Activity Graph

Page 1 of 2 | Next »


[18:49:01] tristanm_: what does "env variables that i provide to mix release" mean?
[18:49:20] tristanm_: are you referring to at the tim you run 'mix release' or when you run the release start script
[19:01:45] tristanm_: mix env should have no effect on 'app start'
[19:02:05] tristanm_: don't know how that would be possible
[19:20:50] tristanm_: you need releases.exs if you want to use env variables at runtime
[19:59:22] tristanm_: build time is read at compile time, which is before the release is built
[20:39:40] tristanm_: if you are using releases.exs then the env var is read at runtime
[20:39:51] tristanm_: so it should effect the operation of the built release when run


[18:19:42] tristanm_: yea. it depends on use case. sometimes I use an integer, sometimes a binary, sometimes a string, just depends
[18:37:11] tristanm_: no, if it is a string with dashes you have to convert from hex
[18:38:21] tristanm_: i guess an example is this module that can work with an integer, a 128 bit binary or a string


[19:55:03] tristanm_: works fine in linux
[20:57:07] tristanm_: what do you mean
[21:17:48] tristanm_: don't rememer, I thoguht it worked
[21:17:53] tristanm_: will check in a minute
[21:44:28] tristanm_: longtomjr: damn, yea, I don't know how this is supposed to work. zeal claims to get them from dash and dash says it officially supports hexdocs
[21:49:17] tristanm_: I guess dash has something special for hexdocs and so it isn't just a feed you can add
[21:51:40] tristanm_: I'd want to know who is hosting the dash feed and why it can't be provided as a regular feed


[16:21:25] tristanm_: as in the .script that becomes a .boot? not that I know of and I doubt it
[16:42:10] tristanm_: ah, is that the elixirforum post?
[16:42:16] tristanm_: about slow boot


[12:52:55] tristanm_: it can depend on the type of work the node is doing, and the host's resources. I think it is usually networking that makes it useful to run multiple erlang nodes on a single host. when a single erlang node can saturate the network


[21:42:52] tristanm_: development is funded by ericsson


[15:22:16] tristanm_: don't htink you'll find an "elegant" solution for that. a script that populates the list is likely the best bet


[16:57:34] tristanm_: is this from dockyard
[17:01:03] tristanm_: always surprised at the money they have to spend on projects :)
[17:01:30] tristanm_: surprised/jealous


[00:56:06] tristanm_: *.net *.split
[01:14:03] tristanm_: has joined #elixir-lang


[15:54:45] tristanm_: so MIX_BUILD_PATH sets where to output instead of _build/dev
[15:55:16] tristanm_: but a) what if I want environment to still be created as a sub dir b) is there a variable to set where ./deps/ is created?
[15:55:32] tristanm_: I need to get everything generated outside of the project's directory
[15:55:42] tristanm_: generated/built to
[16:06:58] tristanm_: anything for deps?
[17:39:49] tristanm_: should I open an issue? can explain my use case in it
[18:33:53] tristanm_: josevalim: ok, great


[00:01:03] tristanm_: hah, yes, I keep saying the elixir team should adopt them :P
[00:01:17] tristanm_: like the Erlang team adopting c0b's erlang images repo
[00:07:09] tristanm_: yea, he's a maintainer on as well


[22:14:33] tristanm_: community mistreatmen? like like people being assholes and acting like he should do what they want?


[20:36:36] tristanm_: "dynamic scheduler", is that an elixir lib?
[20:37:10] tristanm_: horde I think just spread processes out evently across nodes in the case of like a node going down and having to restart those that were lost
[20:41:06] tristanm_: dynamic supervisor doesn't doa nything with that as far as I know. I thought it was more a friendly 'siple one for one' which technicallyou can give it a node to start a process on but it doesn't do any balancing for you
[20:41:25] tristanm_: horde distributed based on a hash ring I think
[20:42:39] tristanm_: yea, there isn't anything like a kubernetes scheduler for erlang/elixir
[20:45:32] tristanm_: oh bleh, I forgot, there is one but it uses global so not good in the cloud, one sec
[20:46:51] tristanm_:
[20:49:40] tristanm_: hehe, yea, distributed erlang is not built for the cloud where nodes come and go constantly
[20:50:29] tristanm_: but could mix the pool strategy of checking the load with something like horde


[15:44:50] tristanm_: format is being bound in the pattern match of the case statement "." <> format


[13:46:35] tristanm_: are you sure you aren't somehow seeing distillery documentation?
[13:46:47] tristanm_: if not then it is just a mistake in the docs
[13:46:54] tristanm_: mix release never creates a tar


[18:42:44] tristanm_: priv/ is for parts of your application that aren't erlang source, so artifacts that need to be deployed with it. not for uploaded files by users


[13:14:56] tristanm_: nox: why can't I update a map in a matchspec
[16:38:13] tristanm_: nox: I'll give you a dollar


[16:44:59] tristanm_: or really anything and compare resource usage, at least against Ruby


[19:08:59] tristanm_: oh interesting, an operator
[19:09:12] tristanm_: kustomize is so much nicer to deal with than helm
[19:30:14] tristanm_: benwilson512: do you use that orb?
[19:33:21] tristanm_: ah ok. and yea, got confused with what all that orb is doing. been planning to create a mix orb since we use circleci for opencensus elixir libs. already published a rebar3 orb


[19:55:36] tristanm_: libc and openssl
[19:58:01] tristanm_: if you include crypto in your release it'll need a compatible openssl installed
[20:15:23] tristanm_: i use a terminal on a tiny scaleway instance for irc
[20:16:00] tristanm_: depending on the datacenter you can get them for under 2 bucks


[21:52:19] tristanm_: josevalim: "using", neither. but I took erl_tidy out of otp to create 'rebar3 format'
[21:54:20] tristanm_: iterate on it faster and then send it back
[21:56:44] tristanm_: josevalim: I never worked on it though, been trying ot get someone to pick it up and drive it forward and then we submit it back to OTP. I don't remember what is still missing/buggy
[21:58:23] tristanm_: josevalim: completely out? why?
[22:04:09] tristanm_: josevalim: might be a good canidate for their first moving of some shit to hex and out of otp
[22:04:29] tristanm_: well I guess they did split out some shit, but not ot hex yet
[22:05:48] tristanm_: completely agree there
[22:06:12] tristanm_: I just want someone to actually maintain it, hehe. we could use a working formatter


[16:44:54] tristanm_: just add the negative
[16:45:14] tristanm_: isn't 2#1110010 band (bnot 2#1001110) it?


[23:02:07] tristanm_: foggyboi: use config/releases.exs for runtime configuration
[23:02:25] tristanm_: you can use System.fetch_env! in there to read in environment variables at runtime


[00:06:27] tristanm_: *.net *.split
[00:11:35] tristanm_: has joined #elixir-lang


[18:53:51] tristanm_: I thought when I tried 'mix release' before it was creating a tarball
[18:54:08] tristanm_: I don't see one now when using the final 1.9 relesae and don't see a task for creating one?


[15:42:27] tristanm_: annoying, I am not :). thanks!
[21:10:28] tristanm_: why blue/green instead of just rolling?
[21:12:04] tristanm_: blue/green isntead of rolling restart I mean, I see you do rolling for the live ugprades
[21:13:30] tristanm_: and basically zero community applications come with appups
[21:29:13] tristanm_: ah nm, saw in the HN comments the issue is millions of websockets


[13:14:20] tristanm_: unless you have a sweet pagerduty ring tone that you want to hear often


[23:26:16] tristanm_: Quit: leaving
[23:29:28] tristanm_: has joined #elixir-lang


[14:38:09] tristanm_: I believe that is what start is. and there should be a daemon optoin as well that is what 'start' is from other release tools
[15:23:39] tristanm_: is built it and better command names, like start :)


[22:29:10] tristanm_: Uniaika_: weird, that is saying it can't read the digraph it stores on disk to track dependencies between source files
[22:29:47] tristanm_: Uniaika_: are you using rebar3 through mix I'm guessing?
[22:30:11] tristanm_: if so try deleteing deps/telemetry/.rebar3 if it exists. that is where the digraph is stored
[22:36:35] tristanm_: curious, did you update erlang?
[22:37:02] tristanm_: just tat this looks the same as an issue someone else had yesterday they said happened after they upgraded
[22:37:25] tristanm_: damn ok. really surprised these Source files can be enoent, haha. should have asked to have the .rebar3 sent to me :(


[12:26:02] tristanm_: Ping timeout: 252 seconds
[12:26:22] tristanm_: has joined #elixir-lang


[14:31:28] tristanm_: gazler: no, that was resolved, the moduels were renamed
[14:33:34] tristanm_: ah right, yes


[23:23:18] tristanm_: jnoon2: there is the pool module
[23:23:46] tristanm_:


[22:47:08] tristanm_: was about to say similar because of the 22 improvements for ordered_set
[22:47:21] tristanm_: curious to see what the comparison is on 22


[18:08:40] tristanm_: looking at a kafka client might help. I have one but its erlang. pretty sure there is a pure Elixir one. I'd suggest using -- used that for my client


[20:09:55] tristanm_: its why MononcQc always warns against drinkingi the koolaid


[16:53:13] tristanm_: I've never gotten that damn thing to work with emacs
[16:53:35] tristanm_: lumpy space princess
[16:53:43] tristanm_: I've tried erlang and rust, maybe some others
[16:53:57] tristanm_: alpha emacs?
[16:54:15] tristanm_: I use master
[16:54:25] tristanm_: but hven't updated in a bit, maybe shoudl try again
[16:55:14] tristanm_: my config breaks anyway since I have it auto update packages on startup, hehe
[16:56:55] tristanm_: eh, rarely start emacs, only on reboots
[16:57:08] tristanm_: na, for erlang
[16:57:23] tristanm_: does it not work with elixir well?
[17:00:51] tristanm_: I've been having to do elixir at work now so should try setting it up
[17:05:54] tristanm_: oh. yea, I use flycheck for rebar3 and not erlang


[17:47:18] tristanm_: josevalim: good question :). I don't know it well enough to say and haven't tried writing as complicated of tests yet with it.
[17:51:38] tristanm_: josevalim: does ex_unit have like data_dir and priv_dir? ct lets you have a specific directory of files you need for a suite in a data dir and you get a priv_dir during running of the suite, which is pretty useful
[17:55:01] tristanm_: josevalim: yea, that works, just easier to find after the fact and removes the need to create/track yourself. data_dir is similar in that sure you can just use any dir and figure out the path, but ?config(data_dir, Config) just makes it simpler
[17:58:34] tristanm_: I've made frequent use of groups too, and init_per_group/end_per_group which i haven't seen a parallel to in exunit. not sure how config works either? ct has the Config passed through. and ordering of tests like you can do with ct groups
[17:59:06] tristanm_: josevalim: a setup can be in describe?
[17:59:35] tristanm_: see, also just confusing to me since it isn't just functions but macros ;)
[17:59:56] tristanm_: and or a data_dir it already exists
[18:00:07] tristanm_: like
[18:00:24] tristanm_:
[18:01:57] tristanm_: getting a path is just annoying if you don't know where you are running from
[18:02:40] tristanm_: and shit I don't use that often, but setting up nodes in a ct spec if nice when needed
[18:05:03] tristanm_: hm, does exunit come with surefire/junit xml output support?
[18:08:41] tristanm_: I get annoyed by eunit's macros, hehe
[18:09:05] tristanm_: sure, useful if you are using generators and shit. but when not using those it just gets confusing


[19:28:03] tristanm_: does elixir just use `--name` and figures out from the passed name if it should use a longname or shortname when enabling net_kernel?


[14:20:16] tristanm_: not sure how config/config.exs was used, but is there a way to set default env variables like are put in .app? that would be good to have for all apps
[14:35:14] tristanm_: defaults when loading application env variables
[14:35:52] tristanm_: a single place to see them as well, instead of being in application:get_env(x, y, Default)
[14:36:41] tristanm_: same use case as any application env varaibles
[14:38:05] tristanm_: oh, not sure how that works. I mean like
[14:40:36] tristanm_: yea. I read in the env and then pass them as arguments to start functions. giving a functional interface for use by other apps but also a way to let the application startup ready to go
[15:26:30] tristanm_: fertapric: ah, yup


[13:52:39] tristanm_: afraid of commitment?
[14:00:48] tristanm_: yea, I don't touch them in relx, so they just wait until someone who uses windows sends patches to keep them up to date with the regular one
[14:01:51] tristanm_: huh, differnet syntax than
[14:02:17] tristanm_: why does ours have so many @!
[14:03:22] tristanm_: is that reverse set -x?
[14:03:42] tristanm_: so it by default prints each line it is excuting/
[14:04:22] tristanm_: actually makes way more sense to do it that way so it can easily be enabled
[14:04:37] tristanm_: I have to do that all the time in the other script to get a print out
[14:05:34] tristanm_: josevalim: oh, make sure you don't have ~/.erlang.cookie files being created willy nilly
[14:07:04] tristanm_: josevalim: right, I recently tracked down a few spots in ours that screwed that up. main one being nodetool escript
[14:07:57] tristanm_: josevalim: how rpc is done basically, and to call release handler
[14:08:29] tristanm_: josevalim: and yea, we have cirrus for relx and rebar3 now
[14:08:49] tristanm_: josevalim: oh, in elixir itself?
[14:09:03] tristanm_: na, probably should
[14:09:29] tristanm_: and the windows one fails for some unknown reason in rebar3
[14:10:46] tristanm_: ah
[14:10:51] tristanm_: heh yea just found that
[14:11:03] tristanm_: hm, maybe I should PR tat to erlang
[14:12:35] tristanm_: would be great to get rid of nodetool, should have thought of this earlier, haha
[14:15:06] tristanm_: yea. has taken longer than travis at times which is surprising since travis is slow as fuck most the time
[14:28:55] tristanm_: stop reminding me I have to do something with working groups


