<     May 2017     >
Su Mo Tu We Th Fr Sa  
    1  2  3  4  5  6  
 7  8  9 10 11 12 13  
14 15 16 17 18 19 20  
21 22 23 24 25 26 _2_7  
28 29 30 31
00:19 kronicdeth joined
00:19 sfbw joined
00:19 <seaword> With ecto, is it possible to issue a delete statement without performing a select? So basically this statement: DELETE FROM post WHERE id = 1;
00:24 jschneck joined
00:27 ssswitch joined
00:29 kronicdeth joined
00:29 zv joined
00:37 cp__ joined
00:37 mmonkeyking joined
00:40 nahtnam joined
00:41 <asonge> seaword: yes
00:41 <seaword> asonge: I figured it out: `Repo.delete_all(from s in Session, where: s.id == ^id and s.user_id == ^current_user.id)`
00:43 __charly__ joined
00:45 kronicdeth joined
00:48 rodolfojcj joined
00:48 zv joined
00:49 gvaughn_ joined
00:56 NeverDie joined
00:56 <Radar> Don't think you need the user_id part of that query... except seaword is now gone :(
00:58 icecreamcohen_ joined
00:59 <asonge> Radar: it might be a good extra measure
01:00 gvaughn_ joined
01:01 the_voice- joined
01:03 mika__ joined
01:05 sevenseacat joined
01:08 ur5us_ joined
01:09 maxbeizer joined
01:10 ur5us joined
01:14 bitmod joined
01:20 sfbw joined
01:22 <mmonkeyking> Is there a way to use Phoenix Channels with vanilla JS/jQuery? App was started with --no-brunch and no other build tool is being used.
01:25 <bitmod> is bitwalker ever on this channel?
01:30 <meh`> is there a way to log rescued errors? Logger.error doesn't like the rescued value
01:31 jmill joined
01:39 tomterl joined
01:42 elgenie joined
01:46 Ayey_ joined
01:51 <bitwalker> bitmod: I'm always passively here, sometimes actively
01:52 <bitwalker> meh`: I would just use Exception.message, which returns the error as a formatted string
01:52 <meh`> bitwalker, nah, I wanted Exception.format/3
01:53 <bitmod> bitwalker: ah ok great! i've seen a lot of the tools you've developed for deploying phoenix, and am just wondering if you're working on an actual deploy tool that massively simplifies the process
01:54 <meh`> also how would I go about doing a `Foo.unquote(name)`?
01:54 <meh`> as in, I want the macro to call `Foo.bar` if `name = :bar`
01:55 zv joined
01:55 <bitmod> i think it would add a huge amount of value to the ecosystem, as a the moment deployment seems to be quite tricky. I would do it myself, but i figured you've got a huge amount of experience in this area and might have started something already
01:55 <meh`> I can't use apply because bar might be another macro
01:55 <meh`> (it is)
01:55 Ayey_ joined
01:56 <bitwalker> bitmod: be right back and I'll answer your question :)
01:57 <bitmod> bitwalker: no worries ;)
01:59 <benwilson512> meh`: you could probably just build the AST of a function / macro call
01:59 <benwilson512> and then call Macro.expand
02:00 <meh`> benwilson512, no unquote fragment magic available?
02:00 <benwilson512> meh`: not that I can think of for that syntax
02:00 <meh`> :(
02:01 <benwilson512> I feel like dynamic macro calling may be something to avoid but
02:01 <benwilson512> dunno about the use case
02:02 hahuang65 joined
02:06 dj_goku joined
02:11 <bitwalker> bitmod: there are already some efforts in that area, namely edeliver and gatling, there is one based on capistrano as well but I can't recall the name off the top of my head, they are all efforts which build on top of distillery as the packaging tool and underlying release handler, but layer on automation for the deploy bits
02:12 <bitwalker> I personally don't have a need for those tools, so I don't think I'm the right person to tackle the development of them, distillery keeps me pretty busy as it is ;)
02:17 jmill joined
02:21 sfbw joined
02:24 <bitmod> bitwalker: oh nice, didn't know about those tools. and yeah, i guess all your projects keep you busy enough xP i was trying to deploy a new phoenix project to DO the other day and had a lot of trouble, eventually i gave up after a couple of days trying to get it to work, and just used nanobox, which is a cool tool. But i thought it would be great if there was a tool like nanobox (maybe not so polished)
02:24 <bitmod> elixir devs could use to easily deploy their apps to the cloud
02:25 <bitmod> maybe that's something i could work on if i get the time
02:29 wong2 joined
02:39 codestorm joined
02:44 hahuang65 joined
02:47 <ssswitch> i had some success deploying to heroku with my elixir app
02:47 <Zarathu> oh, good
02:47 mika__ joined
02:48 <ssswitch> very basic stuff though
02:48 <ssswitch> i used the buildpack on the phoenix docs site
02:49 <Zarathu> meh`: I just did this, sec
02:50 <Zarathu> meh`: rescue e ->; e |> Exception.message |> IO.inspect
02:54 srxa joined
03:00 __charly__ joined
03:03 sfbw joined
03:04 mhutter joined
03:17 griffinbyatt joined
03:18 Lokathor_ joined
03:44 nickjj_ joined
04:00 elgenie joined
04:04 wong2 joined
04:04 codestorm joined
04:04 gvaughn joined
04:11 kronicdeth joined
04:16 proteusguy joined
04:18 hahuang65 joined
04:53 elasticdog joined
04:54 elasticdog joined
04:54 elgenie joined
04:58 jkreeftmeijer joined
04:58 jkreeftmeijer joined
04:59 codestorm joined
05:05 nox joined
05:11 Tica2 joined
05:11 dec0n joined
05:12 cschneid_ joined
05:12 cschneid_ joined
05:18 hahuang65 joined
05:22 manukall joined
05:26 meandi_2 joined
05:27 griffinbyatt joined
05:33 sunaku joined
05:34 Cohedrin joined
05:35 sunaku joined
05:40 srxa joined
05:46 ariejan joined
05:48 gvaughn joined
05:51 srxa_ joined
06:09 micmus joined
06:09 bigkevmcd joined
06:12 solatis joined
06:16 Ayey_ joined
06:19 icanhazbroccoli joined
06:22 craigp joined
06:23 cemilowski joined
06:27 griffinbyatt joined
06:28 tuacker joined
06:38 dignifiedquire joined
06:56 <m1dnight1> Hey guys, my slack bot has a dodgy dependency (Elixir_Slack) and I want my entire app to shut down when the connection fails, so systemd can restart it for me, does that sound reasonable?
06:57 l14n joined
06:57 <m1dnight1> I don't know why it doesn't crash entirely, though.
06:57 <m1dnight1> I see this in my output: "21:02:16.419 [info] Application ohai_irc exited: shutdown", but I don't think the entire VM is actually shutting down, eventhough Im running with the prod environment.
06:57 sunaku joined
06:58 steffkes joined
06:58 <m1dnight1> I recall reading that the prod environment will just kill the entire VM if anything fails
06:58 <m1dnight1> (fails to restart the root supervisor, that is)
06:59 <m1dnight1> https://github.com/m1dnight/slackbot/blob/master/lib/supervisor/connection.ex for reference, this is my root supervisor
06:59 asabil joined
07:06 sammers joined
07:08 mattyw joined
07:19 ariejan joined
07:21 cschneid_ joined
07:21 mark_66 joined
07:25 codestorm joined
07:30 srxa joined
07:31 ur5us joined
07:36 apotry joined
07:44 justelex joined
07:44 akeating joined
07:46 josevalim joined
07:52 elgenie joined
08:00 ariejan joined
08:09 <ariejan> Hi. I'm new to Elixir and currently in the process of writing an Elixir app to connect to slack, process messages and send back responses. There may be outside triggers (like webhooks) that send message to Slack. I've found elixir-slack and I'm able to connect/send/receive messages. The question is, what would be a good way to model this architecturally? A new GenServer to handle Slack interactions? What about callbacks for incoming messages?
08:11 flaviodesousa joined
08:17 ur5us joined
08:20 akeating joined
08:20 akeating joined
08:21 marr joined
08:22 PaReeOhNos joined
08:29 griffinbyatt joined
08:49 stephen_m joined
09:08 hbraun joined
09:21 soud joined
09:22 <soud> hey guys, in phoenix, I'm trying to make a route like this: `/posts/:id.wav`, but it fails with ":identifier in routes must be made of letters, numbers and underscores". is there a way to make a route like that?
09:23 <gazler> soud: No, but you can probably use /posts/:id in combination with https://github.com/mschae/trailing_format_plug for the desired effect.
09:23 mpsag joined
09:23 Guest67 joined
09:26 mika__ joined
09:32 <soud> gazler: thanks, that works for now
09:32 griffinbyatt joined
09:35 lopex joined
09:35 cemilowski joined
09:37 mika__ joined
09:42 srxa_ joined
09:44 __charly__ joined
09:48 akeating joined
09:53 chrismccord joined
09:55 akeating joined
10:03 ariejan_ joined
10:05 mirf joined
10:07 akarambir joined
10:14 gmcabrita joined
10:28 dimitarvp joined
10:33 griffinbyatt joined
10:36 srxa joined
10:48 dimitarvp` joined
10:49 chrismccord joined
10:54 <gazler> micmus: Great blog post on contexts http://michal.muskala.eu/2017/05/16/puting-contexts-in-context.html
10:54 griffinbyatt joined
10:54 <micmus> gazler: thanks!
10:57 <Nicd-> typo in title? puting?
10:58 <micmus> ah! that's terrible :( Thank you, Nicd-
10:59 <PaReeOhNos> micmus: Also noticed a possible missing word? "it might be a good to keep"
11:02 asabil joined
11:02 <micmus> PaReeOhNos: thanks!
11:02 <PaReeOhNos> great post otherwise :)
11:07 aedigix joined
11:28 cristianR_ joined
11:31 cevado joined
11:36 ariejan_ joined
11:44 proteusguy joined
11:46 chrismccord joined
12:04 fmcgeough joined
12:08 fmcgeough_ joined
12:10 asabil joined
12:13 PaReeOhNos joined
12:15 akeating joined
12:21 cschneid_ joined
12:22 ariejan_ joined
12:24 cemilowski joined
12:25 akeating joined
12:36 mika__ joined
12:40 __charly__ joined
12:45 akeating joined
12:46 sfbw joined
12:57 <Ankhers> micmus: Is there anything that I can do to help get that ecto-2 branch merged?
13:02 cdg joined
13:06 <OliverMT> anyone using phoenix_swagger
13:06 <OliverMT> it looks very good on paper
13:06 <OliverMT> but it might be a bit clunky to double-maintain in your controller
13:07 ssswitch joined
13:07 akeating joined
13:09 cristianR_ joined
13:14 kronicdeth joined
13:18 jschneck joined
13:19 codestorm joined
13:20 inoas joined
13:20 <ciawal> OliverMT: I've been looking at it but I have the same concern, I wanted to find a way to remove the duplicate work
13:29 maxbeizer joined
13:30 <OliverMT> I am getting a warning that params is unsafe.. wut?
13:30 <OliverMT> query = if Map.has_key?(params, "x_data_type") do
13:30 <OliverMT> query <> "#{measurement}x.value"
13:30 <OliverMT> end
13:30 <OliverMT> "params is set outside blabla"
13:31 <OliverMT> oh, it *is* actually set inside an if somehwere else in the function, was looking at wrong line.. lol
13:31 <* OliverMT> scuttles back into the corner
13:32 elgenie joined
13:33 <Nicd-> hee hee
13:33 shuaybi joined
13:34 <OliverMT> cleaning up a 750 lines long SQL generator that produces sql strings
13:34 <OliverMT> that I wrote on elixir 1.1
13:34 <OliverMT> the pain is real
13:35 justelex joined
13:35 <Nicd-> I have some code from Elixir 0.1x or something
13:36 zodiak joined
13:44 ejpcmac joined
13:46 jmill joined
13:46 justelex joined
13:50 gvaughn joined
14:02 ejpcmac left
14:03 kronicdeth joined
14:04 akeating joined
14:05 cevado joined
14:06 josevalim_ joined
14:10 akeating joined
14:11 jerel joined
14:14 <iFire> OliverMT: why no ecto?
