I sent a few mails earlier, but they bounced it seems, to now I try again! :)
This morning I was getting errors with memory hungry symptoms. Even with only one or two active users it is using over 80% of available memory. I started debug in the config, but there was no activity. Still the memory is committed. When I try to restart the transport by issuing the "kill pid" the process does not die. I have to issue a "kill -9 pid" before it releases. This never happened before. Right after I freshly start the transport I am able to log on the transport from a client, but then I get this error in the debug: [2006-02-07 09:28:01] INFO :: [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]> :: :: sendPresence :: Session :: {'status': u'Om du kj\xf8per kurvstol fra IKEA, m\xe5 du da flette den sammen selv?', 'show': None, 'self': 'instance', 'fro': '[EMAIL PROTECTED]/msn' <mailto:[EMAIL PROTECTED]/msn%27>, 'avatarHash': '1ff1a923d502ca3574324d71f0b80a53f17dcda1', 'ptype': None, 'priority': None, 'to': u'[EMAIL PROTECTED]' <mailto:[EMAIL PROTECTED]>, 'nickname': u'Brodern', 'payload': [<twisted.words.xish.domish.Element object at 0xb756ddcc>]} [2006-02-07 09:28:01] Traceback - ERROR in checkMessage: 400 [2006-02-07 09:28:02] Traceback - ERROR in checkMessage: maximum recursion depth exceeded in cmp [2006-02-07 09:28:04] INFO :: [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]> :: :: contactPersonalChanged :: msn.msnw.NotificationClient :: {'personal': 'M\xc3\xa6rsjm\xc3\xa6sjef\xc3\xa6ra?', 'self': 'instance', 'userHandle': '[EMAIL PROTECTED]' <mailto:[EMAIL PROTECTED]>} [2006-02-07 09:28:04] INFO :: [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]> :: :: contactStatusChanged :: glue.LegacyConnection :: {'remoteUser': '[EMAIL PROTECTED]' <mailto:[EMAIL PROTECTED]>, 'self': 'instance'} [2006-02-07 09:28:04] INFO :: [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]> :: :: sendPresence :: Session :: {'status': u'M\xe6rsjm\xe6sjef\xe6ra?', 'show': None, 'self': 'instance', 'fro': '[EMAIL PROTECTED]/msn' <mailto:[EMAIL PROTECTED]/msn%27>, 'avatarHash': '51973ca31135011cc71f68f6ad026c22f4dabcee', 'ptype': None, 'priority': None, 'to': u'[EMAIL PROTECTED]' <mailto:[EMAIL PROTECTED]>, 'nickname': u'Calle', 'payload': [<twisted.words.xish.domish.Element object at 0xb756fe0c>]} [2006-02-07 09:28:04] Traceback - ERROR in checkMessage: maximum recursion depth exceeded in cmp After this error I cannot get back on the transport. It just never answers. I thoght the problem lay with poll but I tried enabling poll without luck. *** a few hours later *** After a few restarts I opted to delete the xml file for my own user in the spool dir. I restarted and re-registered on the transport, but at logon I get 5-6 contacts up, then I get a 400 error TB: [2006-02-07 11:46:32] Traceback - ERROR in checkMessage: [2006-02-07 11:46:33] Traceback - ERROR in checkMessage: [2006-02-07 11:46:33] Traceback - ERROR in checkMessage: [2006-02-07 11:46:39] Traceback - ERROR in checkMessage: [2006-02-07 11:46:42] Traceback - ERROR in checkMessage: 400 I also tried to restart the old transport (0.10), to see what happened then. Then my account logged on, then off, then on, then off, on, off, on, off etc.. etc... etc.. Now I reccon this error has got to do with my account, and my account stalls the entire transport for some reason. Latest test I have done is to restart the transport and not log on. I have started the svn transport, and just let the other users use it. So far it seems to work, no error thus far. Any ideas what to do? I have tried to log on with messenger and that works. Any tips greatly appreciated :) James: about the proxy65, I will install this, and thanks for the walktrough, very useful information! -stian -------------- next part -------------- An HTML attachment was scrubbed... URL: http://modevia.com/pipermail/py-transports/attachments/20060207/63f5c753/attachment.html From [EMAIL PROTECTED] Tue Feb 7 12:07:21 2006 From: [EMAIL PROTECTED] (Daniel Henninger) Date: Tue Feb 7 12:07:32 2006 Subject: [py-transports] PyAIM-t and PyICQ-t "twisted.internet.error.ConnectionLost" - why? In-Reply-To: <[EMAIL PROTECTED]> References: <[EMAIL PROTECTED]> Message-ID: <[EMAIL PROTECTED]> I've never found a precise cause for this. It typically stems from the OSCAR servers severing your connection for whatever reason ... either they went down, or there was a problem with network connectivity between PyAIM/ICQ and them, or something else happened. Either way, I don't have anything concrete about them yet. Daniel On Feb 7, 2006, at 4:09 AM, Michael Vogel wrote: > Hi! > > I'm using the latest stable version of PyAIM-t and a working SVN > version of PyICQ-t. > > Since some days I receive the following messages from the transport: > > ------ > AIM connection lost! Reason: [Failure instance: Traceback: > twisted.internet.error.ConnectionLost, Connection to the other side > was lost in a non-clean fashion. > ] > > ------ > ICQ connection lost! Reason: [Failure instance: Traceback: > twisted.internet.error.ConnectionLost, Connection to the other side > was lost in a non-clean fashion. > ] > > > Where lies the problem? The transport automatically reconnects > after that. > > Michael > _______________________________________________ > py-transports mailing list > py-transports@blathersource.org > http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports > >
