Nitin H Vaidya
Wed, 1 Jul 1998 09:07:13 -0500 (CDT)
Like I said, the ErrorModel behavior was unpredictable. It
has been working fine lately, but what you pointed out
is, in fact, a bug.
>> From email@example.com Wed Jul 1 08:32:20 1998
>> I pointed out this bug a few months ago. What is your ns
>> version ?.
The bug appears in the version (from Jan'98) we are using.
We apparently forgot to re-check for that bug (Saad works with
me) ... lack of long-term memory -:)
>> Indeed, the bind is not done properly for the variable errPkt_.
>> Therefore, it happens that errPkt_ is not initialized properly resulting
>> in the following behavior : the first loss may not occur if errPkt_ took
>> accidently a large number at the beginning. Since, it is not initialized,
>> the value you start with is just random.
>> On Wed, 1 Jul 1998, Parag at CMU wrote:
>> > Nitin,
>> > I was checking the errmodel.cc files when i came across this piece of code:
>> > bind("rate_", &rate_);
>> > !!! bind("errPkt_", &errByte_); !!!
>> > bind("errByte_", &errByte_);
>> > bind("errTime_", &errTime_);
>> > bind("onlink_", &onlink_);
>> > !!! bind("enable_", &onlink_); !!!
>> > bind("off_mac_", &off_mac_);
>> > I wonder if the 2nd and the 2nd to last lines are really what they are
>> > supposed
>> > to be?
>> > Let me know if you fixed the bug and if this was the cause.
>> > I am also interested in passing a few new variables from my tcl scripts to
>> > the
>> > errmodel.cc file.... what is the right way to do it?
>> > Will just binding as illustrated above do it?
>> > What are the other places to make changes if any?
>> > thanks,
>> > -Parag
>> > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> > Parag Manihar
>> > Master of Science in Information Networking (MSIN),
>> > Carnegie Mellon University (CMU) .
>> Saad Biaz (Abou Youssef) Graduate Research Assistant
>> Texas A&M University Office : (409) 845-5007
>> Department of Computer Science Home : (409) 862-9135
>> College Station, TX, 77843-3112 Fax : (409) 847-8578