Re: Core dump in fopen

[Andreas Jaeger]

"Rob McDonald" <address@hidden> writes:

> I am working with an application, and I beleive I may have come across a
> problem with glibc.  Unfortunately, the application is quite large, so I
> won't be able to produce a small test case.  However, I will be happy to
> provide as much information as I can, and if there is a test you would like
> me to perform, I will do my best.
>
> During the course of execution, the program core dumps on fopen.  Earlier
> on, the program does many other fopens successfully.  Further, it performs
> the exact same fopen earlier on, without problem.

This looks like some random memory corruption from your application.
The error is definitly not in fopen.

Andreas

> I have observed this failure on the following systems.
>
> RedHat 7.1 libc-2.2.2
> RedHat 7.2 libc-2.2.4
> RedHat 7.3 libc-2.2.5
>
> I also tried my program on a Sun, it ran properly, without dumping core.
>
> SunOS 5.7
>
> I traced the core dump w/ GDB on the RH7.3 machine.  After the dump message,
> I issued the 'up' command twice to track down the source.  The GDB output
> from these operations are as follows.
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x4207add2 in chunk_free () from /lib/i686/libc.so.6
>
> #1  0x4207ac24 in free () from /lib/i686/libc.so.6
>
> #2  0x4206bbe4 in fopen@@GLIBC_2.1 () from /lib/i686/libc.so.6
>
> I am not sure whether this is my problem, or an issue with glibc.  Any help
> in isolating the source of the problem, or resolving the issue will be
> greatly appretiated.
>
> Thanks in advance,
>
>           Rob
>
>
>
> _______________________________________________
> Bug-glibc mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/bug-glibc
>

Andreas
-- 
 Andreas Jaeger
  SuSE Labs address@hidden
   private address@hidden
    http://www.suse.de/~aj