Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2015-05-29

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

All times shown according to UTC.

Time Nick Message
00:00 forrest KyleG: Also might be good to try a straight up equivalence match to see if the grain is equal to EXACTLY that value
00:00 forrest see if you get better output
00:00 KyleG yeah it works when I put the exact value
00:01 forrest tiadobatima: Pretty sure it would compile them all up, I can't remember though, you'd either have to check out the code, or just try it out.
00:01 KyleG I'm basically trying to determine how I can seperate networks in an sls file, so that I can push the appropriate /etc/resolv.conf to the servers
00:01 tiadobatima cool
00:01 tiadobatima thx forrest
00:01 warthog42 forrest: looks like it could be this one:  https://github.com/saltstack/salt/issues/20615
00:01 KyleG based on their hostname IP which will determine which VLAN they're on and which firewall IP they need to go to
00:02 forrest KyleG: Yeah makes sense.
00:02 KyleG maybe I need to switch the logic
00:02 KyleG if grains in
00:02 KyleG rather than if ip in grains
00:02 forrest warthog42: You should add your example to that if you believe it is that one, you have a lot of good data
00:02 forrest KyleG: I usually just check if thing in grains, let me see if I can find an example, I don't think I have one for ip matching
00:04 arif-ali joined #salt
00:05 smcquay joined #salt
00:05 forrest KyleG: You said you already tried the star right? Just replacing the whole IP address with * to see if it matched?
00:06 KyleG * also returns unknown network
00:06 KyleG {% if '*' in grains['fqdn_ip4'] %}
00:06 forrest hmm
00:07 KyleG I also tried {% if grains['fqdn_ip4'] in ['*'] %}
00:07 forrest yeah, I'm not sure then, the syntax seems okay to me, maybe I'm tired and missing something (or just dumb).
00:08 forrest I've got to head out for the night, if you don't figure it out try to come back tomorrow morning, should be more people around.
00:08 forrest hopefully someone will point out something obvious I missed :)
00:08 KyleG gnite :) Thanks for trying
00:09 forrest np, sorry we couldn't get it figured out quickly.
00:09 loz-- joined #salt
00:11 warthog42 ok, yeah looking at saltmod.py looks like it was fixed in state() but not function()  I'll add my data to the bug.  thank you!
00:35 elfixit joined #salt
00:35 scoates joined #salt
00:36 MatthewsFace joined #salt
00:37 mosen joined #salt
00:37 bfoxwell joined #salt
00:59 sunkist joined #salt
01:01 otter768 joined #salt
01:04 scoates joined #salt
01:09 c10 joined #salt
01:12 greenride joined #salt
01:13 greenride left #salt
01:15 mbrgm joined #salt
01:18 bones050_ joined #salt
01:24 __number5__ TIL: archive state in 2014.7.5 don't support archive_user or user (not changing extracted files ownership) :(
01:29 vovcia left #salt
01:36 david_an111 joined #salt
01:37 murrdoc whats a good way to remove a key
01:37 murrdoc but not reject it
01:37 murrdoc like i am rebuilding a minion
01:37 murrdoc it will come back with a new key
01:37 murrdoc if i salt-key -d it then salt just rejects it
01:37 __number5__ remove the key on the minion and master
01:37 murrdoc minion is getting rebuilt
01:38 murrdoc is there no salt-key discard ?
01:38 murrdoc or salt forget about this minion
01:38 __number5__ remove old key from master should be ok
01:38 murrdoc thats salt-key -d + remove key
01:38 murrdoc or just phsyically rm the key
01:40 __number5__ delete the file is more reliable, me think
01:40 __number5__ still not fully understand how cache works yet
01:41 murrdoc i hate it
01:41 murrdoc and i think u for clarifying it
01:41 murrdoc but i hate it
01:41 murrdoc there should be a salt manage discard.minion
01:48 andrej heh
01:48 andrej make a feature request murrdoc
01:48 murrdoc i will probably write it up
01:48 murrdoc really bad
01:49 murrdoc get someone to hate driven programming fix it
01:49 murrdoc MUHAHAHA
01:49 tiadobatima hey folks... not sure I understand the behaviour of the environments with multiple top files... with this config https://gist.github.com/tiadobatima/785f61b39eeba61dd90c
01:50 smcquay joined #salt
01:50 tiadobatima aren't the directories specified in base available for dev as well?
01:51 tiadobatima local:
01:51 tiadobatima Data failed to compile:
01:51 tiadobatima ----------
01:51 tiadobatima No matching sls found for 'zookeeper' in env 'dev'
01:55 andrej Does anyone have an example handy on how to set a custom grain via API?
01:56 andrej Or an IPA, alnost that time of the week
01:56 aurynn I did that not that long ago
01:56 andrej :}
01:56 aurynn I'd have go dig into the code, tho
01:56 aurynn don't remember how
01:56 andrej Heh
01:57 andrej Would you do that? :}
01:57 andrej I'll shout you a beer ;}
01:57 murrdoc damn kid
01:57 troyready joined #salt
01:57 murrdoc github.com/saltstack/salt-contrib
01:57 murrdoc check out the grains dir
01:58 murrdoc send aurynn the beer tho
01:58 aurynn yay beer
01:58 aurynn :D
01:58 smcquay joined #salt
01:59 murrdoc make it a la fin du monde
01:59 andrej Whadda yah mean "send" =D .. we're neighbours, I learnt today ;}
01:59 andrej And I am no kid
01:59 andrej lol
01:59 murrdoc so literal
01:59 andrej The pommie comedy the end of the world? :)
02:00 murrdoc i do want to go to a restaraunt
02:00 murrdoc at the end of the world
02:00 andrej Ah - that one :)
02:03 andrej how old are you, murrdoc ? just wondering whether you calling me kid might be justified ;D
02:04 murrdoc 33
02:04 murrdoc kid
02:04 murrdoc :)
02:05 hasues joined #salt
02:05 hasues left #salt
02:05 andrej LOL
02:05 andrej You're almost young enough to be my son matey ;D
02:07 murrdoc but i have the hubris of an old un
02:07 murrdoc (i dont)
02:07 andrej and none of the modules in salt-contrib seem to be setting grains; they appear to return them
02:07 murrdoc what do u mean 'setting' grains
02:07 andrej I thought hybris was what one looses with age :}
02:07 murrdoc k u are old
02:07 murrdoc only older people get that
02:08 murrdoc what do u mean 'setting' grains
02:08 andrej heh
02:08 andrej I'm 50 ;D
02:08 murrdoc damn
02:08 murrdoc how do u type
02:08 murrdoc one finger i presume
02:08 murrdoc HAHAHA
02:08 andrej totally
02:08 andrej but very quickly
02:08 murrdoc what do u mean 'setting' grains
02:08 murrdoc u want to assign a grain to a minion
02:08 murrdoc there is a module for that
02:08 andrej re setting the gains ... I have a python program that extracts info about a range of minions
02:09 andrej I'd like to add that into to the respective minon's grains as a custom grain
02:09 andrej so I can then use it from a state
02:09 murrdoc where does this program run ?
02:09 andrej on the master
02:09 murrdoc k
02:09 andrej it's a runner at this stage
02:09 murrdoc right
02:10 murrdoc so u can execute 'modules' in it ?
02:10 andrej manually invoked during dev
02:10 andrej dunno :)
02:10 nikogonzo Pretend I have a box with a bunch of nics on different subnets, is there a grain/simple procedure for getting the ip address of whichever interface is connected to the network with the default gw? under linux if it matters
02:10 murrdoc andrej:  u can
02:10 murrdoc so just use __salt__['grains.setval']('key', 'val')
02:10 murrdoc now i must double check
02:11 murrdoc my assertion
02:11 andrej Ooooh ... sweet
02:11 murrdoc what version u running
02:11 murrdoc andrej:
02:12 andrej 2015.05
02:13 murrdoc yeah this is easy
02:14 hybridpollo left #salt
02:15 murrdoc try salt.modules.grain.setvals( {key1: value1}, {key2: value2})
02:15 andrej Will do
02:15 andrej thanks son :}
02:15 murrdoc *snerk*
02:15 __number5__ nikogonzo: grains ip have very details network info
02:15 tiadobatima murrdoc: can I use something like that from a pillar?
02:15 murrdoc what u mean 'from a pillar'
02:16 Furao_ joined #salt
02:16 murrdoc can u do stuff in python pillars ?
02:16 murrdoc course u can
02:16 murrdoc u can do whatever u want
02:16 murrdoc as long as u return a dictionary
02:18 andrej Hmmm ... wait ... how do I make that specific to a minion?  I have a list of dicts with minon names as the keys
02:18 tiadobatima I mean... if I need to make an external API call to get some data necessary for a salt-call run...
02:18 andrej and the values are dicts with interface names & IPs
02:21 andrej ugly, hackish idea ... I use popen to do it a la 'salt minion grains.setval name value' :D
02:22 nikogonzo __number5__: right, so for example if I grains.get ip4_interfaces I get a big list of IPs, but say only bond0:3 is on the network with the default gw. Is there a grain that I'm not seeing that will determine bond0:3 is the interface with the default gw ( among dozens of others )? or do I need to write this myself? I don't mind, just making sure I don't reinvent the wheel before I start
02:23 Aidin joined #salt
02:23 aurynn andrej, ew. don't do that. :)
02:23 aurynn import the module and do it in python
02:23 mgw joined #salt
02:25 favadi joined #salt
02:27 __number5__ nikogonzo: I'm not sure if there are other grains related, you can see all grains available by `grains.items`, and you might want to checkout the network module
02:27 __number5__ nikogonzo: https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.network.html
02:29 andrej aurynn : none of the modules I see that use grain.setval allow me to target a minion directly
02:29 andrej and I can't see how I'd do that
02:29 aurynn hmm
02:37 evle joined #salt
02:42 ITChap joined #salt
02:44 rdavis joined #salt
02:45 ITChap joined #salt
02:50 andrej Aight .. I think I'll wander down the ugly path mentioned earlier
02:51 aurynn andrej, you'll probably have to directly use the event bus, which would be ew
02:59 c10 joined #salt
03:01 otter768 joined #salt
03:09 beauby joined #salt
03:09 andrew_v joined #salt
03:10 andrej aurynn : I was told in another area that what I want can't be done with API, and the recommendation actually was to do exactly what I said above ... use the salt command line tool to address the minions.
03:10 andrej SO ....
03:10 andrej *sigh*
03:10 andrej Ugly as she goes
03:10 aurynn hm
03:11 aurynn this seems ... wrong
03:11 andrej yah
03:11 andrej \I agree
03:11 andrej Quoting Kevin Bloody Wilson:"Super mega fugly"
03:14 nikogonzo maybe someone here knows, http://lxr.free-electrons.com/source/include/uapi/linux/route.h?v=3.13#L50 says RTF_GATEWAY is 0x0002, however, /proc/net/route has 0003 for the default gw. I assume this is because int(0001,16) + int(0002) and a destination of 0.0.0.0 are what ip route and friends uses to determine 'default'?
03:46 Furao joined #salt
03:54 murrdoc joined #salt
03:56 rdavis left #salt
03:58 hybridpollo joined #salt
04:01 hal58th joined #salt
04:01 hal58th_ joined #salt
04:02 hal58th__ joined #salt
04:33 hvn joined #salt
04:33 hvn joined #salt
04:45 blahnana joined #salt
04:47 c10 joined #salt
04:51 TyrfingMjolnir joined #salt
04:51 noway_ joined #salt
04:55 rdas joined #salt
04:59 sneha joined #salt
05:01 blahnana I'm trying to use salt-ssh, and specifically am using it to communicate with a normal user account, which has sudo privileges (NOPASSWD)
05:01 sneha left #salt
05:01 blahnana I've noticed that things like salt-ssh target test.ping don't work, with dmidecode being listed as not found
05:02 blahnana I've tracked it down and used the debugging and it appears that despite using the roster file with sudo: True, sudo is not being used when commands are executed
05:02 sneha joined #salt
05:02 blahnana anyone used salt-ssh in that kind of setup (with a normal user account on the target, with sudo privs) and had success or similar failure?
05:02 blahnana I'm using the ubuntu PPA as my source for salt
05:02 otter768 joined #salt
05:05 sneha hello! I am using ubuntu 12.04 and I want to replace a word in a file with the hostname or minion-id using salt?Can it be possible using grains or script will be better?Please help me out.
05:12 hojgaard joined #salt
05:14 hojgaard_ joined #salt
05:15 markm joined #salt
05:17 joeto joined #salt
05:22 blahnana actually looks to be more that dmidecode has no path
05:23 mohae joined #salt
05:26 blahnana found it finally https://github.com/saltstack/salt/issues/23342
05:27 juanito how can i make sure that a user.absent state executed properly and then trigger an action after that
05:27 juanito im trying this but doesnt work
05:27 juanito http://pastebin.com/E9A2nCjA
05:35 bluenemo joined #salt
05:48 c10 joined #salt
06:05 bfoxwell joined #salt
06:07 colttt joined #salt
06:15 ndrei joined #salt
06:18 badon_ joined #salt
06:22 viq joined #salt
06:24 Furao joined #salt
06:25 AndreasLutro joined #salt
06:26 peters-tx0 joined #salt
06:27 rdas joined #salt
06:29 hvn joined #salt
06:30 Auroch joined #salt
06:32 c10 joined #salt
06:32 Berty_ joined #salt
06:40 illern joined #salt
06:41 CeBe joined #salt
06:44 bastiaan joined #salt
06:44 quist` joined #salt
06:45 flyboy joined #salt
06:45 stephanbuys joined #salt
06:51 stephanbuys joined #salt
06:54 ndrei joined #salt
06:56 malinoff joined #salt
07:01 kawa2014 joined #salt
07:01 al joined #salt
07:02 KermitTheFragger joined #salt
07:02 impi joined #salt
07:03 tmclaugh[work]_ joined #salt
07:03 otter768 joined #salt
07:14 eseyman joined #salt
07:23 fe92 joined #salt
07:24 fredvd joined #salt
07:26 aw110f joined #salt
07:30 quist`` joined #salt
07:31 aelmi_ joined #salt
07:32 aw110f_ joined #salt
07:32 VSpike Can anyone tell me how I can work out which releases (if any) this code is in https://github.com/saltstack/salt/pull/13445/files ?
07:32 Aidin joined #salt
07:34 supersheep joined #salt
07:38 AndreasLutro VSpike: https://github.com/saltstack/salt/commit/d27c6912b905bd48dd43092e7454bde4faaef4e4
07:39 AndreasLutro you just click the commit hash link in "thatch45 merged commit d27c691 into saltstack:develop on Jun 13, 2014"
07:39 Auroch joined #salt
07:39 eliasp VSpike: take the commit ID (e559b33) and ask git which tags/releases contain this commit… "git branch --contains e559b33"
07:39 dthorman joined #salt
07:40 eliasp eh "git tag --contains e559b33", sorry ;)
07:40 eliasp but "branch" works too, just for branches then
07:47 mortis how do i lookup pillars (can i even?) in an ext_pillar-module? im noobing around trying to figure it out.. the module works as intended, but i want to add something like : pillar.get("my_super_pillar") inside the module and cant figure out how
07:47 mortis any pointers?
07:49 mortis im guessing: from salt.modules import pillar
07:49 mortis and then something, but cant get it to work
07:52 Aidin left #salt
07:55 ninkotech__ joined #salt
07:57 fredvd joined #salt
07:57 VSpike AndreasLutro / eliasp : thanks! I'm still confused because it shows it in tags 2014.7 onwards to 2015.5.1 but only in branch develop
07:59 AndreasLutro VSpike: 2014.7 was develop when the PR was created
07:59 VSpike And it seems like no matter what version I select on github, I don't see that code in the file... e.g. https://github.com/saltstack/salt/blob/2014.7/salt/modules/win_pkg.py
08:00 AndreasLutro well, it might have been reverted
08:00 PI-Lloyd joehh: I can't see 2015.5.1 in the Ubuntu PPA, any news?
08:00 sk_0 joined #salt
08:01 VSpike I see output in the log suggesting that code is running in my 2015.5.0 minion, but if i look at https://github.com/saltstack/salt/blob/2015.5/salt/modules/win_pkg.py it's not there either
08:01 VSpike The thing is, I see a bug related to that piece of powershell it's running, but I can't work out if that actual code should in 2015.5.0 or not
08:01 AndreasLutro VSpike: you misunderstand how git works - the *commit* is present in those versions, the code may have been removed since then
08:01 Grokzen joined #salt
08:01 AndreasLutro in fact it has: https://github.com/saltstack/salt/commit/2d11d6545e324ccfcb5d323282328aee03e312e1
08:02 AndreasLutro https://github.com/saltstack/salt/pull/23433
08:03 AndreasLutro was able to find that through the history tab on the win_pkg.py file
08:04 s_kunk joined #salt
08:06 VSpike AndreasLutro: oh, thanks, that's good. I do understand how git works, and it certainly seemed to me like that code had been reverted at some point (although I didn't know how to find out when so thanks for that)
08:06 VSpike but that still doesn't explain why I see output in my minion log suggesting that exact piece of code is running
08:07 VSpike Here's the trace output https://bpaste.net/show/c39de2c79a41
08:07 s_kunk joined #salt
08:08 Berty_ joined #salt
08:08 VSpike I figured either I'm not running the version I think I am, or there's a problem with the build process, or github is not telling me what I think it's telling me
08:08 AndreasLutro what's your master and minion versions?
08:10 stephanbuys1 joined #salt
08:11 AndreasLutro VSpike: as far as I can tell, the powershell stuff is present in all 2014.7 versions except 2014.7.6, and in 2015.5.0 (not 2015.5.1)
08:14 ndrei joined #salt
08:16 soren joined #salt
08:18 dthorman joined #salt
08:24 losh joined #salt
08:24 dthorman joined #salt
08:28 VSpike AndreasLutro: 2015.5.0 both
08:28 VSpike Ahh, I see what you mean
08:29 VSpike Well, that's good then. So if I upgrade to 2015.5.1 my wierd bug should go away :)
08:32 pkimber joined #salt
08:33 balltongu joined #salt
08:35 ksj morning. is there any way to dynamically modify the minion config? I know it seems like a bad idea, but I need to add some lines for mysql config to the end of it and then restart. I'd like to test the config is valid before restarting the minion, obviously
08:38 markm joined #salt
08:41 balltongu joined #salt
08:44 VSpike AndreasLutro / eliasp : many thanks for the help! I understand what I'm seeing in github much better now
08:45 VSpike ksj: If you're talking about mysql creds, you can put those in the pillar as it turns out
08:45 ksj VSpike: oh right, cool, didn't know that
08:45 ksj thanks
08:45 VSpike ksj: no problem. Someone here suggested it might work, and it does for sure.
08:51 permalac joined #salt
08:52 Hazelesque Re; preseeding hosts with salt [private] keys and SSH host keys and such... do you think using Openstack's userdata (config drive/cloud-init etc) would be "safe" enough?
08:52 cberndt joined #salt
08:52 ksj VSpike: can you tell me how to do it? nothing on http://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.mysql.html
08:52 cberndt joined #salt
08:53 keimlink joined #salt
09:00 ndrei joined #salt
09:04 otter768 joined #salt
09:11 supersheep joined #salt
09:14 denys joined #salt
09:14 codehotter in salt ec2.py, aws.query is called in a lot of places with return_root=True, but the rest of the code seems to be assuming return_root=False. Am I stupid? This has been this way since november 2014, and if it was broken, surely someone would have noticed.
09:14 codehotter But... what am I missing?
09:15 codehotter For example, here https://github.com/saltstack/salt/blob/develop/salt/cloud/clouds/ec2.py#L1591
09:16 mage_ any idea for this https://dpaste.de/kmdH/raw ? it happens only the *first* time the python3 package is installed, if I restart the salt minion it works perfectly .. I think I should reload a module (or .. ?) but I don't know what/why ...
09:24 codehotter Anyone have a tip for how to go about understanding this code? I'm 99% sure that it's a bug, but I don't understand how it can really be broken without anyone noticing
09:24 codehotter Is there some mailing list I should post to or a development irc channel I should join?
09:27 c10 joined #salt
09:28 PI-Lloyd joined #salt
09:29 AndreasLutro codehotter: github issues tend to get replied to fairly quickly
09:30 AndreasLutro codehotter: what makes you think the rest of the code "assumes return_root=False"?
09:32 julez joined #salt
09:33 julez joined #salt
09:40 ndrei joined #salt
09:41 codehotter AndreasLutro: I don't understand when you want return_root=True or when you would want return_root=False. return_root=False is the default, but it is explicitly set to return_root=True in a number of places
09:41 codehotter however the code later down  if 'blockDeviceMapping' in rd_data[0]:
09:41 codehotter rd_data[0] would be the requestId, if return_root=True. Right?
09:41 codehotter and that will never have blockDeviceMapping. That would be in rd_data[1][0]
09:44 codehotter I checked the other places that return_root=True in this file, and I don't understand why it is set to true there either.
09:44 codehotter O
09:44 codehotter I'm not that familiar with the amazon api
09:47 AndreasLutro codehotter: I don't know the api at all, but return_root=True would just make it return the root element instead of the first child of the root element
09:48 codehotter OK, but then you do "if 'blockDeviceMapping in rd_data[0]" - that condition will never be true, I think.
09:48 codehotter because rd_data[0] is just the request_Id
09:49 codehotter if you want something that potentially has blockDeviceMapping, you're looking in rd_data[1][0]. If you do return_root=False, rd_data[1] is what it returns in the first place, so then rd_data[0] works.
09:49 codehotter That wasn't the most clear explanation but I hope you understand me
09:50 AndreasLutro the aws.query function seems to do some magic stuff as well so I'm not sure https://github.com/saltstack/salt/blob/develop/salt/utils/aws.py#L259
09:50 codehotter AndreasLutro: https://github.com/saltstack/salt/blob/develop/salt/utils/aws.py#L412
09:51 codehotter items = root[1], if return_root is True: items = root
09:51 AndreasLutro yes, I know
09:51 AndreasLutro but for example
09:51 AndreasLutro https://github.com/saltstack/salt/blob/develop/salt/utils/aws.py#L427-L429
09:52 AndreasLutro depending on how salt._compat.ElementTree works that could be the root element or all elements
09:53 codehotter I did print(json.dumps(rd_data, indent=3)) in the ec2.py
09:53 codehotter then rd_data[0] is requestId
09:53 codehotter if I put return_root=False, it's something that contains blockDeviceMapping
09:53 AndreasLutro then it sounds like you're right
09:54 codehotter What about the other places that have return_root=True?
09:54 codehotter When do you want return_root=True?
09:54 codehotter Do you ever want it?
09:55 codehotter Are those others also broken?
09:55 AndreasLutro shrug
09:55 AndreasLutro I'd open a github issue
09:55 codehotter How can it be broken everywhere without anyone noticing
09:55 codehotter OK.
09:56 codehotter Maybe amazon changed the api recently
09:57 codehotter I will try to check the other places I guess. I wish I was more experienced with python and salt and amazon api V_V I'm just a poor sysadmin not a programmer!
09:59 julez joined #salt
10:00 Aidin joined #salt
10:02 linjan joined #salt
10:03 LeProvokateur Hi, I want to deploy collectd-web from its git-repo to a minion. I activated git as file_server and the github-repo as gitfs-remote. But how do I use it in a state-file?
10:04 ndrei joined #salt
10:04 julez joined #salt
10:09 julez joined #salt
10:11 bluenemo joined #salt
10:12 quist`` joined #salt
10:16 refnode joined #salt
10:17 Eureka__ joined #salt
10:22 ksj can anyone point me to an explanation of when to use salt['pillar.get'] and when to use pillar[] in state files?
10:23 mage_ ksj: when you want to use 'some:keys', for ex: salt['pillar.get']('foo:bar:blah')
10:24 danjj ksj: http://docs.saltstack.com/en/latest/topics/tutorials/pillar.html#pillar-makes-simple-states-grow-easily scroll up a bit from that link
10:24 mage_ "pillar" only is just a dict
10:25 Sacro @seen utahdave
10:25 Sacro !seen UtahDaver
10:25 Sacro grr
10:26 ksj so, if you only want one value, you use pillar['something']['something'], but if you want to return a list of values, you use salt['pillar.get']....or....no hold on, that's not right. dammit the coffee's just not working properly this morning
10:27 ksj I really need to learn more python so I can read the code and see what it's actually doing
10:29 malinoff ksj, imagine this dict: {'key1': {'key2': 'somevalue'}} as your pillar. pillar['key1']['key3'] fill fail since there is no 'key3' in the inner dict, and salt['pillar.get']('key1:key3', 'defaultvalue') will return 'defaultvalue' with no errors
10:30 wnkz joined #salt
10:31 malinoff ksj, although you can write this: {{ pillar['key1'].get('key3', 'defaultvalue') }} but a deeper nesting will result in this: {{ pillar['key1'].get('key2', {}).get('key3', {}).get('key4', {}).get('key5', 'default') }}
10:31 danjj as I was reading the docs I ran across at least one place where it was warned that pillar[] wouldn't work in that particular situation
10:32 allanparsons joined #salt
10:32 danjj as a result, and for consistency, I'm only using pillar.get
10:38 giantlock joined #salt
10:39 ksj malinoff: ok, thanks, I think I get it from that.....maybe. long story short, just stick with salt['pillar.get'], which is ugly, but works
10:40 malinoff ksj, you can apply this to almost every piece of salt :)
10:41 ksj malinoff: yeah, my beautiful, simple yaml state files that worked are now starting to get real ugly and complex with all the jinja hacks necessary to abstract them
10:42 codehotter OK, I finally understand now how amazon api works. Yes, I am idiot, and yes, most of the return=True is correct, and yes, I found a bug, this one isn't correct.
10:42 codehotter I feel much better now ^^
10:48 dendazen joined #salt
10:54 mortis no hints as to how i can access pillars from within python-code?
10:55 eliasp mortis: http://docs.saltstack.com/en/latest/topics/development/dunder_dictionaries.html
10:55 mortis oooh tnx eliasp
10:57 mortis so basically it should be available whenever you import salt in python?
11:00 numkem joined #salt
11:00 malinoff mortis, of course not, it is injected magically into some places
11:01 mortis im trying to play with it in bpython
11:01 mortis trying to figure out how to import salt and access its pillar-data
11:02 mortis im a bit too dumb to find out how
11:02 AndreasLutro that would probably be a difficult task
11:03 Aidin joined #salt
11:03 mortis shouldnt it just be a matter of using the client module somehow?
11:03 AndreasLutro of course not
11:04 AndreasLutro the salt client python code does a ton of bootstrapping before it actually executes anything
11:04 mortis what i eventually want to end up with, is accessing pillar-data inside an ext_pillar-module i have made
11:04 mortis not even sure if thats possible
11:05 mortis thats why im fiddling with it in bpython
11:05 AndreasLutro you could try starting here https://github.com/saltstack/salt/tree/develop/scripts and work your way up the chain
11:05 numkem_ joined #salt
11:05 otter768 joined #salt
11:05 denys joined #salt
11:06 AndreasLutro with a system as large as salt I think I'd just test it live
11:06 ndrei joined #salt
11:07 AndreasLutro not like on a production server, but add the module to salt and debug via salt-call or whatever
11:08 mortis the module works fine, i just dont get the pillardata ....i will debug some
11:10 linjan joined #salt
11:11 hhenkel joined #salt
11:13 numkem joined #salt
11:19 ponpanderer joined #salt
11:21 JDog joined #salt
11:26 evle joined #salt
11:30 michelangelo joined #salt
11:32 eirsyl joined #salt
11:32 c10 joined #salt
11:33 c10 joined #salt
11:41 codehotter salt-run cloud.action reboot=True passes reboot as a boolean, but salt-cloud -f reboot=True passes reboot as a string.
11:41 codehotter Is that a bug?
11:42 codehotter Should I write code to convert to string in my function? Or try to fix salt-cloud or salt-run?
11:42 bhosmer joined #salt
11:42 saifi joined #salt
11:43 codehotter Also should create_image (which creates an ami in the provider, but FROM an instance) be an 'action' which receives an instance name, or a 'function' which receives a provider?
11:46 amcorreia joined #salt
11:47 patrek joined #salt
11:54 Twiglet iggy: that mine issue you were helping me diagnose yesterday was caused by this: https://github.com/saltstack/salt/issues/23835
12:07 julez joined #salt
12:16 DammitJim joined #salt
12:18 eseyman joined #salt
12:19 julez joined #salt
12:19 ndrei joined #salt
12:28 hasues joined #salt
12:29 _mel_ can i have a list like set modules = " moduleA moduleB" and create a file.managed who creates one line per module like  "blacklist moduleA"
12:29 hasues left #salt
12:33 impi joined #salt
12:36 giantlock joined #salt
12:37 julez joined #salt
12:38 julez joined #salt
12:52 zerthimon joined #salt
12:53 iggy Twiglet: you are using multi-masters?
12:53 toastedpenguin joined #salt
12:54 jdesilet joined #salt
12:55 Twiglet nope, just had it in a list
12:55 iggy ahh, yeah, there are other bugs about that too
12:55 Twiglet rather than "master: ip" it was "master:\n - IP"
12:55 ProT-0-TypE joined #salt
12:55 Twiglet Yeah bit of an odd one, no wonder we were both stumped
12:57 iggy I hope they track that down soon
12:57 iggy it seems to cause a lot of problems
12:58 emaninpa joined #salt
13:00 jeremyr joined #salt
13:00 andrew_v joined #salt
13:02 subsignal joined #salt
13:04 mage_ is the Reactor System the right thing to use to manage interactions between states ?
