After upgrading my network to run Quakenet
ircu+asuka1.0.5 and Quakenet Operservice 2.36, I have
experienced an endless loop in Newserv. If Newserv is
already linked to the hub (running asuka 1.0.5), and
then Operservice 2.36 links to that hub too, Newserv
goes into an endless loop and pings out after some time.
Logs from Newserv:
Recv: AB S operserv.network 2 0 1063109777 J10 AUA]] +
:This is operservice 2.36
Recv: AU N O 2 1063109775 operserv operserv.network
+odk B]AAAB AUAAB :This is operservice 2.36
- Add new user: AUAAB O!operserv@operserv.network
(+odk) :This is operservice 2.36
- Adding new activeuser O
+ Added new activeuser O
Recv: AU EB
Recv: AU EA
Send: BM EA
Send: BMAAB P #NewServ :End of burst understood, sent EA.
Recv: AUAAB J #network 1062324909
Handling channel join: #network
At this point Newservs stops responding. Operservice
has one channel (#network) it has been set up to join
automatically on link; apparently this is what causes
the loop. I have not tested without the on-link join,
but I suspect it won't happen in that case.
Apart from denying users access to the Newserv
services, this bug also hogs the cpu of the computer
running Newserv.
Logged In: YES
user_id=474843
Might be because of a missing or failing parser for join time
argument. This will be looked into.