Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2016-04-13

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

All times shown according to UTC.

Time Nick Message
00:00 AdamSewell joined #salt
00:02 flowstate joined #salt
00:08 dlam joined #salt
00:09 jab416171 'string' in 'string' is true, so does 'string' in ['string', 'foo']
00:09 jab416171 so there really shouldn't be *any* difference.
00:09 jab416171 oh, you know what? I think I see what's going on
00:09 RandyT joined #salt
00:10 jab416171 I've actually been testing from salt02 in my example
00:10 jab416171 salt_path[0] will have saltstack, but salt_path[1] will have saltstack-staging
00:11 teatime heh
00:11 teryx510 joined #salt
00:11 nZac joined #salt
00:11 jab416171 stupid hacks
00:12 jab416171 and we had the same problem before the hack, just kept trying to fix it when the underlying problem is the way the loop is working.
00:12 teatime I don't understand
00:13 teatime btw the |sequence filter is nice when you want to ensure a value is a sequence (empty things get upgraded to [], strings get upgraded to single-element lists)
00:14 brianfeister joined #salt
00:16 hoonetorg joined #salt
00:16 _JZ_ joined #salt
00:19 Muchoz joined #salt
00:26 jchen joined #salt
00:26 jchen left #salt
00:30 InAnimaTe joined #salt
00:31 pfallenop joined #salt
00:32 jchen joined #salt
00:32 jchen left #salt
00:34 josuebrunel joined #salt
00:35 armguy_ joined #salt
00:36 hacks jab416171: What did I ever do to you?
00:37 antpa joined #salt
00:41 cyberviking joined #salt
00:45 antpa joined #salt
00:47 amcorreia joined #salt
00:49 bltmiller joined #salt
00:50 arif-ali joined #salt
00:53 ajw0100 joined #salt
00:57 cpowell joined #salt
01:03 edrocks joined #salt
01:05 _armguy_ joined #salt
01:05 hoonetorg joined #salt
01:07 jab416171 heh
01:10 sjorge joined #salt
01:20 favadi joined #salt
01:22 nZac joined #salt
01:22 cliluw joined #salt
01:23 hvn joined #salt
01:25 andrew_v joined #salt
01:26 amcorreia joined #salt
01:27 michelangelo joined #salt
01:34 catpigger joined #salt
01:35 armguy_ joined #salt
01:39 bowhunter joined #salt
01:39 racooper joined #salt
01:40 andrew_v joined #salt
01:41 flowstate joined #salt
01:47 ilbot3 joined #salt
01:47 Topic for #salt is now Welcome to #salt! | Latest Versions: 2015.8.8, 2016.3.0rc2 | SaltConf16: http://saltconf.com/register/ | Support: https://www.saltstack.com/support/ | Logs: http://irclog.perlgeek.de/salt/ | Paste: https://gist.github.com/ (please don't multiline paste into channel) | See also: #salt-devel, #salt-offtopic | Ask with patience as we are volunteers and may not have immediate answers
01:50 _armguy_ joined #salt
01:50 hasues joined #salt
01:50 hasues left #salt
01:52 teatime Were there people here interested in Perl6?  This is a great slidedeck I just came across:  https://speakerdeck.com/sidnet/perl-6-the-butterfly-has-landed
01:54 teatime honestly: ^
01:59 tharkun joined #salt
02:05 armguy_ joined #salt
02:07 hvn joined #salt
02:07 hvn joined #salt
02:08 mapu joined #salt
02:12 hightekvagabond joined #salt
02:12 sjmh kevinquinnyo1 : re: the expansion question ( and I think this is correct ) - the expansion is done at the minion level.  not the master, so there's not an internal function that's going to give you the list.
02:13 sjmh every minion gets every message, and each minion determines whether it needs to pay attention to that message ( by seeing if it matches the target )
02:14 sjmh manage.up is really just doing a test.ping and returning the list of minion's that responded.
02:16 josuebrunel joined #salt
02:21 bltmiller joined #salt
02:23 antpa joined #salt
02:30 josuebrunel joined #salt
02:34 barajasfab joined #salt
02:36 anthpa joined #salt
02:36 bltmiller joined #salt
02:38 cberndt joined #salt
02:40 hemebond sjmh: Which completely contradicts what the documentation says.
02:41 hemebond (unless I'm thinking of one of the other commands that shows you which minions are connected, supposedly without contacting the minion)
02:41 quasiben joined #salt
02:42 flowstate joined #salt
02:43 teryx510 joined #salt
02:47 bltmiller joined #salt
02:49 ageorgop1 joined #salt
02:50 evle joined #salt
02:50 _armguy_ joined #salt
02:51 hvn joined #salt
02:51 hvn joined #salt
02:59 favadi joined #salt
03:04 writtenoff joined #salt
03:05 edrocks joined #salt
03:07 TheoSLC joined #salt
03:11 sjmh hemebond : maybe, but the source from the manage runner is pretty clear.  up runs 'ret = status(output=False).get('up', [])'.  and status runs 'minions = client.cmd('*', 'test.ping', timeout=__opts__['timeout'])'
03:13 hemebond up doesn't mention that it won't contact the minion. I was thinking of the others.
03:13 sjmh list_state I think
03:14 Cogent joined #salt
03:17 teatime I beleive manage.up does ping stuff
03:17 teatime there's other stuf tho
03:17 teatime yeah
03:19 bowhunter joined #salt
03:20 sjmh ah looks like you can use the CkMinion stuff in salt.utils.minions to try and get a best effort of what minions should be responding
03:39 Cogent Hello salt enthusiasts! I'm playing around with saltstack, currently trying to get the dockerng module to work. It complains about not being imported. Is it something I have to install separately?
03:39 hemebond Cogent: Have you been through the tutorial and read the salt formula documentation?
03:40 Cogent To the best of my knowledge yes. Did I not RTFM enough?
03:40 hemebond I believe you :-)
03:41 hemebond We'd need more information though, like the full error pasted somewhere (not in here) so we can see it.
03:41 Cogent the exact error is "Reason: 'dockerng' __virtual__ returned False: 'dockerng' __virtual__ returned False: Docker module could not get imported"
03:41 hemebond Oh
03:41 Cogent upon googling it, it brings up logs of this IRC chat
03:41 hemebond You need to install the Python Docker module.
03:42 hemebond I don't know if it's required on the master or the minion (guessing the minion)
03:42 flowstate joined #salt
03:42 Cogent I surmised that as well, however the docs say nothing about it needing to be installed separately.
03:42 Cogent And which module it is specifically, and where I can install it from, (pip/yum/custom rpm etc...)
03:43 hemebond https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.dockerng.html#installation-prerequisites
03:44 Cogent :-( I did not RTFM. Sorry about that. I guess I'll put "put a link to install page on https://docs.saltstack.com/en/latest/ref/states/all/salt.states.dockerng.html" in the suggestion box
03:45 hemebond Sounds good. Maybe raise an issue on the formula you're using.
03:45 dandelo joined #salt
03:46 Cogent Thank you again for the help hemebond!
03:46 hemebond You're welcome :-) Good luck
03:50 armguy_ joined #salt
03:51 ageorgop joined #salt
03:51 cberndt joined #salt
03:53 anthpa I'm attempting to create an ssl installation state from the tls.create_ca module but receive this error: state 'tls.create_ca' was not found in SLS 'webserver.applications' Reason: 'tls.create_ca' is not available. https://gist.github.com/anthonyscolaro/1daac2d9aa4bb6c821b3c2abfec47816
03:54 anthpa Is there a way to create a state from this module?  Or better question, how can I create a state to deploy self signed certificates for our websites on staging?
03:54 hemebond anthpa: create_ca is an execution module function, not a part of the state module.
03:55 hemebond You can call execution modules from a state.
03:55 hemebond https://docs.saltstack.com/en/latest/ref/states/all/salt.states.module.html
04:00 anthpa Ok great.  I think I've almost got it set up.  Thank you for the quick reply.
04:01 hemebond :thumbsup:
04:01 source47 joined #salt
04:02 hal58th_ joined #salt
04:06 msn joined #salt
04:14 kevinquinnyo joined #salt
04:20 XenophonF joined #salt
04:26 tedbot joined #salt
04:27 tedbot requiring a package is "pkg:", and requiring multiple packages is "pkgs:"....so if requiring an sls is "sls:", how do I require multiple SLS files?
04:27 tedbot "slss:" ?
04:29 teatime no, I wouldn't think so.
04:29 tedbot do I just list separate "sls:"s?
04:29 tedbot ^ I'll try that
04:30 hemebond Are you sure it's pkgs?
04:30 tedbot yep
04:30 teatime tedbot: yeah just specify a list of requires, each one "  - sls: whatever"
04:31 tedbot teatime: awesome, thank you for responding so quickly
04:31 tedbot hemebond: you too, thanks
04:31 teatime package.installed has a pkgs argument, but that is not actually a requisite/requires
04:32 hemebond tedbot: Can you point me to documentation on that please?
04:32 hemebond Ah, that's what I thought.
04:32 Guest55503 joined #salt
04:32 tedbot hemebond: https://docs.saltstack.com/en/latest/ref/states/all/salt.states.pkg.html -- use your browser find to search for 'pkgs:'
04:33 hemebond My understanding was it's just "- module: stateid"
04:33 hemebond tedbot: I found that but it's not used in the require declaration anywhere.
04:34 hemebond That's just the pkgs argument to pkg.installed.
04:35 tedbot hemebond: I can tell you from experience, that using 'pkgs:' in a require statement works
04:35 ubaumann joined #salt
04:35 tedbot I just tried it on a hunch and it worked
04:35 tedbot Salt docs aren't the greatest all the time, hence this room
04:35 tedbot at least, from version 2015.5.3
04:36 hemebond It possibly does, but that's not what those docs say and that mislead you a bit on how require works.
04:36 hemebond I thought there was actual documentation on it, that's all.
04:37 tedbot ^ agreed....sounds like a contrib opp :)
04:37 tedbot though "slss:" looks weird
04:38 hemebond Looks like it's not a standard thing and just something someone threw in :-(
04:42 flowstate joined #salt
04:42 AndreasLutro fails when I try it
04:42 AndreasLutro are you sure you're actually running the state?
04:43 AndreasLutro I've worked on the code for requirements and would be shocked if you could just put "s" at the end of everything for the fun of it
04:43 hemebond lol
04:44 hemebond If it did work, I would expect it to have come from a request like "pkg is confusing when requiring many pkgs" or something.
04:45 AndreasLutro people just need to realize the "pkg: foo" in a require means a pkg.* state with the id/name "foo", not an actual package named "foo"
04:46 AndreasLutro it's probably in the docs, but it is less intuitive
04:47 hemebond There might also be a bit of flexibility there too. When using pkgs it looks like you can require one of the packages by name instead of the state it's listed under.
04:47 hemebond (just going by comments in a particular issue I just read)
04:47 brotatochip joined #salt
04:48 hemebond But it should be module_name: state_id
04:48 AndreasLutro that also sounds wrong
04:49 hemebond I wouldn't have expected it to work.
04:49 hemebond And I haven't tested it.
04:49 hemebond It might have been allowed due to a bug with - name.
04:50 hemebond Still, I think the document states clearly that it's the module name.
04:50 hemebond and the state ID
04:50 teatime maybe it's related to the aggregation that is mainly used by the pkg state(s) now
04:50 AndreasLutro that is a possibility
04:51 bowhunter joined #salt
04:57 av_ joined #salt
04:59 cpowell joined #salt
05:03 ivanjaros joined #salt
05:04 impi joined #salt
05:07 brianvdawson joined #salt
05:07 josuebrunel joined #salt
05:07 edrocks joined #salt
05:09 tedbot AndreasLutro: what fails when you try it? specifying pkgs in a require block?
05:11 source47 joined #salt
05:16 sauvin joined #salt
05:20 eightyeight joined #salt
05:21 graffic joined #salt
05:21 roock joined #salt
05:26 cyberviking joined #salt
05:31 felskrone joined #salt
05:33 favadi joined #salt
05:42 flowstate joined #salt
05:45 dmaiocchi joined #salt
05:45 evilrob joined #salt
05:49 jY joined #salt
05:51 rdas joined #salt
05:52 bryguy joined #salt
05:57 oravirt joined #salt
06:01 jhauser joined #salt
06:03 tedbot left #salt
06:04 dmaiocchi joined #salt
06:13 punkoivan joined #salt
06:14 neilf__ joined #salt
06:16 AndreasLutro tedbot: yes, as well as "slss"
06:18 hvn joined #salt
06:18 hvn joined #salt
06:20 djgerm joined #salt
06:26 ivanjaros3916 joined #salt
06:31 bdrung_work joined #salt
06:33 dyasny joined #salt
06:36 ribx joined #salt
06:39 rdas joined #salt
06:41 flowstate joined #salt
06:45 TyrfingMjolnir joined #salt
06:48 KermitTheFragger joined #salt
06:52 aanriot left #salt
06:53 josue joined #salt
06:54 kawa2014 joined #salt
06:56 teryx5101 joined #salt
07:00 cberndt joined #salt
07:04 TyrfingMjolnir joined #salt
07:04 josue joined #salt
07:07 msn joined #salt
07:08 null1 joined #salt
07:09 tawm04 joined #salt
07:10 edrocks joined #salt
07:10 null1 I am trying to understand the master's cache of minion data. When does any data gets cache in the /var/cache/salt/master/minion/minion_id/
07:12 source47 joined #salt
07:13 Miouge joined #salt
07:14 dyasny joined #salt
07:15 _mel_ Hi. i have a question. i like to do some complex logic to build a nginx config file based on the ip adress of the server. can i somehow easy use some kind of inline python code instead jinja to accomplish that?
07:16 iggy _mel_: yes
07:16 iggy https://docs.saltstack.com/en/latest/ref/renderers/
07:20 fredvd joined #salt
07:23 dgutu joined #salt
07:25 hax404 joined #salt
07:26 _mel_ iggy: and is there some documentation? i like to create a ngine port listening config file based on the ip adress
07:27 teatime _mel_: is it actually complicated?  usually people get by with jinja as much as they can.
07:29 _mel_ teatime: maybe there is a jinja way, but i don't know how. i need to create 4 lines of ip adresses from the interface lo and eth0 but not all of them. and to make it more complicated one config file per vhost
07:30 _mel_ my second thought was to created static files per server and use normal states
07:31 teatime you can combine approaches, and use files that re jinja templates heh
07:31 yuhlw joined #salt
07:31 teatime there are some useful module functions for getting specific ip addrs etc.
07:31 teatime you can filter by CIDR
07:34 _mel_ can i for example create a sample config and use some definition in the first line to create a substittution? this is how it should be: http://pastebin.com/QbZHkpF6
07:35 _mel_ i could replace the first key (vhost-IP Adress) with jinjy, but the second part (server ip) comes from the interface (last octet)
07:35 teatime you should put data that drives states in pillars (or grains)
07:36 _mel_ teatime: yes, the ip adress is in the ipv4 grain but i don't know how to access it with jinja (specially cut of the last octed from the ip)
07:36 teatime check these out as well https://docs.saltstack.com/en/latest/ref/modules/all/salt.modules.network.html#salt.modules.network.ipaddrs
07:36 teatime you can actually call any python method that exists on the object on a jinja value
07:36 teatime so you can use string methods like .split if you need to
07:37 _mel_ can you show me an exanple?
07:38 garphy joined #salt
07:38 _mel_ for example how to get the primary ip adress from network.ip_addrs
07:39 babilen What is the primary IP address?
07:40 _mel_ babilen: if you bind several adresses to a interface only one will be used for creating outgoing conections. this is what i call primary ip adress
07:40 teatime that's not a common definition
07:40 flowstate joined #salt
07:41 teatime there's really no such thing as a primary IP address
07:41 babilen _mel_: network.ip_addrs takes a cidr argument that lets you specify the network. That allows you to get exactly the address you want.
07:41 Rumbles joined #salt
07:41 teatime you can eliminate things like RFC1918 addrs and localhost addr, but then if you still have more than 1 addr you need more info to choose one
07:41 babilen exactly
07:41 babilen (hence my question)
07:41 teatime like knowing what network it should be in, or knowing some DNS name that maps to it
07:42 _mel_ can i reverse lookup the hostname?
07:42 babilen And then, if you know the network, you can pass it to network.ip_addrs and get a single address back
07:42 babilen _mel_: Don't you know the network?
07:43 _mel_ something like this? salt-call network.ip_addrs eth0 255.255.0.0
07:46 _mel_ i use a /16 network for the loadbalancing. the 3. octet is the representation of the  loadbalancer ip, the 4. octet is the representation of the server. so to map 10.207.0.131 -> server 1 (10.207.0.43) gets 10.207.131.43, server 2 (10.207.0.44) gets 10.207.131.44,
07:46 babilen _mel_: well, rather a cidr like 10.0.0.0/8
07:46 babilen So, then use the /16
07:46 devleb joined #salt
07:46 _mel_ babilen: this gives me a list of three adresses.
07:46 babilen Or whatever is appropriate for your network setup
07:46 favadi joined #salt
07:47 _mel_ what about "network.interface eth0" can i use this?
07:47 babilen _mel_: Well, then your network definition is too large (as you have multiple networks in that /16)
07:47 _mel_ babilen: yes, its per design for the loadbalancing
07:48 babilen Can't you trim it down to a /24 or /26 (or somesuch)
07:48 _mel_ not easy. i need to have both ways to access the server. via loadbalancer and direct (for monitoring)
07:49 brianfeister joined #salt
07:49 babilen And yes, you can get the address(es) on eth0 via grains (cf. grains.items), but are you sure that the address will always be on eth0 ? I've seen people create a lot of work for themselves because they hardcoded interfaces everywhere and then, suddenly, the order changed.
07:49 devleb Hey, just starting to get into salt. I use " salt-run manage.status --out json " to get the minion status but im missing information like "last seen" "down since" "up since" e.g ... any pro tips how to get infos like that :-) ?
07:51 Guest79 joined #salt
07:51 _mel_ babilen: i think the eth0 is fixed (at least on wheezy, i don't know how much pottering detroyed debian network interface naming in jessie yet ...)
07:52 babilen ...
07:53 babilen Sorry, not in the mood for systemd fud
07:53 _mel_ yeah, sorry :)
07:54 babilen fwiw, even if you run this within Xen the persistent naming is being ignored and interfaces are simply enumerated by order in the Xen config
07:54 _mel_ so i try to figure out if i can use jinja / python whatever to create the config file or just use static files per host
07:54 _mel_ babilen: in fact i'm using xen (via ganeti)
07:54 babilen so ..
07:55 anthpa I'm trying to call to a list of pillar data but something's not set up properly.  Could someone help? https://gist.github.com/anthonyscolaro/18f56094637d1e6e441ecf0781c7e9b0
07:57 lero joined #salt
07:59 anthpa the nginx formula doesn't see the list of pillar data though I'm able to see when I use pillar.ls or pillar.items.
08:02 Miouge joined #salt
08:02 ronnix joined #salt
08:05 geomacy joined #salt
08:10 beardedeagle joined #salt
08:14 source47 joined #salt
08:18 yuhlw joined #salt
08:20 s_kunk joined #salt
08:20 s_kunk joined #salt
08:22 SpX joined #salt
08:27 kshlm joined #salt
08:28 ravenx joined #salt
08:28 ravenx what is this error:   Warnings: 'fetch_tags' is an invalid keyword argument for 'git.latest'
08:29 ravenx i am running:  Salt: 2015.5.3
08:29 ravenx Warnings: 'fetch_tags' is an invalid keyword argument for 'git.latest'. If
08:29 ravenx you were trying to pass additional data to be used in a template
08:29 ravenx context, please populate 'context' with 'key: value' pairs. Your
08:29 ravenx approach will work until Salt Carbon is out. Please update your
08:29 ravenx state files.
08:30 AndreasLutro it is what it says
08:31 akhter joined #salt
08:32 ravenx hm
08:33 ravenx i only passed it true.
08:33 impi joined #salt
08:33 ravenx since that's what it says in the docs
08:34 AndreasLutro https://docs.saltstack.com/en/2015.5/ref/states/all/salt.states.git.html
08:34 AndreasLutro ctrl + f "fetch_tags" no results
08:35 ravenx ah
08:35 ravenx i didn't click the button on the side bar
08:35 ravenx turns out there are two versions.
08:35 ravenx my bad
08:35 keimlink joined #salt
08:36 MadHatter42 joined #salt
08:37 slav0nic joined #salt
08:41 flowstate joined #salt
08:42 ronnix joined #salt
08:43 Hydrosine joined #salt
08:50 impi_ joined #salt
08:54 dmaiocchi joined #salt
08:58 om joined #salt
09:01 cpowell joined #salt
09:04 ronnix joined #salt
09:12 edrocks joined #salt
09:17 ronnix_ joined #salt
09:19 antpa joined #salt
09:19 antpa I'm able to call to data stored in grain using this but when I move it from grain it pillar, it breaks: https://gist.github.com/anthonyscolaro/18f56094637d1e6e441ecf0781c7e9b0
09:22 AndreasLutro antpa: you don't use "grains" anywhere in that snippet
09:27 antpa AndreasLutro: In the first line, {%- set staging_server_data = pillar.get('staging_servers', {} ) %}, I changed grains to pillar and copied the grain data to the pillar file, domain_datas.sls
09:30 babilen antpa: What does "salt 'yourminion' pillar.get staging_server_data" return on the master?
09:34 antpa nothing, but 'salt-call pillar.get staging_servers' returns the data in that pillar file
09:35 garphy joined #salt
09:35 babilen Right, c&p mistake
09:35 SpX joined #salt
09:36 babilen It appears that you didn't paste the entire SLS or I'm too blind to find "{% for vhost, settings in nginx.vhosts.managed.items() %}" in the SLS
09:37 babilen prod/nginx/ng/vhosts_config.sls in particular seems to be missing from your paste
09:37 brianfeister joined #salt
09:38 LondonAppDev__ joined #salt
09:39 AirOnSkin joined #salt
09:39 nomad_fr_ joined #salt
09:40 AbyssOne_ joined #salt
09:40 antpa babilen: It's using the nginx-formula.
09:40 trave_ joined #salt
09:40 antpa babilen: https://github.com/saltstack-formulas/nginx-formula
09:40 egil_ joined #salt
09:41 ajolo_ joined #salt
09:41 flowstate joined #salt
09:41 antpa perhaps it's an issue of referencing the domains_data.sls (one pillar file) from nginx.forumula.init.sls (another pillar file)?
09:42 babilen antpa: Okay, so what does "salt-call pillar.get nginx" return?
09:42 N-Mi joined #salt
09:42 N-Mi joined #salt
09:42 TyrfingM1olnir joined #salt
09:42 ]V[_ joined #salt
09:43 dkrae1 joined #salt
09:43 murkey_ joined #salt
09:43 asoc_ joined #salt
09:43 saltsa joined #salt
09:43 arapaho_ joined #salt
09:43 om2 joined #salt
09:44 kaushal_ joined #salt
09:44 ramblinpeck_ joined #salt
09:44 antpa It returns everything in my pillar file up to the my the first line of my statements shown in the file excerpt I put in nginx.ng.pillar.init.sls
09:45 antpa babilen: I can paste the entire file there if you'd like to see it.
09:45 pprkut_ joined #salt
09:45 babilen antpa: And "salt-call pillar.get nginx:ng:vhosts:managed" ?
09:46 babilen Well, right now I'm trying to work with the information you provide ..
09:46 dayid_ joined #salt
09:46 babilen But I'm also off to lunch now. See you later!
09:46 antpa babilen: It returns None.  Because the vhosts data is enclosed in that for statement which is probably the culprit.
09:47 wych_ joined #salt
09:47 Vaelater1 joined #salt
09:47 babilen https://github.com/saltstack-formulas/nginx-formula/blob/master/pillar.example -- do you use this pillar structure?
09:47 antpa babilen: Enjoy.  I've updated gist to show the entire file.
09:47 babilen Can't tell without seeing your data
09:47 renoirb_ joined #salt
09:48 alexlist joined #salt
09:48 babilen antpa: aaaah!
09:48 antpa babilen: Yes, I'm pretty familiar with the pillar structure.  Again, it's working fine when I store the content of domans_data.sls in my /etc/minion/grains and change         {%- set staging_server_data = pillar.get('staging_servers', {} ) %}  to         {%- set staging_server_data = grains.get('staging_servers', {} ) %}
09:48 babilen Okay, that helps
09:48 crashmag joined #salt
09:48 babilen You can't reference pillars in pillars
09:48 ThomasJ joined #salt
09:48 antpa babilen: Ok, that's the issue then
09:49 babilen So "pillar.get('staging_servers', {} )" in there returns {}
09:49 babilen antpa: Sorry, it really wasn't clear what you did
09:49 geomacy joined #salt
09:49 antpa babilen: No worries.  I thought it might of been an issue but wasn't sure.
09:49 llua joined #salt
09:49 antpa babilen: I can always do better at explaining myself :)
09:50 babilen I wondered how the "{%- set staging_server_data = pillar.get('staging_s ..." bit is related to your error
09:50 netcho joined #salt
09:50 netcho hello everyone
09:50 antpa babilen: I just need to rethink how I can store this data in pillar instead of grain.
09:50 antonw joined #salt
09:51 netcho i'm testing salt-cloud on aws and provisioned 2 minion instances
09:52 netcho salt-cloud commands are working but salt ones are not
09:53 netcho i.e. salt-cloud --action=get_tags my_first_minion returns correct value but salt '*' test.ping says no reposnse
09:54 Nazca joined #salt
09:54 sagerdearia joined #salt
09:54 jab416171 joined #salt
09:54 whytewolf joined #salt
09:54 joshin joined #salt
09:54 joshin joined #salt
09:55 jfindlay joined #salt
09:55 gmoro joined #salt
09:56 netcho i provisioned them with salt-cloud --profile=base_ec2_private my_first_minion command
09:56 alvinstarr joined #salt
09:57 netcho keys are also accepted]
09:57 punkoivan joined #salt
09:57 TyrfingMjolnir joined #salt
09:59 netcho looks like minion does not know where the master is, also tried manual change of the minion config and put master ip address insteat of dns name, still nothing
09:59 teryx510 joined #salt
10:01 Hydrosine joined #salt
10:08 ronnix joined #salt
10:10 netcho anyone?
10:15 source47 joined #salt
10:22 Berty_ joined #salt
10:22 Berty_ left #salt
10:23 JohnnyRun joined #salt
10:26 antpa joined #salt
10:30 ravenx are your ports open?
10:31 sjmh_ joined #salt
10:32 atmosx_alt joined #salt
10:34 asoc joined #salt
10:34 atmosx joined #salt
10:34 whitenoise joined #salt
10:34 Miouge joined #salt
10:36 babilen ravenx: netcho left
10:36 s_kunk joined #salt
10:36 s_kunk joined #salt
10:36 Guest79 joined #salt
10:36 netcho joined #salt
10:36 APLU joined #salt
10:38 viq joined #salt
10:38 Dev0n joined #salt
10:38 dgutu joined #salt
10:39 kows joined #salt
10:40 ravenx babilen: netcho?
10:40 ravenx oh
10:40 netcho ?
10:40 ravenx netcho: i just asked if your ports were opened
10:40 flowstate joined #salt
10:40 netcho ravenx: what ports? 22 is open only from master instance
10:41 ravenx right but for your minion to connect to master
10:41 fxhp joined #salt
10:41 ravenx i think it needs 4505 and 4506
10:41 netcho on master?
10:41 ravenx yes
10:42 ravenx though if your keys are accepted, i think it is opened already
10:42 ravenx some one more experienced than i can comment on that.
10:43 cro joined #salt
10:45 mowntan joined #salt
10:45 netcho ravenx: i opened all ports just for test and still nothing
10:45 TomJepp joined #salt
10:46 bantone joined #salt
10:46 rhand joined #salt
10:47 alvinstarr joined #salt
10:47 impi joined #salt
10:49 JohnnyRun joined #salt
10:49 dh__ joined #salt
10:50 AndreasLutro netcho: have you checked your minion logs for clues?
10:53 kevops joined #salt
10:54 netcho AndreasLutro: this is strange, looks like there is no minion istalled at all, wtf
10:55 ronnix joined #salt
10:56 teryx510 joined #salt
10:56 bdrung_work joined #salt
11:00 notnotpeter joined #salt
11:02 cpowell joined #salt
11:03 oravirt joined #salt
11:03 asyncsrc1 joined #salt
11:04 ivanjaros joined #salt
11:06 eightyeight joined #salt
11:10 amcorreia joined #salt
11:14 edrocks joined #salt
11:17 sagerdearia joined #salt
11:20 MikaT joined #salt
11:21 quasiben joined #salt
11:22 teryx510 joined #salt
11:25 netcho ok, so i created a new minion and added master ip to hosts file and now it works, why doesn't it work when i change it in minion conf? i
11:26 brianfeister joined #salt
11:27 LondonAppDev joined #salt
11:27 netcho shouldn't it know on bootstrap? i mean the ip of salt-master it was bootsrapped from?
11:28 dh joined #salt
11:30 Hydrosine joined #salt
11:31 renoirb joined #salt
11:32 thehaven_ joined #salt
11:34 watersoul_ joined #salt
11:35 inre joined #salt
11:35 jhujhiti_ joined #salt
11:36 pmcg_ joined #salt
11:36 preachermanx joined #salt
11:36 whatevsz_ joined #salt
11:37 lorengordon joined #salt
11:37 teryx510 joined #salt
11:37 Antiarc joined #salt
11:37 khorben joined #salt
11:37 Ludo- joined #salt
11:37 vilitux joined #salt
11:37 Hipikat joined #salt
11:37 cwright joined #salt
11:37 honestly joined #salt
11:38 mTeK joined #salt
11:38 jfred joined #salt
11:38 diegows joined #salt
11:38 Pie_Mage joined #salt
11:38 jcastle joined #salt
11:38 Garo_ joined #salt
11:38 muep_ joined #salt
11:41 g3cko joined #salt
11:41 emid joined #salt
11:41 foundatron joined #salt
11:42 Vaelater1 joined #salt
11:42 kbaikov joined #salt
11:43 marcinkuzminski joined #salt
11:44 elektrix joined #salt
11:45 seblu joined #salt
11:46 kaictl joined #salt
11:49 ry joined #salt
11:49 oida joined #salt
11:49 mdupont joined #salt
11:50 sauvin joined #salt
11:50 danemacmillan joined #salt
11:52 joshin joined #salt
11:52 joshin joined #salt
11:53 jamesp9 joined #salt
11:57 jamesp9 When creating and PostgreSQL user with "postgres_user.present" do I use "user" or "db_user" to specify the postgres root account name
11:58 kawa2014 joined #salt
11:58 ravenx is it possible to run salt modules as a different user?
11:58 AndreasLutro jamesp9: "user" should be the system user the postgres daemon is running under - usually "postgres"
11:58 ravenx such as supervisord.status
11:59 AndreasLutro jamesp9: the "name" arg is what the actual database user should be named, not sure what "db_user" is
12:00 jamesp9 AndreasLutro: I should mention I'm attempting to create db user on an AWS RDS Postgres instance
12:00 jamesp9 in this case the system username is set as "root"
12:01 AndreasLutro oh, no experience with that
12:05 jamesp9 AndreasLutro, I get a "CommandExecutionError: User 'postgres' is not available" error
12:06 jamesp9 so it seems it's trying to user "postgres" user despite setting "user" in the state
12:07 AndreasLutro jamesp9: can you get a stack trace?
12:07 cliluw joined #salt
12:07 jamesp9 what's the preferred pastebin to around here ?
12:09 evle joined #salt
12:11 ronnix joined #salt
12:11 AndreasLutro anything that's not pastebin
12:11 jamesp9 haha
12:11 jamesp9 sorry I don't use them very often, not sure what around
12:11 AndreasLutro gist.github.com, bpaste, dpaste
12:12 ravenx can anyone help me with my module and non root user problem?
12:13 ravenx is it possible to run salt modules as a different user?
12:13 ravenx such as supervisord.status
12:13 ravenx cuz i have my salt-minion running as root, and the sudo_user as a non priv user
12:13 AndreasLutro ravenx: no, you rely on the module function having a user argument
12:13 ravenx but i can't run supervisord.status from my command line @ my master server.
12:13 ravenx AndreasLutro: i can pass it a user argument?
12:13 AndreasLutro I don't know, why don't you check?
12:13 ravenx i am
12:13 ravenx i'm checking on irc
12:14 AndreasLutro check the docs first :)
12:14 ravenx aah
12:14 ravenx oaky
12:14 jamesp9 AndreasLutro, https://gist.github.com/anonymous/560b7a817eb14bb296f969fd41b28c3b
12:14 teryx510 joined #salt
12:15 AndreasLutro jamesp9: which salt version are you on?
12:17 source47 joined #salt
12:17 AndreasLutro jamesp9: nevermind, I'll assume 2015.8.8.2. try the "runas" arg
12:18 jamesp9 yes 2015.8.8.2
12:20 CeBe joined #salt
12:20 jamesp9 hmm that gives me 'runas' is an invalid keyword argument for 'postgres_user.present'
12:21 AndreasLutro hrmph right
12:21 AndreasLutro okay
12:21 AndreasLutro remove the runas and user args and specify postgres as db_user then I guess
12:22 west575 joined #salt
12:22 AndreasLutro "user" gets passed to cmd.run as "runas", "db_user" gets passed to the postgres command as --username {db_user} as far as I can tell
12:23 jamesp9 yeah it's confusing
12:24 geomacy joined #salt
12:25 akhter joined #salt
12:26 jamesp9 I'm gonna call it a night and take a look in the morn again
12:26 jamesp9 AndreasLutro, thanks for your assistance
12:26 ravenx i have been getting this:  IOError: [Errno 13] Permission denied: '/var/log/salt/minion'
12:26 ravenx but my salt is running as root, only the sudo_user has changed.
12:26 ronnix joined #salt
12:27 hoonetorg joined #salt
12:27 teryx510 joined #salt
12:27 ravenx does the sudo_user need access to that file
12:27 ravenx or just the root
12:29 TooLmaN joined #salt
12:30 Rumbles joined #salt
12:32 nZac joined #salt
12:35 dmaiocchi_ joined #salt
12:35 dmaiocchi__ joined #salt
12:44 ronnix joined #salt
12:45 nZac joined #salt
12:46 numkem joined #salt
12:50 gh34 joined #salt
12:52 bowhunter joined #salt
12:54 WKNiGHT joined #salt
12:54 Fabbe joined #salt
12:57 catpig joined #salt
12:57 subsignal joined #salt
13:01 bbradley joined #salt
13:01 dyasny joined #salt
13:01 toastedpenguin joined #salt
13:02 cyborg-one joined #salt
13:02 edrocks joined #salt
13:02 quix joined #salt
13:03 bbradley joined #salt
13:03 cpowell joined #salt
13:07 hoonetorg joined #salt
13:07 cpowell joined #salt
13:07 edrocks joined #salt
13:12 xmj left #salt
13:15 brianfeister joined #salt
13:16 josue joined #salt
13:18 nZac joined #salt
13:23 josue joined #salt
13:27 jhauser joined #salt
13:27 subsigna_ joined #salt
13:28 mapu joined #salt
13:33 traph joined #salt
13:33 dynamicudpate joined #salt
13:34 Arendtse1 joined #salt
13:34 jgelens_ joined #salt
13:35 teryx5101 joined #salt
13:35 AbyssOne joined #salt
13:35 sjmh joined #salt
13:36 gnord_ joined #salt
13:36 murkey joined #salt
13:36 invalidexceptio- joined #salt
13:38 the_lale1u joined #salt
13:38 MTecknol1gy joined #salt
13:38 honestly_ joined #salt
13:38 diegows_ joined #salt
13:38 guanophobic joined #salt
13:38 nahamu_ joined #salt
13:39 hunmaat joined #salt
13:39 hunmaat joined #salt
13:39 iggy__ joined #salt
13:40 Taytay joined #salt
13:40 muep__ joined #salt
13:40 jfroot joined #salt
13:41 thallera1 hi would it be possible to add session only config options to mysql_user and mysql_grants. we have problems with replicating users on our mysql servers and would manually run SET sql_log_bin = 0; before adding users which is cumbersom
13:42 stooj_ joined #salt
13:43 scc_ joined #salt
13:43 manfred joined #salt
13:43 Vaelatern joined #salt
13:44 wych joined #salt
13:44 garphy` joined #salt
13:48 Gareth joined #salt
13:48 flowstat_ joined #salt
13:49 scoates joined #salt
13:49 tkharju joined #salt
13:50 NaPs joined #salt
13:50 SheetiS joined #salt
13:50 mavhq joined #salt
13:53 kaptk2 joined #salt
13:56 Muchoz joined #salt
13:59 Rumbles joined #salt
13:59 hblz joined #salt
13:59 jerredbell joined #salt
14:01 shiriru joined #salt
14:02 dmaiocchi joined #salt
14:02 _JZ_ joined #salt
14:02 hblz left #salt
14:05 feld joined #salt
14:06 armguy joined #salt
14:08 wych joined #salt
14:08 hasues joined #salt
14:09 hasues left #salt
14:09 dmaiocchi joined #salt
14:14 flowstate joined #salt
14:15 mohae joined #salt
14:16 guerby joined #salt
14:17 jscott joined #salt
14:17 tharkun joined #salt
14:17 Rumbles joined #salt
14:18 garphy joined #salt
14:18 ZiLi0n joined #salt
14:18 source47 joined #salt
14:20 ZiLi0n Hello everyone, I have recently started to use salt-call as masterless minion. I am wondering how I can use the pillar, is there a way to specify to the salt-call command a directory or sls file which contains all the pillar data?
14:21 teatime joined #salt
14:22 gtmanfred you just use the same pillar_root stuff that you would with mastered minion i believe
14:25 garphy joined #salt
14:25 kus joined #salt
14:29 ravenx hi, when i put "user" for my salt-minion to run as an unpriv user
14:29 ravenx and start the salt-minion daemon as root
14:29 ravenx it still shows that it is running as root, an idea/
14:29 ravenx any ideas?
14:33 kevinquinnyo joined #salt
14:34 ZiLi0n gtmanfred thanks, I don't have a top.sls though for pillar, I am not using one to map states to minions, instead I am using state.sls and the sls file directly (excellent suggestion from this channel :)). I was wondering if the same thing can be done for pillar, no top.sls to match pillar data to minions, and specifiy a file instead
14:35 west575 joined #salt
14:35 armguy_ joined #salt
14:36 jhujhiti joined #salt
14:39 rubenb joined #salt
14:39 racooper joined #salt
14:39 honestly joined #salt
14:39 MTecknology joined #salt
14:42 __number5__ joined #salt
14:43 garphy joined #salt
14:43 debian112 joined #salt
14:45 geomacy joined #salt
14:45 TheoSLC joined #salt
14:48 ravenx anyone here?
14:49 guerby joined #salt
14:51 ZiLi0n ravenx sorry I can't help much, I haven't played with permissions and minion enough
14:51 ravenx ZiLi0n: darn. that's alright thanks anyways.
