Hej List,
may be one can help. A customers server has some strange errors in the
firebird.log:


There are two patterns showing up:
First is:
8<===============================CUT HERE=============================>8
dbserver Wed Jul 11 14:33:14 2012
Operating system call pthread_create failed. Error code 11


dbserver Wed Jul 11 14:33:14 2012
Database: /path/to/data/base.GDB
lock manager error
operating system directive pthread_create failed
Resource temporarily unavailable
8<===============================CUT HERE=============================>8

Second is:
8<===============================CUT HERE=============================>8
dbserver Wed Jul 11 14:34:46 2012
Operating system call pthread_create failed. Error code 11


dbserver Wed Jul 11 14:34:46 2012
SRVR_multi_thread: shutting down due to unhandled exception


dbserver Wed Jul 11 14:34:46 2012
Could not start first worker thread - shutdown server


dbserver Wed Jul 11 14:34:46 2012
SRVR_multi_thread: forcefully disconnecting a port
8<===============================CUT HERE=============================>8


One point not to forget:
First thing i have seen:
a lot of segfaults like
8<===============================CUT HERE=============================>8
Jul 9 12:14:04 dbserver kernel: fb_inet_server[27033]: segfault at 8 ip
0000003430c07659 sp 00007fffa18e4f30 error 6 in
libpthread-2.12.so[3430c00000+17000]
8<===============================CUT HERE=============================>8

That error was fixed by reserving the assigned 8 GB memory to the vm.

Some more background informations:
- OS is CEntOS 6.2 64 bit
- Firebird is 2.5.1
- way of installation is unknown (rpm or tar.gz). If rpm, source is also
unknown, maybe the Firebird one, may be epel
- Server is running in an vmware environment
8 GB RAM reserved
- number of connections made from the application to the database:
roundabout 400

The sad part: i neither do have direct access to the server nor the
customers IT Crew. And on top of that: IT Crew is somewhat clumsy, when
it comes to unix.

tia, Marcus