[22:28:39] tristanm_: I recently had to do a number of fixes to relx's run script to get it to work smoothly when a container was run as a random user
[22:28:53] tristanm_: so it may be that the start scirpt creates a log dir when it doesn't need to
[22:29:13] tristanm_: that was one of the ones in the relx script. that and it was creating ~/.erlang.cookie a number of times it didn't need to


[14:14:37] tristanm_: pssh, I did mine like a year ago


[15:15:39] tristanm_: put it in a database
[17:06:00] tristanm_: has joined #elixir-lang


[22:41:14] tristanm_: a genserver already is a loop


[16:07:51] tristanm_: how did you miss that nox
[20:10:18] tristanm_: does 1.8.1 use the otp logger by default or is it anoptional backend?
[20:22:09] tristanm_: grr, no slides or video of Andrew's talk that benchmarked lager and otp logger yet
[20:23:03] tristanm_: can't remember the areas otp's had issues and wonder if it is where elixir logger outperforms
[21:39:35] tristanm_: best laptop
[21:46:20] tristanm_: OliverMT: did you apply the patch?
[21:47:05] tristanm_: damn. I have 2 and both have been fine
[21:50:40] tristanm_: I'm 33, so still young and care free


[18:23:02] tristanm_: to add generics
[18:33:47] tristanm_: protocols seem nice
[18:38:31] tristanm_: cond is great. wish nox would write the damn EEF :P
[18:39:16] tristanm_: words are hard


[15:51:17] tristanm_: they moved nameservers
[15:51:21] tristanm_: so dns is propagating


[16:20:29] tristanm_: and in postgres tey are basically (or exactly?) the same, so always use text unless you really need to be sure you restrict the content of a column to N chars for some reason


[00:45:03] tristanm_: it is not wrong


[00:51:55] tristanm_: has joined #ruby
[00:56:01] tristanm_: Read error: Connection reset by peer
[06:56:18] tristanm_: has joined #ruby
[07:29:31] tristanm_: Ping timeout: 252 seconds
[08:01:53] tristanm_: has joined #ruby
[08:06:39] tristanm_: Ping timeout: 260 seconds
[13:00:27] tristanm_: has joined #ruby
[13:05:01] tristanm_: Ping timeout: 252 seconds


[00:35:38] tristanm_: has joined #ruby
[00:40:58] tristanm_: Ping timeout: 252 seconds
[01:57:34] tristanm_: has joined #ruby
[02:02:55] tristanm_: Ping timeout: 252 seconds
[06:18:57] tristanm_: has joined #ruby
[06:24:10] tristanm_: Ping timeout: 252 seconds
[09:34:57] tristanm_: has joined #ruby
[10:18:28] tristanm_: Ping timeout: 252 seconds
[10:26:07] tristanm_: has joined #ruby
[11:00:57] tristanm_: Ping timeout: 276 seconds
[12:51:27] tristanm_: has joined #ruby
[12:56:19] tristanm_: Ping timeout: 252 seconds
[12:57:47] tristanm_: has joined #ruby
[13:05:07] tristanm_: Ping timeout: 252 seconds
[13:34:45] tristanm_: has joined #ruby
[13:40:04] tristanm_: Ping timeout: 260 seconds
[15:15:35] tristanm_: has joined #ruby
[15:20:58] tristanm_: Ping timeout: 252 seconds
[15:28:08] tristanm_: has joined #ruby
[15:59:28] tristanm_: Ping timeout: 252 seconds
[20:38:47] tristanm_: has joined #ruby
[20:43:34] tristanm_: Ping timeout: 260 seconds
[21:31:26] tristanm_: has joined #ruby
[21:35:33] tristanm_: Read error: Connection reset by peer
[22:09:57] tristanm_: has joined #ruby
[22:32:39] tristanm_: Ping timeout: 260 seconds