14:14 <iFire> OliverMT: I should cleanup my repo SQL generator build on ecto D:. I got sidetracked at some point.
14:15 <iFire> I got a bit sad that I couldn't find people to test it.
14:15 <iFire> OliverMT: is phoenix swagger usable now?
14:17 <shuaybi> I am writing a web app in elixir that connects to phoenix via websocket client library https://github.com/Aircloak/phoenix_gen_socket_client. Question: the client loses connectivity every 60 seconds and has to reconnect. Is there a way to ping the server to prevent disconnect? Does Phoenix server have a channel topic for heartbeat. I see in the phoenix code a topic called "phoenix" and event called "heartbeat". But when I try t
14:18 <shuaybi> join error on the topic phoenix: %{"reason" => "unmatched topic"}
14:18 cevado joined
14:19 mika__ joined
14:19 junsuijin joined
14:20 <Nicd-> it's not a topic
14:20 <Nicd-> it's the automatic heartbeat that should be handled by the client
14:21 justelex joined
14:22 <dimitarvp> didn't Phoenix support websockets? (which imply automatic reconnect from the client side?)
14:22 justelex_ joined
14:24 <Nicd-> yes
14:27 NeverDie joined
14:32 elgenie joined
14:34 RexM joined
14:35 bigkevmc_ joined
14:38 jschneck joined
14:40 al-maisan joined
14:41 jkreeftm_ joined
14:41 JEG2 joined
14:50 <chrismccord> shuaybi the client is responsible for sending the heartbeats
14:51 <chrismccord> it does so with a "phoenix" topic and "heartbeat" event
14:56 ariejan_ joined
14:58 mattyw joined
15:01 tuacker joined
15:02 jschneck joined
15:04 <sasajuric> shuaybi: when starting the client, you should include `transport_opts: [keepalive: interval_in_milliseconds]`in the 4th argument of GenSocketClient.start_link
15:05 <sasajuric> I guess it's our bad this isn't documented.
15:10 wong2 joined
15:10 monir joined
15:12 cdg joined
15:14 <shuaybi> chrismccord - I tried doing that by invoking GenSocketClient.push(transport, "phoenix", "heartbeat", state) every 20 seconds through a timer. However I still disconnect in 60 seconds. Is there any pre-requisite to being able to make that call? Do you have to join some channel first?
15:14 cschneid_ joined
15:16 <chrismccord> shuaybi : no, it is a special event as far as phoenix is concerned
15:20 <sasajuric> shuaybi: See my answer above. You don't need to push heartbeat messages manually. Just provide the keepalive option, which will under the hood use WS ping-pong frames. That definitely works for us, as we have some idle connections running for months without problems.
15:22 <shuaybi> sasajuric - your suggestion worked! thanks so much!! Added the following as 4th argument to the GenSocketClient.start_link method: [transport_opts: [keepalive: 20000]]
15:22 <shuaybi> sasajuric - please document it!
15:22 <sasajuric> Will do, or maybe even make it a default, so you don't need to worry about it.
15:24 <shuaybi> sasajuric - yes default it to 30s since I think default timeout on phoenix is 60s
15:24 JamesA joined
15:25 <shuaybi> sasajuric - just for my understanding, why did chris's suggestion not work with your framework? I made a call to GenSocketClient.push(transport, "phoenix", "heartbeat", state) every 20 seconds through a timer.
15:26 rozap joined
15:27 <sasajuric> shuaybi: not really sure, I'd need to debug this. Regardless, normally you shouldn't need to send "phoenix" messages anyway.
15:28 <shuaybi> ok got it. thanks again!
15:30 cevado joined
15:32 stephen_m joined
15:33 jkreeftmeijer joined
15:35 rkazak joined
15:36 v0j joined
15:43 ejpcmac joined
15:44 <OliverMT> iFire: define usable :p
15:45 akeating joined
15:47 ejpcmac left
15:50 icecreamcohen joined
15:53 akeating joined
15:53 rozap joined
15:54 davidw joined
15:54 wsieroci joined
15:55 bigkevmcd joined
15:57 ejpcmac joined
15:57 meh` joined
15:57 amclain joined
15:58 zodiak_ joined
16:12 zodiak__ joined
16:23 josevalim_ joined
16:29 PaReeOhNos joined
16:33 pesnk joined
16:34 elgenie joined
16:37 wsieroci joined
16:39 craigp joined
16:41 cevado joined
16:51 rkazak joined
16:51 cevado joined
16:55 maxbeizer joined
16:55 hahuang65 joined
16:59 refriedchicken joined
16:59 squallstter joined
17:02 wsieroci joined
17:03 isorehalav joined
17:03 srxa_ joined
17:05 jeffweiss joined
17:09 Cohedrin joined
17:11 unrahul joined
17:13 superherointj joined
17:16 PaReeOhNos joined
17:21 sfbw joined
17:29 codestorm joined
17:31 hahuang65 joined
17:31 hahuang65 joined
17:31 <the_voice_> In an umbrella application I understand how to use phoenix to run a short lived microservice that sends a single response using Task.async. What I am trying to figure out to do now though is how to have a long lived microservice update the phoenix application on certain events. Would I use pubsub?
17:32 hahuang65 joined
17:32 <benwilson512> the_voice_: Task.async is a process, you can call it a microservice I guess
17:32 <benwilson512> but as far as googling your question goes, long lived process is gonna get you more answers
17:32 <benwilson512> and a long lived process is a genserver
17:32 <the_voice_> yes I am using genserver
17:32 kronicdeth joined
17:32 <the_voice_> but you have genserver.call, genserver.cast etc.. where the phoenix application would request information
17:33 <benwilson512> well generally you hide GenServer.call/cast function calls inside functions of the genserver module
17:33 InternetFriend joined
17:33 <the_voice_> I don't see the reverse, where the genserver would continueously update the phoenix application on events that happen
17:33 <benwilson512> well
17:33 <benwilson512> what does "update the phoenix application" mean?
17:34 <benwilson512> you've got clients connected via websocket?
17:34 <benwilson512> if you've got events happening in a genserver and you want to let clients know in real time you'll want to use channels yea
17:34 <benwilson512> and you'd broadcast to a topic from within the genserve
17:34 <the_voice_> Channels I get, that's not waht I am talking abotu
17:35 <the_voice_> I have an umbrella application and insdie my apps directory I have the phoenix app, let
17:35 <the_voice_> s call it frontend
17:36 <the_voice_> and then I have a genserver for instance, facebook_webhooks, which has one job, to accepts webhooks from facebook
17:38 InternetFriend joined
17:39 srxa joined
17:40 akeating joined
17:41 Cohedrin joined
17:42 <the_voice_> facebook_webhooks has it's own supervision tree, with each child having :simple_one_for_one strategy as they can be dynamically generated
17:43 sunaku joined
17:43 InternetFriend joined
17:43 <the_voice_> and from the phoenix server I can call FacebookWebhook.start(%Config{}) which will create a long lived genserver whose one job is to listen for webhooks. teh problem is I need that genserver to update the phoenix app when those webhooks come in
17:44 <the_voice_> I use gproc to create unique names for each one etc..
17:44 <the_voice_> Does that make sense?
17:44 Ioyrie joined
17:44 mkwiek joined
17:44 c-rack_ joined
17:45 Scorchin joined
17:45 pantech joined
17:45 dch joined
17:46 darach joined
17:46 jerme_ joined
17:47 asonge joined
17:47 jdqx joined
17:47 icecreamcohen joined
17:48 sfbw joined
17:49 isorehalav joined
17:54 chrismccord joined
17:55 wsieroci joined
17:59 cemilowski joined
17:59 jmill joined
18:02 <the_voice_> hmm, I see that I don't actually need to use gproc
18:05 sunaku joined
18:06 wsieroci joined
18:07 squallstter joined
18:08 sfbw joined
18:09 lexmag joined
18:11 jschneck joined
18:12 peterpp joined
18:16 OtherAllan joined
18:16 dch_ joined
18:18 sfbw joined
18:22 milmazz joined
18:24 <Ankhers> the_voice_: Is there a reason you don't just handle the web hooks within the phoenix application?
18:28 Cohedrin joined
18:32 <the_voice_> Yah, I have many services, it's not just facebook, so I want to split it up into many small applications
18:33 <the_voice_> That's the idea, I may resort back to what you say if I don't find a relatively simple solution to what I am dealing with now :D
18:35 <Ankhers> the_voice_: Can you give an example of what you mean by "I need that genserver to update the phoenix app"?
18:35 <Ankhers> What kind of update are you trying to do?
18:36 elgenie joined
18:37 <the_voice_> so for instance if I want to broadcast to a channel
18:37 <the_voice_> or update a repo on certain events that happen inside the genserver
18:41 __charly__ joined
18:42 <Ankhers> And you would rather do what?
18:43 refriedchicken joined
18:47 bitmod joined
18:48 smt joined
18:52 PaReeOhNos joined
18:57 Ven joined
18:59 rom1504 joined
19:09 smt_ joined
19:10 junsuiji1 joined
19:11 cevado joined
19:12 zodiak_ joined
19:15 codestorm joined
19:21 PaReeOhNos joined
19:22 <superherointj> Hi. I have made a Phoenix App and it works when I direct access it. But when I proxy requests from NGINX to it it shows 404. I have seen the logs and there are some differences regarding the ¨host¨ field. I wonder causes it to not work for requests from Phoenix.
19:22 <superherointj> *from NGINX. Sorry.
19:22 <superherointj> I use NGINX to easy SSL.
19:23 cevado joined
19:26 smt joined
19:28 <superherointj> The location is same ¨/¨, difference is just source ¨host¨.
19:28 <superherointj> host: "www.domainname.com" vs host: ""
19:29 <superherointj> Got this info from IO.inspect conn (Plug connection)
19:29 <superherointj> Both requests show as ¨[info] GET /¨
19:29 <superherointj> Which is supposed to be same location.
19:30 <Ankhers> Did you follow http://www.phoenixframework.org/docs/serving-your-application-behind-a-proxy?
19:30 <superherointj> Thanks!
19:30 <superherointj> I have not seen this page before.
19:31 <Ankhers> I presume that should work. I haven't followed it before though.
19:37 elgenie joined
19:39 icecreamcohen joined
19:40 Cohedrin joined
19:42 Ven_ joined
19:51 Na joined
19:58 Ven joined
20:01 zv joined
20:01 PaReeOhNos joined
20:13 cdg joined
20:17 chrismccord joined
20:18 dch_ joined
20:18 rom1504 joined
20:21 gazler_ joined
20:25 rkazak joined
20:25 sunaku joined
20:26 bananenmannfrau joined
20:26 cevado joined
20:27 Ven_ joined
20:29 ur5us joined
20:29 icanhazbroccoli joined
20:34 sfbw joined
20:34 superherointj joined
20:35 maxbeizer joined
20:36 nbg joined
20:39 tonyc joined
20:48 Ven_ joined
20:52 Uniaika joined
20:52 ryryryry joined
20:53 tonyc joined
20:53 <superherointj> Ankhers, I have tried several combinations so far no luck. I can access other resources in other paths that aren´t root GET ¨/¨. That is the weird part.
20:53 <superherointj> This problem seems to affect only the ¨/¨ PATH.
20:58 __charly__ joined
20:58 <Ankhers> Unfortunately I cannot lend much help here. It has been a number of years since I setup NGINX.
20:59 <superherointj> Do you use SSL in Phoenix?
20:59 <Ankhers> Yes.
20:59 <superherointj> I could drop NGINX and do this.
20:59 <superherointj> I only use NGINX for SSL atm.
20:59 <Ankhers> But I am currently deploying to AWS EC2 Container Service.
21:00 <superherointj> I deploy to a local DockerHost that has a Valid IP.
