server进程， core dump之前有fork行为

Program received signal SIGSEGV, Segmentation fault.
core dump 的stack 如下:
#0 0x40b36a38 in _int_malloc () from /lib/tls/libc.so.6
#1 0x40b37f74 in malloc () from /lib/tls/libc.so.6
#2 0x401388fe in operator new () from /usr/lib/libstdc++.so.5
#3 0x40138a3d in operator new[] () from /usr/lib/libstdc++.so.5
#4 0x085afbdd in BRSClient::write (this=0x48d1a930, data=@0xbfffd618, len=143)
at BRSClient.cc:391
。。。。
相应的代码：
386 BString logStr ;
387
388 char * tmp = new char[143];
389 delete [] tmp;
390
391 char * tmp2 = new char[144];
392 delete [] tmp2;

初步怀疑是进程在该步操作之前，进程heap被损坏， 但不排除其他可能性.
实在找不到问题所在..

代码为移植代码，非本人所写，代码茫茫， 无处查询，望各位大侠指点一二..

难倒原来的代码就是new了马上delete?希望发全一些

操作系统是什么