg env=None,
11:16 oleg backup='',
11:16 oleg show_diff=True,
11:16 oleg create=True,
11:16 oleg contents=None,
11:16 oleg contents_pillar=None,
11:16 oleg **kwargs):
11:16 oleg '''
11:16 oleg If you attempt to call this functon with extra arguments like "nameservers" or "searchpath" - you would receive exception
11:16 oleg Actual salt implementation silently hide this problem.
11:17 techdragon Ahh, i can see some use for that. I wouldn't know if theres a hidden reason why thats not yet been done however. Im fairly new to making modifications in the core of the code, haven't needed to change much of anything for my use so far.
11:17 oleg Exactly
11:17 oleg I propose to solve problem on core level
11:17 oleg By modification of actual way of function call
11:18 oleg As result in any place where you call - directly or indirectly - some python function, you would receive warnings or errors about invalid arguments
11:18 oleg I suppose, it is easy to support
11:18 oleg Warnings by default
11:18 fllr joined #salt
11:18 oleg And error if you set some option in config - like "strict function call"
11:18 oleg Do you understand what I proposed? :)
11:19 techdragon A few other tickets are mentioned in 5378 i linked you to earlier. You may want to read them and comment about your idea in the right one. Since its related to the goal of the parser/linter in that it ensures better state related diagnostic output. If the input is bad, salt should produce more informative information why in most cases since the current information can be extremely terse at times.
11:19 oleg Exactly!
11:19 oleg I read ticket which you linked, it slighty different
11:20 oleg So, ok, I am goind to create new ticket with explanation of problem / proposed solution
11:20 techdragon Let me see if i can find the ticket I'm thinking of.
11:20 oleg Of course, take your time
11:21 techdragon https://github.com/saltstack/salt/issues/802 this much older ticket may be a better place for either linking your new ticket if you feel like making one still, or to add to the conversation.
11:23 oleg Ticket 802 more big than my proposal....
11:23 oleg I understand why this old ticket is not closed yet
11:23 oleg Hard to support
11:23 oleg My proposal really simpler and would work in most of cases
11:23 oleg Do you agree? :)
11:24 techdragon since your talking about better handling the specific sub part you explained earlier, perhaps make a new ticket with the information and link it into 802 since it may help with the overall work to get 802 done, its a rather big ticket and may be best solved in smaller steps.
11:24 oleg Yeap, exactly
11:24 oleg Ok, I am filling ticket
11:25 oleg I also will try to provide patch for this smaller proposal
11:27 techdragon Awesome. Thanks for taking the time to do all that work. Hope the core guys see it as useful.
11:29 oz_akan_ joined #salt
11:34 oleg https://github.com/saltstack/salt/issues/9572
11:34 oleg Damn. How I can quote some code block in ticket description?
11:36 techdragon github use ```` to delineate bigger code blocks. try using that on a line by itself at the start and end of your code block
11:37 derelm joined #salt
11:39 oleg Thank you for proposal, I am trying
11:40 oleg It works, thank you very much!
11:40 oleg https://github.com/saltstack/salt/issues/9572
11:41 oleg Now looks nive from my point of view :
11:41 oleg :)
11:41 viraptor joined #salt
11:41 oleg Is anybody from salt core team here?
11:41 BenCoinanke_a joined #salt
11:44 redmin joined #salt
11:47 oleg anybody? core team? :) Do not ignore me please :D
11:53 BenCoinanke joined #salt
11:57 namtab joined #salt
12:02 bhosmer joined #salt
12:04 dccc joined #salt
12:08 sroegner_ joined #salt
12:11 anuvrat joined #salt
12:13 bhosmer_ joined #salt
12:18 fllr joined #salt
12:30 oz_akan_ joined #salt
12:55 redmin joined #salt
13:02 holys joined #salt
13:05 juso joined #salt
13:18 fllr joined #salt
13:30 oz_akan_ joined #salt
13:31 william_20111 joined #salt
13:31 oleg Is anobody from salt core team here?
13:31 Teknix joined #salt
13:31 neilf_ joined #salt
13:31 crane joined #salt
13:31 bwq joined #salt
13:32 aarontc joined #salt
13:32 vortec joined #salt
13:32 scalability-junk joined #salt
13:33 munhitsu joined #salt
13:43 zooz joined #salt
13:49 favadi joined #salt
13:52 jeddi oleg:  they normally keep US business hours .. should be coming online soon I guess.
13:57 sroegner_ joined #salt
13:59 anuvrat joined #salt
14:00 oleg Thanks, jeddi
14:01 bhosmer_ joined #salt
14:04 alekibango joined #salt
14:06 redmin joined #salt
14:10 oleg By the way, what the right way to manage repositories list under SaltStack?
14:11 oleg I mean, /etc/apt/sources.list.d/something or /etc/yum.conf.d/something.repo
14:11 tempspace joined #salt
14:13 jperras joined #salt
14:13 djinni` joined #salt
14:15 oeuftete joined #salt
14:18 fllr joined #salt
14:27 toastedpenguin joined #salt
14:31 oz_akan_ joined #salt
14:37 MrTango joined #salt
14:42 patrek oleg, take a look at http://docs.saltstack.com/ref/states/all/salt.states.pkgrepo.html
14:55 psyl0n joined #salt
15:12 bhosmer joined #salt
15:18 redmin joined #salt
15:18 fllr joined #salt
15:32 oz_akan_ joined #salt
15:34 joehh joined #salt
15:34 rojem joined #salt
15:36 elfixit joined #salt
15:46 sroegner_ joined #salt
15:50 bhosmer joined #salt
15:54 viraptor joined #salt
15:59 nahamu https://github.com/nshalman/salt/commit/ed8583eca6dcc04412710ba77bc8e3166ca296d9
16:00 nahamu I need to test that a bit more to make sure I didn't typo anything when I did some cleanup.
16:00 oleg patrek: thanks, can you provide some example to me, how to add some repository (for instance, saltstack or postgresql) thought this API?
16:00 oleg I mean, repo URL, import gpg keys, so on
16:00 nahamu If I wanted to get that commit into both develop and the 2014.1 branch, what's the best procedure?
16:03 JasonSwindle joined #salt
16:04 neko joined #salt
16:08 xinkeT joined #salt
16:18 fllr joined #salt
16:19 holys joined #salt
16:28 vu joined #salt
16:33 oz_akan_ joined #salt
16:35 fllr joined #salt
16:36 redmin joined #salt
16:40 nahamu xinkeT: I respun the 2014.1-rc1 esky build with some minor tweaks and also generated and uploaded a tarball version
16:42 ahammond joined #salt
16:42 psyl0n joined #salt
16:59 oz_akan_ joined #salt
17:01 oz_aka___ joined #salt
17:02 anuvrat joined #salt
17:03 oz_aka___ joined #salt
17:10 bhosmer joined #salt
17:22 sroegner_ joined #salt
17:22 Psi-Jack What the heck does this error actually mean: State files.managed found in sls supervisor is unavailable
17:25 timoguin joined #salt
17:26 sroegner_ Psi-Jack: i believe it tries to tell you that the name of the state would have to be file.managed
17:26 Psi-Jack Ahhhh
17:27 Psi-Jack Yep. That was it. LOL
17:27 Psi-Jack Thank you, sroegner_. :)
17:29 root joined #salt
17:38 bhosmer_ joined #salt
17:38 Guest40154 Hi. Just have another stupid question here. Don't worked with saltstack for a while, and havint troubles with a manifest, http://pastebin.com/WAFtmqL9, if someone could tell how to fix it, it would be cool, thx. Pls. sorry for my english.
