Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2014-07-05

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

All times shown according to UTC.

Time Nick Message
00:14 zain_ joined #salt
00:18 ramishra joined #salt
00:35 analogbyte joined #salt
00:47 bhosmer joined #salt
00:56 bhosmer joined #salt
00:57 diegows joined #salt
01:26 zain_ joined #salt
01:42 joehh ~.
01:45 joehh !.
01:45 krow joined #salt
01:47 ipalreadytaken joined #salt
02:19 ramishra joined #salt
02:23 ipalreadytaken joined #salt
02:36 bhosmer joined #salt
02:42 fllr joined #salt
02:48 oz_akan_ joined #salt
02:49 ekristen joined #salt
02:53 akitada joined #salt
03:02 jslatts joined #salt
03:08 logix812 joined #salt
03:15 ramishra joined #salt
03:24 ipalreadytaken joined #salt
03:27 tkharju joined #salt
03:33 catpiggest joined #salt
03:47 tkharju joined #salt
03:52 fejjerai joined #salt
04:06 philipsd6 joined #salt
04:12 Tween joined #salt
04:14 ramishra joined #salt
04:21 ipalreadytaken joined #salt
04:23 vlcn left #salt
04:24 bhosmer joined #salt
04:29 kivihtin joined #salt
04:29 thayne_ joined #salt
04:36 agronholm joined #salt
04:38 rjc joined #salt
04:38 rjc joined #salt
04:47 gildegoma joined #salt
04:47 jpaetzel joined #salt
05:12 krow joined #salt
05:18 oz_akan_ joined #salt
05:41 Outlander joined #salt
05:47 ml_1 joined #salt
05:58 Outlander joined #salt
06:00 oz_akan_ joined #salt
06:12 bhosmer joined #salt
06:15 ramishra joined #salt
06:19 oz_akan_ joined #salt
06:21 schimmy joined #salt
06:21 bhosmer joined #salt
06:22 krow joined #salt
06:33 Ryan_Lane joined #salt
06:37 logix812 joined #salt
07:03 ramishra joined #salt
07:13 ramishra_ joined #salt
07:20 oz_akan_ joined #salt
07:21 m____s joined #salt
07:24 auser joined #salt
07:30 E1NS joined #salt
07:35 chiui joined #salt
07:36 Kelsar joined #salt
07:46 felskrone joined #salt
07:48 DaveQB joined #salt
08:00 bhosmer joined #salt
08:20 oz_akan_ joined #salt
08:22 ramishra joined #salt
08:24 pjs joined #salt
08:26 dcmorton joined #salt
08:27 ahammond joined #salt
08:28 redondos joined #salt
08:31 ramishra_ joined #salt
08:40 toddejohnson joined #salt
08:45 acabrera joined #salt
08:49 oz_akan_ joined #salt
08:51 oz_akan__ joined #salt
09:20 jalaziz joined #salt
09:22 bauser joined #salt
09:25 ramishra joined #salt
09:46 ramishra joined #salt
09:48 bhosmer joined #salt
09:52 oz_akan_ joined #salt
10:16 ramishra joined #salt
10:52 oz_akan_ joined #salt
10:58 ramteid joined #salt
11:09 bhosmer joined #salt
11:11 jhauser joined #salt
11:17 ramishra joined #salt
11:17 kermit joined #salt
11:24 kermit joined #salt
11:36 bhosmer_ joined #salt
11:41 TheThing joined #salt
11:52 Hell_Fire joined #salt
11:53 oz_akan_ joined #salt
11:56 mapu joined #salt
12:17 krow joined #salt
12:18 ramishra joined #salt
12:19 bhosmer joined #salt
12:26 bhosmer joined #salt
12:42 diegows joined #salt
12:54 oz_akan_ joined #salt
13:16 seeg joined #salt
13:17 seeg hello
13:18 seeg suppose I have a file_root /srv/salt/a and /srv/salt/b, how can I assure that first all states from 'a' are loaded before 'b' ?
13:18 ramishra joined #salt
13:24 bhosmer joined #salt
13:27 elfixit1 joined #salt
13:33 thayne_ joined #salt
13:50 mgw joined #salt
13:55 oz_akan_ joined #salt
14:10 luminous seeg: I'm making a guess, but you could test.. in file_roots, define one before the other
14:12 seeg I did but it doesn't work :)
14:19 ramishra joined #salt
14:25 luminous seeg: use different environments?
14:25 luminous rsync them together in order that works for you?
14:25 * luminous shrugs
14:25 Kenzor joined #salt
14:26 luminous usually salt respects order of stuff in this sort of situation
14:26 oncallsucks joined #salt
14:31 analogbyte how come merge request #9539 still has not made it into the debian packages? joehh? I am also waiting for #10761 to make the cut, but that one was at least merged somewhat more recent, and not last year...
14:32 mateoconfeugo joined #salt
14:37 happytux joined #salt
14:48 yomilk joined #salt
14:51 oncallsucks joined #salt
14:55 oz_akan_ joined #salt
15:12 bhosmer joined #salt
15:16 babilen analogbyte: That commit is only in develop. I wouldn't expect the Debian package to  deviate too much from official versions.
15:17 babilen (same for #10761)
15:19 babilen I mean it is okay to patch code if later releases *of the same branch* will incorporate that commit, but cherry-picking commits from different branches is IMHO neither justified nor a good idea
15:20 ramishra joined #salt
15:21 jnials joined #salt
15:22 jnials joined #salt
15:23 acabrera joined #salt
15:29 malinoff joined #salt
15:36 quenode joined #salt
15:56 Luke_ joined #salt
15:59 m1crofarmer joined #salt
16:03 jnials joined #salt
16:08 Luke_ joined #salt
16:10 FarrisG joined #salt
16:21 ramishra joined #salt
16:29 jnials joined #salt
16:32 Luke_ joined #salt
16:35 Luke_ joined #salt
16:48 luminous babilen: the salt git manifesto is pretty clear that only in the most dire of cases will a bug be backported, but they would cherry-pick in those cases
16:52 elfixit1 joined #salt
16:52 jnials joined #salt
16:57 oz_akan_ joined #salt
16:58 Gareth 3
16:58 Gareth erm
17:00 babilen luminous: Yes, but the *Debian* packages shouldn't deviate from upstream too much. I mean it is IMHO okay to integrate bug fixes that already made it into 2014.1 into an earlier release, but not to cherry pick stuff from develop into 2014.1 iff that wasn't done by upstream already.
17:00 babilen luminous: Where can I find this git manifesto though?
17:01 babilen luminous: I think that we are essentially on the same page there :)
17:01 bhosmer joined #salt
17:04 jnials joined #salt
17:12 happytux_ joined #salt
17:14 ajw0100 joined #salt
17:21 ramishra joined #salt
17:34 mgw joined #salt
17:40 dalexander joined #salt
17:43 schimmy joined #salt
17:43 krow joined #salt
17:56 jnials joined #salt
17:58 oz_akan_ joined #salt
18:01 luminous babilen: yea, I think the point is that that type of thing is on a case-by-case basis. debian did what you said they 'should' not for heartbleed..
18:02 luminous babilen: RE the manifesto, it's in the docs somewhere. I forget where.. it's somewhere in the developer/contiributor docs
18:05 bfwg joined #salt
18:06 luminous any ideas on how to use salt as a replacement for celery? I'm thinking it's possible, but I haven't mapped out all the todos, has anyone done this?
18:06 luminous not like anyone is around on a holiday weekend like this..
18:08 agronholm luminous: what do you do with celery that could be replaced with salt?
18:11 krow joined #salt
18:11 ndrei joined #salt
18:14 techdragon joined #salt
18:14 Corey luminous: I tried replacing celery with salt. It resulted in a fscking horrible salad.
18:16 agronholm salt is a configuration management tool, celery is a task queue
18:16 agronholm how can you replace one with the other?
18:19 luminous agronholm: salt is not a cfm.. no, look at the docs closely.. it's a framework for async tasks
18:20 luminous Corey: I would love to hear more about your story.. what did you do, what failed you? how did it get so bad?
18:20 Corey luminous: I was kidding. :-)
18:20 luminous yea, I tend to ignore puns
18:20 luminous haha
18:20 Corey luminous: Boy, is this the wrong project for you then. ;-)
18:20 Corey Every component is a pun.
18:21 * luminous shrug
18:21 luminous I don't even notice half the time, I don't usually care about those sorts of things
18:21 luminous I don't know why.. it's just what happens
18:22 luminous I don't want to replace celery as much as just not need it in the future
18:22 ramishra joined #salt
18:22 luminous funny that you're around Corey, we must be two of the only people not caring about this holiday :P
18:22 TaiSHi What holiday?
18:23 luminous awesome, 3's a crowd!