14:52 tkharju joined #salt
14:53 babilen ravenx: Not saying that you can't do it, but I'd consider it to be rather unusual to run salt-minion as an unprivileged user.
14:53 babilen I mean, in the end, you run everything as root anyway.
14:53 ZiLi0n gtmanfred I have created a top.sls file under pillar_root, and now salt-call can find the pillar data. It is good, I was just wondering if there was a way to input a data pillar file right into salt-call avoiding the top.sls mapping
14:54 teatime babilen: I could see a use-case where you want to use salt to manage only things that are available to a specific unprivileged user, and limit your loses in case it gets compromised.
14:54 babilen Sure
14:55 teatime but for general config mgmt, yeah, kinda gotta be root.
14:55 teatime ZiLi0n: you can specify it on the command-line.
14:55 toddnni joined #salt
14:55 teatime ZiLi0n: but if it's more than a couple of values, that gets unmanagable obviously.
14:56 ZiLi0n teatime, right, I was looking for to input a file
14:56 ravenx babilen: i do want to.  but my sysadmin discovered that on the master you can do thsi:
14:56 ravenx salt '*' cmd.run 'cat /etc/shadow'
14:56 ravenx and he wasn't happy
14:56 ravenx babilen: this `salt-user` account whihc hosts the salt-master is going to be given to devs so they can deploy their own stuff.
14:57 Obrit joined #salt
14:57 babilen ravenx: https://docs.saltstack.com/en/latest/ref/clientacl.html
14:57 * teatime sighs @ your sysadmin having to 'discover' that the privlileged config mgmt tool gives arbitrary system access.
14:57 babilen ravenx: I mean .. if you want to use something that can do everything and are then surprised that you can do everything :)
14:58 babilen teatime: exactly
14:58 ravenx :(
14:59 quix joined #salt
14:59 ravenx i'm trying to make use of the user:  and the sudo_user: fields in the minion file in /etc/salt
14:59 ravenx been struggling quite a bit
14:59 teatime there's a doc somewhere in the docs about running salt as unpriv user
14:59 teatime it's mostly about what log etc. files it needs permissions for
14:59 ravenx https://docs.saltstack.com/en/latest/ref/configuration/nonroot.html
14:59 ravenx this one
14:59 ravenx but it doesn't help me with my cause
15:00 ravenx cuz either my minions time out, or the minion still gets started as root.
15:00 teatime also, forgive me if this is obvious, but if you give the minion *full* sudo access, you haven't fixed the original problem.
15:01 teatime you'll have to lock down specific commands or whatever that the minion can run via sudo, if it needs sudo.
15:01 ravenx i gave my salt minion an account, under sudo_user, which has very limited ability.
15:01 ravenx that account can only do things within it's own realm of /home/special-user
15:01 ravenx i have checked that the master cannot:  'salt '*' cmd.run 'cat /etc/shadow' '
15:02 ravenx the problem i am having is that, everytime i restart the salt-minion, my /etc/salt/, /var/cache/salt/, /var/log/salt, /and /var/run/salt gets resetted to root owner
15:02 ravenx and i gotta chown everything again
15:02 teatime as for why 'user:' gets ignored, I dunno; sorry, have not tried to use it myself.
15:02 ravenx yeah, that is something i haven't figured out.
15:03 ravenx technicaly, if, after setting that field, and starting it as root, it should register in ps aux | grrep as the user i specified
15:03 teatime you could also though be using the init system to actually start the process as the right user.  if it doesn't need root access at the beginning (for something like port binding like a web server), that would be better anyway.
15:03 teatime ravenx: not sure of that either; not necessarily, it depends how it's designed.
15:04 brianfeister joined #salt
15:04 teatime but, probably.
15:04 gtmanfred ZiLi0n: it cannot, you do need a top.sls file for your pillars, the only other way to do it would be to pass pillar='{}' to the state.sls (which you can also use state.apply instead)
15:04 ravenx teatime: that's a good idea.
15:04 ravenx the init file, i mean.
15:05 spuder joined #salt
15:05 garphy joined #salt
15:05 _armguy_ joined #salt
15:07 nZac joined #salt
15:08 DammitJim joined #salt
15:09 om3 joined #salt
15:10 Vaelatern joined #salt
15:11 ravenx oh god it's an upstart script
15:11 salty joined #salt
15:11 teatime I beleive that would only be used on an upstart system.
15:11 ravenx i'm on ubuntu
15:12 ronnix joined #salt
15:12 ravenx but most of our things are done via service
15:12 salty i want to install upath
15:12 salty upath:   npm.installed:     - require:       - pkg: npm
15:12 salty Hey, how do i install an npm package through sls?
15:12 onlyanegg joined #salt
15:12 Miouge joined #salt
15:15 msheiny joined #salt
15:17 pfallenop joined #salt
15:18 fredvd joined #salt
15:20 bluenemo joined #salt
15:20 armguy_ joined #salt
15:21 salty_solution joined #salt
15:21 salty_solution Hello all. :)
15:25 PeterO joined #salt
15:25 brotatochip joined #salt
15:29 ajw0100 joined #salt
15:30 kwpk joined #salt
15:30 source47 joined #salt
15:31 Ahlee Neat. EVery day we generate on average 91,876 JIDs, which results in on average 9,140,968 state runs
15:31 Ahlee just 30 day average, tis all i keep
15:31 Ahlee shall need to start graphing these
15:32 teatime I love graphs
15:32 kwpk Hello - I am running a masterless minion with salt configuration files and pillar data. Is it possible to override the pillar data with salt-call much like you do with salt? I have tried this: salt-call --local apply.state pillar={ "foo":"bar" } but it doesnt override the pillar data
15:32 salty I have an issue with npm salt state. im installing a package "upath" and it tells me on the master that its installed but it doesn't on the minions
15:32 salty its like this with every npm package
15:33 salty i need to run npm install "package" on the minions...
15:33 kwpk apply.state == state.apply)
15:35 minot joined #salt
15:35 _armguy_ joined #salt
15:36 jerredbell1 joined #salt
15:36 cpowell_ joined #salt
15:37 Edgan_ joined #salt
15:37 ronnix joined #salt
15:38 Antiarc_ joined #salt
15:38 colttt_ joined #salt
15:38 R0N_ joined #salt
15:39 diegows joined #salt
15:39 duckfez_ joined #salt
15:39 MikaT_ joined #salt
15:40 newjersey joined #salt
15:41 guanopho1ic joined #salt
15:43 doglike joined #salt
15:43 doglike joined #salt
15:43 dandelo_ joined #salt
15:43 aarontc_ joined #salt
15:43 scc joined #salt
15:44 rodr1c_ joined #salt
15:44 dandelo joined #salt
15:44 evilrob_ joined #salt
15:46 Rumbles joined #salt
15:46 flowstat_ joined #salt
15:47 terratoma joined #salt
15:49 akhter joined #salt
15:49 workthrick joined #salt
15:50 trave joined #salt
15:50 minesskyline joined #salt
15:50 quasiben1 joined #salt
15:50 quasiben2 joined #salt
15:51 kawa2014 joined #salt
15:52 andrew_v joined #salt
15:52 VSpike Just trying to start using the PHP formula, with the ng states. I notice my opcache.ini file has a bunch of settings in it, and the formula doesn't seem to manage this file AFAIK
15:53 armguy_ joined #salt
15:53 VSpike What's my best way to get these settings in?
15:53 quasiben joined #salt
15:53 sauvin joined #salt
15:53 racooper joined #salt
15:53 cyberviking joined #salt
15:54 UtahDave joined #salt
15:54 Ch3LL_ Vspike: you can use a file.managed state to manage the file
15:56 Fabbe joined #salt
15:56 VSpike Ch3LL_: Ok, ta. Was just checking there wasnt some clever way in the formula to inject extra generic settings to any file
15:57 nidr0x joined #salt
16:01 Ch3LL_ VSpike: i'm not familiar with the formula itself, but if opcache.ini is a typical file to manage you could submit a PR to the formula to improve it
16:01 kwpk Is specifying "pillar" on the command line for salt-call supported? This is for a masterless minion
16:01 teatime it's rare for a formula to do exactly all the things you want the way you want
16:02 teatime forking them is the norm
16:02 teatime every CM system has this issue
16:05 Ch3LL_ kwpk: yes i just tested this: sudo salt-call state.sls test pillar="{'test': 'test-salt-call'}" and it worked
16:05 armguy_ joined #salt
16:05 _armguy_ joined #salt
16:07 bowhunter joined #salt
16:08 mpanetta joined #salt
16:09 dfinn joined #salt
16:09 beardedeagle joined #salt
16:10 ronnix_ joined #salt
16:12 aw110f joined #salt
16:13 onlyanegg joined #salt
16:14 akhter joined #salt
16:15 alemeno22 joined #salt
16:17 kwpk Ch3LL: thanks! I had my ' and " swapped - I changed this, however no such luck. Here is what I am running with verbatim: salt-call --local state.apply pillar="{ 'qbVars': { 'user':'foo','fullName':'User' } }"
16:18 writtenoff joined #salt
16:18 grumm_servire joined #salt
16:18 kwpk Ch3LL: I have nested params - from what I read online, that should be the way to set them
16:20 armguy_ joined #salt
16:21 aw110f_ joined #salt
16:22 quix joined #salt
16:24 ageorgop joined #salt
16:30 Ch3LL_ kwpk: just tested with this: sudo salt-call state.sls test pillar="{'test': { 'test1': 'test-salt-call1', 'test2': 'test-salt-call2'}}" and it worked.
16:30 kevinquinnyo so the reactor system has no way of knowing what the event tag was?  it just receives that event 'data' dict?
16:30 Ch3LL_ kwpk: what version of salt are you on
16:30 jab416171 after issuing 'pkg.install salt' from the master on cent7 minions, and then running: salt -C "not G@saltversion:2015.8.8.2" test.ping, the cent7 minions are still showing up. they only go away after manually restarting the salt-minion daemon. This is not true for our cent6 minions. Is this a known issue? Should I file a bug?
16:31 kevinquinnyo for instance if i have a reactor, i can't do something slightly different depending on what the event tag that was fired?
16:32 toastedpenguin joined #salt
16:34 tharkun joined #salt
16:34 source47 joined #salt
16:35 crashmag joined #salt
16:35 armguy_ joined #salt
16:36 systeem joined #salt
16:36 renoirb joined #salt
16:38 ZiLi0n gtmanfred thanks!
16:39 heaje joined #salt
16:42 source47 joined #salt
16:42 andrew_v joined #salt
16:43 hightekvagabond joined #salt
16:43 jtand joined #salt
16:43 edrocks joined #salt
16:45 LondonAppDev joined #salt
16:46 beardoLU__ /nick beardo
16:46 beardoLU__ left #salt
16:46 beardo joined #salt
16:49 UtahDave kevinquinnyo: there's a 'tag' variable available, too
16:53 brianfeister joined #salt
16:53 kevinquinnyo UtahDave: oh perfect!
16:53 UtahDave :)
16:53 kevinquinnyo that wil help me keep my reactors DRY, so they can do different, related things depending on the event
16:56 andrew_v joined #salt
16:57 impi joined #salt
17:04 brianvdawson joined #salt
17:10 jfroot joined #salt
17:14 hal58th joined #salt
17:20 _armguy_ joined #salt
17:27 kevinquinnyo can a pillar be an alias to another pillar?
17:28 kevinquinnyo i know that's weird, but i use an ext_pillar that pulls content from an internal rest api, and i prefer the format it supplies, but a module i'm using prefers it as a different name, and it's hardcoded in the module
17:29 djgerm left #salt
17:30 akhter joined #salt
17:31 UtahDave kevinquinnyo: no, not yet, unfortunately
17:31 iggy__ you can use jinja includes or includes
17:32 kevinquinnyo ok thanks
17:32 UtahDave true, but you can't just point to an existing pillar variable
17:32 kevinquinnyo i might look into that iggy, but for now i just created the alias in the rest api since i have control over that
17:32 iggy__ https://docs.saltstack.com/en/latest/topics/pillar/#including-other-pillars
17:32 akhter joined #salt
17:32 iggy__ you can set a key on the include
17:33 iggy__ hmm, not sure how that'll work with file_pillar vs ext_pillar
17:33 iggy__ if you find out, let us know ;)
17:34 jab416171 hmm, did anyone see my message?
17:34 iggy__ everybody uses ubuntu
17:35 teatime I'm becoming increasingly annoyed with Jinja
17:35 jab416171 haha
17:35 armguy_ joined #salt
17:35 UtahDave teatime: I always suggest minimizing the amount of jinja
17:36 jab416171 moar jinja!
17:36 jab416171 after issuing 'pkg.install salt' from the master on cent7 minions, and then running: salt -C "not G@saltversion:2015.8.8.2" test.ping, the cent7 minions are still showing up. they only go away after manually restarting the salt-minion daemon. This is not true for our cent6 minions. Is this a known issue? Should I file a bug?
17:36 teatime for one, some of its design decisions are more appropriate for web templating than salt
17:36 UtahDave jab416171: I just read your message.
17:36 jab416171 all the minions were on 2015.8.7
17:36 jab416171 and the masters
17:37 UtahDave it's a design decision of Centos7.  packages don't restart themselves, generally
17:37 teatime for two, it seems to take an opinionated stance in a couple of cases and not include functionality that would be easy to implement.. actually I think that's just an example of my former statement
17:37 UtahDave on upgrades
17:37 jab416171 that seems
17:37 jab416171 odd
17:37 flowstate joined #salt
17:37 iggy__ teatime: you don't have to use it
17:37 UtahDave that's how the redhat world works.  Install software, but it doesn't start up by default.
17:37 iggy__ teatime: a lot of people prefer mako (or just straight python) renderers
17:37 teatime iggy__: practically speaking, I do.
17:37 jab416171 sure, but if it's already running, why shouldn't it restart to get the new version?
17:38 teatime because it's still easier / clearer / less work than not
17:38 teatime although I could try mako.  it seemed to be giving beardedeagle fits, perhaps due to having less attention paid to it in salt than jinja gets.
17:38 onlyanegg joined #salt
17:38 iggy jab416171: because you're expected to restart services on your own during your maintenance window
17:39 beardedeagle tis true
17:39 beardedeagle no end of issues
17:39 iggy (I'm guessing anyway)
17:39 teatime iggy: you wouldn't upgrade a package well in advance of the maintenece window, and then restart the service within it.
17:39 teatime not in 98% of cases anyway.
17:39 jab416171 so a followup command could look something like:  salt -C "G@osmajorrelease:7 and not G@saltversion:2015.8.8.2" service.restart salt-minion
17:39 iggy I would
17:39 teatime it's a recipie for disaster.
17:39 iggy we do
17:40 jab416171 iggy, what if the machine gets rebooted or someone restarts the service ahead of the window?
17:40 teatime beardedeagle: so you would not personally suggest that path, then?  :)
17:40 iggy case in point... the recent libc vuln... we applied updates immediately, and restarted non-production impacting services right away... waited for maint window to restart production impacting services
17:41 beardedeagle I would suggest as many people use mako as possible. that way it can get more attention
17:41 beardedeagle you can't fix things by not using them
17:41 iggy jab416171: then the service get's restarted? I'm not sure what you're saying the problem with that would be
17:45 disbound joined #salt
17:48 jab416171 you have a lack of parity at that point
17:48 jab416171 maybe the two versions don't interoperate
17:48 jab416171 or something
17:48 jab416171 it's a counterargument to upgrading before the window
17:49 teatime I just would not want to run into the executable looking for runtime data that has changed or moved, or lazy-loading shared libs that are now not availble (because replaced w/ differentv ersion) etc.
17:49 jab416171 yeah
17:49 punkoivan joined #salt
17:49 flowstate joined #salt
17:49 jab416171 ever do a "yum update" and then lsof | grep deleted?
17:50 teatime I could only do that kind of upgrade after a bunch of checks / validation of assumptions, which would probably take longer than just waiting for the window.
17:50 flowstate joined #salt
17:50 DammitJim anyone dealing with badlock and samba?
17:50 punkoivan left #salt
17:54 iggy all my samba servers are exposed to untrusted networks
17:54 UtahDave ouch
17:56 onlyanegg joined #salt
17:56 morissette joined #salt
17:56 ronp_usa joined #salt
18:01 iggy there was a blog post a while back about heartbleed or one of those... couple of salt one-liners to take care of it
18:02 iggy also, if I ever meet the people that keep naming these damned vulns, I'll drag them into an alley and shank them and make it look like a robbery
18:05 nZac joined #salt
18:05 beardedeagle lol
18:05 brianvdawson left #salt
18:05 armguy_ joined #salt
18:06 mavhq why? easy to remember
18:07 DammitJim LOL
18:07 impi joined #salt
18:07 DammitJim iggy, you don't have samba servers exposed to untrusted networks!
18:08 edrocks joined #salt
18:08 geomacy joined #salt
18:09 DammitJim iggy, are you by any chance running ubuntu?
18:10 notnotpeter joined #salt
18:10 XenophonF joined #salt
18:16 barajasfab joined #salt
18:16 baweaver joined #salt
18:19 lemur joined #salt
18:20 nZac joined #salt
18:20 _armguy_ joined #salt
18:21 edrocks joined #salt
18:21 honestly iggy just doesn't trust anyone on his network
18:22 iggy I don't actually have any samba servers
18:22 iggy but yes, I run Ubuntu
18:22 hasues joined #salt
18:22 hasues left #salt
18:22 honestly iggy has set his network up so every packet comes by his desk and if he doesn't confirm it it gets dropped
18:23 iggy I just route everything through so many hops that most people just give up
18:24 s_kunk joined #salt
18:24 s_kunk joined #salt
18:26 M-liberdiko joined #salt
18:31 honestly The maximum TTL in an IP packet is 1024 isn't it
18:32 honestly oh it's only 255
18:36 sjmh_ joined #salt
18:36 tercenya_ joined #salt
18:37 punkoivan joined #salt
18:37 edrocks joined #salt
18:37 voileux_ joined #salt
18:38 gtmanfred ZiLi0n: no problem :)
18:39 pmcg joined #salt
18:40 Miouge joined #salt
18:40 ZiLi0n hello, I am trying to download a file using http, using a masterless minion. I have been trying to use file.manage, but I am getting error : content-lenght too long... yeah, it is a 2GB file... any suggestions?
18:40 ]V[ joined #salt
18:40 murkey joined #salt
18:40 honestly ZiLi0n: cmd.run wget?
18:40 sk_0 joined #salt
18:40 R0N joined #salt
18:41 eseyman joined #salt
18:41 armguy_ joined #salt
18:41 eliasp joined #salt
18:41 yidhra joined #salt
18:42 brianfeister joined #salt
18:42 ZiLi0n honestly yeah I guess I can do that, looking for a way to tell wget not to download a file it already exists or only if it newer
18:42 munhitsu_ joined #salt
18:42 sarlalian joined #salt
18:43 honestly does the server actually tell you how new the file is?
18:43 honestly I'm sure there's a way to do this with curl or wget
18:44 honestly http://blog.yjl.im/2012/03/downloading-only-when-modified-using.html
18:44 Dev0n joined #salt
18:44 samed joined #salt
18:44 samed joined #salt
18:44 scc joined #salt
18:44 nikogonzo joined #salt
18:45 foundatron joined #salt
18:45 devtea joined #salt
18:46 kevinquinnyo honestly: is the server you are pulling the file from setting that header?
18:46 hillna joined #salt
18:46 khorben joined #salt
18:47 honestly 20:43:14 < honestly> does the server actually tell you how new the file is?
18:47 honestly way ahead of you
18:47 kaictl joined #salt
18:47 kevinquinnyo heh
18:47 honestly most servers should respect If-Modified-Since if it's not explicitly disabled
18:47 Bucciarati joined #salt
18:47 honestly I'd be very surprised if that wasn't on by default in apache or nginx
18:48 Vivek joined #salt
18:48 kevinquinnyo i dont think it is
18:48 kows joined #salt
18:48 czchen joined #salt
18:48 kevinquinnyo do curl -I whatever-server.com/whatever-file.html
18:48 kevinquinnyo and see if the header is being set
18:49 kevinquinnyo i imagine the server has to stat() the file on every request, and snag the file's mtime, so it's probably a small performance hit and not on by default
18:49 kevinquinnyo but i dont know
18:50 lkannan joined #salt
18:50 honestly well you can't even use sendfile without having an fd to the file
18:50 honestly so you're going to have to do that anyway...
18:51 teatime plus caching
18:51 eightyeight joined #salt
18:51 kevinquinnyo ohh wait that's a client-side http header... i get it
18:52 arcleo_ joined #salt
18:52 Fiber^ joined #salt
18:52 Qwazerty joined #salt
18:52 monokrome joined #salt
18:52 ecdhe joined #salt
18:53 honestly and it's not like http servers tend to be tuned for maximum performance in the default config anyway :P
18:53 egil_ joined #salt
18:54 oida joined #salt
18:54 FreeSpencer joined #salt
18:54 FreeSpencer joined #salt
18:54 nobrak joined #salt
18:54 nobrak joined #salt
18:55 JohnnyRun joined #salt
18:56 M-MadsRC joined #salt
18:58 geomacy joined #salt
18:58 jtand joined #salt
18:58 bowhunter joined #salt
18:59 ajw0100 joined #salt
19:01 quix joined #salt
19:02 phtes joined #salt
19:03 devtea joined #salt
19:07 ivanjaros joined #salt
19:08 _armguy_ joined #salt
19:09 SteamWells joined #salt
19:11 shawnbutts joined #salt
19:12 kutenai joined #salt
19:15 djural joined #salt
19:21 baweaver joined #salt
19:22 Guest79 joined #salt
19:23 scoates joined #salt
19:24 RalfJ joined #salt
19:26 repl1cant joined #salt
19:31 angvp hey guys, quick question
19:31 angvp why 2015.8.8.2 is not on PyPI ?
19:35 Gabemo joined #salt
19:36 cliffstah joined #salt
19:36 chrismoos joined #salt
19:36 jtand joined #salt
19:37 UtahDave angvp: Hm. not sure. let me check
19:37 pmcnabb joined #salt
19:37 liskl joined #salt
19:38 brianfeister joined #salt
19:38 ageorgop joined #salt
19:42 Eureka70_ joined #salt
19:42 hightekvagabond joined #salt
19:50 punkoivan1 joined #salt
19:50 punkoivan1 joined #salt
19:51 nkuttler_ joined #salt
19:53 R0N_ joined #salt
19:53 baffle_ joined #salt
19:53 armguy_ joined #salt
19:53 Vivek joined #salt
19:54 Ludo__ joined #salt
19:54 `chris_ joined #salt
19:54 XenophonF joined #salt
19:54 kevinquinnyo so i have groups of clusters that i manage, for instance web1-1001-foo.com, web2.1001.foo.com, etc.  Now if i do orchestrations, is there a way for me to have a general target glob of say 'web*' but only pass to it a single cluster group?
19:54 kevinquinnyo so the orchestration would match, but only on a subset of servers with say '1001' in their minion id
19:55 debian112 joined #salt
19:55 bowhunter joined #salt
19:55 VSpike Hi. I'm trying to install PHP FPM using the php formula, and I was trying to use PHP 7. I seem to be getting a mixture of PHP5 and PHP7 packages no matter what I do. https://bpaste.net/show/06bab2571eed is my pillar
19:55 VSpike Does that look correct, as far as it goes?
19:56 GothAck joined #salt
19:56 VSpike Ah, I should wait for the netsplit to end before asking :)
19:57 i90rr joined #salt
19:57 VSpike ohh 7.0
19:58 aurynn_ joined #salt
19:58 Dev0n joined #salt
19:58 baweaver joined #salt
19:59 brucewang joined #salt
19:59 terratom1 joined #salt
20:00 VSpike Also, it seems to be automatically install apache2, which I definitely don't want as I'm using nginx
20:00 mapu_ joined #salt
20:00 MindfulMonk joined #salt
20:00 rideh- joined #salt
20:02 Karunamon joined #salt
20:02 AndreasLutro formulas never cease to amaze
20:03 VSpike oh? :)
20:03 kevinquinnyo is there any way to pass something to an orchestration?  a variable that it can use in the tgt ?
20:03 arcleo_ joined #salt
20:03 Guest59521 joined #salt
20:03 R0N joined #salt
20:03 Miouge joined #salt
20:03 Guest70044 joined #salt
20:03 UtahDave joined #salt
20:03 terratoma joined #salt
20:03 __number5__ joined #salt
20:03 mapu joined #salt
20:03 jcastle joined #salt
20:03 Ludo- joined #salt
20:03 bantone joined #salt
20:03 mowntan joined #salt
20:03 jfindlay joined #salt
20:03 wendall911 joined #salt
20:03 tiadobatima joined #salt
20:03 ekristen joined #salt
20:03 dreese joined #salt
20:03 rome_390 joined #salt
20:03 baffle joined #salt
20:03 keekz joined #salt
20:03 rideh joined #salt
20:03 ablemann joined #salt
20:03 TOoSmOotH joined #salt
20:03 Horgix joined #salt
20:03 artemz joined #salt
20:03 sab3r joined #salt
20:03 thebwt joined #salt
20:03 JoeJulian joined #salt
20:03 twiedenbein joined #salt
20:03 `chris joined #salt
20:03 aurynn joined #salt
20:03 afics joined #salt
20:03 AndreasLutro kevinquinnyo: yeah, pillars are basically arbitrary variables in orchestration
20:04 kevinquinnyo whew
20:04 kingscott joined #salt
20:04 kevinquinnyo ok cool i think i can make that work then
20:04 kingscott Anyone know how to use the salt.output.progress command? I'm trying to be able to view in real time a the progress of a cmd.run command to my minions.
20:04 afics joined #salt
20:05 VSpike Hmm. This formula seems completely broken for php 7
20:09 OliverMT joined #salt
20:11 arcleo_ joined #salt
20:11 Miouge joined #salt
20:11 UtahDave joined #salt
20:11 jcastle joined #salt
20:11 bantone joined #salt
20:11 wendall911 joined #salt
20:11 tiadobatima joined #salt
20:11 ekristen joined #salt
20:11 dreese joined #salt
20:11 rome_390 joined #salt
20:11 keekz joined #salt
20:11 ablemann joined #salt
20:11 TOoSmOotH joined #salt
20:11 artemz joined #salt
20:11 thebwt joined #salt
20:11 JoeJulian joined #salt
20:11 twiedenbein joined #salt
20:13 catpig joined #salt
20:14 Miouge_ joined #salt
20:14 UtahDave1 joined #salt
20:16 UtahDave So the 2015.8.8-2 is just a small package patch, that's why it's not on pypi
20:16 bantone joined #salt
20:17 ekristen joined #salt
20:17 ablemann joined #salt
20:17 edrocks joined #salt
20:17 dreese joined #salt
20:19 fredvd joined #salt
20:20 baweaver joined #salt
20:20 keekz joined #salt
20:21 twiedenbein joined #salt
20:21 JoeJulian joined #salt
20:21 artemz joined #salt
20:22 TOoSmOotH joined #salt
20:22 VSpike No matter what I do, this PHP formula seems determined to install PHP5 packages instead of PHP7 ones
20:23 rome_390 joined #salt
20:24 thebwt joined #salt
20:24 jcastle joined #salt
20:26 mgresser joined #salt
20:28 arcleo_ joined #salt
20:28 repl1cant anyone know if there is a sleep value for pool_size in /etc/salt/cloud?
20:30 iggy VSpike: welcome to the world of barely maintained software (php and formulas)
20:31 wendall911 joined #salt
20:33 tiadobatima joined #salt
20:34 beardedeagle if it isn't on https://docs.saltstack.com/en/latest/topics/cloud/misc.html#connection-timeout you'd be better suited searching the code to see if it is in there and undocumented.
20:35 beardedeagle (or in some rare cases, documented but not in the code)
20:36 Horgix joined #salt
20:37 _JZ__ joined #salt
20:37 boredatwork joined #salt
20:37 evidence joined #salt
20:38 donmichelangelo joined #salt
20:38 lempa joined #salt
20:38 repl1cant hrm, k. i may need to hack something then
20:38 _armguy_ joined #salt
20:38 repl1cant issue is EC2 only allows for X number of API calls in a given time
20:39 foundatron_ joined #salt
20:39 repl1cant so for something fast like deleting nodes from a map file it starts to puke after the 20th call
20:39 Guest79 joined #salt
20:39 hlub_ joined #salt
20:39 notnotpe_ joined #salt
20:39 VSpike_ joined #salt
20:39 repl1cant loginc would be execute against pool_size->wait for given time->execute agaiainst pool_size->etc
20:39 etw_ joined #salt
20:39 repl1cant s/loginc/logic
20:39 chamunks- joined #salt
20:40 NVX joined #salt
20:40 repl1cant you start getting a bunch of {'Message': 'Request limit exceeded.', 'Code': 'RequestLimitExceeded'}
20:41 repl1cant pool_size fixed for deploying 100+ nodes, but not for deleting
20:41 flebel_ joined #salt
20:41 __alex_ joined #salt
20:41 titilambert joined #salt
20:41 Vivek___ joined #salt
20:41 ze-_ joined #salt
20:42 adrienr_ joined #salt
20:42 quix_ joined #salt
20:42 tawm04_ joined #salt
20:42 stooj joined #salt
20:42 duckfez joined #salt
20:42 feld` joined #salt
20:42 feld` joined #salt
20:42 Patch___ joined #salt
20:43 jhujhiti_ joined #salt
20:43 muep_ joined #salt
20:43 VSpike iggy: heh. Which one is barely maintained? PHP itself, or the PHP formula for Salt?
