Logs for jdev@conference.jabber.org

[00:04:13] * ralphm left the chat.
[00:05:25] * tato joined the chat.
[00:32:34] * lolo joined the chat.
[00:32:34] * lolo left the chat.
[00:53:15] * stpeter joined the chat.
[00:53:19] * stpeter left the chat.
[00:59:59] * Lance left the chat.
[01:13:44] * Lance joined the chat.
[01:15:19] * jcea left the chat.
[01:35:33] * ralphm joined the chat.
[01:37:10] * deryni joined the chat.
[02:11:52] * Flow left the chat.
[02:27:38] * Lance joined the chat.
[02:33:02] * Lance left the chat.
[02:35:31] * Lance joined the chat.
[02:37:29] * Lance left the chat.
[02:46:24] * waqas left the chat.
[02:50:18] * Tobias joined the chat.
[02:52:05] * tato left the chat.
[02:55:44] * Tobias left the chat.
[03:00:07] * ralphm left the chat.
[03:58:25] * ermine joined the chat.
[04:32:48] * Lance left the chat.
[04:46:01] * Lance joined the chat.
[05:09:35] * Asterix joined the chat.
[05:33:32] * Lance left the chat.
[05:48:02] * Asterix left the chat.
[05:51:26] * timothee.jaussoin joined the chat.
[06:25:15] * Alex joined the chat.
[06:42:50] * Neustradamus left the chat.
[07:04:52] <> There doesn't seem to be, to me. There's a stream feature advertising starttls, but that's entirely different.
[07:11:48] * Lance joined the chat.
[07:47:36] * Lance left the chat.
[08:02:12] * 0xAFFE joined the chat.
[08:02:46] * Tobias left the chat.
[08:15:17] * Zash joined the chat.
[08:31:40] * Lance joined the chat.
[08:31:54] * Tobias joined the chat.
[08:33:47] * ermine left the chat.
[08:36:56] * ermine joined the chat.
[08:58:05] * dwd joined the chat.
[09:19:57] * Flow joined the chat.
[09:26:23] * jabberjocke joined the chat.
[09:32:54] * leo joined the chat.
[09:34:04] * Lance left the chat.
[09:36:04] * leo left the chat.
[09:56:22] * Hermitifier left the chat.
[10:12:37] * Tobias left the chat.
[10:12:41] * Tobias joined the chat.
[10:36:29] * Tobias left the chat.
[10:36:41] * Tobias joined the chat.
[10:46:20] * aaa joined the chat.
[10:46:35] * aaa left the chat.
[10:54:10] * jcea joined the chat.
[11:02:05] * timothee.jaussoin left the chat.
[11:47:15] * timothee.jaussoin joined the chat.
[12:19:50] * KevWalke left the chat.
[12:42:09] * stpeter joined the chat.
[12:42:23] * stpeter left the chat.
[12:53:32] * lloyd.watkin joined the chat.
[13:18:50] * jabberjocke left the chat.
[13:32:29] <> And because everyone should be able to share in the joy, not just dwd, here's an artist's impression of what the SREAM-SHA-1 mech might look like:
[13:32:30] <> C:sha-1 S:Louder C:Sha-1 S:LOUDER C:SHA-1 S: LOUDER! C: SHA ONE!!!!!111onebbq
[13:44:50] * Alex left the chat.
[13:48:20] <> *SCREAM-SHA-1
[13:48:27] <> Excellent time to typo.
[13:52:01] * lloyd.watkin left the chat.
[13:54:33] <> Kev: I already feared you were trying to suggest CREAM-SHA-1
[13:54:50] <> There was a mail on jdev@ where someone mentioned SCREAM-SHA-1
[13:56:39] * timothee.jaussoin left the chat.
[13:56:59] * timothee.jaussoin joined the chat.
[14:04:20] <> Must be very secure.
[14:05:11] <> You must assume people will eavesdrop on the connection, though.
[14:06:17] <> I'
[14:06:22] <> It's a possibility.
[14:06:34] <> what's the status on the log file syncing?
[14:08:27] <> MUC logs?
[14:08:30] <> Someone needs to do it.
[14:08:32] <> That's about it.
[14:09:46] <> yup
[14:09:48] <> MUC logs
[14:09:57] <> or IPv6 @ j.o?
[14:10:49] * timothee.jaussoin left the chat.
[14:26:34] * jabberjocke joined the chat.
[14:44:00] * Zash left the chat.
[14:45:03] * darkrain_ joined the chat.
[14:45:17] <> I think we can safely enable IPv6 on j.org in the same way as we did before. The service is already reachable over IPv6, we just didn't add the IPv6-only host record to the SRV set yet.
[14:45:34] <> I don't see a reason not to.
[14:45:50] <> I'll add the records in a minute.
[14:46:22] <> I connect to it over IPv6 all the time at the moment.
[14:48:16] * xnyhps left the chat.
[14:48:35] <> Updated the SRV record.
[14:48:51] <> +s
[14:49:00] <> Ta.
[14:49:12] <> \o/ wooooo
[14:49:42] <> Great news for all people who are on an IPv6-only network ;)
[14:50:17] * xnyhps joined the chat.
[14:50:25] <> you mean for the guy
[14:54:37] <> :)
[14:55:24] <> There's a guy?
[14:55:52] <> somewhere i bet
[14:56:44] * stpeter joined the chat.
[14:56:50] * stpeter left the chat.
[14:56:50] * stpeter joined the chat.
[15:02:40] * Tobias left the chat.
[15:16:23] * waqas joined the chat.
