Logs for jdev
[05:46:30] * jkhii left the chat.
[05:54:08] * scippio left the chat.
[06:23:53] * mlundblad_laptop joined the chat.
[06:46:12] * Zash joined the chat.
[06:52:14] * ermine joined the chat.
[07:02:07] * Alex joined the chat.
[07:13:25] * Guus joined the chat.
[07:27:13] * dbanes left the chat.
[07:29:51] * teo1 left the chat.
[07:29:52] * teo1 joined the chat.
[07:40:01] * dwd joined the chat.
[07:42:13] * Tobias joined the chat.
[07:56:47] * nabatt joined the chat.
[08:01:05] * jprieur joined the chat.
[08:01:07] * jprieur left the chat.
[08:08:47] * smoku joined the chat.
[08:12:31] * smoku left the chat.
[08:12:32] * smoku joined the chat.
[08:20:23] <Tobias> is there pubsub subscription which automatic unsubscribe when you go offline?
[08:22:24] * luca tagliaferri joined the chat.
[08:22:33] * petermount joined the chat.
[10:03:11] * Tobias left the chat.
[10:03:18] * louiz’ left the chat.
[10:03:24] * louiz’ joined the chat.
[10:19:23] * scippio joined the chat.
[10:56:07] * louiz’ left the chat.
[10:56:49] * smoku left the chat.
[10:56:52] * smoku joined the chat.
[10:57:57] * luca tagliaferri left the chat.
[10:58:43] * luca tagliaferri joined the chat.
[11:00:21] * smoku left the chat.
[11:00:23] * smoku joined the chat.
[11:01:37] * smoku left the chat.
[11:01:39] * smoku joined the chat.
[11:12:26] * smoku left the chat.
[11:12:28] * smoku joined the chat.
[11:18:28] * smoku left the chat.
[11:18:30] * smoku joined the chat.
[11:20:36] * smoku left the chat.
[11:20:39] * smoku joined the chat.
[11:30:57] * smoku left the chat.
[11:30:59] * smoku joined the chat.
[11:55:20] * mlundblad_laptop left the chat.
[12:02:43] * mlundblad_laptop joined the chat.
[12:03:20] * mlundblad_laptop left the chat.
[12:03:32] * mlundblad_laptop joined the chat.
[12:05:50] * mlundblad_laptop left the chat.
[12:06:23] * mlundblad_laptop joined the chat.
[12:10:20] * Roti joined the chat.
[12:11:47] * Roti left the chat.
[12:30:11] * scippio left the chat.
[12:30:13] * scippio joined the chat.
[12:35:50] * Guus left the chat.
[12:35:52] * Guus joined the chat.
[12:47:14] * bear joined the chat.
[12:48:51] * Guus left the chat.
[12:53:21] * mlundblad_laptop left the chat.
[13:00:14] * mlundblad_laptop joined the chat.
[13:05:21] * mlundblad_laptop left the chat.
[13:05:53] * mlundblad_laptop joined the chat.
[13:14:44] * louiz’ joined the chat.
[13:25:17] * teo1 left the chat.
[13:39:50] * kzwin joined the chat.
[13:40:58] * deryni left the chat.
[13:47:21] * kzwin left the chat.
[13:52:11] * scippio left the chat.
[13:52:13] * scippio joined the chat.
[13:55:05] * Guus joined the chat.
[14:06:21] * Guus left the chat.
[14:06:25] * Alex left the chat.
[14:10:28] * Tobias joined the chat.
[14:10:40] * smoku left the chat.
[14:10:42] * smoku joined the chat.
[14:15:53] * Lance Stout joined the chat.
[14:15:54] * Lance Stout left the chat.
[14:16:16] * Lance Stout joined the chat.
[14:16:16] * Lance Stout left the chat.
[14:19:36] * Lance Stout joined the chat.
[14:21:47] * smoku left the chat.
[14:21:48] * smoku joined the chat.
[14:28:45] <Tobias> is there a pubsub node configuration that unsubscribes users as they go offline automatically?
[14:31:44] <Tobias> that would require presence subscription somehow, right?
[14:42:34] * deryni joined the chat.
[14:52:51] * mlundblad_laptop left the chat.
[14:54:55] * johnny left the chat.
[14:55:28] * teo1 joined the chat.
[14:55:40] <smoku> Tobias, you might use direct presence
[14:56:33] <Tobias> direct presence would require the client to send direct presence..if it however signs off in an unclean way that's not possible
[14:57:29] <smoku> not really. it's servers job to track direct presences and send direct presence offline when user disconnect
[14:57:43] * nabatt left the chat.
[14:57:43] <Tobias> is it? that might be useful then
[14:59:21] <Tobias> i wonder if there's an existing configuration feature in pubsub for it or if that would be custom out of the definitions in
XEP-0060
[15:08:06] <dwd> I don't know of one. But I'd support such an addition - outside of XEP-0060 itself, mind.
[15:17:36] * hawke joined the chat.
[15:19:35] * MattJ joined the chat.
[15:34:06] <louiz’> what is the state of E2E, currently?
[15:35:17] <louiz’> And in discussions about E2E, the word Cryptanalis is often used. What does this mean (in that context), exactly? I'm not
sure to get it…
[15:35:50] <Zash> louiz’: Breaking it.
[15:36:04] <dwd> louiz’, Cryptanalysis, basically, means "someone who knows Lots About This Stuff figuring out how good it is"
[15:36:16] <Tobias> i guess they mean that some cryptography and security competent people look over the protocol
[15:36:36] <dwd> louiz’, With a TLS based system, we assume that people have some pretty good ideas of whether it's broken. Less so if somehitng
new(ish) gets used.
[15:37:45] <louiz’> hm, ok
[15:37:52] <louiz’> so, it's far from being ready?
[15:38:41] <louiz’> "After all, the point of using good crypto is to *avoid* (successful) cryptanalysis. " for example, I don't understand this
sentence
[15:38:59] <smoku> goal achieved!
[15:39:41] <dwd> louiz’, Humour. There are two common uses of "cryptanalysis", which is ensuring a system cannot be broken, or breaking it.
[15:40:00] <louiz’> Ooh, ok, they (on the ietf mailing list) seem to missuse the "cryptanalis" word, instead of "security analysis"
[15:40:13] <louiz’> dwd, yep, ok, I re-read the post and I now get it :p
[15:41:53] <louiz’> and for the "so, it's far from being ready?"…
[15:45:43] <dwd> louiz’, I'd agree with you.
[15:46:04] * jkhii joined the chat.
[15:47:10] <louiz’> But wasn't there a "deadline" or something, at wich date a complete E2E draft should be finished?
[15:47:24] <dwd> louiz’, Very probably.
[15:47:24] <Zash> It's done when it's done!
[16:12:54] * zanchin joined the chat.
[16:12:59] * petermount left the chat.
[16:18:21] * tofu left the chat.
[16:18:32] * tofu joined the chat.
[16:18:52] * Neustradamus left the chat.
[16:48:04] * tofu left the chat.
[16:48:09] * tofu joined the chat.
[16:48:43] * tofu left the chat.
[16:48:57] * tofu joined the chat.
[16:58:49] * smoku left the chat.
[17:01:35] * smoku joined the chat.
[17:04:36] * smoku left the chat.
[17:15:49] * Lance Stout left the chat.
[17:24:34] * stpeter joined the chat.
[17:24:36] * tofu left the chat.
[17:24:38] * tofu joined the chat.
[17:24:50] * cisox joined the chat.
[17:26:17] * luca tagliaferri left the chat.
[17:29:33] <cisox> So, I'm working on a web XMPP client. I'm using Ejabberd as the backend. I'm using BOSH, and creating the connection in a
script before I load the page, and then attaching with Strophe. On page unload, I store the connection in a cookie, and then
return that instead of creating a connection in the script. Is there a good way to verify that the connection I read from
the cookie is valid still? I tried just sending a roster query request, but on successful connections I end up having to wait
the BOSH wait period (60 seconds) and that won't work for me.
[17:33:02] <Zash> cisox: It should just fail somehow. Try waiting longer than the server. And set proper expire times on the cookie
[17:34:31] <Zash> and "<MattJ> In general, developing an XMPP client against ejabberd is a little like developing a website in Internet Explorer
:)" ;)
[17:34:43] <MattJ> /me slaps Zash
[17:34:47] <Zash> :P
[17:34:56] <MattJ> How long have you sat on that? :)
[17:35:29] <Zash> 2010-08-09 :D
[17:35:59] <Zash> /me laughs manicaly
[17:36:10] * tofu left the chat.
[17:37:47] <Zash> cisox: .. so test with mutiple servers ;)
[17:38:09] * tofu joined the chat.
[17:38:16] <MattJ> An XMPP ping should suffice... if there isn't a BOSH way I'm missing
[17:38:22] <MattJ> but I haven't read the XEP for a while
[17:38:56] <Zash> Shouldn't Strophe just throw some error?
[17:40:28] <MattJ> Erm
[17:40:34] <MattJ> Yes, it would actually
[17:40:49] <MattJ> since attaching opens a request, and the server would reply instantly if the session wasn't valid
[17:41:01] <MattJ> Thanks for prompting me to think :)
[17:41:05] <Zash> :)
[17:43:54] <Zash> How should you handle someone opening the page in two tabs?
[17:44:00] <Zash> Locking?
[17:44:40] <Tobias> you'd fail to login most likely
[17:44:46] <Tobias> in one of them
[17:44:56] <Tobias> because the rid would be invalid
[17:44:59] <Zash> Tobias: Not if you store sid+rid+jid, and resume
[17:45:36] <Tobias> even then i think
[17:45:37] <Zash> Tobias: When I poked around with that it caused every request to end instantly (since there was another)
[17:46:12] <cisox> It would be going in both tabs unless you paused one tab and then took up the other and then resumed with the new rid set.
You could always just fail over to a new connection in the new tab and have two resources going.
[17:46:22] <Tobias> i mean the rid has incremented and isn't correct anymore (the one in the cookie) as soon as you use it
[17:46:59] <Tobias> doesn't the BOSH CM expect the next rid to be lastrid+1?
[17:47:15] <cisox> You normally have a window of 5
[17:47:35] <cisox> It expects the rid to be within that window, which it would quickly leave with a normal XMPP session.
[17:47:58] <Tobias> another way would be removing the cookie as soon as you successfully logged in
[17:48:44] <cisox> And then be sure to save the connection on unload so that if you were refreshing you could resume the same session.
[17:49:08] <Zash> Hm, I wonder if I didn't just save the rid on each request
[17:50:06] <cisox> If you are using Strophe, be sure to pause the connection before saving that info so that the rid doesn't change afterwards.
[17:52:31] <Zash> Anyways, if you know how long the server allows the connection to be paused, just have the cookie expire at that time, or
store that timestamp and compare when resuming
[18:00:22] * cisox left the chat.
[18:01:59] * johnny joined the chat.
[18:12:52] * tofu left the chat.
[18:12:57] * tofu joined the chat.
[18:21:29] * mlundblad joined the chat.
[18:31:23] * mlundblad left the chat.
[18:31:32] * mlundblad joined the chat.
[18:52:40] * tofu left the chat.
[18:52:44] * tofu joined the chat.
[19:00:11] * kzwin joined the chat.
[19:01:03] * kzwin left the chat.
[19:10:53] * Treebilou left the chat.
[19:11:28] * Treebilou joined the chat.
[19:12:07] * Alex_G joined the chat.
[19:12:23] * Alex_G left the chat.
[19:34:28] * kwin joined the chat.
[19:35:40] <kwin> forgive me if my question not in the right room: is jabberd2 considered to be actively developed?
[19:36:03] <johnny> not the last time i saw
[19:36:06] <MattJ> Well
[19:36:12] <johnny> that's why i switched to ejabberd, and then prosody
[19:36:14] <MattJ> It depends on your definition of "actively"
[19:36:17] <stpeter> :)
[19:36:31] <johnny> kwin, prosody is highly recommended
[19:36:34] <dwd> I have seen development of it, at times.
[19:41:42] <kwin> i looked into ejabberd but it's configuration was very foreign to me but i'm going to look into again because it's supposed
to come with set of useful api
[19:42:46] <dwd> kwin, API for what?
[19:43:08] <Tobias> API for configuration obviously
[19:43:10] <kwin> general purpose: my application is database replication.
[19:43:33] <dwd> Tobias, Now, now.
[19:44:07] <kwin> not convinced xmpp is good for this but i'm following orders
[19:44:07] <dwd> kwin, Well, in that case, I'd suggest you want to act as a client or component, and then the server is probably immaterial.
[19:44:52] <dwd> kwin, Depends on the volume of data, and the type of it. If it's medium to low volume, and principally textual in nature (ie,
wrappable in XML easily), XMPP may well be interesting.
[19:45:46] <dwd> kwin, Oh, the other case is if you want to replicate over the internet, and/or to multiple peers simultaneously.
[19:46:12] <dwd> kwin, But quite honestly, if you just want 1:1 replication of binary data over a LAN, it's the wrong tool for the job.
[19:47:13] <kwin> there may be more than 1 peer but not many.
[19:49:00] <kwin> change of topic: pidgen behind firewall; jabberd2 out there; firewall has outgoing 5222 open. no success connecting. (log
shows connect and immediate disconnect) i changed port to 22; then it works. why
[19:50:03] <dwd> Does jabberd2 show any connect to 5222?
[19:50:11] <kwin> (log shows connect and immediate disconnect)
[19:50:59] <kwin> i also changed port to 80 but that didn't work either.
[19:51:32] <kwin> something to do with how firewall (not under my control) is handling port 5222,80 packets compared to 22
[19:52:27] <dwd> Port 80 may well act funny, as it's presumed to be HTTP.
[19:53:06] <dwd> Alternately, it might have decided that port 22 (SSH) is "special". Either way, this seems a little unhelpful.
[19:58:22] <deryni> What error does pidgin give you? What does the Help->Debug Window show during the disconnection?
[20:02:43] <kwin> lost connection with server: connection reset by peer
[20:03:31] <kwin> debug window "(16:01:50) connection: Connection error on 0x179c070 (reason: 0 description: Lost connection with server: Connection
reset by peer)"
[20:04:13] <dwd> deryni, http://www.pigeonstreet.com - just thought I'd mention.
[20:05:00] <kwin> pidgen and other clients can connect to this server at 5222 if not through this firewall.
[20:05:28] <deryni> That sounds like the firewall is terminating the connection for some reason. Does it happen when pidgin tries to start tls?
[20:05:53] <deryni> dwd: Neat.
[20:06:49] <kwin> no indication that tls was even attempted.
[20:07:06] * Treebilou left the chat.
[20:07:45] <deryni> kwin: What xmpp traffic precedes the disconnection in the pidgin Help->Debug Window
[20:07:45] <deryni> ?
[20:09:19] <kwin> (16:01:50) proxy: Connecting to a.b.c:5222 with no proxy
(16:01:50) proxy: Connection in progress
(16:01:50) proxy: Connecting to a.b.c:5222.
(16:01:50) proxy: Connected to a.b.c:5222.
(16:01:50) jabber: Sending (kwin@a.b.c/here): <?xml version='1.0' ?>
(16:01:50) jabber: Sending (kwin@a.b.c/here): <stream:stream to='a.b.c' xmlns='jabber:client' xmlns:stream='http://etherx.jabber.org/streams'
version='1.0'>
[20:10:00] <dwd> deryni, They spelt pidgin wrong, though.
[20:10:43] <deryni> kwin: That's it? It disconnects immediately at that point?
[20:10:50] <dwd> kwin, Your firewall is trying to proxy the connection and doesn't understand the application protocol spoken. That would be
my guess.
[20:11:01] <kwin> yes.. after that follows what i posted earlier
[20:11:17] <dwd> kwin, Unless it's actually deliberately trying to block XMPP, which I'v also seen, but only by mobile operators.
[20:11:48] <deryni> Yeah, it definitely could be what dwd says or similar firewall stupidity. You could see if an old-style SSL connection on
5223 works.
[20:21:15] * Lance Stout joined the chat.
[20:21:16] * Lance Stout left the chat.
[20:21:21] * Lance Stout joined the chat.
[20:21:21] * Lance Stout left the chat.
[20:21:49] <kwin> it seems then that unless firwall vendor rewrite their firware or whatever i'm stuck with port 22
[20:23:16] * kwin left the chat.
[20:24:01] <Zash> tunnel everything over ssh then?
[20:24:24] * Lance Stout joined the chat.
[20:39:33] * hawke left the chat.
[20:40:02] * hawke joined the chat.
[21:06:03] * ermine left the chat.
[21:11:56] * deryni left the chat.
[21:15:12] * Lance Stout left the chat.
[21:27:46] * Kevmin joined the chat.
[21:28:07] * Kevmin left the chat.
[21:34:44] * louiz’ left the chat.
[21:34:47] * louiz’ joined the chat.
[21:41:31] * Robert joined the chat.
[21:44:14] * Florob joined the chat.
[21:58:37] * mlundblad left the chat.
[22:02:07] * stpeter left the chat.
[22:04:57] * Lance Stout joined the chat.
[22:04:57] * Lance Stout left the chat.
[22:06:42] * Lance Stout joined the chat.
[22:12:16] * Robert left the chat.
[22:40:31] * Zash left the chat.
[22:47:28] * Lance Stout left the chat.
[23:20:11] * jkhii left the chat.
[23:23:07] * teo1 left the chat.
[23:23:09] * teo1 joined the chat.
[23:47:48] * hawke left the chat.
[23:50:37] * Tobias left the chat.
[23:51:46] * Lance Stout joined the chat.
[23:51:46] * Lance Stout left the chat.
[23:51:46] * Lance Stout joined the chat.
[23:51:46] * Lance Stout left the chat.
[23:53:55] * Lance Stout joined the chat.
[23:54:42] * Lance Stout left the chat.
[00:08:33] * louiz’ left the chat.
[00:08:36] * louiz’ joined the chat.
[00:12:05] * louiz’ left the chat.
[00:12:08] * louiz’ joined the chat.
[00:15:45] * Lance Stout joined the chat.
[00:15:45] * Lance Stout left the chat.
[00:20:21] * Lance Stout joined the chat.
[00:20:25] * Lance Stout left the chat.
[00:20:25] * Lance Stout joined the chat.
[00:20:25] * Lance Stout left the chat.
[00:25:21] * Lance Stout joined the chat.
[00:26:39] * Lance Stout left the chat.
[01:13:05] * MattJ left the chat.
[01:13:16] * Florob left the chat.
[02:26:34] * louiz’ left the chat.
[02:26:38] * louiz’ joined the chat.
[03:22:24] * Neustradamus joined the chat.