Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2018-04-26

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

All times shown according to UTC.

Time Nick Message
00:00 onslack joined #salt
00:00 cgiroua_ joined #salt
00:05 cgiroua joined #salt
00:08 cgiroua_ joined #salt
00:27 dendazen joined #salt
00:34 sh123124213 joined #salt
00:40 MTecknology Dangit!! I just created an incompatibility by assuming two zlib.crc32 functions were equivalent
00:40 al joined #salt
00:41 al joined #salt
00:42 MTecknology Anyone interested in helping me reinvent this for py2?
00:43 MTecknology str(abs(zlib.crc32(bytes(s, 'utf-8'))))
00:44 xet7 joined #salt
00:50 MTecknology got it! :)
00:52 MTecknology str(zlib.crc32(bytes(s, 'utf-8')) & 0xffffffff)   ... that seems rather obtuse and ugly
01:22 noraatepernos joined #salt
01:24 renoirb joined #salt
01:29 xet7 joined #salt
01:42 AngryJohnnie joined #salt
01:57 ilbot3 joined #salt
01:57 Topic for #salt is now Welcome to #salt! <+> Latest Versions: 2017.7.5, 2018.3.0 <+> Support: https://www.saltstack.com/support/ <+> Logs: http://irclog.perlgeek.de/salt/ <+> Paste: https://gist.github.com/ <+> See also: #salt-devel, #salt-offtopic, and https://saltstackcommunity.herokuapp.com (for slack) <+> We are volunteers and may not have immediate answers
02:03 pcn Edgan: do you mean grains.get?
02:08 corbe joined #salt
02:11 mbologna1 joined #salt
02:12 whiteing_ joined #salt
02:12 cmichel_ joined #salt
02:12 shiranaihito joined #salt
02:12 rubenb_ joined #salt
02:16 viq[m] joined #salt
02:16 ThomasJ|m1 joined #salt
02:16 viq[m] joined #salt
02:16 agustafson joined #salt
02:16 hexa joined #salt
02:16 preludedrew joined #salt
02:16 tzero joined #salt
02:16 noraatepernos joined #salt
02:16 sayyid9003 joined #salt
02:17 nledez joined #salt
02:17 nledez joined #salt
02:17 shiranaihito joined #salt
02:21 shiranaihito joined #salt
02:22 adongy joined #salt
02:25 xet7 joined #salt
02:25 justanotheruser joined #salt
02:44 zulutango joined #salt
03:01 xet7 joined #salt
03:46 justan0theruser joined #salt
04:12 inad922 joined #salt
04:12 zulutango joined #salt
04:15 FL1SK joined #salt
04:35 Pjusur joined #salt
04:38 dendazen joined #salt
04:43 inad922 joined #salt
04:55 dendazen joined #salt
05:03 mbologna joined #salt
05:10 sauvin joined #salt
05:15 inad922 joined #salt
05:17 druonysus joined #salt
05:32 darioleidi joined #salt
05:36 briner joined #salt
06:15 cbosdonnat joined #salt
06:20 lompik joined #salt
06:22 fabos joined #salt
06:24 fabos left #salt
06:35 toanju joined #salt
06:44 awerner joined #salt
06:52 jose1711_ joined #salt
06:55 jose1711 joined #salt
06:56 jose1711 hello, in sls.. i first do {% set var = '123' %}. then i do include: -.anothersls. why .anothersls does not recognize var?
06:58 sh123124213 joined #salt
06:59 briner joined #salt
07:00 awerner jose1711: the set var is a jinja-level statement, the include is sls-level. so when the include is encountered, all {% %}, {{}} and the like are already parsed
07:00 jose1711 understood. so what is the correct way how to do this?
07:01 jose1711 i still want the split my huge sls file into smaller chunks but make use of a few global vars (defined n the master sls)
07:03 awerner jose1711, https://docs.saltstack.com/en/latest/topics/best_practices.html#modularity-within-states especially search for map.jinja and {% from ... import ... %}
07:05 leyal_ joined #salt
07:08 leyal_ hi - i have a question - in my /etc/salt/proxy config file multiprocessing define to be False - but when i run the proxy by systemctl i can see 2 proxy process - is that expected behaviour ?
07:08 leyal_ root@ubuntu-xenial:/srv/pillar# ps auxww | grep junos  root      1878  3.2  2.6 151148 55160 ?        Ss   07:03   0:00 /usr/bin/python /usr/bin/salt-proxy --proxyid=junos root      1882 29.0  5.7 685072 117244 ?       Rl   07:03   0:00 /usr/bin/python /usr/bin/salt-proxy --proxyid=junos root      1944  0.0  0.0  11284  1088 pts/0    S+   07:03   0:00 grep --color=auto ju
07:11 jose1711 awerner: thank you, bottom line for me: both master and partial sls files need to reference configuration
07:11 Hybrid joined #salt
07:14 awerner jose1711: yes, I for my self moved all defaults and variables to a defaults.yml, and used load_yaml in the map.jinja file, then adapt it in the map.jinja according to distribution/os
07:23 xet7 joined #salt
07:28 orichards joined #salt
07:31 rollniak joined #salt
07:35 briner joined #salt
07:37 satmd joined #salt
07:37 satmd good morning
07:39 jrenner joined #salt
07:41 rollniak Hello :)
07:44 colttt joined #salt
07:51 xet7 joined #salt
07:58 Waples_ joined #salt
08:15 briner_ joined #salt
08:17 xet7 joined #salt
08:17 DanyC joined #salt
08:31 Elsmorian joined #salt
08:38 Elsmorian joined #salt
08:39 demize joined #salt
08:40 Ricardo1000 joined #salt
08:48 exarkun joined #salt
08:49 Processus42 joined #salt
08:54 xet7 joined #salt
09:13 Elsmorian joined #salt
09:15 hoonetorg joined #salt
09:22 rtr63gdh[m] joined #salt
09:22 sxar joined #salt
09:22 freelock joined #salt
09:22 viq[m] joined #salt
09:22 benasse joined #salt
09:22 atmoz joined #salt
09:22 Tenyun[m] joined #salt
09:22 toofoo[m] joined #salt
09:22 hoverbear joined #salt
09:22 jerrykan[m] joined #salt
09:22 benjiale[m] joined #salt
09:23 aboe[m] joined #salt
09:23 ThomasJ|m joined #salt
09:23 gomerus[m] joined #salt
09:23 alj[m] joined #salt
09:23 glock69[m] joined #salt
09:23 systeem[m]1 joined #salt
09:23 gomerus[m]1 joined #salt
09:23 fujexo[m] joined #salt
09:24 sceptiQ_ left #salt
09:25 sceptiQ joined #salt
09:25 sceptiQ o/
09:25 sceptiQ if i use a formula like apache and some informations that are needed to configure apache are already set in an other pillar how can i avoid maintaining the same information on different places?
09:27 briner joined #salt
09:27 hemebond sceptiQ: I use a jinja file to share data between multiple pillar files.
09:28 sceptiQ hemebond do you have an howto or something to read with some examples? maybe (:
09:29 hemebond Create a jinja file with a variable in it (a big dict, in my case "g") and then include it with {%- from "globals.jinja" import g %}
09:30 sceptiQ thx, i ll try!
09:35 dem joined #salt
09:37 Elsmoria_ joined #salt
09:39 om2 joined #salt
09:43 colttt joined #salt
09:44 nledez joined #salt
09:44 nledez joined #salt
09:44 Elsmorian joined #salt
09:51 briner joined #salt
10:00 xet7 joined #salt
10:02 xet7 joined #salt
10:11 briner joined #salt
10:15 sceptiQ hemebond may i have an example how to access these 'g' vars in the pillar?