13:04 bastiaan joined #salt
13:05 mage_ for example I have a state venv.python.foo avec I would like to restart a daemon every time this venv is regenerated/upgraded
13:05 mage_ I know I can do that with watch/watch_in but it's less flexible
13:06 otter768 joined #salt
13:06 mage_ in other words: is the Reactor System a pub/sub-like system?
13:09 primechuck joined #salt
13:09 iggy as long as you only pub/sub one thing at a time
13:09 ndrei joined #salt
13:10 mage_ and when to use watch/watch_in and the Reactor System ?
13:10 mgw joined #salt
13:11 Auroch joined #salt
13:16 ProT-0-TypE joined #salt
13:16 cpowell joined #salt
13:21 dyasny joined #salt
13:22 drawsmcgraw mage_: That sounds like a good use case for Reactor to me.
13:23 cpowell joined #salt
13:29 jhauser joined #salt
13:29 impi joined #salt
13:32 racooper joined #salt
13:34 cpowell joined #salt
13:35 cpowell_ joined #salt
13:37 wnkz joined #salt
13:40 wnkz joined #salt
13:41 sunkist joined #salt
13:42 wnkz joined #salt
13:43 wnkz joined #salt
13:43 Tyrm joined #salt
13:45 Hydrosine so i have a cronjob which i like to rm, but salt doesn't match on the cronjob. It doesn't have an identifier
13:45 Hydrosine bash /home/script_apply_asd >/dev/null 2>&1
13:46 Hydrosine thats what the cronjob looks like^^
13:46 Hydrosine i tried enclosing in "" and '' and also cron.rm_job and cron.rm
13:46 wnkz_ joined #salt
13:51 Hydrosine nvm i'm going to do a crontab -r :))
13:53 wnkz joined #salt
13:54 OliverMT joined #salt
13:55 OliverMT I am trying to service.running on a file I have placed in /etc/init.d/ on ubuntu, but salt keeps saying service is not found, even though I have tried both require_in from file or require: - file from service... anyone got a hint?
13:56 OliverMT aha, found the cause just as I wrote this https://github.com/saltstack/salt/issues/2908 :P
13:56 dendazen joined #salt
13:57 wnkz joined #salt
13:57 mage_ if I have a foo: ... bar: ... in an sls file (so 2 definitions) is there a way to only execute "foo" with state.sls ?
13:59 mage_ for ex: https://gist.github.com/silenius/423aae95d056683a8e45
13:59 peters-tx joined #salt
14:00 mage_ I have this so I can regenerate each venv with a state.sls venvs.python.foo or state.sls venvs.python.bar
14:00 peters-tx joined #salt
14:00 mage_ but I wonder if there is a way to achieve this without having those two extra files in my example
14:01 LGSilva joined #salt
14:03 Twiglet is where a way to get a list of online servers matching a glob from a minions?
14:03 Twiglet kinda like {% for server in salt['mine.get']('rabbit-*') %}
14:04 Twiglet oh wait, ignore the "from a minion" there, derping hard today
14:04 mage_ I found salt '*' state.sls core exclude="[{'id': 'id_to_exclude'}, {'sls': 'sls_to_exclude'}]" but it's really ugly ...
14:05 gladiatr joined #salt
14:07 LGSilva hello. I need to write a ini file based on pillar data but I'm getting ID xxx in SLS xxx is not a dictionary
14:08 LGSilva here is the pillar and the jinja template: http://pastebin.com/fGq8910N
14:10 iggy LGSilva: that's not valid yaml
14:10 iggy and it's kind of hard to tell what's what there
14:11 iggy protip: gist supports multiple files per paste
14:11 LGSilva iggy: the pillar data is not valid?
14:11 iggy so you don't have to cram all your files into 1 paste making it unreadable
14:12 LGSilva let me try gist
14:12 DammitJim I'm reading conf_master-svnfs_trunk
14:12 gladiatr joined #salt
14:12 DammitJim and it says that the user running the salt master will need read access to the repo
14:13 DammitJim how do I do that if the salt master is running as root?
14:13 DammitJim I have a user configured with read only capabilities in subversion
14:14 LGSilva iggy: https://gist.github.com/anonymous/dbdf75754e6bbf179a9a
14:15 iggy LGSilva: what about the state file? (you can just update that existing gist
14:15 perfectsine joined #salt
14:15 iggy the error message says your state file is horked... not the pillar or jinja template
14:20 murrdoc joined #salt
14:22 LGSilva iggy: you're right. I was missing the source file in the state file (duh!) thank you for the help
14:22 iggy ^5
14:22 iggy go forth and help others
14:26 schuckles joined #salt
14:28 DammitJim is there a good doc on how to configure the salt master with svn?
14:35 murrdoc iggy:  so orchestrate has this tgt thing yeah
14:36 murrdoc so it lets you run a bunch of states across minions
14:36 DammitJim I can see that for git, one can set authentication with git_user
14:36 murrdoc what if i wanted to orchestrate one minion
14:36 DammitJim but there is no such option for svn?
14:38 sunkist joined #salt
14:39 nobrak hi folks, using salt-ssh (2015.5.1) in a virtualenv trying to run test.ping, i get an error about no _ssl module. however can verify it's there in my virtualenv. pastebin containing what i've run through: http://pastebin.com/K3V01cPG
14:39 julez joined #salt
14:40 dendazen https://www.eventbrite.com/e/saltstack-enterprise-training-vilt-or-in-person-at-saltstack-tickets-16448027493
14:40 dendazen this salt training
14:40 dendazen is it regular salt stack training or it is training for salt-stack  enterprise?
14:41 DammitJim when using a source control backend for files, does the minion check out the file or the master checks it out and then pushes it to the minion?
14:42 Pulp joined #salt
14:45 DammitJim or do I have to set up my project on SVN as read only for everyone?
14:45 overyander joined #salt
14:46 sandah joined #salt
14:46 iggy murrdoc: tgt: one-minion
14:47 iggy but then that's just a highstate, not really an orch
14:47 racooper joined #salt
14:47 iggy dendazen: they are basically the same (minus the web interface)
14:49 giantlock joined #salt
14:49 dendazen So if i do not intend to use enterprise version, i still can take this training?
14:49 ProT-0-TypE joined #salt
14:50 premera joined #salt
14:50 iggy dendazen: 99% sure the answer is yes
14:50 dendazen Thanks.
14:50 iggy but they will likely try to sell it while you're there ;)
14:50 dendazen yeah, they did send us a quote.
14:51 dendazen it gets pretty expensive arounf 125$ a node.
14:51 dendazen per year
14:51 iggy so $10/month/node
14:51 iggy cheaper than most
14:52 iggy it'd be nice if some infrastructure came with that
14:52 iggy like git hosting for the backends or something
14:52 murrdoc yeah but i dont know waht the one-minion is
14:52 dendazen yeah, still if you have around 500 nodes it is 62,500$ a year.
14:52 iggy then we wouldn't have to also fork over for github
14:53 murrdoc i want the reactor to call orchestrate for the one minion that called the reactor event
14:53 schuckles joined #salt
14:53 iggy they might have better luck pricing like rightscale does... % of your cloud bill
14:53 DammitJim murrdoc, did you say you use git with salt?
14:53 iggy murrdoc: that's perfectly acceptable
14:53 dendazen we use git with sal
14:53 dendazen salt
14:54 murrdoc but how do i pass down the minion id to the orchestrate
14:54 DammitJim dendazen, does that mean that your master checks out files from git or the minion checks out stuff from git?
14:55 mpanetta joined #salt
14:55 dendazen master checksout from git.
14:55 dendazen minion gets stuff from master
14:55 murrdoc iggy:  but how do i pass down the minion id to the orchestrate
14:56 iggy {{ data['id'] }}
14:56 murrdoc gtfo
14:56 DammitJim dendazen, so you have gitfs_remotes defined in master?
14:56 dendazen yes.
14:57 DammitJim how do you provide authentication credentials for git?
14:57 iggy murrdoc: http://docs.saltstack.com/en/latest/topics/reactor/ it's in the second example
14:57 dyasny joined #salt
14:58 murrdoc the most valid rtfm ever
14:58 murrdoc i owe u a beer
14:58 murrdoc so i can pass down the {{ data.id }} to orchestrate from the reactor.sls
14:58 iggy vodka please
14:58 SheetiS joined #salt
14:59 Norrland Hm, if I specify pillars in "/srv/pillar/docker.sls" for my docker state. Can I override those pillar values in say "/srv/pillar/dev.sls" ?
14:59 murrdoc i ll get u when u come to la
14:59 iggy murrdoc: correct
14:59 unni joined #salt
15:00 murrdoc done
15:00 Gareth morning morning
15:00 unni Hi There,  I am trying to add a reactor to execute on new minions with a prefix and is not working. Where can I see the logs of reactor? Please see my conf https://gist.github.com/anonymous/2a0e2ee261ff2aaaa9d4
15:03 dendazen joined #salt
15:05 dendazen ssh keys
15:05 DammitJim dendazen, thanks! oh really? so, the check out is done via scp basically?
15:06 DammitJim hhmmmm... I don't think I can do this with svn :(
15:06 dendazen well, it’s git checkout via pull
15:07 otter768 joined #salt
15:07 DammitJim is that like doing a git clone?
15:11 Grokzen joined #salt
15:12 dendazen joined #salt
15:13 debian112 joined #salt
15:18 iggy it's actually a fetch w/o a checkout (vs a clone which is basically fetch+checkout)
15:18 iggy it accesses the git objects directly
15:19 DammitJim man, am I really going to have to set up a git server instead of using subversion?
15:20 DammitJim I can't find a way to specify a different user other than the one salt master is using (which is root)
15:24 SheetiS joined #salt
15:25 iggy DammitJim: you can't just put the info in the url?
15:25 DammitJim iggy, I'm in the process of testing that according to: https://github.com/saltstack/salt/issues/6763
15:25 DammitJim you know, even though you don't have the direct answer (since you don't use svn), just the fact that you answered is good support to keep me going
15:26 iggy well, I would have had something sooner, but I was digging through the salt fileserver code and pysvn code to see what options were supported
15:27 iggy apparently, you can also set a username per server in ~/.subversion/servers
15:28 murrdoc iggy:  so how do i call an orchestrate job for one user ?