18:23 TaiSHi Oh it's (or was) 4th of July in the northern part of America
18:23 luminous yes, which is the whole weekend around these parts
18:23 TaiSHi Yeah
18:24 TaiSHi 9th of july is a holiday for us, and it's on
18:24 toastedpenguin joined #salt
18:24 TaiSHi Well, that day in the middle I always forget how to pronounce
18:24 TaiSHi It's called like that terrible terrible thing
18:24 TaiSHi WED
18:25 luminous wednesday?
18:25 luminous "whens-day"
18:25 TaiSHi According to my autocorrect
18:25 TaiSHi If I write it with non-capital W, it's incorrect
18:26 TaiSHi Trust me, I tried for a couple mins
18:26 luminous haha
18:26 luminous I do without auto-correct, I'd waste too much cpu cycles :P
18:26 luminous Corey: is that better?
18:26 ckao joined #salt
18:27 TaiSHi Sec, let me give you something
18:27 TaiSHi instantrimshot.com
18:42 scoates joined #salt
18:43 chamunks joined #salt
18:49 bhosmer joined #salt
18:49 krow joined #salt
18:52 mway joined #salt
18:54 mway Anyone seeing any wonkiness installing latest develop with salt-bootstrap? It's hanging indefinitely, the last line logged is "INFO: Running install_ubuntu_restart_daemons()"... any thoughts?
18:55 TaiSHi mway run it with -D
18:57 luminous mway: unless you are just playing around, I would seriously consider not touching latest 'stable', but run what you've tested the fsck out of (across all your use cases, formula, modules, etc)
18:58 luminous it's safer, more productive, and fewer headaches to assume the latest stable is buggy. salt is too big, does too much, and is too extensible to ensure 100% perfection
18:58 oz_akan_ joined #salt
18:58 fxhp joined #salt
18:59 ajolo joined #salt
19:01 TaiSHi luminous: lates develop you mean ?
19:02 mway @luminous Yeah, definitely not looking to use it in a real capacity like that. I don't generally use the bootstrap installer, just testing some bugfixes etc.
19:06 mway @TaiSH, sorry, ran afk - here's what I'm seeing: http://pastebin.com/2AVgdGjB
19:07 davet joined #salt
19:08 TaiSHi I do use develop in production :P until Helium comes out
19:08 luminous TaiSHi: no, latest stable
19:09 TaiSHi Seriously latest develop is stabler than latest stable?
19:09 TaiSHi mway: dig what's happening on the vm in the meantime
19:09 luminous whatever 2014.1.x we're on right now, this is the release, this is not to be trusted
19:09 luminous mway: sure, but now see how you are spending time on this instead of testing
19:09 luminous that's lost efficiency
19:09 TaiSHi Seriously 2014.1.x is not to be trusted?
19:10 luminous I would not, no
19:10 TaiSHi And what would you use ? Develop ?
19:10 luminous right now, I'm using 0.17.x
19:10 luminous it's known to be working in my environment
19:11 luminous and if I need to, I pull down states/modules from more recent releases, but those are tested
19:11 TaiSHi Yeah I need a lot of features introduced in helium
19:11 luminous blind updates to lastest release ('stable') will screw you
19:11 TaiSHi Like merge
19:11 luminous that's why you have _states and _modules
19:11 luminous you can pull in whatever you want
19:11 TaiSHi yeah but it's a serious pain to introduce all the merge stuff I need
19:11 TaiSHi I could try tbh
19:11 luminous it's a balancing act
19:12 TaiSHi But right now I can't, also not doing auto-deployment as of now
19:12 TaiSHi All the deploys are manual so...
19:12 luminous my point is this: don't blindly update, run what you have tested end to end
19:12 luminous and lock that version in
19:12 luminous you will save yourself a lot of time
19:13 mway I mean, it's not a bad point, and you're not wrong, and I will probably end up doing ad hoc updates for testing then. Still doesn't help with figuring out why the installer is hanging. :p
19:13 luminous a good handful of people in here will tell you the same
19:13 mway Even if it would be irrelevant at that point.
19:13 luminous mway: neither does that debugging help you test your original update
19:13 luminous the flip side is that the project needs people to test and find bugs
19:14 luminous so not everyone can do this, but that is the point to the salt jenkins setup
19:14 TaiSHi luminous: How about deb releases ?
19:14 * luminous shrug
19:14 luminous what do you mean?