10:19 Processus42 joined #salt
10:26 gmoro joined #salt
10:27 xist joined #salt
10:27 ghimiryu joined #salt
10:29 ghimiryu_ joined #salt
10:35 onslack <alex> for some reason, during orchestration, a highstate run targets minions that are not connected
10:36 onslack <alex> but `saltutil.sync_all` with the same compoung matcher (G@service:spark-cluster-{{ opts.pillarenv }} and G@role:master) only targets the wanted minion
10:36 hemebond sceptiQ: {{ g.smtp.host }}
10:39 xet7 joined #salt
10:43 om joined #salt
10:43 toanju joined #salt
10:54 freelock joined #salt
10:54 rtr63gdh[m] joined #salt
10:54 Tenyun[m] joined #salt
10:54 sxar joined #salt
10:54 atmoz joined #salt
10:54 jerrykan[m] joined #salt
10:54 hoverbear joined #salt
10:54 benasse joined #salt
10:54 viq[m] joined #salt
10:54 toofoo[m] joined #salt
10:54 benjiale[m] joined #salt
10:54 aboe[m] joined #salt
10:54 ThomasJ|m joined #salt
10:54 alj[m] joined #salt
10:54 gomerus[m] joined #salt
10:54 gomerus[m]1 joined #salt
10:54 systeem[m]1 joined #salt
10:54 fujexo[m] joined #salt
10:54 glock69[m] joined #salt
11:00 xet7 joined #salt
11:01 cablekevin joined #salt
11:03 rollniak joined #salt
11:39 esharpmajor joined #salt
11:57 briner joined #salt
12:05 chutzpah joined #salt
12:08 DammitJim joined #salt
12:11 xet7 joined #salt
12:15 Nahual joined #salt
12:29 xet7 joined #salt
12:30 jerematic joined #salt
12:40 rollniak joined #salt
12:41 briner joined #salt
12:51 cablekevin joined #salt
13:02 Deliant joined #salt
13:08 michiel left #salt
13:08 pewpew joined #salt
13:09 Elsmorian joined #salt
13:23 renoirb joined #salt
13:34 AngryJohnnie joined #salt
13:39 racooper joined #salt
13:40 ddg joined #salt
13:43 nona joined #salt
13:50 nixjdm joined #salt
13:52 gh34 joined #salt
13:52 keimlink joined #salt
14:02 alvinstarr joined #salt
14:06 Yamakaja joined #salt
14:08 cgiroua joined #salt
14:11 _val__ can you tell a service to start in a same definition as pkg.installed
14:11 _val__ ?
14:13 AngryJohnnie joined #salt
14:16 jose1711 _val__: example?
14:16 _val__ jose1711: input output?
14:16 _val__ err sorr.
14:16 _val__ second
14:17 _val__ jose1711: https://dpaste.de/jDrG/raw
14:22 crux-capacitor joined #salt
14:25 zer0def _val__: you can have multiple function calls from *different* state modules under the same id
14:27 _val__ zer0def: mind to explain? the id is different. I'm using filenames directly instead
14:27 zer0def _val__: https://ghostbin.com/paste/knzdb , but for example, you can't have multiple `pkg` or `service` states under the same id; with that said, you can use the `names` keyword (instead of `name`) to (ab)use stating multiple states from the same module and id
14:29 zer0def a particular state stanza is basically `<state module>.<state module function>`, so this should clear up terminology
14:33 pf_moore joined #salt
14:44 CrummyGummy joined #salt
14:53 AngryJohnnie joined #salt
14:56 spiette joined #salt
15:05 DammitJim joined #salt
15:06 DanyC joined #salt
15:10 crux-capacitor hi everyone! what could cause a minion's copy of the master's public key to no longer be valid?
15:15 stooj joined #salt
15:16 djinni` joined #salt
15:19 crux-capacitor it seems like lots of our minions woke up this morning and had an invalid master public key
15:22 briner joined #salt
15:29 exarkun joined #salt
15:33 tiwula joined #salt
15:35 sh123124213 joined #salt
15:43 jamtoast joined #salt
15:49 spiette joined #salt
15:58 AngryJohnnie joined #salt
16:15 Edgan pcn: Kind of. Looks like get lets you explicitly state a delimiter, but item works with a : in master mode. I am going to try get.
16:19 Edgan pcn: get works with salt-ssh, but given that item works with salt-master, it should be fixed. Looks also like get and item are two like but different implementations of the same idea. get can only get one value, so grain.get os oscodename only gives you os, but grain.item os oscodename will give you both.
16:22 DanyC joined #salt
16:23 DanyC_ joined #salt
16:27 dendazen joined #salt
16:34 xet7 joined #salt
16:59 andi- joined #salt
17:15 JacobsLadd3r joined #salt
17:20 noraatepernos joined #salt
17:36 JacobsLadd3r joined #salt
17:46 JacobsLadd3r joined #salt
17:47 DanyC joined #salt
17:57 JacobsLadd3r joined #salt
18:05 AngryJohnnie joined #salt
18:10 rollniak joined #salt
18:16 toanju joined #salt
18:21 sh123124213 joined #salt
18:49 noraatepernos joined #salt
18:54 briner joined #salt
19:00 AngryJohnnie joined #salt
19:06 noraatepernos Is there a way to run highstate on matching minions in a series?  I don’t want to take them all off the load balancer at once.
19:07 babilen noraatepernos: You can specify the batch size with "-b 1" / "-b 25%" or similar options
19:08 babilen https://docs.saltstack.com/en/latest/topics/targeting/batch.html
19:08 noraatepernos babilen: that is amazing.
