[reSIProcate] TransportSelector Route Bug

[Alan Hawrylyshen]

On Jun 14, 2004, at 10:19, Jason Fischl wrote:

> Yes. This is right. Apparently, somebody added in the following code 
> (also
> incorrectly) into TransportSelector:
>       else if (msg->exists(h_Routes) && !msg->header(h_Routes).empty())
>       {
>          // put this into the target, in case the send later fails, so 
> we
> don't
>          // lose the target
>          msg->setForceTarget(msg->header(h_Routes).front().uri());
>          DebugLog (<< "Looking up dns entries (from route) for " <<
> msg->getForceTarget());
>          result = mDns.lookup(msg->getForceTarget(), handler);
>       }

Hmm.. Since my name appears on a few of these lines -- I'm wondering 
what, specifically needs to be done here and elsewhere to fix the 
problem.  I remember discussing this with you on the phone and not 
really coming to a decent conclusion.  I'm concerned about how this is 
all going to work. So what are the proposed solutions?

This sounds like a problem in more than one location.

A

a l a n a t j a s o m i d o t c o m