17:40 Psi-Jack manifest? Salt != puppet :)
17:40 Guest40154 sorry, state)
17:42 sroegner_ Guest40154:
17:42 sroegner_ Guest40154: looking at it
17:43 sroegner_ you don't need the base keyword in init.sls
17:43 Guest40154 removing it produces the same result
17:45 Guest40154 oh, not the same. Name pkgrepo.managed in sls states.mongodb is not a dictionary and Name pkg.latest in sls states.mongodb is not a dictionary
17:46 sroegner_ Guest40154:
17:46 sroegner_ Guest40154: seeing it
17:47 cym3try joined #salt
17:47 matanya joined #salt
17:47 sroegner_ Guest40154: sec
17:47 Guest40154 sroegner_: ok)
17:53 sroegner_ https://gist.github.com/sroegner/8271518
17:54 sroegner_ Guest40154: this is how I would build that - you definitely had problems in your data structure
17:55 sroegner_ it doesn't work still because of some URL problem there - i don't have much experience with Ubuntu
17:56 sroegner_ and order within the state file usually doesn't need to be declared
17:57 fllr joined #salt
17:59 vu joined #salt
18:01 vu left #salt
18:08 vu joined #salt
18:11 Guest40154 sroegner_: thanks, it works with dist instead of precise)
18:16 sroegner_ Guest40154: took forever to download here
18:19 bhosmer joined #salt
18:46 matanya joined #salt
18:52 bhosmer joined #salt
19:02 martoss joined #salt
19:15 micko joined #salt
19:17 seth joined #salt
19:17 seth is it possible to reference builtin grains inside of a custom grain?
19:20 beach joined #salt
19:24 indymike joined #salt
19:26 bhosmer joined #salt
19:29 bdf_ joined #salt
19:30 andersb_ joined #salt
19:30 kedo39_ joined #salt
19:30 xinkeT joined #salt
19:30 druonysuse joined #salt
19:30 druonysuse joined #salt
19:30 grep_away joined #salt
19:30 nebuchad` joined #salt
19:31 mattikus` joined #salt
19:31 martoss joined #salt
19:31 rhand joined #salt
19:31 matanya_ joined #salt
19:34 indymike joined #salt
19:35 brucewang joined #salt
19:35 Ixan joined #salt
19:37 shennyg joined #salt
19:37 seb` joined #salt
19:40 teepark joined #salt
19:40 Ixan joined #salt
19:40 Tropolis joined #salt
19:40 faulkner joined #salt
19:40 pexio joined #salt
19:42 teepark joined #salt
19:46 indymike joined #salt
19:46 fishpen01 joined #salt
19:47 __number5__ joined #salt
19:47 Kraln- joined #salt
19:51 fishpen02 joined #salt
19:51 sine_nitore joined #salt
19:52 stantont joined #salt
19:53 djinni`_ joined #salt
19:53 ubuntu1 joined #salt
19:56 pexio_ joined #salt
19:57 cym3try joined #salt
19:57 matanya_ joined #salt
19:57 cym3try joined #salt
19:59 micko joined #salt
20:00 defunctzombie_ joined #salt
20:02 EvaSDK joined #salt
20:02 EvaSDK joined #salt
20:02 bdf joined #salt
20:02 tallpaul_ joined #salt
20:02 mollerst1and joined #salt
20:03 dnsn_ joined #salt
20:03 xzarth joined #salt
20:03 cyrusdav- joined #salt
20:04 cbloss_ joined #salt
20:04 pviktori_ joined #salt
20:04 tamentis joined #salt
20:04 Ahlee_ joined #salt
20:04 darrend_ joined #salt
20:04 mattikus joined #salt
20:05 fishpen0 joined #salt
20:05 pmcg_ joined #salt
20:06 vu_ joined #salt
20:06 codysoyland_ joined #salt
20:06 brucewang joined #salt
20:06 indymike joined #salt
20:06 TomasNunez1 joined #salt
20:06 main__ joined #salt
20:08 wunki joined #salt
20:08 pexio joined #salt
20:08 bersace joined #salt
20:08 zz_cro joined #salt
20:08 pexio joined #salt
20:09 crazysim joined #salt
20:10 Voziv_ joined #salt
20:10 nahamu joined #salt
20:10 nahamu__ joined #salt
20:11 jeffrubi` joined #salt
20:11 darrend joined #salt
20:12 indymike joined #salt
20:13 ajw0100 joined #salt
20:13 jeffrubic joined #salt
20:13 bdf_ joined #salt
20:19 indymike joined #salt
20:19 cyrusdavid joined #salt
20:22 Tekni joined #salt
20:23 basepi_ joined #salt
20:23 goncalo_ joined #salt
20:24 joelperras joined #salt
20:25 kamyl_ joined #salt
20:25 andersb joined #salt
20:25 _ioni joined #salt
20:25 pviktori joined #salt
20:26 Hefeweizen joined #salt
20:26 psyl0n joined #salt
20:28 crazysim joined #salt
20:29 nahamu joined #salt
20:29 aurigus_ joined #salt
20:30 yml_ joined #salt
20:30 analogbyte joined #salt
20:31 chadhs joined #salt
20:31 andersb joined #salt
20:31 bdf joined #salt
20:31 APLU joined #salt
20:32 jms2 joined #salt
20:34 ubuntu1 joined #salt
20:34 Daviey joined #salt
20:35 keekz_ joined #salt
20:35 z3uS| joined #salt
20:35 akoumjian_ joined #salt
20:35 litheum joined #salt
20:35 d10n joined #salt
20:35 ajw0100 joined #salt
20:36 kevinbrolly_ joined #salt
20:37 indymike joined #salt
20:37 nineteen1ightd joined #salt
20:37 matanya_ joined #salt
20:37 analogby1e joined #salt
20:37 analogby1e left #salt
20:37 micko__ joined #salt
20:37 juice- joined #salt
20:37 fuser joined #salt
20:37 ahammond_ joined #salt
20:37 Nazca joined #salt
20:38 flebel joined #salt
20:38 Guest42344 joined #salt
20:38 techdragon_away joined #salt
20:38 indymike joined #salt
20:38 mattikus joined #salt
20:38 jmccree_ joined #salt
20:38 Tropolis joined #salt
20:38 Guest64230 joined #salt
20:39 timoguin joined #salt
20:39 sroegner_ joined #salt
20:39 toastedpenguin joined #salt
20:39 kermit joined #salt
20:39 logix812 joined #salt
20:39 mikkn joined #salt
20:39 utahcon joined #salt
20:39 Damoun_ joined #salt
20:39 ClausA joined #salt
20:39 ktenney_gone joined #salt
20:39 harobed__ joined #salt
20:39 worstadmin joined #salt
20:39 patrek joined #salt
20:39 philipsd6 joined #salt
20:39 ninkotech joined #salt
20:39 benturner joined #salt
20:39 londo__ joined #salt
20:39 coolj joined #salt
20:39 tsantero joined #salt
20:39 babilen joined #salt
20:39 vbabiy joined #salt
20:39 lv_ joined #salt
20:39 mau_ joined #salt
20:39 ahale joined #salt
20:39 kyusan_ joined #salt
20:39 Linuturk joined #salt
20:39 rope_ joined #salt
20:39 AlcariTheMad joined #salt
20:39 dcolish joined #salt
20:39 andrej joined #salt
20:39 emilisto joined #salt
20:39 eightyeight joined #salt
20:39 cwright joined #salt
20:39 yota joined #salt
20:39 jdenning joined #salt
20:39 lahwran_ joined #salt
20:40 vu_ joined #salt
20:42 nn0101 joined #salt
20:42 lahwran joined #salt
20:42 lahwran joined #salt
20:42 cewood joined #salt
20:43 lahwran joined #salt
20:43 beach Hey guys, any reason for getting 'Failed to authenticate, is this user permitted to execute commands?' for no apparent reason... I've even reinstalled the salt-master but it's screwed.
