Core dump while call void*std::__node_alloc<true,0>::_M_allocate(unsigned l

Hi,

My process runs on multi-CPU solaris 10 platform.

But some times it was coredumped and the pstack file always has:

void*std::__node_alloc<true,0>::_M_allocate(unsigned long)

For details, please refer to the below:

terminated by signal SEGV (Segmentation Fault).

ffffffff789c9ad0 kill (b, ffffffff7fffc830, ffffffff7fffc550, ffffffff78aeb20c, 520, 0) + 8

ffffffff789c9030 __sighndlr (b, ffffffff7fffc830, ffffffff7fffc550, ffffffff7cdbe670, 0, 0) + c

ffffffff789bd7a4 call_user_handler (ffffffff7de02000, ffffffff7de02000, ffffffff7fffc550, ffffffff, 0, 0) + 3e0

ffffffff7916a27c void*std::__node_alloc<true,0>::_M_allocate(unsigned long) (28, 4, 27, ffffffff79319d90, ffffffff79321d48, 100000102153708) + 54

I donot know what is the root cause!

What is your opinions?

Thanks!

BR,

Eric

[895 byte] By [wliangya] at [2007-11-27 2:24:54]