15:28 murrdoc uh one minion
15:28 mbrgm left #salt
15:29 iggy didn't we already figure that out?
15:30 * evilrob needs to look at the orchestration bits.  He's just been highstating boxes
15:30 DammitJim if I set fileserver_backend: - svn and - roots
15:30 DammitJim does that mean that the master will check out stuff from svn and put it in roots?
15:31 iggy DammitJim: no (at least not with git)
15:31 DammitJim or does that mean that if I run an sls on a minion, that the master will "checkout" the file and run it instead of reading it from roots?
15:32 murrdoc iggy:  i get how it works through reactor to orchestrate
15:32 ageorgop joined #salt
15:32 murrdoc actually let me rtfm
15:34 bharper joined #salt
15:34 iggy DammitJim: not entirely sure with svnfs, but gitfs actually reads the git objects directly (and never actually does a checkout)... the only place the file ever gets fully restored is in the minion cache
15:40 DammitJim man, why is the spacing so messed up in the master file?
15:40 DammitJim like one has to have a space before file_roots: for example?
15:40 DammitJim or did I configure something wrong?
15:40 iggy you configured something wrong
15:41 DammitJim ugh
15:41 iggy it's standard yaml
15:41 DammitJim maybe because the first instance already had a space
15:41 jalbretsen joined #salt
15:44 codehotter Can I use salt-cloud to deploy a virtual machine WITHOUT installing salt on it?
15:44 codehotter ec2
15:46 iggy codehotter: set the bootstrap script to /bin/false or something
15:46 iggy probably true
15:46 iggy since you don't want salt-cloud freaking out because the bootstrap script failed
15:49 DammitJim I'm getting  No matching sls found for 'tomcat7.tomcat7' in env 'base' when trying to use svnfs
15:50 iggy salt-call cp.list_master and see what you see
15:51 iggy salt-run -l debug fileserver.update
15:51 DammitJim https://gist.github.com/anonymous/e3858962b0be23f46ff0
15:51 DammitJim that's my master
15:53 DammitJim salt-run said this: https://gist.github.com/anonymous/62ad5ddc9ac713b82897
15:53 DammitJim and salt-call said: [ERROR   ] This master address: 'salt' was previously resolvable but now fails to resolve! The previously resolved ip addr will continue to be used
15:53 DammitJim [WARNING ] Master hostname: salt not found. Retrying in 30 seconds
15:53 iggy protip: gist supports multiple files per gist and multiple revisions
15:53 DammitJim sorry, maybe that last one should have been a gist
15:54 DammitJim or were you talking to protip to run those commands? LOL
15:54 iggy trying to save myself the extra memory for the multiple chrome tabs
15:55 iggy the gitfs backend chattier
15:58 murrdoc joined #salt
16:02 desposo joined #salt
16:02 stoogenmeyer joined #salt
16:03 zerthimon joined #salt
16:03 throwaway joined #salt
16:05 drawsmcgraw joined #salt
16:05 toastedpenguin joined #salt
16:12 gladiatr joined #salt
16:15 ek6 how many gitfs_remotes are folks running... just following the guidance for salt-formulas seems like it could lead to quite a few and ive been very conservative pushing that number up to date.
16:15 mgw joined #salt
16:15 iggy ek6: ~16
16:16 murrdoc 0
16:16 ek6 no 'stability' issues?
16:16 * murrdoc rolls with debs
16:16 iggy no more than without
16:16 ek6 that was to you iggy...i think its quite clear murrdoc has stability along with other 'issues'
16:16 ek6 ok thanks
16:17 iggy (to be precise, each environment has it's own master, dev has 0 gitfs remotes, qa/prod have ~15)
16:18 ek6 thats fine...i just dont want to risk issues with the couple very important gitfs mounts to add on a bunch of formulas that would be used infrequently
16:19 DammitJim man, this svn business is not fun
16:20 DammitJim https://gist.github.com/anonymous/62ad5ddc9ac713b82897
16:20 ajw0100 joined #salt
16:20 DammitJim why isn't salt looking in my svn repo?
16:20 DammitJim I have it defined in the master as fileserver_backend: -svn -roots
16:21 DammitJim when I restart salt-master, I don't get any errors in the master log
16:21 iggy do you see anything in /var/cache/salt/master/svnfs ?
16:22 drawsmcgraw DammitJim: Are you able to try out some of the module functions on the command line for some sanity checks? https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.svn.html
16:22 DammitJim there is no svnfs folder in /var/cache/salt/master :(
16:22 drawsmcgraw Could try just checking out a repo to see if the master at least has the configs right.
16:23 ek6 everyone else getting expired cert warning from all the saltstack.com sites?
16:23 [7hunderbird] joined #salt
16:23 ajw0100_ joined #salt
16:24 DammitJim so, I just run: sudo salt '*' svn.info ?
16:24 DammitJim why is there a "path/to/repo" ?
16:24 DammitJim I didn't specify a path to repo
16:24 DammitJim anywhere
16:25 DammitJim or is path/to/repo something like svn://myserver.com ?
16:25 drawsmcgraw ek6: N-no. That link to modules.svn I just posted had no problems in Chromium for me.
16:26 drawsmcgraw DammitJim: mmm... it looks like it's the path on disk to a checkout
16:26 baweaver joined #salt
16:26 drawsmcgraw Can you try just checking out a repo?
16:26 DammitJim drawsmcgraw, if I run salt '*" svn.info, I get: "svn.info" is not available
16:26 drawsmcgraw D:
16:26 drawsmcgraw of course...
16:26 DammitJim ok
16:26 drawsmcgraw This is the same problem you had yesterday, yeah?
16:26 DammitJim let me make a snapshot of my vm
16:26 drawsmcgraw The module isn't loading
16:26 DammitJim well, today I'm trying to load the module from svn vs locally
16:27 drawsmcgraw I mean Salt's refusing to load its SVN module because it seems to be detecting that there is no 'svn ' python module
16:27 eseyman left #salt
16:28 DammitJim I did an: sudo apt-get install python-svn
16:28 DammitJim does it matter that I had already checked out the code w/o salt? with just regular svn co ?
16:29 DammitJim can I do 'master' instead of '*' ?
16:29 DammitJim I don't want to check out code to the minions!
16:29 DammitJim or is that what you are asking me to do?
16:30 drawsmcgraw hrm....
16:30 drawsmcgraw That the repo already exists shouldn't matter if you're trying  svn.info()
16:30 drawsmcgraw In fact, I'd figure it's a requirement
16:30 whytewolf is python-svn pysvn?
16:30 drawsmcgraw And, the way you're running your commands, yeah, the minion is the one that's going to be doing the checkout
16:31 drawsmcgraw whytewolf: good question... I'm not really sure exactly what package needs installing here... Wonder if the code can tell us..
16:31 enarciso joined #salt
16:31 DammitJim but isn't this supposed to work like: the master checks out the sls and pushes it to the minion?
16:31 DammitJim I need to know this; otherwise, I need to start opening up firewall ports
16:31 spookah joined #salt
16:32 whytewolf salt.fileserver.svnfs calls for pysvn
16:32 DammitJim http://blog.mindfab.net/2014/04/saltsubversion-backend.html
16:32 DammitJim drawsmcgraw and whytewolf , I followed the above link
16:32 DammitJim but this still doesn't explain if the master is doing the checkout or the minions
16:33 drawsmcgraw Looks like it just needs the 'svn' command to be available on the command line: https://github.com/saltstack/salt/blob/develop/salt/utils/__init__.py#L440
16:33 whytewolf DammitJim: master checks out and pushes to minion. the minions do not need svn installed
16:33 DammitJim now, I installed python-svn on the master, not the minion
16:33 drawsmcgraw Yeah the minion is the one doing the work
16:33 drawsmcgraw Quick sanity check -> what exactly are we trying to accomplish here?
16:34 whytewolf drawsmcgraw: svnfs, not svn module
16:34 DammitJim I'm trying to use svn to hold all my sls files and conf files
16:34 drawsmcgraw I'm glibly assuming you just want to do an svn checkout somewhere but I've been wrong before.
16:34 drawsmcgraw AH
16:34 drawsmcgraw Thank you
16:34 DammitJim I don't even know if I need to do an svn checkout
16:34 bluenemo joined #salt
16:34 drawsmcgraw No, you shouldn't have to do *anything* for this to work. It should just checkout and cache it
16:34 dendazen joined #salt
16:35 writtenoff joined #salt
16:35 DammitJim I guess from reading, I assumed that by setting up a fileserver_backend as svn, that the files could be checked in to the svn server and then when the file is needed by the master to perform a state.sls, it would just pull it from svn
16:35 DammitJim but maybe I'm wrong
16:35 drawsmcgraw DammitJim: So if you configure it, stop the Master, then run the master in the foreground (salt-master -l debug), *nothing* comes out of the logs?
16:35 murrdoc Damnit Jim … test it out!
16:35 ferbla joined #salt
16:35 murrdoc :D
16:35 DammitJim murrdoc, test what out? ;) I've been doing all kids of tests
16:36 murrdoc Dammit Jim!
16:36 DammitJim dammit!
16:36 murrdoc Jim!
16:36 DammitJim ok drawsmcgraw I started salt-master with debug
16:36 aparsons joined #salt
16:38 drawsmcgraw Nothing interesting?
16:38 DammitJim https://gist.github.com/anonymous/843bb4a9055d3d401392
16:38 DammitJim I don't think it is even trying to use svn
16:40 DammitJim This is my master configuration: https://gist.github.com/anonymous/e3858962b0be23f46ff0
16:40 KyleG joined #salt
16:40 KyleG joined #salt
16:40 whytewolf are you getting any errors in your logs?
16:41 whytewolf about loading svn
16:41 DammitJim in /var/log/salt/master ?
16:41 DammitJim no
16:41 DammitJim or what log are you referring to?
16:41 whytewolf yeah thats what i was refering to
16:42 DammitJim no, there is nothing in there
16:42 huddy joined #salt
16:42 DammitJim like I said... it's as if the master isn't even aware that svn is a fileserver source
16:43 DammitJim if I remove - roots from fileserver_backend, I get a weird error
16:43 iggy just check it out
16:44 DammitJim iggy, what do you mean?
16:45 iggy stop trying to use svnfs and just make it a static checkout (in /srv/salt or wherever)
16:45 DammitJim you mean... every time I need to run something, to run an svn co manually?
16:46 iggy well, you could setup a commit hook that poked the salt master and told it to do a pull
16:46 DammitJim if I remove - roots, I get this when I start the salt-master: https://gist.github.com/anonymous/a50434f9cc1f8f575a42
16:47 DammitJim or should I look into just using git?
16:47 DammitJim maybe git is more mature?
16:47 iggy certainly more used
16:51 DammitJim hhhmmmmm
16:53 tomh- joined #salt
16:55 forrest joined #salt
16:55 TranquilityBase joined #salt
16:58 Berty_ joined #salt
16:59 cruatta joined #salt
16:59 cruatta joined #salt
17:00 cruatta_ joined #salt
17:01 Brew joined #salt
17:03 forrest joined #salt
17:05 TyrfingMjolnir joined #salt
17:07 zerthimon joined #salt
17:08 otter768 joined #salt
17:10 anotherZero joined #salt
17:17 drawsmcgraw DammitJim: Totally walked away, there. Sorry. If you're able to use Git then abso-freakin'-lutely use Git.
17:17 drawsmcgraw The cases where I see Subversion being used are poor people stuck on legacy projects that are unable to migrate to Git
17:19 murrdoc DammitJim:  http://helb.github.io/goodbye-sourceforge/
17:21 jngd joined #salt
17:23 rap424 joined #salt
17:27 DammitJim lol
17:27 DammitJim I am poor
17:27 DammitJim I was just trying to reuse some of the existing infrastructure we have
17:28 DammitJim and not have to set up a new server, then configure backups for that server, etc
17:28 schuckles joined #salt
17:29 whytewolf why not put git on the same server as subversion? neither one takes up a ton of resources
17:30 RedundancyD joined #salt
17:33 DammitJim you know how it is... it takes forever to get permission to put stuff on servers
17:34 mrbigglesworth joined #salt
17:34 whytewolf i work at a bank, and don't have that much trouble.
17:35 murrdoc forrest:  I'd give salt the middle finger at that point.
17:35 murrdoc HAHAHAHA
17:35 tiadobatima question: the default minion config comes with this comment: "# By default, the Salt fileserver recurses fully into all defined environments
17:35 tiadobatima # to attempt to find files. "
17:35 DammitJim whytewolf, this place is very "interesting"
17:36 DammitJim I'm going to set up a git server on the side
17:36 tiadobatima But I'm not seeing this behaviour
17:36 zerth joined #salt
17:36 forrest murrdoc: I would
17:36 forrest DammitJim: You shouldn't put them on the same box.
17:36 whytewolf interesting is how they phrase it on the job description when they are looking for new human sacrarfices?
17:37 forrest DammitJim: What happens when you need to retire SVN? People will still be going 'BUT THE SERVER IS UP, OMG WHY DOES IT NOT WORK!'
17:37 forrest DammitJim: Just spin up some super cheap VM if you guys are in the cloud.
17:38 DammitJim forrest, you don't understand LOL... this is a very "interesting" place
17:38 DammitJim but yes, I can spin up VMs as I wish
17:38 DammitJim so, that's what I'll do
17:38 tiadobatima in this config, https://gist.github.com/tiadobatima/785f61b39eeba61dd90c I get an error saying zookeeper formula (defined in dev's top sls) is not found
17:38 whytewolf good point forrest i forgot about whiney devs.
17:38 DammitJim but I think I got a little distracted this morning and I need to get back to configuration management vs source control
17:39 forrest interesting or not, multiple apps on a single server BE DAMNED!
17:39 forrest DammitJim: config management versus source control?
17:39 DammitJim forrest, I wasn't going to put git on the same svn server
17:39 DammitJim I mean... create the SLSs for setting up my servers
17:40 DammitJim instead of worrying about source control for the SLSs
17:40 DammitJim and files
17:40 DammitJim iggy, had a point... I should check out stuff manually for now... one step at a time
17:40 forrest yeah, you can easily switch over to gitfs or something later.
17:41 DammitJim yes, sorry for the rambling... I'm sure I'm not the only one with issues
17:42 asoc I am trying to wrap my head around a reactor sls files. Basically what I need to do is to kick off a two different processes on different servers then after the second server is done go and kill the first process. Is there a more robust way rather than just adding a scheduled job in the future to go clean up the first process?
17:42 iggy tiadobatima: there's a bug open about that... can't remember the #
17:42 stoogenmeyer joined #salt
17:42 asoc I am using local.cmd.run to kick off the processes but I can't seem to see how to wait for the second process to finish before going something else.
17:43 tiadobatima iggy: https://github.com/saltstack/salt/issues/12483#issuecomment-64181598 ?
17:44 TyrfingMjolnir joined #salt
17:45 forrest KyleG: Did you figure that IP thing out last night?
17:45 iggy tiadobatima: looks like the one I was thinking of, yeah
17:45 murrdoc KyleG:  iggy that same ng shit is in nginx too
17:46 murrdoc i hate it
17:46 MatthewsFace joined #salt
17:46 forrest murrdoc: Complain to cheus
17:46 forrest ;)
17:46 iggy yeah, I looked at the nginx formula one time
17:46 iggy slit my wrists
17:47 iggy woke up in a cave a couple days later and decided to just open code an nginx state
17:47 KyleG forrest: Not yet, but I feel like I might be able to use split('.') to accomplish it
17:48 forrest KyleG: Ugh, that should just work though :\ Can you paste that gist again?
17:48 forrest Ahh it's still up
17:48 forrest iggy: https://gist.github.com/kylegato/0ee67e92451363396c05 You see anything there as to why the matching isn't working that I missed last night?
17:49 Gareth Hrm. Just encountered a bug in 2015.5 that I thought I had fixed.
17:50 iggy forrest: without more context, it looks like it's doing what it's supposed to
17:51 forrest iggy: Well you can see the grain listed, but when the state right below runs, it doesn't match that IP. I thought the wildcard matching needed would just be a .* on the end of the IP, but apparently that didn't work
17:52 iggy forrest: oh, yeah, the grain is a list, and you are trying to string comparison
17:52 forrest iggy: I'm not trying it, KyleG is. Shit I forgot that grain is a list
17:52 KyleG aye
17:52 bradthurber joined #salt
17:53 iggy so maybe something like {% if '172.16.20' in grains['fqdn_ip4']|first %}
17:53 dimeshake joined #salt
17:53 forrest iggy: Good idea.
17:54 iggy or {% if '172.16.20' in grains['fqdn_ip4']|join %}
17:54 KyleG you geniuses
17:54 KyleG {% if '172.16.20' in grains['fqdn_ip4']|first %}
17:54 KyleG seems to have worked
17:55 Pixionus So using salt and cmd.run I cat a file and >> it to a new one and for some reason it grew by 115%
17:55 Pixionus anyone know what this extra bit of info might be?
17:55 ndrei joined #salt
17:56 mgw joined #salt
17:56 ageorgop joined #salt
17:56 debian112 anyone using: salt.states.file.replace
17:58 debian112 any pointers with this: http://paste.debian.net/187538/
17:58 iggy replace uses the re module
17:59 forrest basepi: Is Jacob Hammons around today?
17:59 forrest basepi: Or did you not go into the office
17:59 iggy it's probably seeing that as a re match... try escaping the ()
17:59 impi joined #salt
18:00 murrdoc sad
18:00 murrdoc there isnt a pillar.setval
18:01 debian112 iggy: on the pattern or repl?
18:01 iggy murrdoc: wheel module?
18:01 c4urself hey all, any pointers for copying the contents of a directory (not in salt config) into another dir via a state?
18:01 iggy debian112: pattern
18:01 basepi forrest: I am in office today,,
18:01 basepi But not at the moment.
18:02 basepi Can't recall whether he was or not.
18:02 iggy strip club for lunch on Fridays?
18:02 forrest basepi: Okay, I didn't see him in IRC. Can you ask him to hop in here if he's around?
18:02 forrest basepi: When you get back
18:02 basepi Will do.
18:03 c4urself specific use-case is the easy-rsa ubuntu pkg which basically installs a bunch of useful utils in /usr/share/easy-rsa that i'd like to copy into a vpn dir
18:04 theologian joined #salt
18:05 iggy c4urself: file.copy won't work?
18:05 c4urself iggy: it looks like it uses shutil.copy which only accepts individual files as the source
18:05 supersheep joined #salt
18:05 forrest c4urself: I'd just use a cmd.run that copies the content
18:06 c4urself forrest: good call, got caught up into looking into file.copy and forgot the simple way out
18:07 iggy or module.run with salt.modules.file.copy
18:07 psojourner joined #salt
18:07 c4urself thanks guys
18:07 forrest Another good idea iggy. I am still confused why there isn't just a recursive copy
18:07 forrest np c4urself
18:09 asoc Is there a way inside one reactor sls file to define multiple jobs on different servers happen in a sequence?
18:10 jlhammons joined #salt
18:10 whytewolf reactor fireing off an orchenstrate?
18:10 ek6 yup
18:11 KyleG default gateway would be a great default grain to have available :}
18:12 jlhammons forrest: you around?
18:12 iggy KyleG: grains aren't supposed to change often (although they get computed every request for some reason)
18:13 KyleG every request?
18:13 KyleG wow
18:13 forrest jlhammons: Yes
18:13 KyleG so me extending the grains system is no more efficient than me using a cmd.run in a state sls file to grab the data...
18:13 asoc whytewolf: Thanks I think that is the nudge I needed to move forward.
18:13 iggy KyleG: there's network.routes
18:14 KyleG That's for setting routes not getting them isn't it?
18:14 forrest jlhammons: For the salt docs, are you deploying to whatever system was previously hosting the docs? Both iggy and I have noticed that the new version of the docs load the page in a way where you can't immediately do a ctrl/cmd f search.
18:14 iggy and network.default_route
18:14 iggy it fetches by default
18:14 KyleG "    'network.default_route' is not available."
18:15 KyleG On 2014.1.13 (Hydrogen)
18:15 iggy it's in 2015.5
18:15 KyleG ah
18:15 iggy oh flying spaghetti monster...
18:15 iggy 2014.1
18:15 jlhammons forrest: the system was moved, it is hosted on wpengine. It could the way the page is structured though
18:15 * iggy puts KyleG on /ignore
18:16 KyleG lol
18:16 iggy jlhammons: the right side also doesn't scroll anymore
18:16 KyleG I used to upgrade salt very often
18:16 forrest jlhammons: Okay cool, iggy are you still noticing the issue I described regarding the search problem?
18:16 KyleG but a few bugs bit me in the ass too hard
18:16 jlhammons iggy: I think that is a caching issue, can you do a hard refresh?
18:16 forrest iggy: ? Scrolls for me
18:16 iggy okay, refresh fixed it
18:17 iggy should cache bust the js/css or something
18:17 forrest jlhammons: I didn't want to file an issue regarding the search thing because it's not really a problem with the site itself.
18:17 forrest jlhammons: Wasn't sure if it was just an issue with how content was being serviced/cached via CDN or something
18:17 jlhammons iggy: great suggestion
18:17 jlhammons forrest: do you mind reminding me what the search issue is :/ :)
18:19 forrest jlhammons: It's just what I described above. So when I load the docs I almost exclusively immediatey do a ctrl/cmd f search to find the appropriate item I want to look at. With how the page is currently loading you can do a search for let's say file.replace, it won't find it, you have to hit ctrl/cmd f again.
18:20 DammitJim silly questions
18:20 forrest so I'll google file.managed salt, go to the page, then ctrl/cmd f for file.replace, it won't grab it that first time. Seems more common in firefox than in chrome
18:20 DammitJim I just "cloned" a VM that had a minion
18:20 DammitJim it automatically showed up on my master as an Accepted key
18:20 iggy it's like it's sending it out in chunks and only doing the first chunk
18:20 MeggieBot joined #salt
18:20 DammitJim does that mean that both VMs (original and clone) have the same key?
18:21 DammitJim what is suggested to do in these cases? re-issue a key?
18:21 forrest yeah that's a good way to describe it iggy, it only happens if you're very fast on your ctrl/cmd f search
18:21 jlhammons forrest: thanks - would you mind submitting an issue with your browser info and if you see the issue on each page? That way I can track it better
18:21 forrest DammitJim: Delete the data on the new minion (key and such), and rejoin the master. Might have to delete it off the master as well
18:22 forrest jlhammons: Will do, I've noticed it on every page of the docs so far.
18:23 iggy DammitJim: it'd be like running 2 copies of the minion on the same host... it'll lead to all kinds of squirrelyness
18:23 MeggieBot joined #salt
18:23 jlhammons forrest: thanks
18:23 iggy until you remove the minion_id and key file from the new one
18:23 kylebusch joined #salt
18:23 forrest jlhammons: NP, sorry for the confusing description, it's tough to describe and I didn't want to go back and forth in an issue comment chain.
18:23 kylebusch left #salt
18:23 iggy jlhammons: I still think the right menu could use a bigger scroll handle
18:23 DammitJim ok, I'm deleting everything in /etc/salt/pick
18:24 forrest iggy: lol
18:24 DammitJim pki
18:24 DammitJim now what?
18:24 iggy DammitJim: /etc/salt/minion_id
18:25 iggy jlhammons: forrest: fwiw, I think it's a browser rendering thing... a curl fetch of everything seems to be _very_ fast
18:25 DammitJim if I start up salt-minion, will it auto regenerate the keys then?
18:25 iggy DammitJim: yes
18:26 forrest iggy: I agree.
18:26 forrest iggy: I'll link you the issue once I make it.
18:26 iggy danke
18:26 saltycharles joined #salt
18:27 forrest iggy: But yeah, I'm almost 100% sure it's a rendering problem. You can notice it as the menu on the right loads, it's almost like they are tied together in some way, so the page data for the appropriate section doesn't load until the relevant data is loaded in that menu.
18:27 jlhammons iggy: I checked and the width is adjustable, so I can make it bigger. Mind making an issue to track?
18:28 iggy rgr
18:29 forrest sorry to drag you back in here jlhammons ;) I figure it's probably nearing lunch time
18:29 jlhammons forrest: anytime :), I really appreciate the feedback
18:30 forrest Cool
18:33 sdm24 joined #salt
18:34 quique joined #salt
18:37 Tyrm joined #salt
18:37 forrest iggy: https://github.com/saltstack/salt/issues/24253
18:38 debian112 so apparently its the whitespace in the pattern: http://paste.debian.net/187578/
18:38 debian112 between if (data
18:39 quique joined #salt
18:39 elektrix hopefully an easy question. I have a salt state, and I want to require an sls that's in a different module. What's the format for the require? - sls: module.state
18:39 elektrix ?
18:39 KyleG it's like
18:39 KyleG include:
18:40 KyleG - module.state
18:40 debian112 any idea how to include whitespaces or do I try to ignore it
18:40 elektrix ok...I'm guessing it's failing because I'm not calling highstate, just doing salt-call state.sls othermodule.state_that_requires_the_other_state production?
18:41 KyleG white space in sls file names?
18:41 KyleG really?
18:43 LeProvokateur I would like to use the saltstack logo in a talk, where do I find one under CC, ideally scalable?
18:43 CeBe joined #salt
18:45 nobrak hi folks, using salt-ssh (2015.5.1) in a virtualenv trying to run test.ping, i get an error about no _ssl module. however can verify it's there in my virtualenv. pastebin containing what i've run through: http://pastebin.com/K3V01cPG
18:48 murphyslawbbs joined #salt
18:53 elektrix figured it out...forgot to include the state file in the state file where I had the require
18:54 elektrix that's what I get for not reading doc close enough
18:55 CeBe joined #salt
18:55 murrdoc yeah dude
18:57 ajw0100 joined #salt
18:58 psojourner joined #salt
19:01 hybridpollo joined #salt
19:02 cruatta joined #salt
19:02 CeBe joined #salt
19:03 stanchan joined #salt
19:04 jale2ice joined #salt
19:04 jale2ice joined #salt
19:04 jale2ice joined #salt
19:06 supersheep joined #salt
19:07 saltycharles joined #salt
19:09 otter768 joined #salt
19:09 rallytime joined #salt
19:10 jale2ice Hello, Anyone using the api to preseed a minion? I'm getting an exception when running the code in the documentation. https://gist.github.com/anonymous/68a33367f5d50f86e4af
19:12 vieira joined #salt
19:19 N-Mi joined #salt
19:20 cpowell joined #salt
19:20 viq joined #salt
19:23 mrbigglesworth joined #salt
19:24 giantlock joined #salt
19:25 bhosmer joined #salt
19:26 Berty_ joined #salt
19:27 hybridpollo joined #salt
19:29 soren joined #salt
19:31 forrest murrdoc: Sorry I just realized my comment to your php-ng suggestion sounded harsher than I meant. I think at some point we should break them out, but I don't think we can do so right now.
19:36 denys joined #salt
19:39 Tahm joined #salt
19:42 murrdoc its ok bro
19:42 murrdoc i ll survive
19:45 tkharju joined #salt
19:49 mrbigglesworth joined #salt
19:50 bhosmer joined #salt
19:51 solidsnack joined #salt
20:02 supersheep joined #salt
20:04 dendazen joined #salt
20:05 Dev0n joined #salt
20:06 sporkd2 joined #salt
20:11 baweaver joined #salt
20:11 forrest murrdoc: I was mostly just poking fun at how confusing the ng formulas are ;0
20:11 forrest *;)
20:13 murrdoc its ok
20:13 murrdoc me and jim figured it out
20:13 murrdoc its cool forrest
20:13 forrest Cool
20:14 sporkd2 hey guys, I can't figure out why salt-cloud is saying aws cannot validate my credentials. if I use the same id: and key: from the command-line on the master I can access aws without issue (2015.5.0)
20:14 murrdoc jim bean that is
20:15 forrest lol
20:16 dstokes hey guys. is there any way to query the job list for a failed states output? something like salt-run jobs.lookup_jid <id> Result=false?
20:16 forrest sporkd2: https://github.com/saltstack/salt/issues/22498 Looks like that issue to me, please add details to the issue/bump it if you have anything to provide.
20:17 bradthurber joined #salt
20:18 soren joined #salt
20:19 denys joined #salt
20:20 MatthewsFace joined #salt
20:24 murphysl1wbbs joined #salt
20:25 solidsnack joined #salt
20:25 solidsnack joined #salt
20:26 sporkd2 @forrest I'm not even getting an instance created
20:28 cheus joined #salt
20:28 forrest sporkd2: I'm not sure then, my familiarity with how salt-cloud does the authentication is kind of limited, manfred are you around and able to comment? aws creds work fine on the command line, but fail with salt-cloud.
20:30 sporkd2 no, reading further into that gist it seems to be the same, bummer this is happing in 2015.5.0
20:30 smcquay joined #salt
20:30 cheus @forrest / @iggy: Yeah, yeah complain as you will. I've actually addressed some of the complexity patterns in more recent iterations (not published) but it's the philosophical battle that I know is divided.
20:30 cheus At the end of the day some people just like establishing api's between their map files and states that split where the data lives into sls, map, and pillar, makes hard assumptions about use cases, and really isn't DRY at all. ;-)
20:31 forrest cheus: We know, it's just fun to poke fun about it. We do need to consider the repo structure though. I don't have a good solution, and we've talked about this before :\
20:32 ajw0100 joined #salt
20:32 cheus @forrest: Aye. I'd say deprecate them and wait to see if I or Arnold or someone else can agree on the structure of a 2.0 formula style which I do think leans the way of pure-python with some abc fundamentals.
20:33 murrdoc woah
20:33 murrdoc thats bad
20:33 forrest cheus: deprecate the ng stuff you mean?
20:33 murrdoc pure python states is terribad
20:33 linjan joined #salt
20:33 murrdoc its like dsl's arent a thing
20:34 cheus @forrest: Yes? I dunno -- I'm not aware of their overwhelming popularity and Seth had us put them in the .ng namespace for a reason -- prototypical.
20:34 forrest cheus: I don't know either. If I did I would have proposed some sort of solution already :)
20:34 forrest Going pure python DOES give me some flashbacks to chef though
20:34 forrest just people writing pure ruby
20:35 forrest spewing their awfulness all over the repos
20:35 pcn cal
20:35 murrdoc yeah its all bad
20:35 pcn Whoops
20:35 cheus Yes but you know what? Sometimes you just need os.path.join or the ability to transform pillar data in order to keep it DRY.
20:35 forrest might be better in this case, I don't use pure python with salt
20:35 murrdoc you dont
20:35 murrdoc i havent needed it yet
20:35 murrdoc and if it is needed … put it in salt
20:36 murrdoc instead of formuals
20:36 forrest cheus: I feel like you work somewhere with terrible infrastructure :(
20:36 forrest The workarounds you've done with some of the ng stuff is cool, but I hate to think what created that need, lol
20:37 forrest murrdoc: Different strokes for different needs I guess, I can't blame cheus since I don't know what he's dealing with.
20:39 ropes I'm just wondering, how common is it to use jinja macros for templating state definitions to save on verbosity? it seems like it just adds another layer of debugging and might lead to misery with extensive usage, but I don't have experience there, so wondering what the community has to say..
20:40 cheus forrest: Not these days (I'm with another organization these days) but it's not that it's a matter of the fact that take a simple case where you've got services that are interacting with files on a filesystem. Somewhere you create the folder where the files live but you also need your service files managed by other means. Now I'm either stuck with having to include '/some/folder/dir' in my pillar and '/some/folder/dir/filexxx' which isn't DRY
20:40 cheus since one's inside the other or I do a -bad- '/some/file/dir' ~ '/' ~ 'filexxxx' in jinja
20:40 iggy ropes: I've never used one
20:41 murrdoc forrest:  heres the thing, you can right ridiculous tricky code, but if people are going to learn baout salt from salt-formulas , then lets not
20:41 ropes no: 1  limited:0  yes: 0
20:41 ropes iggy: ty
20:42 iggy there are some in different formulas, the salt-formula one is the only one I can really see a good reason for
20:42 forrest murrdoc: Oh yeah, that's been my perspective since forever, but I can understand the perspective. I just don't care about some repeats if it's easier to understand what is happening
20:42 cheus forrest: We wrote a lot of formulas and re-deployed some of them in more contexts than I can count. So I probably deployed the nginx formula to support at least 50 different end-user applications each with quirks. At some point I was either going to have to expose every part of every salt function like a salt-map api or just make everything pillar-override-able
20:42 iggy and honestly, it'd better be done elsewhere, but I'm not fixing it
20:43 lexter joined #salt
20:43 forrest cheus: Totally understandable
20:44 ropes iggy: I've been burned enough times by the salt compilation system to want to avoid further complication
20:44 spookah joined #salt
20:44 cheus The pattern is a bit more refined in my old age -- the current jinja/yaml one, I mean. But I've been letting Arnold's approach steep. And I, for one, would much rather debug python errors than jinja compilation ones.
20:46 iggy that's the one thing I can agree on
20:47 cpowell joined #salt
20:47 ropes cheus: that aligns with my thoughs as well
20:47 schuckles joined #salt
20:52 UtahDave joined #salt
20:53 forrest Hey UtahDave.
20:53 UtahDave jacksontj: you around?
20:53 UtahDave hey, forrest!
20:58 stoogenmeyer joined #salt
21:00 linjan joined #salt
21:00 sdm24 How do I pass jinja and/or grains/pillar information in the command line? Basically, my states are broken into 2 envs: Linux and Windows. I want to target all minion and run a state 'foo' (one version is linux.foo, the other is windows.foo). Can I do "salt '*' state.sls {{ grains['kernel }}.foo, or something equivalent?
21:01 sdm24 I vaguely remember seeing a way to do this but  I can't find it in the docks
21:01 Gareth UtahDave: hey hey
21:01 UtahDave hola, Gareth!
21:02 Gareth UtahDave: hows it going?
21:02 UtahDave pretty good!  it's been a super busy awesome week!
21:03 UtahDave how are you?
21:04 Gareth pretty good :)
21:04 Gareth Glad it's Friday.
21:05 MatthewsFace joined #salt
21:05 UtahDave yeah
21:06 Heartsbane is it beer o'clock yet?
21:07 nahamu It is in Boston.
21:09 MohShami joined #salt
21:09 otter768 joined #salt
21:13 murrdoc joined #salt
21:15 brandk joined #salt
21:16 baweaver joined #salt
21:17 linjan joined #salt
21:20 arnoldB hehe
21:21 arnoldB guys, is there any info when the enterprise  demo recording will be available?
21:21 stoogenmeyer joined #salt
21:22 iggy it didn't happen
21:24 TyrfingMjolnir joined #salt
21:25 forrest jlhammons: You still around?
21:26 g3cko joined #salt
21:27 jlhammons forrest: yep
21:27 forrest jlhammons: Did you see this already today? https://percy.io/
21:28 murrdoc http://arstechnica.com/security/2015/05/crypto-flaws-in-blockchain-android-app-sent-bitcoins-to-the-wrong-address/
21:30 jlhammons forrest: I didn't, that looks really useful. Would be nice to have in place for automated doc builds
21:31 forrest yeah I was thinking the same
21:38 primechuck joined #salt
21:42 jlhammons forrest: I requested an invite
21:42 forrest jlhammons: Awesome
21:45 sdm24 does anyone know how/if you can store a hashed password into a pillar?
21:45 hasues joined #salt
21:45 hasues left #salt
21:46 MohShami hey guys, is it possible in the top file to have the "*" section apply to all servers and have other sections that expand on it?
21:46 iggy thats how it works
21:46 brandk joined #salt
21:47 MohShami iggy: if I add a section for a server it overrides whatever is in the "*" section
21:47 iggy nope, top file is additive
21:47 MohShami iggy: I remember it wasn't, seems I got that wrong, let me check
21:47 MohShami thanks a million
21:51 voxxit joined #salt
21:52 MohShami_ joined #salt
21:55 fllr joined #salt
21:56 hal58th_1 joined #salt
21:56 hal58th_2 joined #salt
21:56 hal58th_3 joined #salt
21:58 murrdoc salt needs some 'validation' states
21:58 murrdoc like 'count number of processes of name blah'
21:58 stevednd joined #salt
21:59 MohShami joined #salt
22:06 steve_ joined #salt
22:13 stevednd joined #salt
22:14 solidsnack joined #salt
22:19 smcquay joined #salt
22:24 debian112 what to check with everyone on this file.replace: http://paste.debian.net/187644/
22:24 debian112 the pattern is not matched
22:25 debian112 when I add the space after if and before (data
22:25 debian112 tried this but nothing: "if\s*(\s*data.status !== 'Active')"
22:25 debian112 any idea?
22:25 sdm24 regex?
22:26 sdm24 never mind I should've read what you typed first haha
22:27 debian112 need some help on that
22:27 sdm24 https://regex101.com/
22:28 sdm24 copy your file contents (or at least whats relevant) in the bottom part, and then play with the different regex until you select what you want
22:30 TyrfingMjolnir joined #salt
22:31 sdm24 and is that pattern what is actually in the file, or do you want an IF statement so that if that pattern is different, replace it?
22:32 debian112 yeah
22:32 debian112 I want the if as well
22:32 sdm24 ok
22:34 murrdoc joined #salt
22:35 mrbigglesworth joined #salt
22:36 marcusguttenplan joined #salt
22:36 sdm24 https://gist.github.com/sdm24/faaf7b81ce84894c337c
22:36 sdm24 try that perhaps?
22:37 oo joined #salt
22:38 sdm24 So that, for example, for the line " if (data.status !== Active ", it will replace it
22:44 steve_ joined #salt
22:44 TaiSHi murrdoc: wake up, not weekend for you
22:48 forrest debian112: Use the python regex to ensure it's getting interpreted correctly.
22:49 murrdoc TaiSHi:  go away , baitin
22:50 TaiSHi wut? speak properly you wannabe southamerican!
22:51 murrdoc there is a movie called idiocracy
22:52 murrdoc its a documentary about the future
22:52 murrdoc beautiful movie, check it out, that line will make sense
22:52 iggy wannabe south american... lol
22:53 iggy I still call him "brown guy in a yellow shirt" in my head
22:53 murrdoc for an indian dude, i get called a bunch of things
23:03 spiette joined #salt
23:06 dendazen joined #salt
23:10 otter768 joined #salt
23:14 joehoyle joined #salt
23:15 solidsnack joined #salt
23:15 joehoyle Hey, I'm trying to get the JSON output, the docs say to use --static to not just get a line of JSON for each minion, however this is the output I'm getting:
23:15 joehoyle https://gist.github.com/joehoyle/39a1c7d5121d57ef5a81
23:16 joehoyle from what I understood about the docs, this is not meant to happen?
23:16 spiette joined #salt
23:20 ingwaem joined #salt
23:31 stevednd joined #salt
23:34 SheetiS joined #salt
23:42 baweaver joined #salt
23:44 murrdoc heres a fun one
23:44 murrdoc i need highstate to run till it passes
23:46 smcquay joined #salt
23:48 cruatta_ joined #salt
23:50 brandonk joined #salt
23:51 bfoxwell joined #salt
23:51 detachedHEAD joined #salt
23:55 warthog42 joined #salt
23:57 Dev0n joined #salt

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