19:09 noraatepernos Thanks
19:09 babilen yw
19:21 AngryJohnnie joined #salt
19:21 pbuell joined #salt
19:30 xet7 joined #salt
19:39 dendazen joined #salt
20:08 tzero joined #salt
20:10 noraatepernos Is it unusual to create an sls file that acts as a “macro” of sorts?  So I don’t have to write such a terse argument to state.apply?  The case is just git.latest and restarting processes under supervisord.
20:10 noraatepernos What I mean is that this sls wouldn’t be referenced anywhere; top.sls, etc.
20:11 noraatepernos I expected state.apply to find required dependencies but it wants me to specify them all.
20:12 briner joined #salt
20:16 LostSoul joined #salt
20:16 Heartsbane joined #salt
20:16 Heartsbane joined #salt
20:19 MTecknology noraatepernos: in most cases, you're better off not using garbage like supervisord.. but that's a general opinion and not specific to salt. In what way have you "turned the sls file into a macro"?
20:21 noraatepernos MTecknology: The arguments one may pass to state.apply can be sls file names.  However, you also have to pass *any* dependencies manually in your argument by means of a comma separated list.  I don’t get why salt doesn’t just find the requisites for me.
20:21 noraatepernos Why do you hate supervisor so much?
20:22 MTecknology because it's a hacky wrapper to the init system
20:23 LostSoul joined #salt
20:23 MTecknology noraatepernos: Without sharing what you're actually doing, it's probably going to be hard to answer. At least I'm not following your descrpition at all.
20:34 pewpew joined #salt
20:42 garphy`aw joined #salt
20:43 noraatepernos MTecknology: salt ‘*’ state.apply [sls files that git.latest, update dependencies, restart processes under supervisor]
20:44 noraatepernos So, with that, state.apply allows one to specify a comma delimited list of sls files.  What I expected salt to do was resolve the requisites for me. It does not.  I have to provide this long list making sure everything is included otherwise salt complains that it can’t find requisites.  Terse.
20:45 noraatepernos I mean I want something terse…like --resolve-requisites heh
20:46 MTecknology yup, that's expected, since salt has no way of knowing what files have what content
20:46 MTecknology * until they're pulled down
20:46 MTecknology You're telling it what files to pull/parse/execute.
20:46 xet7 joined #salt
20:47 MTecknology If you really don't like that, then you can use state.highstate
20:47 noraatepernos MTecknology: So the “macro” sls I was talking about was a way to make a concise sls…like deploy.sls that just “includes” what I need.
20:47 noraatepernos MTecknology: Yeah it was really just for the purpose of expediency.
20:48 MTecknology are you trying to treat salt like a remote execution platform?
20:48 noraatepernos Well I mean it has run.cmd so it is already right?
20:48 MTecknology no?
20:48 noraatepernos Umm.  Define “remote execution platform"
20:49 noraatepernos Don’t minions just observe a zeromq topic and wait for stuff to do?
20:49 MTecknology remote execution platform... vs. configuration management
20:50 MTecknology -> pushing lots of various bits to whatever server needs those bits at the time vs. managing the state of files/packages/services and ensuring they're always in the desired/correct state
20:50 noraatepernos https://docs.saltstack.com/en/2017.7/topics/tutorials/modules.html
20:52 noraatepernos https://docs.saltstack.com/en/latest/ref/states/all/salt.states.supervisord.html apples/oranges
20:52 noraatepernos I guess we can all use salt differently.
20:53 MTecknology we can... but you're the one having issues with how salt works
20:57 noraatepernos Thanks for your help.
20:57 noraatepernos left #salt
21:00 richard joined #salt
21:05 bd joined #salt
21:07 oida joined #salt
21:26 xet7 joined #salt
21:31 jeremati_ joined #salt
21:32 stooj joined #salt
21:37 ymasson joined #salt
21:37 dxiri joined #salt
21:47 megamaced joined #salt
21:50 xet7 joined #salt
22:09 xet7 joined #salt
22:39 druonysus joined #salt
22:40 cliluw joined #salt
22:53 XenophonF joined #salt
22:59 Felgar joined #salt
23:07 xet7 joined #salt
23:26 zulutango joined #salt
23:29 Deliant joined #salt
23:33 xet7 joined #salt
23:36 capn-morgan joined #salt
23:41 oida joined #salt
23:45 oida joined #salt

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