[15:19:48] * graham joined the chat.
[15:22:08] <> If my buddies server dies, how / when do I discover that he's no longer online?
[15:22:51] <> The client / server can't send the unavailable. Does my server remember which server my buddies are using? Or send some presence probes?
[15:23:24] <> graham: Some servers have been known to resend probes every once in a while to detect that
[15:23:29] <> I think jabber.org does that
[15:23:38] <> Most servers don't
[15:24:04] <> waqas: If the server doesn't, I don't discover the presence change until I try sending a message?
[15:26:22] <> You don't discover presence changes until the remote server sends (or fails to send) a presence. That happens when the remote user changes their presence, or when your server sends a probe. Your server sends a probe when you login, or some servers send periodic probes, etc.
[15:27:19] <> Thank you
[15:30:00] * Tobias joined the chat.
[15:33:51] * Zash joined the chat.
[15:39:22] <> PSA would also solve this, and better, but I don't know of any implementations in either servers or clients.
[15:39:51] <> PSA?
[15:40:30] <> Presence-state-annotations
[15:40:49] <> waqas: http://xmpp.org/extensions/xep-0310.html
[15:41:33] <> Thanks, I somehow missed ever seeing that one
[15:44:04] <> PSA sounds like exactly what I need
[15:44:20] <> I'd started sketching something along those lines
[15:45:03] <> Our specific problem is n clustered servers. If one of them goes down, clients will re-connect to a different one. If they don't specify a resource we generate it, so we now have two connections for that user.
[15:45:48] <> The servers could track who is on which node, and use PSA when that node goes away
[15:46:32] <> Well, in clustered servers you're fine.
[15:46:43] <> Because the remaining nodes will be able to tell you that the user went offline.
[15:46:57] <> Scratch that, no they won't.
[15:47:31] <> If each node knows which sessions the other nodes have ..
[15:47:36] <> yes
[15:47:50] <> But then, which node is responsible for sending notifications?
[15:47:54] <> Then they can use PSA to tell their clients about it.
[15:48:05] <> Then could all send node-local only
[15:48:23] <graham> Hiya - I put that rapidxml fork onto my github; http://github/dwd/rapidxml I think.
[15:49:17] <> Zash: They can't, because they don't know if it's them or the other node that's 'offline' :)
[15:50:04] <> I was thinking we pretend there's no network partitions
[15:50:13] <> so just deal with server crash
[15:50:26] <> I suspect that's by far the common case, actually.
[15:51:05] <> I've seen far more network partition / split brain issues than crashes.
[15:51:19] <> Servers are all in the same data center, same rack probably, so I'd hope we're mostly partition free
[15:52:44] <> There's nothing quite like optimism :)
[15:53:01] <> Kev: Are there any notes on why PSA was deferred? i.e technical problems implementing it, or just lack of momentum?
[15:53:22] <> Deferred doesn't mean problems, it means it wasn't updated or advanced to Draft in a period of time.
[15:53:25] <> graham: Same reason any XEP is deferred.
[15:53:40] <> Which often implies a lack of momentum, but not always.
[15:53:46] <> In this case, it did (I'm the author).
[15:53:53] <> ok, thanks
[15:54:42] <> Largely because the reason for it was trying to solve issues I was seeing at work with MUC across dodgy networks, and those are better sorted with FMUC.
[15:55:36] <> There are plenty of cases that PSA's good for, I just didn't have as great a need as for FMUC.
[15:55:51] * Tobias left the chat.
[15:56:00] <> What is FMUC?
[15:56:03] <> 289
[15:56:06] <> Federated MUC.
[15:57:10] <> Allowing MUCs to join each other, so you can connect to a local node, continue working in the case of netsplits, and reduce bandwidth between the nodes.
[15:57:46] * Tobias joined the chat.
[15:58:12] <> oh, that's good to know about, thanks
[16:00:51] <> It's not the best approach for the Internet, but for the sort of environments we need to supply for, it's quite handy.
[16:06:46] * naw joined the chat.
[16:07:39] * Lance joined the chat.
[16:12:52] * Tobias left the chat.
[16:13:49] * Tobias joined the chat.
[16:17:24] * tato joined the chat.
[16:39:10] * Tobias left the chat.
[16:40:06] * tato left the chat.
[16:41:57] * tato joined the chat.
[17:07:29] * 0xAFFE left the chat.
[17:16:11] * Zash left the chat.
[17:20:33] * Zash joined the chat.
[17:33:13] * ralphm joined the chat.
[17:40:17] * stpeter left the chat.
[18:00:27] * tato left the chat.
[18:03:29] * xnyhps joined the chat.
[18:21:51] * Zash left the chat.
[18:24:31] * Lance left the chat.
[18:32:56] * ermine left the chat.
[19:05:37] * Neustradamus joined the chat.
[19:17:58] * naw left the chat.
[19:19:54] * Lance joined the chat.
[19:33:14] * stpeter joined the chat.
[19:33:23] * stpeter left the chat.
[19:33:23] * stpeter joined the chat.
[19:34:11] * Flow left the chat.
[21:48:00] * Florob joined the chat.
[21:50:22] * Florob left the chat.
[21:59:12] * stpeter left the chat.
[22:18:58] * xnyhps left the chat.
[22:51:42] * graham left the chat.
[23:29:34] * Tobias joined the chat.
[23:44:24] * Tobias left the chat.
[23:46:31] * Tobias joined the chat.
[23:48:12] * Flow joined the chat.
[23:57:59] * jcea left the chat.