*chuckle* Thanks for reminding me about this! I meant to put in a "is this session active?" check and I forgot. Will fix soon.
Daniel On Dec 1, 2005, at 5:27 PM, Norman Rasmussen wrote: > I managed to break it again :-) > > http://www.darkskies.za.net/~norman/jabber/pyaimt_failure.log.gz > > To reproduce: > - edit config file and change login.oscar.aol.com to an unresolvable > address (this just forces the issue, instead of waiting for dns to be > slow and timeout) > - try and login > - execute adhoc command > > Results: > - traceback is immediately output > - transport 'disappears' from server, and never comes back. > > On 11/29/05, Daniel Henninger <[EMAIL PROTECTED]> wrote: >> =D Well . . . that's not really going to get to it's destination, is >> it? Fixed in SVN. >> >> Daniel >> >> On Nov 29, 2005, at 4:23 AM, Norman Rasmussen wrote: >> >>> woot. no crashes on search anymore. >>> >>> Just the error iq is a bit wierd: >>> >>> <route to='aim.jabber.darkskies.za.net' >>> from='aim.jabber.darkskies.za.net'><iq >>> to='aim.jabber.darkskies.za.net' xmlns='jabber:client' >>> from='aim.jabber.darkskies.za.net' id='ae9fa' type='error'><error >>> code='503' type='cancel'><service-unavailable/></error></iq></route> >>> >>> I think 'to' should be set to the user trying to run the search :-) >>> >>> On 11/28/05, Norman Rasmussen <[EMAIL PROTECTED]> wrote: >>>> On 11/28/05, Daniel Henninger <[EMAIL PROTECTED]> wrote: >>>>> What do you think about the password? Originally, I had a >>>>> separate >>>>> config option for sasl username and password, but if I ask for a >>>>> separate password from the "secret", then "secret" is never >>>>> used. So >>>>> I figured, why am I bothering with that? Why don't I just use >>>>> "secret" as the sasl password. What do you think about this? I >>>>> could change "useSASL" into a "saslUser" or something like that >>>>> and >>>>> change the check to be "is it empty or not". >>>> >>>> yes, this is exactly what I'm doing with the yahoo transport, (if >>>> username != blank, enable JD2 and use username to auth) >>>> >>>>> Hrm. Well ok. I'm used to a bazillion things showing up in disco >>>>> and so I didn't even think that it would be "prettier" if it were >>>>> tucked away. *ponder* >>>> :-) >>>> >>>>>> mmm, was the searching while offline bug, pretty easy to >>>>>> duplicate. >>>>>> The exception seems to cause the transport to disconnect and >>>>>> reconnect, but it seems to forget to re-auth, etc. >>>>> >>>>> I think, if it's not using the jabber 2 component protocol, >>>>> that it >>>>> does reconnect properly after a crash like that. I must have >>>>> missed >>>>> something. Either way, I'll look into it. >>>> if you want tcpflow dumps, it's easy to make them :-) >>>> >>>>> BTW, I find it amusing that I left "jabber2.vorpalcloud.org" in >>>>> main.py in the jabber component stuff. ;) Dooh. >>>> trying to 'lock down' access eh :-P >>>> >>>> -- >>>> - Norman Rasmussen >>>> - Email: [EMAIL PROTECTED] >>>> - Home page: http://norman.rasmussen.co.za/ >>>> >>> >>> >>> -- >>> - Norman Rasmussen >>> - Email: [EMAIL PROTECTED] >>> - Home page: http://norman.rasmussen.co.za/ >>> _______________________________________________ >>> py-transports mailing list >>> py-transports@blathersource.org >>> http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports >>> >>> >> >> _______________________________________________ >> py-transports mailing list >> py-transports@blathersource.org >> http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports >> > > > -- > - Norman Rasmussen > - Email: [EMAIL PROTECTED] > - Home page: http://norman.rasmussen.co.za/ > _______________________________________________ > py-transports mailing list > py-transports@blathersource.org > http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports > >
