[omniORB] core dump on Solaris using omnithreads

[Duncan Grisby]

On Monday 1 October, vjo@infosynergi.no wrote:

> Thanx for the tip. Only thing is that I get the same problem with
> 3.0.1 as well...  Also using Suns C++ compiler, Sun WorkShop 6
> update 2, gave the same result.
> 
> Running my application without using multithreading everything works fine.
> 
> A typical backtrace on the core file from gdb looks like this
> 
> (gdb) bt
> #0  0xfeb456f0 in _smalloc () from /usr/lib/libc.so.1
> #1  0xfeb45734 in malloc () from /usr/lib/libc.so.1
> #2  0xff3241f0 in c::c () from /usr/lib/libCrun.so.1

It's dying in malloc(). The usual cause for that is that you have
corrupted the heap somewhere. If you have Purify, try running your
code under that.

If you don't have Purify, njamd works well on Linux. It claims to work
on Solaris too. See

  http://fscked.org/proj/njamd.shtml

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --