Home
ejabberd
ejabberd@conference.jabber.ru
Среда, 23 марта 2011< ^ >
badlop установил(а) тему: www.ejabberd.im english room | March releases: exmpp 0.9.6, ejabberd 3.0.0-alpha-2, and Erlang/OTP R14B02
Конфигурация комнаты
Участники комнаты

GMT+3
[00:18:37] MBR вошёл(а) в комнату
[00:37:51] Cory Radcliff вошёл(а) в комнату
[00:53:13] deryni вошёл(а) в комнату
[00:58:32] Cory Radcliff вышел(а) из комнаты
[01:10:21] MBR вышел(а) из комнаты
[02:02:21] scippio вышел(а) из комнаты: Replaced by new connection
[02:10:50] BcTpe4HbIu вышел(а) из комнаты
[02:40:56] <Sérgio> greetings
[02:41:27] <Sérgio> i need your advise about a xmpp client for android platform, for "technical" users
[02:41:37] <Sérgio> what do you recommend ?
[02:42:15] <Sérgio> preferably some client with muc support
[02:47:09] <badlop> i think my mobile doesn't even has java
[02:47:32] <badlop> you can try to ask in jdev or jabber rooms at conference.jabber.org
[02:47:49] <badlop> * in 'jdev' or 'jabber' rooms
[02:57:16] badlop вышел(а) из комнаты
[07:08:45] Cory Radcliff вошёл(а) в комнату
[07:10:22] <Cory Radcliff> Hello! I have a noob question. I'm writing a mod, and I totally get why I want to use the behavior gen_mod. Some of the plugins to Ejabberd also incorporate gen_server, which I also understand as client/server. When Ejabberd is operating as your server, what makes the determination to when you want to use gen_server and when you want to just use gen_mod?
[07:11:17] Cory Radcliff вышел(а) из комнаты
[07:11:57] Cory Radcliff вошёл(а) в комнату
[07:12:25] <Cory Radcliff> adium sure doesn't like this server
[07:14:37] Cory Radcliff вышел(а) из комнаты
[07:19:49] Cory вошёл(а) в комнату
[07:20:43] <Cory> Ok, Psi is better
[07:21:31] <Cory> So, any direction for my question? I'm still new with Erlang
[07:21:45] <deryni> What was adium doing badly/wrong?
[07:27:38] <Cory> oh. it was crashing after about 30 seconds from joining this muc
[07:27:52] <deryni> Ah.
[07:28:09] <Cory> and i didn't see my posts
[07:28:48] <deryni> I don't have a good answer to your question but they define different behaviours so if one behaviour/contract works better for your plugin than the other I'd say that's likely to be your determining factor. But I haven't worked on ejabberd plugins in years.
[07:31:09] <Cory> ok
[07:31:14] <Cory> Thanks
[07:31:35] <Cory> I suppose there's probably some more information to be gathered by reading the code.
[07:31:43] <Cory> Anyone here working on ejabberd 3?
[07:31:44] <deryni> There should be documentation on gen_server on the erlang site somewhere I think.
[07:31:52] <Cory> oh yeah I read it
[07:32:22] <Cory> I know what they do, I'm just not sure why you would use one vs both in a ejabberd module. I figure there's a good reason that someone with experience would point out
[07:32:32] <deryni> I see.
[07:32:38] <Cory> I programmed my first bit of erlang today =)
[07:32:44] <deryni> If you stick around you can ask badlop, he might be able to tell you better.
[07:33:10] <Cory> ok
[07:33:31] <Cory> I've seen his name come up in a lot of archives =)
[07:36:10] <Cory> oh he works for process one maybe
[07:36:50] <deryni> Yeah.
[07:43:40] <Cory> Well, looks like he's online when I'm sleeping (west coast) so i'll ask all my newbie questions and go to sleep.
[07:45:29] <Cory> We are building a massive backend (millions and billions of users expected*), and openfire and tigase just didn't keep up. Ejabberd is build for *clustering*. Can a single *cluster* of say 10 machines handle 10 x the number of concurrent users of 1 machine? Or is there a limit to the size of a single domain as a result of the limit to the size of a cluster
[07:46:45] <Cory> I am considering building a federation of machines to handle the horizontal load, but I'm not sure if that's necessary if I can build a single uber cluster (lol) that will deal with a huge number of concurrent (and active) users
[08:45:03] Cory вышел(а) из комнаты
[09:07:43] Cory вошёл(а) в комнату
[09:11:07] Cory вышел(а) из комнаты: Computer went to sleep
[11:05:52] Jonas вошёл(а) в комнату
[11:30:20] lemenkov вошёл(а) в комнату
[11:49:42] Jonas вышел(а) из комнаты
[12:03:19] MBR вошёл(а) в комнату
[12:33:35] Jonas вошёл(а) в комнату
[12:52:01] scippio вошёл(а) в комнату
[14:04:01] scippio вышел(а) из комнаты
[14:42:10] MBR вышел(а) из комнаты
[14:42:33] scippio вошёл(а) в комнату
[14:46:22] wiik.andreas вышел(а) из комнаты
[15:38:54] scippio вышел(а) из комнаты
[16:15:51] BcTpe4HbIu вошёл(а) в комнату
[16:53:12] deryni вышел(а) из комнаты
[17:08:51] Holger вышел(а) из комнаты: leaving
[17:09:44] Holger вошёл(а) в комнату
[17:19:59] Cory вошёл(а) в комнату
[17:24:23] Cory вышел(а) из комнаты
[17:24:48] Cory вошёл(а) в комнату
[17:41:55] Cory вышел(а) из комнаты: Computer went to sleep
[17:49:21] deryni вошёл(а) в комнату
[18:51:36] lemenkov вышел(а) из комнаты
[19:08:20] MiGri вышел(а) из комнаты
[19:12:13] MiGri вошёл(а) в комнату
[19:39:45] MiGri вышел(а) из комнаты
[19:40:35] MiGri вошёл(а) в комнату
[19:51:19] lemenkov вошёл(а) в комнату
[20:04:57] Jonas вышел(а) из комнаты
[20:21:23] Cory вошёл(а) в комнату
[20:45:25] badlop вошёл(а) в комнату
[20:55:12] BcTpe4HbIu вышел(а) из комнаты
[20:58:29] <badlop> Cory: there's a cost in clustering: you must configure some mnesia tables to be replicated over all nodes, at least 'session'
[21:01:18] BcTpe4HbIu вошёл(а) в комнату
[21:23:08] <Cory> ok. So each mnesia table takes up so much memory, therefore a limit to the theoretical size
[21:24:13] Cory вышел(а) из комнаты: Logged out
[21:24:55] Cory вошёл(а) в комнату
[21:27:41] Cory вышел(а) из комнаты
[21:38:59] BcTpe4HbIu вышел(а) из комнаты
[22:05:10] BcTpe4HbIu вошёл(а) в комнату
[22:09:31] BcTpe4HbIu вышел(а) из комнаты: Replaced by new connection
[23:03:10] scippio вошёл(а) в комнату
Powered by ejabberd Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!