Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2017-07-26

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

All times shown according to UTC.

Time Nick Message
00:01 alexlist joined #salt
00:03 debian112 joined #salt
00:06 DoomPatrol dxiri: ha softlayer, interesting company to sat the leasy
00:07 DoomPatrol s:sat:say
00:07 dxiri it is :)
00:12 mosen joined #salt
00:28 N-Mi_ joined #salt
00:29 AvengerMoJo joined #salt
00:37 dxiri_ joined #salt
00:51 raspado joined #salt
01:02 onlyanegg joined #salt
01:05 dxiri joined #salt
01:23 raspado joined #salt
01:29 dxiri joined #salt
01:33 raspado joined #salt
01:40 dober joined #salt
01:40 dober joined #salt
01:52 ilbot3 joined #salt
01:52 Topic for #salt is now Welcome to #salt! <+> Latest Versions: 2016.11.6, 2017.7.0 <+> Support: https://www.saltstack.com/support/ <+> Logs: http://irclog.perlgeek.de/salt/ <+> Paste: https://gist.github.com/ <+> See also: #salt-devel, #salt-offtopic <+> We are volunteers and may not have immediate answers <+> The call for speakers for SaltConf17 is now open: http://tinyurl.com/SaltConf17
02:01 dxiri joined #salt
02:07 zerocoolback joined #salt
02:10 dendazen joined #salt
02:15 zerocoolback joined #salt
02:17 zerocool_ joined #salt
02:17 shadoxx_ joined #salt
02:17 oeuftete_ joined #salt
02:19 bantone_ joined #salt
02:19 Ahlee_ joined #salt
02:19 Kruge joined #salt
02:19 notakai joined #salt
02:19 LotR joined #salt
02:20 darvon_ joined #salt
02:21 dxiri joined #salt
02:24 Ludo- joined #salt
02:26 dober joined #salt
02:26 dober joined #salt
02:53 om3 joined #salt
03:03 dxiri joined #salt
03:19 vishvendra joined #salt
03:25 Ni3mm4nd joined #salt
03:31 justan0theruser joined #salt
03:32 dxiri joined #salt
03:32 tapoxi joined #salt
03:32 beardedeagle joined #salt
03:33 beardedeagle what happened to the docs for tornado in the 2017.7.0 docs?
03:33 beardedeagle they are just...gone
03:34 Guest73 joined #salt
03:36 justanotheruser joined #salt
03:36 om2 joined #salt
03:38 beardedeagle missing in develop too
03:39 beardedeagle actually looks like the docs for tornado need to be revamped for all versions as I look at the code.
03:53 ivanjaros joined #salt
04:00 jeddi joined #salt
04:11 om2 joined #salt
04:18 Ni3mm4nd joined #salt
04:18 deep-book-gk_ joined #salt
04:19 deep-book-gk_ left #salt
04:20 fxhp joined #salt
04:25 onlyanegg joined #salt
04:25 cyteen joined #salt
04:35 cyborg-one joined #salt
04:44 svij3 joined #salt
04:47 om2 joined #salt
04:56 inad922 joined #salt
05:07 _JZ_ joined #salt
05:10 golodhrim|work joined #salt
05:11 tobstone joined #salt
05:19 Bock joined #salt
05:23 dxiri joined #salt
05:26 onlyanegg joined #salt
05:52 dxiri joined #salt
05:56 preludedrew joined #salt
05:58 do3meli joined #salt
06:00 donmichelangelo joined #salt
06:05 armyriad joined #salt
06:05 ivanjaros joined #salt
06:06 aldevar joined #salt
06:41 lubyou joined #salt
06:58 om2 joined #salt
07:14 brokensyntax joined #salt
07:19 Hybrid joined #salt
07:26 pualj joined #salt
07:26 onlyanegg joined #salt
07:37 inad922 joined #salt
07:41 Rumbles joined #salt
07:52 dxiri joined #salt
07:55 Ricardo1000 joined #salt
08:00 om2 joined #salt
08:03 mikecmpbll joined #salt
08:16 haam3r_ Has anybody used the new mattermost runner? I'm getting a KeyError when using the master config to specify the hook and url.
08:18 pualj_ joined #salt
08:24 jhauser joined #salt
08:26 CrummyGummy joined #salt
08:28 alexlist joined #salt
08:29 jhauser joined #salt
08:30 Mattch joined #salt
08:37 socket-_ joined #salt
08:38 giomandaz joined #salt
08:42 Dabuti joined #salt
08:43 Dabuti Hello guys, I have a salt master server under FreeBSD and looks like there is no package release of the latest salt version (2017)
08:43 Dabuti is this normal ?
08:45 peuter joined #salt
08:46 high_fiver joined #salt
08:49 puter joined #salt
08:52 Ni3mm4nd joined #salt
08:53 pbandark joined #salt
08:56 _KaszpiR_ joined #salt
08:58 N-Mi_ joined #salt
09:07 Dabuti it is bullshit releasing a new version not for all platforms...
09:07 Dabuti having a salt-master setup on a FreeBSD machine is just useless because of that
09:14 phobosd__ joined #salt
09:15 snakegums_ joined #salt
09:21 poseur joined #salt
09:21 descrepes joined #salt
09:21 graffic_ joined #salt
09:21 nledez joined #salt
09:21 nledez joined #salt
09:21 astronouth7303 joined #salt
09:21 haam3r_ Seems I am also getting the same KeyError when using the runner module inside a reactor sls
09:21 ople_ joined #salt
09:21 ople_ joined #salt
09:21 skr0bul joined #salt
09:22 jeddi joined #salt
09:22 msellers__ joined #salt
09:22 relidy_ joined #salt
09:23 JPT_ joined #salt
09:23 godlike joined #salt
09:23 godlike joined #salt
09:23 Sacro joined #salt
09:23 kalessin joined #salt
09:25 Ssquidly left #salt
09:26 TooLmaN joined #salt
09:27 onlyanegg joined #salt
09:28 tom29739 joined #salt
09:30 godlike joined #salt
09:30 godlike joined #salt
09:32 impi joined #salt
09:34 JPT joined #salt
09:45 gnomethrower joined #salt
09:45 do3meli left #salt
09:50 inad922 joined #salt
10:11 om2 joined #salt
10:20 viq Dabuti: I believe FreeBSD packages are done by FreeBSD, not by saltstack
10:34 Guest73 joined #salt
10:35 inad922 joined #salt
10:39 vishvendra1 joined #salt
10:47 marcinkuzminski joined #salt
10:48 joe_n joined #salt
10:58 doradus joined #salt
11:02 ntropy Dabuti: see http://www.freshports.org/sysutils/py-salt/, you might want to contact the maintainer and offer help ;)
11:02 Psy0rz joined #salt
11:03 ntropy Dabuti: are you having any specific issues with the 2016.11.6 version?  im running that on my master and things are fine
11:03 ntropy on FreeBSD that is
11:11 kiorky joined #salt
11:17 zerocoolback joined #salt
11:28 onlyanegg joined #salt
11:40 dendazen joined #salt
11:50 xet7 joined #salt
12:02 DammitJim joined #salt
12:05 tacoboy joined #salt
12:09 dxiri joined #salt
12:11 numkem joined #salt
12:12 pualj_ joined #salt
12:22 om2 joined #salt
12:22 Dabuti ntropy: Yes, I have issues if the minion has the version 2017.x\
12:23 Dabuti And yeah I know there are packages done by FreeBSD, but I am using the saltstack repository as it says in the documentation
12:23 Dabuti I just think it is a really bad practice to launch a release without having the package on any/some platforms
12:29 ssplatt joined #salt
12:34 debian112 joined #salt
12:39 dxiri joined #salt
12:41 ecdhe joined #salt
12:42 jdipierro joined #salt
12:43 GMAzrael_ joined #salt
12:46 samodid joined #salt
12:47 justanotheruser joined #salt
12:50 justanotheruser joined #salt
12:58 Ahlee_ joined #salt
13:02 ronnix joined #salt
13:15 Cottser joined #salt
13:17 Naresh joined #salt
13:18 beardedeagle joined #salt
13:21 GMAzrael joined #salt
13:26 hemebond left #salt
13:27 GMAzrael joined #salt
13:29 onlyanegg joined #salt
13:34 tapoxi joined #salt
13:35 zer0def joined #salt
13:39 dxiri joined #salt
13:41 squishypebble joined #salt
13:41 racooper joined #salt
13:52 cgiroua joined #salt
13:54 cswang joined #salt
13:59 Ricardo1000 joined #salt
14:01 XenophonF Dabuti: you are welcome to submit a patch to the FreeBSD ports team
14:01 XenophonF the (volunteer) package maintainer for sysutils/py-salt is christer.edwards@gmail.com
14:02 XenophonF nice guy
14:02 XenophonF he usually tests the release for a few days before submitting the update to the ports team
14:03 XenophonF in the meantime it's probably best practice to pin your minions to specific major Salt releases (which is what I do)
14:05 om2 joined #salt
14:05 XenophonF alternatively, you can create, build, and install your own package locally
14:05 XenophonF e.g., fork the freebsd-ports repo on github and use poudriere to build packages from your fork
14:06 XenophonF that's how i handle development and testing for the ports I maintain
14:08 noobiedubie joined #salt
14:08 OlleS joined #salt
14:09 Sarph joined #salt
14:09 dxiri joined #salt
14:09 XenophonF here's a precis of how you'd submit an update - https://www.freebsd.org/doc/en_US.ISO8859-1/books/porters-handbook/port-upgrading.html
14:10 XenophonF PM me if you want help, or ask on #freebsd-ports
14:20 onlyanegg joined #salt
14:20 GMAzrael joined #salt
14:33 mikea joined #salt
14:33 kshlm joined #salt
14:33 mikea Anyone aware of some pre-packaged salt packages for Solaris 10?
14:34 mikea We're trying to build python and all of the prereqs for salt ourselves on this oracle dumpster fire, but it's turning out to be quite the pain in the ass. :-)
14:36 tapoxi anyone know of an ovirt salt-cloud driver?
14:36 tapoxi I could use libvirt but them I'm only talking to one compute node at a time
14:38 raspado joined #salt
14:40 Dabuti thanks XenophonF for the hints, I wish I could work on that patch, but currently I have not enough time for it, anyway I'll try this weekend.
14:41 ivanjaros joined #salt
14:45 Ni3mm4nd joined #salt
14:52 btorch joined #salt
14:55 XenophonF oh man Solaris 10? there might be packages in one of the IPS repos floating around out there
14:56 XenophonF mikea: as a long time sunos admin, you have my deepest sympathy
14:57 mikea XenophonF, yeah, I started on solaris but moved to Linux years ago, but now I've recently inherited a few hundred solaris boxes..
14:57 XenophonF OpenCSW has a really, really old version of Salt :(
14:57 mikea There's an E3500 running Solaris 8, I told them to screw off
14:57 XenophonF holy cow
14:59 sarcasticadmin joined #salt
14:59 XenophonF read it and weep - https://www.opencsw.org/qa/package/py_salt/
14:59 XenophonF I don't know where else you'd look for Solaris packages outside of the Illumos community.
14:59 mikea yeah, the guys here have avoided opencsw and blastwave
15:00 mikea illumos has a good python 2.7 package
15:00 alasdair joined #salt
15:00 mikea I was just hoping that there was already a "This is the goto for salt on Solaris 10" :-)
15:01 XenophonF sadly no
15:01 XenophonF install python and setuptools and pip install salt
15:02 XenophonF then use rc.local or whatever it was on Solaris 10 to start the daemon at boot
15:03 XenophonF I wonder if salt-ssh wouldn't be a better fit for you.
15:04 tkojames joined #salt
15:07 promorphus joined #salt
15:08 preludedrew joined #salt
15:13 GMAzrael joined #salt
15:13 tacoboy joined #salt
15:14 Heartsbane joined #salt
15:15 sp0097 joined #salt
15:24 cablekevin joined #salt
15:24 jbrouwers joined #salt
15:25 KingOfFools Can I run state function with salt-run?
15:26 cablekevin joined #salt
15:31 woodtablet joined #salt
15:32 nebuchadnezzar joined #salt
15:33 aldevar left #salt
15:35 Guest73 joined #salt
15:37 promorphus joined #salt
15:40 twooster i think you want salt-call --local state.apply
15:41 twooster KingOfFools: ^
15:41 high_fiver joined #salt
15:45 tiwula joined #salt
15:51 keith4 joined #salt
15:51 Guest73 joined #salt
15:53 bowhunter joined #salt
15:59 tiwula joined #salt
16:00 fatal_exception joined #salt
16:01 astronouth7303 `salt-run` is a master thing
16:02 RandyT joined #salt
16:04 tacoboy joined #salt
16:06 GMAzrael joined #salt
16:06 svij3 joined #salt
16:07 alasdair left #salt
16:07 alasdair joined #salt
16:08 Guest73 joined #salt
16:12 dwfreed joined #salt
16:13 beardedeagle joined #salt
16:13 seanz joined #salt
16:16 wendall911 joined #salt
16:28 chutzpah joined #salt
16:29 promorphus joined #salt
16:36 chutzpah joined #salt
16:39 keith4 I have a very odd problem. 3 identical debian stretch VMs, running the salt minion from the debian repo (2016.11.2)... state.highstate makes one of them completely unresponsive. master reports "Minion did not return. [No response]", ssh session dies, the console no longer accepts input. only way out is a (virtual) power cycle
16:40 ChubYann joined #salt
16:41 KingOfFools after updating to 2017.7.0 i'm getting empty pillar with 'salt['pillar.get']('pillar item')' in state which I run with salt-run state.orch. Any thoughts?
16:42 astronouth7303 keith4: what states are you using?
16:43 astronouth7303 also, what's your VM vendor?
16:43 astronouth7303 KingOfFools: known bug https://github.com/saltstack/salt/issues/42403 will be fixed in 2017.7.1
16:44 Slimmons joined #salt
16:44 GMAzrael joined #salt
16:44 keith4 astronouth7303: vmware esxi. mostly just pkg.install, with a few file.managed
16:44 astronouth7303 keith4: does ESXi report extremely high CPU?
16:44 KingOfFools astronouth7303: u_u.
16:44 AvengerMoJo joined #salt
16:45 KingOfFools astronouth7303: thanks tho
16:45 astronouth7303 KingOfFools: i know your pain, though. I ended up downgrading the production gear.
16:45 keith4 astronouth7303: looks like 100% of 1 core, yes
16:46 keith4 ... that was a very specific question ;-)
16:46 Slimmons Running salt '*' state.sl doathing pillar='{filepath: '"C:\Program Files\Aprogramdir"'}'      returns, Pillar data must be formatted as dictionary.  I'm assuming the problem is with the quotes and space in program files....I've tried tons of variations.  Any ideas?
16:46 Slimmons I just need to pass in pillar data that has double quotes and a space in it, like the file path above.
16:46 astronouth7303 keith4: i'm also running debian on ESXi and I get that behavior randomly. I didn't think it was tied to salt actions, though. No, I don't have any solutions or much insight, sorry.
16:47 astronouth7303 Slimmons: depends on your shell escaping rules
16:47 keith4 astronouth7303: I have about 75 jessie VMs, with no problems. 5 or 6 stretch VMs also without this problem. very odd.
16:48 astronouth7303 keith4: yeah, i've got about 5 stretch VMs, all but one (the one i spun up late last week) have succumbed to it at some point or another.
16:48 Slimmons astronouth7303: I'm sending it to a windows minion, so from what I looked up, the single quotes around the double quotes is what it wanted
16:48 censorshipwreck joined #salt
16:48 Slimmons I mean, it wants the double quotes around the file path with a space in it, and for that, it needed the single quotes
16:48 keith4 astronouth7303: damn. let me know if you figure anything out?
16:48 astronouth7303 Slimmons: what it's looking for in the argv is `pillar=<valid yaml>`. Windows escaping rules can be weird though. :/
16:49 astronouth7303 keith4: same for you.
16:49 J0hnSteel joined #salt
16:49 doradus joined #salt
16:54 keith4 astronouth7303: have you seen this? https://github.com/saltstack/salt/issues/35470
16:54 astronouth7303 oh thanks!
16:56 Slimmons In case anyone in the future finds this conversation, the answer was to pillar='{filepath: "\"C:\\Program Files\\Aprogramdir\""}'
16:56 dxiri joined #salt
16:56 Slimmons so, normal escapes, but double quotes inside double quotes
16:57 Slimmons and double windows slashes.
16:57 Slimmons Thanks for pointing me towards valid yaml astro
16:57 astronouth7303 keith4: the weird thing is that it happens to me at random, i can't attribute it to anything in particular. I mean, I'm running mine and test.ping regularly, but if that can trigger 100% CPU...
16:58 astronouth7303 Slimmons: you're welcome!
16:58 keith4 astronouth7303: I can reboot the VM, test.ping over and over and over with no problem. run a single state.highstate (which results in no changes!), and it dies
16:58 AvengerMoJo joined #salt
16:59 astronouth7303 yeah, most of my minions get highstated on a regular basis (deploys from CI), but it hasn't come up in a few days
17:00 KingOfFools astronouth7303: damn, looks 2017 minions are not compatible with 2016 master, so i have to downgrade minions too. Beautiful ;D
17:00 KingOfFools looks like*
17:00 astronouth7303 KingOfFools: oh, yeah. that's a thing.
17:01 fatal_exception joined #salt
17:01 KingOfFools astronouth7303: I'm proud of myself that i just upgraded 1 minion for tests. lol
17:01 astronouth7303 KingOfFools: if you're using the saltstack repos, i think you can just tell apt to install a specific version and it'll be happy
17:02 KingOfFools astronouth7303: i'm using local repo which i update myself. So yea, yum downgrade helps alot. THanks
17:03 dwfreed joined #salt
17:15 jmedinar joined #salt
17:16 jmedinar Hi all... I just updated the master to the latest version... and cannot start it with the following error [ https://pastebin.com/M8biCq7r ]
17:16 jmedinar any idea?
17:20 jmedinar I found this [ https://github.com/giampaolo/psutil/blob/master/psutil/_pslinux.py ]... should I manually place it along the failing script?
17:20 impi joined #salt
17:20 astronouth7303 jmedinar: no, it has a bunch of other stuff it references
17:21 astronouth7303 what OS/distro are you running and how did you update salt-master?