20:44 beach sudo salt '<somename*' state.highstate
20:45 indymike joined #salt
20:46 lahwran joined #salt
20:46 lahwran joined #salt
20:47 lahwran joined #salt
20:47 giantlock joined #salt
20:49 lahwran joined #salt
20:50 lahwran joined #salt
20:50 lahwran joined #salt
20:50 pcarrier joined #salt
20:52 markm joined #salt
20:52 lahwran joined #salt
20:52 MTecknology joined #salt
20:53 lahwran joined #salt
20:53 lahwran joined #salt
20:55 aarontc joined #salt
20:56 crazysim joined #salt
20:56 lahwran joined #salt
20:56 lahwran joined #salt
20:57 lahwran joined #salt
20:59 lahwran joined #salt
21:00 lahwran joined #salt
21:00 lahwran joined #salt
21:02 lahwran joined #salt
21:03 lahwran joined #salt
21:03 lahwran joined #salt
21:05 lahwran joined #salt
21:06 lahwran joined #salt
21:06 lahwran joined #salt
21:07 rlarkin joined #salt
21:07 beach I created a new master with the same config/state and it's working fine. What happened to my original master????
21:08 lahwran joined #salt
21:09 lahwran joined #salt
21:10 lahwran joined #salt
21:11 indymike joined #salt
21:11 jdenning joined #salt
21:12 lahwran joined #salt
21:12 lahwran joined #salt
21:13 lahwran joined #salt
21:14 bhosmer joined #salt
21:15 lahwran joined #salt
21:15 lahwran joined #salt
21:16 lahwran joined #salt
21:17 jfzhu_us joined #salt
21:18 lahwran joined #salt
21:18 lahwran joined #salt
21:19 lahwran joined #salt
21:21 lahwran joined #salt
21:22 lahwran joined #salt
21:22 lahwran joined #salt
21:25 lahwran joined #salt
21:25 lahwran joined #salt
21:26 lahwran joined #salt
21:28 lahwran joined #salt
21:28 lahwran joined #salt
21:29 lahwran joined #salt
21:29 jfzhu_us Does a require_in guarantee that the state which has the require_in to run before the state that is listed under the require_in?
21:31 lahwran joined #salt
21:31 lahwran joined #salt
21:32 lahwran joined #salt
21:34 lahwran joined #salt
21:35 lahwran joined #salt
21:35 lahwran joined #salt
21:38 lahwran joined #salt
21:38 lahwran joined #salt
21:39 lahwran joined #salt
21:39 nebuchadnezzar joined #salt
21:41 lahwran joined #salt
21:42 beach freaking reboot solved the problem... WTF?!
21:42 lahwran joined #salt
21:42 lahwran joined #salt
21:42 grep_awesome joined #salt
21:43 grep_awesome joined #salt
21:45 lahwran joined #salt
21:45 lahwran joined #salt
21:46 lahwran joined #salt
21:46 jcockhren beach is over there spazzin. lol
21:48 lahwran joined #salt
21:48 beach totally... been running into brick walls with salt. Weird things keep cropping up with my salties. GitFS is totally punching me in the nutz
21:48 lahwran joined #salt
21:49 jcockhren give me a min. I'll help you out with that
21:49 lahwran joined #salt
21:51 lahwran joined #salt
21:51 lahwran joined #salt
21:52 lahwran joined #salt
21:54 lahwran joined #salt
21:55 lahwran joined #salt
21:55 viq joined #salt
21:55 viq joined #salt
21:57 lahwran joined #salt
21:57 lahwran joined #salt
21:57 scalability-junk joined #salt
21:57 pyykkis joined #salt
21:58 lahwran joined #salt
22:00 lahwran joined #salt
22:00 lahwran joined #salt
22:01 lahwran joined #salt
22:03 lahwran joined #salt
22:03 lahwran joined #salt
22:04 lahwran joined #salt
22:06 lahwran joined #salt
22:07 lahwran joined #salt
22:08 lahwran joined #salt
22:10 lahwran joined #salt
22:10 lahwran joined #salt
22:11 lahwran joined #salt
22:13 lahwran joined #salt
22:14 lahwran joined #salt
22:14 TomasNunez joined #salt
22:15 lahwran joined #salt
22:17 lahwran joined #salt
22:17 lahwran joined #salt
22:17 robbyt joined #salt
22:18 lahwran joined #salt
22:18 darless joined #salt
22:18 ioni joined #salt
22:20 lahwran joined #salt
22:20 lahwran joined #salt
22:21 lahwran joined #salt
22:22 elithrar joined #salt
22:23 lahwran joined #salt
22:23 Daviey joined #salt
22:24 lahwran joined #salt
22:24 lahwran joined #salt
22:25 forresta joined #salt
22:27 lahwran joined #salt
22:27 lahwran joined #salt
22:28 lahwran joined #salt
22:30 darless Hello all. I created a saltstack formula and put it in my account. https://github.com/MadRussian/apt-cacher-formula. What are the next steps for linking it to saltstack-formulas?
