[OpenAFS-devel] pr_Initialize falis when called within a simple application

[John Hayes]

Ok,

I'll have to check that out. Just for reference when I use klog linked
agains the libraries with debugger code the following output is produced:
jhh@jhh:~/work/projects/afs_support> klog jhh
Password:
in rx1
got mem!
in rx2
in rx3
in rx4
leaving rx!
in rx1
got mem!
in rx2
in rx3
in rx4
leaving rx!
in rx1
got mem!
in rx2
in rx3
in rx4
leaving rx!

As compared to the output produced when I call my test application:
hh@jhh:~/work/projects/afs_support> ./afstest jhh apw test test
R = 0
in rx1
got mem!
in rx2
in rx3
in rx4
leaving rx!
got conn error!
in newTC - calling rxnc
in rx1
got mem!
in rx2
in rx3
in rx4
leaving rx!
returned from calling rxnc
got conn error!
in newTC - calling rxnc
in rx1
got mem!
in rx2
Segmentation fault (core dumped)

My test program reports a connection error and then calls for a new
connection twice before failing wile the klog call displays no such
performance.

Thanks,
John Hayes



>> The debug output lists "in rx1" 3X before failing, and that after
>> printing
>> "in rx2"". It never completes the third "in rx3" message before
>> producing
>> the segfault.
>>
>> The debugger messages being displayed around the vicinity of the error
>> take place within the module "rx/rx.c" on line number 758 (that is with
>> the inclusion of some debug print statements) right here:
>> printf ( "in rx2\n" );
>>    RXS_NewConnection(securityObject, conn);    <------
>> printf ( "in rx3\n" );
>>
>> Now this line ( RXS_NewConnection(securityObject, conn) ) ) appears to
>> be
>> a  macro which I haven't tried to make much sense out of yet.
>
> It chooses which security method to call the new connection function in.
> Choices are rxnull or rxkad. I can't remember but I think the only part of
> the ka suite which uses rxkad explicitly is password changing.