17:22 jmedinar I updated with a simple yum update
17:22 AlexLau joined #salt
17:22 jmedinar redRed Hat Enterprise Linux Server release 7.3 (Maipo)
17:22 astronouth7303 are you using the saltstack repos?
17:22 jmedinar yes
17:22 samodid joined #salt
17:22 KingOfFools Can someone explain how to run salt.cmd(https://docs.saltstack.com/en/2016.11/ref/runners/all/salt.runners.salt.html) from state file?
17:23 astronouth7303 KingOfFools: ... what are you trying to do?
17:24 KingOfFools astronouth7303: I'm running state file with salt-run state.orch and there i want to run execution module on master.
17:24 jmedinar here's my repo [ https://pastebin.com/DxQGhiVu ]
17:24 astronouth7303 KingOfFools: then you want to run it from the orchestrate file. States run on minions and you can't back up to the master from them.
17:25 ronnix joined #salt
17:26 astronouth7303 jmedinar: hmmmm... I don't know redhat/yum, but i would double check that python-psutil installed correctly and the .rpm in the repo has the files
17:26 KingOfFools astronouth7303: but there's no difference between state files and orch files. They are both sls extension and can sit in same directories.
17:26 fritz09 joined #salt
17:26 astronouth7303 KingOfFools: the only similarity is that they're Jinja+Yaml
17:27 astronouth7303 orch files execute on the master and i think have a different set of what's accessible (don't quote me on that)
17:27 alau joined #salt
17:27 astronouth7303 state files are executed on the minion and use the state modules
17:27 jmedinar I double checked the repo from the official website seems to have changed...
17:27 jmedinar I am fixing it
17:28 KingOfFools astronouth7303: yes, with 'state file' i mean sls file. ofc i use runner "style" there.
17:28 astronouth7303 KingOfFools: orch example: https://github.com/whytewolf/salt-phase0-orch/blob/master/orch/sys/salt/update.sls
17:28 KingOfFools astronouth7303: and in that file i want to run just one thing for now, cmd.run
17:29 astronouth7303 run it where?
17:29 KingOfFools astronouth7303: on master
17:30 KingOfFools astronouth7303: something like this, but it doesnt work, im getting an error: https://paste.debian.net/plain/978305
17:31 astronouth7303 yeah, runners are a different namespace from execution and state modules
17:32 KingOfFools astronouth7303: what do you mean?
17:33 astronouth7303 https://docs.saltstack.com/en/latest/ref/runners/all/index.html vs https://docs.saltstack.com/en/latest/ref/wheel/all/index.html vs https://docs.saltstack.com/en/latest/ref/states/all/index.html vs https://docs.saltstack.com/en/latest/ref/modules/all/index.html
17:33 astronouth7303 one of the things hardest things i've found when learning salt is keeping track of all these scopes and which bits refer to which scopes
17:34 KingOfFools astronouth7303: yes, and thereis runner module which called salt. And it has function cmd. Im trying to run that.
17:34 astronouth7303 so orchestrate uses states, but on the master, the `salt` global in jinja refers to execution modules
17:34 ronnix joined #salt
17:34 alau joined #salt
17:35 astronouth7303 almost all orchestration files use pretty much just https://docs.saltstack.com/en/latest/ref/states/all/salt.states.saltmod.html
17:35 astronouth7303 although https://docs.saltstack.com/en/latest/ref/states/all/salt.states.cmd.html might work? I've generally avoided it
17:36 astronouth7303 it's also fairly recommended to run a minion on the master to deal with these kinds of things
17:36 KingOfFools astronouth7303: https://docs.saltstack.com/en/latest/ref/runners/all/salt.runners.salt.html#module-salt.runners.salt
17:36 KingOfFools astronouth7303: i'm trying to use that
17:36 jmedinar I have re-installed the whole salt in the master and the problem persist
17:37 jmedinar will try now re-installing psutil
17:37 astronouth7303 yes
17:37 astronouth7303 but orchestrate runs states on the master, so using the cmd state directly (instead of jumping through 2 things) might work
17:38 astronouth7303 at least, i think it does? I've honestly been salt for like a month and a half
17:40 alasdair joined #salt
17:40 jkjk joined #salt
17:41 jmedinar FYI -- I uninstalled psutil ... and reinstall with PIP... then run a pip check and found 3 dependencies lost
17:41 jmedinar [root@prdpcrdhydra01(HYDRA) yum.repos.d]$ pip check psutil
17:41 jmedinar cloud-init 0.7.5 requires oauth, which is not installed.
17:41 jmedinar cloud-init 0.7.5 requires pyserial, which is not installed.
17:41 jmedinar cloud-init 0.7.5 requires argparse, which is not installed.
17:41 alau joined #salt
17:41 jmedinar after installing those dependencies the error is gone
17:42 adc_ joined #salt
17:42 astronouth7303 what does cloud-init... nevermind. glad you got it working.
17:43 jmedinar I have no idea
17:43 jmedinar but thanks!
17:45 onlyanegg joined #salt
17:45 AvengerMoJo joined #salt
17:45 inad922 joined #salt
17:53 ronnix joined #salt
17:58 Edgan joined #salt
18:03 samodid joined #salt
18:05 ronnix joined #salt
18:05 XenophonF joined #salt
18:06 AvengerMoJo joined #salt
18:10 AlexLau joined #salt
18:12 yidhra joined #salt
18:15 ronnix joined #salt
18:15 aneeshusa joined #salt
18:27 druonysus joined #salt
18:31 AlexLau joined #salt
18:32 ecdhe joined #salt
18:33 valkyr2e joined #salt
18:37 ivanjaros joined #salt
18:38 censorshipwreck joined #salt
18:38 AlexLau joined #salt
18:42 Sokel joined #salt
18:42 wendall911 joined #salt
18:43 Sokel In some of my states, I have something like this: {% if grains['osmajorrelease'] == '6' %}. Even if the grain on the system actually is '6', what's within the if statement is NOT executed. did something change in 2017.7.0 for the osmajorrelease grain?
18:44 astronouth7303 Sokel: https://docs.saltstack.com/en/latest/topics/releases/2017.7.0.html#grains-changes
18:45 colabeer joined #salt
18:46 cachedout joined #salt
18:50 high_fiver joined #salt
18:51 high_fiver joined #salt
18:51 jdipierro joined #salt
18:53 cachedout left #salt
18:53 cachedout joined #salt
18:53 cachedout Hi friends! One of your friendly Salt maintainers here. I wanted to drop in and mention that SaltStack is hiring for a few positions and I'd surely be grateful if y'all would pass them around and if you or somebody you know is interested, please drop us a note.
18:53 cachedout https://saltstack.com/careers/
18:56 astronouth7303 cachedout: remote?
18:57 cachedout ashmckenzie, for the right person, yes
18:58 viq cachedout: "Project Coordinator" seems to have formatting issues in the second half
18:58 viq (at least with javascript disabled"
18:58 cachedout Yeah, I already caught that and have admonished the person who wrote it with great prejudice.
18:58 inad922 joined #salt
18:58 viq :D
18:59 beardedeagle cachedout: damn, if only I could stand to code in javascript enough to want to make it a job
19:00 cachedout beardedeagle, Hah. It takes a certain type of person. ;]
19:01 cachedout Do have a look at the services and support positions though. No JS required. :]
19:04 beardedeagle Hate to sound like a snob, but I work as an SRE now, wouldn't want to step back to a support type position. I'll be the first in line for a Engineering position on the python side of your house though.
19:05 Eugene It is not snobbish to say No to a pay cut ;-)
19:06 beardedeagle I get paid pennies to maintain a pretty massive infrastructure and write code right now, so it would be hard to pay me less lol
19:06 cachedout it pays more considerably more than you'd expect but i totally understand of course. ;]
19:07 beardedeagle pennies = what I do vs. what what I do is worth
19:07 cachedout i don't like the "support" title that's on it. i just lost a political fight on that one.
19:08 cachedout but it's really providing some services work and advanced troubleshooting to some very large infrastructures.
19:08 cachedout but yes, it's not an SRE gig. that's certianly true. :]
19:08 beardedeagle I would argue SRE for that role then vs engineering being SDE
19:08 beardedeagle ah ok
19:08 cachedout yes, the title should be changed. i have a meeting later today in which i will bang my shoe on the table about it again. :]
19:09 beardedeagle werd, "support" might scare of potential hires for sure
19:09 cachedout i completely agree
19:09 cachedout now that i've made my appeal on that
19:10 cachedout for those of you that DO want to do some coding, i have a major project that needs help :]
19:10 cachedout we have to review THIS: https://github.com/saltstack/salt/pull/42572
19:10 cachedout help is very welcome
19:11 astronouth7303 keep in mind, SaltStack Inc's business model is B2B, so support is in support of client's engineering teams
19:12 cachedout and interfacing with our engineering team
19:12 KingOfFools beardedeagle: but thats support work, no?
19:12 astronouth7303 it's a far cry from "have you tried turning it off and on again"
19:13 beardedeagle I thought there was a pep for python 3 for unicode literals that was approved?
19:13 beardedeagle let me dig
19:13 astronouth7303 yes there was
19:13 astronouth7303 like, 3.2 or 3.3
19:14 beardedeagle 3.3
19:15 beardedeagle KingOfFools: That's a grey area in my eyes, yeah you are supporting a customers engineering team but you aren't _directly_ support per say since you are a level two. Kinda like ops in front of dev. Idk
19:16 beardedeagle which tbh, you can't really have a strictly ops team these days. it's all devops which is where SRE's live
19:16 beardedeagle so I could argue either way if I had to be honest
19:16 Ni3mm4nd joined #salt
19:16 sjorge joined #salt
19:17 beardedeagle astronouth7303: ah, here we go: https://www.python.org/dev/peps/pep-0414/#exclusion-of-raw-unicode-literals
19:18 aldevar joined #salt
19:19 astronouth7303 but you generally use u-strings and something like six to try to obtain consistent behavior between py2 and py3
19:19 Guest73 joined #salt
19:21 beardedeagle astronouth7303: unfortunately there is no way around the performance hit iirc though
19:22 astronouth7303 which performance hit?
19:22 astronouth7303 you mean that unicode strings are bigger?
19:24 pabloh007fcb joined #salt
19:24 mikecmpbll joined #salt
19:25 astronouth7303 if you mean the dict performance, i thought that got fixed in 3.6?
19:25 pabloh007fcb left #salt
19:26 pabloh007fcb joined #salt
19:26 zam joined #salt
19:28 pabloh007fcb joined #salt
19:28 mavhq joined #salt
19:31 maddecent_ joined #salt
19:31 pabloh007fcb Hey guys, quick question. Is it possible to  pass arguments to the docker_image.present state. my current non salt docker build process uses the docker --build-arg
19:31 Bryson joined #salt
19:32 beardedeagle maybe? would have to dig through the changelog to see what changes have occured in 3.6+. But if you want to work with strings in py2 and 3 you usually use __futures__ and future.utils.
19:33 GMAzrael_ joined #salt
19:34 maddecent_ left #salt
19:34 maddecent_ joined #salt
19:34 beardedeagle but again, pretty sure that incurs a penalty for both a function call and then a decode. I'll have to dig into the PR to see what terminalmage did though.
19:34 astronouth7303 future.utils depends on the use of python future (as opposed to six, eight, nine, or probably a few others), but yeah
19:34 astronouth7303 you should be codecing at the edges anyway
19:36 GMAzrael_ joined #salt
19:42 oida_ joined #salt
19:43 eykd joined #salt
19:46 viq cachedout: how remote are you possibly considering?
19:49 beardedeagle in case I wasn't online to see a response last night, did anyone else notice the docs for the tornado rest api are gone in 2017.7* and dev? And are incomplete in 2016.*
19:50 beardedeagle cachedout: also, what was the story for that pr pointed at? seems huge.
19:53 whytewolf I should apply for the support position. seems people think i already do it anyway :P
19:54 beardedeagle whytewolf: top keks
19:55 Hybrid joined #salt
19:56 beardedeagle whytewolf: if you could fast track to an engineering position from support that wouldn't be a bad gig then I guess.