21:00 <Uniaika> I never use Phoenix for SSL
21:00 <Uniaika> I think it's too risky
21:00 <superherointj> Humm...
21:00 <superherointj> Uniaika, what do you do?
21:00 <Uniaika> that, and allowing beam.smp to open port 80
21:00 <Uniaika> Nginx
21:01 <Uniaika> nginx with hardened SSL configuration, such as the one in https://bettercrypto.org/static/applied-crypto-hardening.pdf
21:01 ur5us joined
21:01 <Uniaika> and I trust nginx for that kinda shit way more than phoenix, or the BEAM
21:02 josevalim joined
21:02 <Ankhers> Sorry, I misunderstood. My AWS load balancer is the thing that handles SSL. I have my phoenix app force SSL though.
21:03 <superherointj> This my current config for NGINX that doesnt work with Phoenix when using the domain name. https://gist.github.com/superherointj/1ae1bcaacfd3242bf52f1f8c3d1bba65
21:03 <superherointj> I also pasted Phoenix log there. IO.inspect conn
21:03 <Uniaika> Ankhers: oh, well, if your app is on AWS' internal network I guess it's better if you do SSL with it as well
21:04 <Uniaika> but then again
21:04 <superherointj> I am using a local server running Docker Host.
21:05 <Uniaika> I don't use AWS, but if your application is on the same host as your load balancer, just drop SSL between them
21:05 <Uniaika> if someone start spying on lo0, you've got worse problems :P
21:05 <Ankhers> It isn't a huge problem right now. It is just in staging.
21:06 <superherointj> Only I have access to this server.
21:08 <superherointj> I use a docker container for NGINX and another docker container for Phoenix App.
21:11 notdaniel joined
21:12 kronicdeth joined
21:18 apotry joined
21:22 ejpcmac left
21:26 elgenie joined
21:33 super joined
21:37 cevado joined
21:40 <meh`> is there a way with the console Logger backend to ignore a specific level?
21:42 fredsir joined
21:45 proteus-guy joined
21:49 codestorm joined
21:51 InternetFriend joined
21:57 PaReeOhNos joined
22:03 pesnk joined
22:07 maxbeizer joined
22:14 Machete joined
22:14 cevado joined
22:15 Machete left
22:19 sunaku joined
22:21 isorehalav joined
22:23 <gvaughn> who remembers the name of that web framework José worked on before Phoenix? Not Chicago Boss, that was/is an Erlang framework.
22:24 pesnk joined
22:24 <Uniaika> gvaughn: I only know of CB, Zotonic and Cowboy
22:24 isorehalav joined
22:24 <asonge> gvaughn: dynamo?
22:24 <asonge> though that was in elixir
22:24 <gvaughn> asonge: Yes! That's what I was trying to remember. Thanks!
22:25 <asonge> i remember using it :D
22:25 ktec joined
22:26 <gvaughn> you deserve a medal for actually using it :-D
22:26 papachan joined
22:27 ktec left
22:33 <asonge> i didn't use it any real products, i just played with it...but the experience wasn't any worse than wiring up erlydtl and mochiweb.
22:35 Sgeo joined
22:38 ktec joined
22:38 Reshi joined
22:48 kronicdeth joined
22:48 arrel joined
22:49 hahuang65 joined
22:49 ktec left
22:51 <sorentwo> I've walked by the "Chicago Boss" bar for 8 years and it just struck me last month that it must have been the namesake that inspired Evan Miller.
22:52 ktec joined
22:53 dch_ joined
22:57 isubasti joined
22:59 <gvaughn> sorentwo: That's a real place?!? :-)
23:03 cemilowski joined
23:05 kronicdeth joined
23:08 cevado joined
23:13 PaReeOhNos joined
23:14 kronicdeth joined
23:15 griffinbyatt joined
23:22 tonyc joined
23:33 cschneid_ joined
23:40 jushur joined
23:45 kronicdeth joined
23:46 NeverDie_ joined
23:49 NeverDie joined
23:52 isubasti joined
23:53 refriedchicken joined
23:57 MarioBranco joined