22:31 lahwran joined #salt
22:32 lahwran joined #salt
22:32 lahwran joined #salt
22:33 ajw0100 joined #salt
22:34 jcockhren beach: still there?
22:34 beach yep
22:34 forresta darless, I've created the repo here: https://github.com/saltstack-formulas/apt-cacher-formula
22:35 beach been messing around with https method and it connects, but I get GitPython exception caught while fetching: len(["23132e220ad0116e6a309451955085a70dd9c005\tnot-for-merge\tbranch 'master' of https://bitbucket.org/fesaltmaster/fesalt\n"]) != len(['POST git-upload-pack (298 bytes)', '   8630619..23132e2  master     -> origin/master'])
22:35 lahwran joined #salt
22:35 beach I've also been playing with ssh but I get  GitPython exception caught while fetching: len([]) != len(['Permission denied (publickey)
22:35 darless forresta: so will I be able to just push to it?
22:35 lahwran joined #salt
22:35 forresta no just make a pull request on it
22:36 jcockhren beach: you're just trying to do gitfs backed states?
22:36 darless forresta: gotcha
22:36 lahwran joined #salt
22:36 beach well... I am now. I started out at first to use gitfs to source a managed file.
22:36 forresta I've added a README now so you should be able to just refresh that page, and do the pull request darless
22:36 forresta forgot when I added it.
22:37 jcockhren beach: ti works best for me when I using the git@url for gitfs remotes
22:37 beach my latest has a top.sls in my git repository
22:38 beach thats what I have no
22:38 beach nwo
22:38 beach bah now
22:38 lahwran joined #salt
22:38 jcockhren does it work now for you?
22:38 beach gitfs_remotes:   - git@bitbucket.org:fesaltmaster/fesalt.git
22:39 jcockhren ok cool
22:39 lahwran joined #salt
22:39 oz_akan_ joined #salt
22:39 beach but I'm getting a permission denied on the public key. - I've added it to the ssh-agent but no bueno
22:39 lahwran joined #salt
22:40 jcockhren ok. the key has to be the one that the salt process is running under. not the use that runs the salt commands
22:40 jcockhren by default that's root.
22:40 beach right I have the keys in /root/.ssh
22:41 jcockhren and you have the pub key uploaded to bitbucket?
22:41 jcockhren under a deployment key?
22:41 beach now that I'm thinking about... shoot I just generated that new .pub - I need to add that to the bucket
22:41 jcockhren there you go
22:42 lahwran joined #salt
22:42 lahwran joined #salt
22:43 lahwran joined #salt
22:43 beach running highstate now...
22:43 beach looks the same: GitPython exception caught while fetching: len([]) != len(['Permission denied (publickey)
22:44 jcockhren you restart the salt-master process?
22:44 beach no
22:44 forresta darless, sorry I wasn't thinking, you might need to fork off the repo, then move your content into, my bad.
22:44 beach trying that
22:45 lahwran joined #salt
22:45 beach same deal
22:45 beach so frustrating
22:45 lahwran joined #salt
22:45 jcockhren beach: can you clone that repo as the process owner?
22:46 jcockhren switch to root and do a normal git clone
22:46 lahwran joined #salt
22:47 beach can't su to root -- running in AWS
22:48 beach I just did ssh -T hg@bitbucket.org and it worked fine
22:48 lahwran joined #salt
22:48 jcockhren beach: are you using mercurial or git?
22:48 lahwran joined #salt
22:49 lahwran joined #salt
22:50 jcockhren beach: sudo -u root -H ....
22:50 beach git
22:51 jcockhren beach: later on. you should run as a non-priviledge user. just Fyi
22:51 lahwran joined #salt
22:52 clearclaw Can someone help me quickly with overstates and creating cloud nodes?
22:52 lahwran joined #salt
22:52 clearclaw I don't seem to be able to get overstates to do cloud.profile...
22:52 lahwran joined #salt
22:53 jcockhren beach: also are the private key have the correct permissions? chmod and such.
22:53 jcockhren ssh protocals fail when the private key doesn't have to a chmod that isn't restrictive enough
22:54 clearclaw 0600 recommended.
22:54 beach -rw-r--r-- 1 root root  391 Jan  5 22:29 id_rsa.pub
22:54 beach okay
22:54 lahwran joined #salt
22:55 lahwran joined #salt
22:55 lahwran joined #salt
22:56 beach the private has always been: -rw------- 1 root root 1766 Jan  5 22:29 id_rsa
22:57 jcockhren can you close as a root user?
22:57 jcockhren s/close/clone
22:58 lahwran joined #salt
22:58 lahwran joined #salt
22:59 beach I can clone fine - it does ask me for my passphrase
22:59 lahwran joined #salt
22:59 beach for the key that is
23:00 beach not sure I know what you mean by s/close/clone
23:02 lahwran joined #salt
23:02 jcockhren beach: some other guys would stab me for this but... how about a pub/priv key combination that doesn't require a password?
23:02 lahwran joined #salt
23:02 jcockhren that'll remove your need for ssh-agent
23:02 lahwran joined #salt
23:02 jcockhren b/c you won't be there the re-renter passwords for the root account on reboots or anything else
23:03 bhosmer joined #salt
23:03 beach well. isn't there a env variable that you can store it... Anyway... I could try that. dunno what I'll do if that works though
23:04 viraptor joined #salt
23:07 jcockhren beach: no. that's for ssh forwarding. that what's stored in env is the socket
23:07 jcockhren also, ssh fowarding is enabled on the ssh server side. I'm sure if butbucket does that.
23:08 beach okay tried the empty passphrase and it seems to be the same thing. Tested a clone too and it worked without asking
23:09 beach huh.... I did restart the master and try again for grins... no *new error message. But the state didn't seem to do anything...
23:09 beach let me check the cache
23:09 beach empty
23:09 jcockhren beach: remove the cache, restart the master. that'll force a reclone
23:10 jcockhren by "cache" I mean the gitfs subdirectory
23:10 beach man... last time I didn't that it dorked up the master... I'll give it a shot
23:10 beach yeah, got it
23:13 beach new cache folder, but it's empty
23:13 beach created a new log, and it's clean with the exception of this: Caught signal 15, stopping the Salt Master
23:14 beach but I've been getting those forever
23:26 beach so... something to report. I changed back the order of the fileserver_backend to git, roots and now I'm getting a states type error
23:26 beach cache is still empty, but seems more promising
23:28 beach holy cat piss...
23:30 elfixit joined #salt
23:31 beach it's working.... BUT I have no security :(
23:42 beach can someone explain to me what the security risks are of no having a passphrase associated to your id_rsa key for ssh????
23:42 beach if you're only using it for git, and nothing else
23:43 ajw0100 joined #salt
23:46 kedo39_ if someone has read access to your file
23:46 kedo39_ then they can do whatever they want with it
23:46 kedo39_ ie: they can push whatever they want to your git repo
23:51 jslatts joined #salt

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