20:00 astronouth7303 can confirm, whytewolf is fantastic support and was invaluable when i was just getting started
20:02 DammitJim joined #salt
20:03 _aeris_ joined #salt
20:05 seffyroff joined #salt
20:08 tobstone joined #salt
20:08 cro joined #salt
20:10 whytewolf beardedeagle: from what i have heard before, the support position actually pays close to what i make at $bank already
20:10 beardedeagle whytewolf: it's not so much the money that I care about, so much as the role
20:10 GMAzrael joined #salt
20:11 whytewolf true, going from engineer level 5 to support would look strange on my resume
20:12 cachedout viq: We'd prefer north america time zones
20:13 cachedout beardedeagle, yes, it is huge. :] it is to solve issues related to unicode and python 3.
20:14 beardedeagle cachedout: yeah, I have given it a quick eyeball so far.
20:14 ecdhe joined #salt
20:19 khaije1 joined #salt
20:19 khaije1 joined #salt
20:19 khaije1 joined #salt
20:21 aldevar left #salt
20:23 viq cachedout: thanks. Not there, me.
20:26 khaije1 Hey all, I'm seeking guidance. I'm creating VMs on vcenter using salt-call state.single cloud.present "..." profile="..."
20:26 GMAzrael joined #salt
20:26 khaije1 but the newly cloned VMs still have the same IP as the template instead of the ip specific in the pillar cloud profile config
20:30 GMAzrael joined #salt
20:30 astronouth7303 khaije1: you're setting a static IP in the VM template? or are you using DHCP?
20:30 GMAzrael joined #salt
20:31 khaije1 static IP in the template, and a different one in the profile. Hey @astronouth7303
20:32 astronouth7303 do you have a state to actually apply that IP?
20:32 khaije1 Nope, I thought it was implicit in the profile config
20:32 astronouth7303 what profile?
20:33 khaije1 the salt cloud profile (for example: https://docs.saltstack.com/en/latest/topics/cloud/config.html#lxc
20:34 khaije1 better example: https://docs.saltstack.com/en/latest/topics/cloud/vmware.html#profiles
20:34 astronouth7303 oh, huh.
20:34 khaije1 you can see the 'ip' designation in the Network interface specification about halfway down the example
20:34 astronouth7303 it might depend on having vmware tools installed, or some kind of vmware dhcp?
20:34 censorshipwreck joined #salt
20:35 astronouth7303 salt cloud more-or-less just hands that over to the vmware driver
20:35 astronouth7303 (i'm on vmware myself, but i'm managing IPs through salt or DHCP, because my instances are more dynamic)
20:35 khaije1 Yeah, I did find I had to install vmware tools earlier in my troubleshooting. It helped me get further but at this point it's already there
20:36 Hybrid joined #salt
20:36 onlyanegg joined #salt
20:36 khaije1 I kinda wish dhcp were an option
20:36 astronouth7303 i would dig into vmware network options and what that means on the vmware side
20:37 khaije1 yeah, situations like this I'm sure the problem is somewhere on my side, just hard to isolate and correct it
20:38 khaije1 I'm out of time for now but will check back later on in the evening. Cheers astronouth7303
20:38 * khaije1 steps away
20:38 astronouth7303 i'm sure it'll be obvious after you figure out what that option actually does and what VMware does with it
20:38 socket-_ Is there any guidance on how to do a silent install of salt-minion on windows server 2016. I tried the bootstrap.ps1 and it fails to install python, i tried using the exe with the flags /S /master=salt /minion-name=hostname but it still prompts me to click yes before install. Any advice?
20:42 lordcirth_work socket-, click? There's a GUI?
20:42 viq lordcirth_work: he's trying to do unattended install
20:42 lordcirth_work viq, right, I'm just puzzled why there's a GUI for a script
20:43 viq Oh. I thought you're telling him to click ;)
20:43 lordcirth_work I wasn't telling him to click, I was asking whether "click yes" implied a GUI
20:43 socket-_ the bootstrap script doesnt seem to work on server 2016, the gui is when i run the exe via command line
20:43 socket-_ the exe prompts for a click, not the script, but the script doesnt seem to work
20:44 socket-_ looking for a solution, or a 3rd option to try
20:44 lordcirth_work Perhaps you should debug the .ps1 and submit a bug report
20:44 viq lordcirth_work: yeah, sorry, I misunderstood
20:45 lordcirth_work viq, its fine, I was pretty vague
20:46 beardedeagle I've never had an issue with bootstrap.ps1, course I maintain my own copy.
20:48 cro joined #salt
20:49 pabloh007fcb does anyone know if there are examples on how to use the new SALT.MODULES.DOCKERMOD in states?
20:49 pabloh007fcb I've seen examples for the CLI
20:50 socket-_ I think i must have downloaded the wrong bootstrap. I found another one that seemed to work
20:50 bowhunter joined #salt
20:50 beardedeagle The provided bootstrap files have never worked for me. Might be a GoDaddy thing though.
20:51 beardedeagle so lucky for me I get to maintain a modded version of both.
20:52 socket-_ i was using: salt-windows-bootstrap/master/Bootstrap.ps1 at first, but then i switched to: salt-bootstrap/develop/bootstrap-salt.ps1 and now it works
20:53 whytewolf pabloh007fcb: https://docs.saltstack.com/en/develop/ref/states/all/salt.states.docker.html#module-salt.states.docker
20:53 mquin joined #salt
20:53 whytewolf pabloh007fcb: https://docs.saltstack.com/en/develop/ref/states/all/salt.states.docker_container.html#module-salt.states.docker_container https://docs.saltstack.com/en/develop/ref/states/all/salt.states.docker_image.html#module-salt.states.docker_image https://docs.saltstack.com/en/develop/ref/states/all/salt.states.docker_network.html#module-salt.states.docker_network
20:53 whytewolf https://docs.saltstack.com/en/develop/ref/states/all/salt.states.docker_volume.html#module-salt.states.docker_volume
20:54 astronouth7303 socket-_: yeah, salt-windows-boostrap is for dev, and appears to be somewhat old
21:01 pabloh007fcb Thank you whytewolf
21:10 btorch is salt-api ok with CherryPy (11.0.0) ?
21:10 btorch I'm trying to set it up to try out this molten webui
21:30 lordcirth_work Anyone use a vuln scanner on their networks?  Any recommendations?
21:31 viq openvas, openscap, lynis, hubblestack
21:31 lordcirth_work viq, thanks
21:31 viq depends what you mean by "vuln scanner"
21:33 lordcirth_work viq, hopefully software I can install on a container and have it constantly monitor systems.  Agent or not.
21:33 astronouth7303 viq: I use cherrypy
21:33 whytewolf ... cherrypy isn't a vuln scanner :P
21:34 whytewolf it was btorch asking about cherrypy
21:35 astronouth7303 Oops, sorry. I blame irccolors
21:35 astronouth7303 Irccloud colors
21:36 fatal_exception joined #salt
21:37 beardedeagle saw the meetup for hubblestack, was amazing
21:38 whytewolf hubblestack is pretty awesome.
21:39 lordcirth_work looking it up, it does look great
21:40 lordcirth_work Yep, definitely installing that in dev tomorrow
21:44 AdamSundance joined #salt
21:46 btorch yeah I'm trying to setup salt-api with rest_cherrypy but no having much luck :)
21:46 AdamSundance left #salt
21:47 beardedeagle it's pretty easy. assume you are following this: https://docs.saltstack.com/en/latest/ref/netapi/all/salt.netapi.rest_cherrypy.html
21:47 btorch yeah I'm following that but the debug logs keep throwing py.warnings
21:48 btorch like 'log_file' is obsolete. Use 'log.error_file' instead even though I have log.error_file within my /etc/salt/master.d/api.conf
21:48 beardedeagle wat?
21:48 beardedeagle oh
21:49 beardedeagle yeah it's supposed to be log_error_file
21:49 beardedeagle not log.error_file
21:49 beardedeagle afaik
21:49 btorch yeah that's waht the docs say but I kept getting those warnings
21:50 beardedeagle give me second, code is source of truth in all things
21:51 beardedeagle yeah netapi module is using log.error_file and looking for log_error_file
21:51 beardedeagle 'log.error_file': self.apiopts.get('log_error_file', ''),
21:51 beardedeagle gist your api.conf file
21:53 beardedeagle btorch: ^
21:54 btorch https://pastebin.ca/3846857
21:54 beardedeagle yeah that's wrong
21:54 btorch well I had it like the docs before and I kept getting those errors
21:55 beardedeagle gist full error
21:55 btorch doesn't matter how I label it, it keeps on outputing those error
21:55 btorch give me a sec
21:55 beardedeagle log_access_file: /var/log/salt/api , log_error_file: /var/log/salt/api you mean?
21:55 btorch yep it was like that before
21:58 censorshipwreck joined #salt
21:58 frew so I have a weird thing where a package should be installed, via the `pkg.latest` state, but it's not, and furthermore there is no evidence that it failed either
21:58 frew I have the full output of a highstate with debug on
21:58 frew I am happy to share bits and peices, but it's enourmous
21:59 whytewolf just gist the section around where the pkg.latest is.
21:59 frew yeah ok just a sec
22:00 beardedeagle what happens if you remove the two log options and let it log to the master.log?
22:00 btorch but I'm having more issues that just that, this should be a pretty simple setup but not even the external_auth setup for pam is working :)
22:00 frew uhhh
22:00 frew whytewolf: I'm gonna show this but I wouldn't be surprised if the problem is the first thing I a seeing
22:00 beardedeagle ok btorch, to be fair, external auth is fuuuuuuuuuuuuk to set up for salt-api. at least in my experiance.
22:01 frew whytewolf: https://gist.github.com/frioux/27f89251546110b9b733bd20ef330b9a
22:01 btorch I thought that was required though
22:01 btorch beardedeagle: at least that's what it seemed like from the docs
22:01 beardedeagle no
22:02 beardedeagle here is a really simple conf
22:02 whytewolf frew: humm, the double spaces shouldn't be an issue.
22:02 beardedeagle btorch: https://gist.github.com/beardedeagle/649b90645a6001da219c8bb2d07b9b13
22:02 frew nah I was wondering about `names`, which I've never seen before
22:02 beardedeagle you don't even have to do ssl if you don't want. i wouldn't suggest that myself though
22:03 whytewolf frew: names is a global thing. it creates a seperate state for each item named within names
22:03 frew ah ok.
22:03 frew so a handy shortcut basically.
22:03 whytewolf yeap
22:03 frew so after this there is a DEBUG line of the OrderedDict etc
22:03 btorch beardedeagle: yeah I disabled SSL on mine and also the collect_stats since that was breaking with 500
22:03 frew and that is the last time that any of these packages show up in the log
22:04 whytewolf strange
22:04 beardedeagle btorch: the only reason to use external auth is if you have dc controllers you want to wire access for salt-api up to
22:04 frew yeah it's really weird.
22:04 whytewolf here is the documentation on names https://docs.saltstack.com/en/latest/ref/states/highstate.html#names-declaration
22:04 beardedeagle ...err that was redundant but you get the point
22:04 frew thanks!
22:05 frew whytewolf: fwiw I see other rendered states that do not use names but also have the same behaviour (not actually being installed)
22:05 frew so I guess I shouldn't have even mentioned that
22:06 whytewolf what version is the salt minion?
22:06 beardedeagle btorch: yeah the collect_stats is a known issue
22:06 beardedeagle I filed an issue about that awhile back
22:07 beardedeagle https://github.com/saltstack/salt/issues/31873
22:07 frew 2016.11.3, and it's masterless, btw
22:08 whytewolf humm, masterless shouldn't effect that. but that does help with not bothering mentioning master :P
22:08 frew :)
22:10 whytewolf so, does it work if you call the state directly? also maybe step up debugging to trace or all
22:10 frew I'll increase it to all
22:11 frew it's hard to run directly, but I'll do waht I can
22:11 frew (this is in built images, once I get ahold of a server all the salt stuff is gone)
22:12 whytewolf oh fun.
22:12 frew I can make it not be that way.
22:12 frew just work.
22:13 whytewolf maybe instead of pkg.latest try pkg.installed with - version: latest [since it is going away after everything is done. you only worry about the current install anyway]
22:13 btorch beardedeagle: yeah this thing is all weird :) https://pastebin.ca/3846861
22:14 frew a fix like that will make me nervous, but we'll see
22:14 btorch beardedeagle: Its strange it says  cherrypy: unknown  though, not sure if that is due to the pip install and not a package install
22:15 btorch beardedeagle: it creates that api.log file but nothing goes into it, instead it keeps on using the api log file
22:15 whytewolf understandably.
22:18 pabloh007fcb Hello all, can someone take a look at my docker state image build. trying to see if this is the right way of building the docker image https://gist.github.com/pabloh007/9a2ff25744606bbf9bccbeea0129cab3
22:18 beardedeagle oh, didn't notice you were using cherrypy 11
22:18 beardedeagle not sure if salt-api has been tested against anything newer than the 4.x branch
22:19 beardedeagle that would also explain why it shows unknown
22:19 btorch I tried 11 cause the 3.2.2 was giving that same error :)
22:19 btorch let me try 4.x
22:20 beardedeagle try 3.2.3
22:20 btorch I did
22:21 btorch I may have to try this tomorrow, even after removing eauth, this curl cmds are still returning 401 Unauthorized<
22:22 debian1121 joined #salt
22:23 beardedeagle werd, I should be on. let me know. As far as I can tell the cherrypy code is correct.
22:23 btorch cool thanks
22:24 beardedeagle Though...why they chose cherrypy over...*anything* else is still a mystery to me
22:25 whytewolf because 4 years ago when that was written it was the hot toy everyone was using.
22:26 beardedeagle even 4 years ago cherrypy was a poor choice for just writting rest apis
22:27 whytewolf they currently are trying to move to tornado
22:27 beardedeagle yeah, which is a better choice
22:27 beardedeagle still wouldn't be my top pick
22:28 whytewolf what would be your top pick?
22:29 beardedeagle probably aiohttp
22:29 beardedeagle since I assume the reason they are moving to tornado is for the async
22:30 whytewolf well that and they are moving to tornado for pretty much all communications. the tcp transport is also done in tornado
22:30 beardedeagle right, makes since to standardize on a single technology if you plan to use it elsewhere as well
22:31 beardedeagle sense* even
22:32 beardedeagle tornado just isn't capable of serving as much as aiohttp afaik
22:32 beardedeagle though all benchmarks should be taken with a grain of salt
22:32 beardedeagle *ahem* get it?
22:32 * whytewolf taps his fingers on the glass.
22:32 whytewolf got it
22:33 ssplatt joined #salt
22:34 whytewolf i think support is another issue. aiohttp looks like it is only 3.4+[since it uses asyncio] which as salt normally tries when ever possable to use system python is a tall order
22:35 sarcasticadmin joined #salt
22:36 snakegums joined #salt
22:36 beardedeagle someday we will be python 3 only. someday.
22:36 whytewolf i hope for that day as well. but that day is not today
22:37 beardedeagle course by the time that happens python 4 will have came out, the gil will be gone, and optional static typing will be a thing.
22:37 beardedeagle re: when pigs fly?
22:37 whytewolf this has nothing to do with the aeronotical capability of bacon
22:38 beardedeagle everything comes back to bacon
22:38 beardedeagle everything
22:38 whytewolf i miss bacon
22:46 frew whytewolf: ugh, so doing `state.apply zr.perl` seems to be installing the actual packages
22:47 whytewolf so it works when called directly
22:47 frew yeah
22:47 whytewolf any chance it just doesn't think it is targetted for that minion?
22:48 frew it wouldn't show up in the highstate output if it weren't matched in top.sls right?
22:49 whytewolf true... what status do the states show as in that output?
22:49 frew when I do state.highstate the states don't show up in the output at all
22:49 frew er in the final state output
22:50 whytewolf ... so that sounds like they are not being considered for targetting.
22:50 snakegums_ joined #salt
22:50 whytewolf do they show up in a state.show_highstate
22:51 frew let's see.
22:52 frew yeah they do
22:53 frew https://gist.github.com/frioux/5ffdf133debe2c36139fed5e35ec44bc
22:54 ivanjaros joined #salt
22:54 whytewolf okay ... but they don't show up in the output after a highstate. i ... i don't even know where to go with that.
22:54 preludedrew joined #salt
22:55 whytewolf is zr.perl zr/perl/init.sls or zr/perl.sls ?
22:56 frew the latter
22:57 whytewolf okay, and both don't exist right?
22:57 whytewolf [just trying out any crazy idea at this point]
22:57 frew heh lemme make sure
22:58 frew no they don't both exist :)
22:59 whytewolf shouldn't be an issue do the states still show on a lowstate? state.show_lowstate
22:59 frew let's see
23:00 frew yeah they show up
23:00 frew https://gist.github.com/frioux/f093a0104bad1f54c645d7438afb07b7 for example
23:02 whytewolf yeap. kind of figured they would be there. just i don't understand. no error. they show up in highstate and lowstate. but they don't run in the actual highstate. I'm actually stumped.
23:02 cyborg-one joined #salt
23:03 whytewolf i have never seen that kind of behavour out of salt
23:03 frew yeah it's really strange
23:04 hemebond joined #salt
23:05 frew I'll try to bisect next I guess
23:05 frew by cutting top.sls down more and more
23:05 whytewolf yeah.
23:06 frew I'll let you know if I find anythign
23:26 frew well, fwiw I found that if I reduce my top.sls to exactly the bit that pulls in the packages in question, they do get installed
23:26 frew still working on finding what makes them not get installed
23:39 hashwagon joined #salt
23:44 kiltzman joined #salt
23:47 frew whytewolf: ok I think I figured out the surface level cause
23:47 frew a state is failing (Which I know, this is a work in progress) but that is causing salt to not continue processing other stuff
23:48 frew normally salt doesn't do that in my experiecne
23:48 whytewolf look for failhard on a state
23:49 whytewolf https://docs.saltstack.com/en/latest/ref/states/failhard.html#state-level-failhard
23:49 whytewolf or failhard set in the minion config file
23:51 frew ok
23:51 frew looking
23:52 frew got it
23:52 frew thanks
23:52 frew that's it
23:56 whytewolf can't believe i didn't think about failhard before. but it is so rare to see in a run
23:57 frew yeah
23:58 frew it is set in a really low level early build, because we want AMIs to either work or not, no inbetween
23:58 frew and I didn't realize that
23:58 frew anyway, we learned something :)
23:58 frew or I did and you remembered

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