20:44 Cottser- joined #salt
20:45 winterblack joined #salt
20:46 dustywusty_ joined #salt
20:46 riceandbeans left #salt
20:46 JPau1 joined #salt
20:46 Vye joined #salt
20:46 \ask joined #salt
20:46 gnord_ joined #salt
20:47 armyriad joined #salt
20:48 devtea joined #salt
20:48 SubOracle joined #salt
20:48 rogst joined #salt
20:49 punkoivan1 left #salt
20:49 cb joined #salt
20:50 hacks joined #salt
20:50 MeltedLux joined #salt
20:51 mpanetta joined #salt
20:52 darix joined #salt
20:52 mpanetta joined #salt
20:52 scooby2 joined #salt
20:52 patarr joined #salt
20:52 hoonetorg joined #salt
20:53 oida joined #salt
20:53 sagerdearia joined #salt
20:53 tkharju joined #salt
20:53 mdupont joined #salt
20:53 lungaro joined #salt
20:53 cyteen_ joined #salt
20:54 fxhp joined #salt
20:55 mowntan joined #salt
20:56 armguy_ joined #salt
20:56 kaptk2 joined #salt
20:58 alvinstarr joined #salt
20:58 sk_0 joined #salt
21:02 iggy VSpike: both
21:02 Rumbles joined #salt
21:03 jab416171 can I do: /path/to/foo/abc: file.directory: makedirs:True and then /path/to/bar: file.managed: require: - file: /path/to
21:04 toastedpenguin joined #salt
21:08 jhauser joined #salt
21:10 BogdanR joined #salt
21:10 _armguy_ joined #salt
21:12 KingJ joined #salt
21:12 justanotheruser joined #salt
21:12 quix joined #salt
21:13 ajw0100 joined #salt
21:17 cyborg-one joined #salt
21:18 _armguy_ joined #salt
21:18 martoss joined #salt
21:18 inre joined #salt
21:18 cberndt joined #salt
21:20 jtand joined #salt
21:21 subsignal joined #salt
21:22 pid1 joined #salt
21:22 irctc708 joined #salt
21:23 armguy_ joined #salt
21:26 flowstate joined #salt
21:30 UtahDave jab416171: you have to require another state that's managing the directory you care about
21:33 jhujhiti joined #salt
21:37 rray joined #salt
21:37 rray left #salt
21:38 _armguy_ joined #salt
21:38 teatime jab416171: you can do something very close to what you said.  you can also use require_in in your files and clean=true on your directory to ensure only state-managed files can exist in it
21:39 baweaver joined #salt
21:39 onlyanegg joined #salt
21:40 flowstate joined #salt
21:41 pmcg_ joined #salt
21:41 subsigna_ joined #salt
21:43 `chris joined #salt
21:43 beardedeagle wonder why they say install the entirety of "Development Tools" on RHEL based os's instead of just gcc, gcc-c++, python-devel
21:44 teatime why who says?
21:44 teatime is that in the building salt docs?  if so probably just laziness :)
21:45 sarlalian joined #salt
21:46 VSpike I guess I can't do a require_in: to an sls that is an included formula sls :/
21:46 artemz joined #salt
21:46 flebel joined #salt
21:46 JPaul joined #salt
21:46 teatime no I don't think you can require_in .sls files
21:47 teatime but you can require_in to any named state no matter where it gets defined
21:47 nobrak joined #salt
21:47 nobrak joined #salt
21:47 teatime (getting the other state included is a different matter)
21:48 Cottser joined #salt
21:48 beardedeagle well I assume these docs are made for non technical people as well so a groupinstall is probably easier
21:49 lungaro joined #salt
21:49 bowhunter joined #salt
21:50 duckfez joined #salt
21:50 Vivek joined #salt
21:50 hasues joined #salt
21:51 kevinquinnyo in an orchestration, if you target a glob, like db* and run a state, and the glob evaluates to db1, db2, and db3, is the state run asynchronously?
21:51 kevinquinnyo on all 3 of those
21:51 gtmanfred yes
21:51 hasues left #salt
21:51 gtmanfred it is run on all 3 at the same time
21:52 kevinquinnyo is there any way to override that behavior
21:52 gtmanfred just like you called salt db\* state.highstate on there
21:52 gtmanfred you can do batch: 1
21:52 kevinquinnyo in an orchestration
21:52 gtmanfred and say only run it on one at a time
21:52 kevinquinnyo ahh
21:52 kevinquinnyo nice
21:52 gtmanfred you should be able to*
21:52 kingscott joined #salt
21:52 kevinquinnyo thanks
21:52 gtmanfred i wrote it so you could at some point
21:52 gtmanfred assuming it isn't broken(i don't think there are tests for it)
21:52 kevinquinnyo i'll try it
21:52 gtmanfred but yeah, you should just be able to throw batch: 1 on there
21:52 jhauser joined #salt
21:52 MindfulMonk joined #salt
21:53 gtmanfred kevinquinnyo: https://github.com/saltstack/salt/pull/26355
21:53 saltstackbot [#26355]title: add batch to orchestrate runner | closes #12556
21:53 gtmanfred kevinquinnyo: it won't be available until 2016.3
21:53 gtmanfred so you would need to be on the rc2
21:53 kingscott new to salt here, is it necessary to set the master.pub fingerprint as the master_finger in the minion config file? or can you just accept the minion's key from the master?
21:53 gtmanfred kingscott: you can just accept the minion key
21:54 joren joined #salt
21:54 hoonetorg joined #salt
21:54 kingscott is that as secure? or is it best practice to add it to the minion?
21:55 gtmanfred i have never used master_finger, so i am not sure what it does
21:55 Eugene Strictly, no. In practice, yes that's fine.
21:55 west575 joined #salt
21:55 gtmanfred i would assume, setting master_finger is slightly more secure
21:55 pid1 joined #salt
21:55 gtmanfred but in my experience, it really isn't needed
21:56 Eugene I think salt-cloud preloads the master's public key when spinning up an instance, but I'd have to double-check that.
21:56 kingscott I have to add about 80 minions, I don't really want to add the public key to each.
21:57 Eugene Are you minionizing your existing infrastructure? And it's all on a LAN that you control or at least trust a bit?
21:57 kingscott correct.
21:57 Eugene Just accept the minions and move on
21:58 kingscott That's what i'm thinking as well. Thanks for the help.
21:59 jfindlay joined #salt
22:00 repl1cant joined #salt
22:01 sab3r joined #salt
22:03 druonysus joined #salt
22:03 druonysus joined #salt
22:03 tawm04 joined #salt
22:05 alemeno22 joined #salt
22:06 ajw0100 joined #salt
22:08 LostSoul joined #salt
22:08 aw110f joined #salt
22:11 pid1 joined #salt
22:12 ggoZ joined #salt
22:14 beardedeagle @Eugene: you are correct about salt-cloud
22:15 druonysus_ joined #salt
22:17 riftman joined #salt
22:17 stbenjam joined #salt
22:17 salty_solution joined #salt
22:20 jab416171 I just did require /path/to/foo/abc
22:23 josuebrunel joined #salt
22:30 Eureka703 joined #salt
22:32 onlyanegg joined #salt
22:33 nethershaw joined #salt
22:42 zenlot6 joined #salt
22:42 mowntan joined #salt
22:42 flowstate joined #salt
22:42 mowntan joined #salt
22:42 twork_ joined #salt
22:44 akhter joined #salt
22:44 nZac joined #salt
22:45 kevinquinnyo no idea how i'm going to orchestrate this http://severalnines.com/blog/how-bootstrap-mysqlmariadb-galera-cluster
22:46 kevinquinnyo i need to detect the status across a cluster of 3 or more db servers, then based on the status combination, do a particular orchestration, one at a time in some cases
22:46 gtmanfred write a custom runner
22:47 gtmanfred that is how I would do it, then you can run it from the master, and not have to use the state.orchestrate stuff directly
22:47 ribx joined #salt
22:47 kevinquinnyo hmm yeah i guess i could do that
22:48 kevinquinnyo yeah, that's a good idea, because i think it's going to be easiest just writing it in pure python instead of trying to work with the orchestration flow
22:48 kevinquinnyo thanks again
22:53 drawsmcgraw joined #salt
22:53 monokrome joined #salt
22:53 hal58th joined #salt
23:00 Fabbe joined #salt
23:01 bowhunter joined #salt
23:02 nidr0x joined #salt
23:06 nidr0x joined #salt
23:08 drawsmcgraw So if I have an engine...
23:09 drawsmcgraw And that Engine throws a stacktrace... I only see what-would-be the last line in that stacktrace appear in the Salt Master logs
23:09 baweaver joined #salt
23:09 drawsmcgraw Is there any way to get the full stack trace of my Engine?
23:09 onlyanegg joined #salt
23:14 orion joined #salt
23:14 orion What version of pygit2 does salt require?
23:16 bowhunter joined #salt
23:17 tvinson joined #salt
23:19 baweaver joined #salt
23:19 sagerdearia joined #salt
23:29 nethershaw joined #salt
23:34 watersoul joined #salt
23:34 mosen joined #salt
23:35 kevinquinnyo1 joined #salt
23:42 flowstate joined #salt
23:43 hemebond orion: https://docs.saltstack.com/en/latest/topics/tutorials/gitfs.html
23:44 hemebond "The minimum supported version of pygit2 is 0.20.3"
23:44 orion thanks
23:44 orion left #salt
23:45 watersoul joined #salt
23:46 nZac joined #salt
23:46 Sammichmaker joined #salt
23:48 ajw0100 joined #salt
23:48 sagerdearia joined #salt
23:50 snc joined #salt
23:54 my50c 3/clear
23:54 kevinquinnyo1 is it possible for me to use one of my custom salt modules within a custom runner?

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