Problems running hlds (Halflife)

Discuss usability issues, general maintenance, and general support issues for a grsecurity-enabled system.

Problems running hlds (Halflife)

Postby michael » Mon May 19, 2003 8:05 am

Hello,

I get the following messages when running and trying to connect to the Halflife server:

May 18 23:40:56 spring kernel: UDP: bad checksum. From 68.82.64.18:3583 to 213.133.100.139:27015 ulen 19
May 18 23:47:04 spring kernel: PAX: From 80.142.234.118: terminating task: /home/halflife/dod/engine_amd.so(engine_amd.so):764, uid/euid: 1001/1001, EIP: 00000001, ESP: 5D067834
May 18 23:47:04 spring kernel: PAX: bytes at EIP: <invalid address>.
May 18 23:47:04 spring kernel: grsec: From 80.142.234.118: attempted resource overstep by requesting 4096 for RLIMIT_CORE against limit 0 by (engine_amd.so:764) UID(1001) EUID(1001), parent (bash:27164) UID(1001) EUID(1001)

What is RLIMIT_CORE? Has anyone an idea howto solve that problem?
Im using kernel version 2.4.20 with grsecurity 1.99e patch.

Bye,

Michael
michael
 
Posts: 1
Joined: Mon May 19, 2003 8:00 am

Re: Problems running hlds (Halflife)

Postby PaX Team » Mon May 19, 2003 10:10 am

michael wrote:May 18 23:47:04 spring kernel: PAX: From 80.142.234.118: terminating task: /home/halflife/dod/engine_amd.so(engine_amd.so):764, uid/euid: 1001/1001, EIP: 00000001, ESP: 5D067834
this looks like a NULL (function) pointer dereference or something similar and hence an application bug.
What is RLIMIT_CORE? Has anyone an idea howto solve that problem?
Im using kernel version 2.4.20 with grsecurity 1.99e patch.
RLIMIT_CORE is a resource limit that controls coredumping. try 'ulimit -c' in a shell and you'll see what it's set to by default on your system (a 0 means that core file dumping is turned off). as for fixing the problem, someone will have to debug the code to see where/why the NULL dereference occurs, you should probably contact the author of engine_amd.so.
PaX Team
 
Posts: 2310
Joined: Mon Mar 18, 2002 4:35 pm


Return to grsecurity support