19:15 TaiSHi Well I used to use ubuntu's packages
19:15 TaiSHi Until I needed a lot of stuff from Helium
19:16 luminous I use .debs, but I lock the apt rpeo
19:17 luminous .debs are great for simplicity, if you don't have a .deb availalbe, create the formula to build one for you
19:17 luminous either way, aim to automate everything and you'll continue to save time
19:19 TaiSHi I'll do that to my to-do list
19:19 TaiSHi I need the stablest stuff when auto-provisioning for my sites goes live
19:23 luminous that type of investment is very worthwhile, so far as I've found
19:23 luminous I've also found reclass to be incredibly helpful
19:23 ramishra joined #salt
19:23 luminous if anything, you're pillar config gets SUPER SIMPLE
19:23 luminous *your
19:24 luminous and then, if you build formula to expect ALL details, even stuff like service.running >> service.{{ status }} are all defined in pillar (reclass), your life is even better
19:24 luminous that's a big nugget right there. I can't go into more detail now, but if you look deep into that you'll find what I mean.
19:25 * luminous back to work
19:28 luminous TaiSHi: I guess one more detail to add there.. for configs that are key/value files, define them in pillar (reclass) and make your config an iteration over that pillar.
19:29 luminous that's million dollar advice right there :P
19:29 TaiSHi Not sure I fully understood it
19:29 TaiSHi But hey, let me grab that million dollar
19:30 TaiSHi I'm using formulas for the most of it, and most stuff is iterated I think
19:30 TaiSHi I'd need to dig on the reclass thignie
19:32 luminous yea, copy/paste that into a text file for later if you need
19:32 happytux joined #salt
19:32 luminous reclass is reclass.pantsfullofunix.net (or .org) I forget
19:32 luminous and it's up on github
19:32 luminous you can use it with salt-minion even
19:32 luminous it's a huge win
19:33 luminous I don't know why salt hasn't imported reclass ability to merge dictionaries
19:33 luminous pillar isn't usable (to me) without reclass
19:33 * luminous really back to work
19:33 TaiSHi Go back to work, I'm sure we will discuss this further on
19:34 TaiSHi It ding'd my interest
19:34 TaiSHi God I have
19:34 TaiSHi 25 minutes to clean my entire apartment
19:38 davet1 joined #salt
19:43 NV joined #salt
19:43 bhosmer joined #salt
19:53 krow joined #salt
19:59 oz_akan_ joined #salt
20:02 m1crofarmer joined #salt
20:04 m1crofarmer_ joined #salt
20:07 krow joined #salt
20:13 azylman joined #salt
20:14 meganerd joined #salt
20:21 m1crofarmer joined #salt
20:24 ramishra joined #salt
20:30 TheRealBill_here joined #salt
20:31 bhosmer joined #salt
20:32 TaiSHi luminous: ping, if you have any examples laying around for reclass I'd appreciate them
20:35 m1crofarmer_ joined #salt
20:54 Luke_ joined #salt
20:55 Luke_ joined #salt
21:00 oz_akan_ joined #salt
21:02 ajolo joined #salt
21:07 m1crofarmer joined #salt
21:11 bhosmer joined #salt
21:25 ramishra joined #salt
21:58 diegows joined #salt
22:01 oz_akan_ joined #salt
22:14 ajolo_ joined #salt
22:14 vexati0n joined #salt
22:15 yomilk joined #salt
22:22 Hell_Fire joined #salt
22:25 bhosmer joined #salt
22:25 ramishra joined #salt
22:36 luminous TaiSHi: in what sense? I can surely paste something
22:46 taterbase joined #salt
22:48 jalaziz joined #salt
22:51 yomilk joined #salt
22:52 therealGent joined #salt
22:52 therealGent So is salt-formulas on github like puppetforge?
23:01 oz_akan_ joined #salt
23:06 mateoconfeugo joined #salt
23:07 babilen therealGent: you can think of it that way if you like
23:07 therealGent babilen, is there something puppet can do that salt can't? Apart from GUI?
23:08 therealGent for example, is salt a good tool for fresh machine provisioning?
23:08 babilen therealGent: Probably and vice versa
23:08 babilen therealGent: It is a good tool for that. Excellent in fact.
23:13 therealGent So for most companies, they would either use salt or puppet right?
23:15 babilen Both are valuable tools, but there are certainly other approaches
23:16 therealGent Why would one use both?
23:17 babilen I don't, but historic reasons come to mind (infrastructure that was implemented in puppet has not yet been replaced with tasy salt)
23:17 babilen *tasty
23:26 ramishra joined #salt
23:52 yomilk joined #salt
23:53 chamunks joined #salt
23:55 krow joined #salt

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