Im not sure, but I think this error can be corrected by applying this patch
that Ive described here:
http://mailman.isi.edu/pipermail/ns-users/2004-September/044531.html
(also see my nest post about FHMIP for ns2.31)
Pedro Vale Estrela
-Original Message-
From: shayma sinan [mailto:[EMAIL PROTECTED]
Sent: quinta-feira, 22 de Março de 2007 2:05
To: [EMAIL PROTECTED]
Subject: Re: FHMIP - anyone got the FHMIP extension to work? (coded by
Robert Hsieh)
Hi,
I am also facing a problem in running FHMIP package under ns2.28
i got the same error mentioned here
http://mailman.isi.edu/pipermail/ns-users/2005-August/051595.html
without modifying (simula.tcl)
hope you can find somebody to help... good luck!
regards,
Shayma
Pedro Vale Estrela <[EMAIL PROTECTED]> wrote:
Hi all,
Previously, I've asked in the NS2 list if anyone had a copy of Robert
Hsieh's FHMIP NS2 extension for NS2.1b7; with the reply, I've updated my NS2
mobility page with the available source code:
http://tagus.inesc-id.pt/~pestrela/ns2/mobility.html#_Toc147660119
Currently, I'm trying to upgrade and put this extension to work.
I've already got it compiling in NS2.31, but when I run the __unmodified__
example (simula.tcl), I got this failed assertion:
BS Setting up encapsulate mechanism [1.0.1]->[2.0.0]
7.093759 HA 1.0.0 updated (FA 2.0.0), packet 32
7.150404 BS 3.1.0 received reply from BS 1.0.0, packet 32
mipmh->coa_ 2.0.0 nodeaddr 3.1.0
ns: mobile/mip-reg.cc:261: virtual void MIPBSAgent::recv(Packet*, Handler*):
Assertion `miph->coa_ == nodeaddr' failed.
Aborted (core dumped)
This happens in this part of the code:
case MIPT_REG_REPLY:
{
#ifndef MIP_QUIET
printf("%f BS %s received reply from BS %s, packet %i\n",
Scheduler::instance().clock(),
Address::instance().print_nodeaddr(addr()),
Address::instance().print_nodeaddr(iph->saddr()), ch->uid());
// JCW
#endif
assert(miph->coa_ == nodeaddr); // crashes here
tcl.evalf("%s get-link %d %d", name_, nodeaddr, miph->haddr_);
//
// XXX hacking mobileip. all this should go away
// when mobileIP for sun-wired model is no longer reqd.
//
...
Thus, my questions are:
a) has somebody got an working FHMIP extension?
b) Is it necessary to modify simula.tcl or mip-reg.cc use another simulation
scenario, as to avoid this bug?
If yes, please send me the modified code and I'll upgrade it to ns2.31, and
give credit to the person in my NS2 mobility page.
Pedro Vale Estrela
_
Web email has come of age. Don't settle for less than the All
New
Yahoo! Mail.
