Bad page state in process apache2 (Server reboots frequently)

Zeest · 05-16-2007, 10:47 PM

Code:

Kernel Version	2.6.18-4-686 (SMP)
Distro Name     Debian 4.0
Model	        Intel(R) Pentium(R) 4 CPU 2.40GHz

I've installed the default apache 2, mysql, php using aptitude.

The server load is not at all high, around 1-1.5 max. The apache processes running are at an avg 60-80. It has 1GB ram and Disk Swap usage is 0kB

My server keeps rebooting every now and then. These are the messages I get in the ssh session just before the reboot..

Code:

Message from syslogd@xxx at xxx
kernel: Bad page state in process 'apache2'
kernel: page:c17bc0a0 flags:0xc0000000 mapping:00800040 mapcount:0 count:0
kernel: Trying to fix it up, but a reboot is needed
kernel: Backtrace:

Sometimes the server just dies out on me. Can't connect through ssh, have to restart it using APC

This is the log from the last such crash

Code:

kernel: BUG: unable to handle kernel paging request at virtual address badab8fc
kernel:  printing eip:
kernel: c01572d3
kernel: *pde = 00000000
kernel: Oops: 0000 [#1]

To add to that it also loses track of time on next reboot, until ntpd synchronises the time.

Could this be some hardware problem? Like faulty ram or something? Is there any program that can detect whether a ram is faulty?

Any help greatly appreciated.

osor · 05-17-2007, 08:47 PM

Try running memtest86 for a few hours (you run it from a boot disk) to check your ram.

Zeest · 05-17-2007, 11:13 PM

I only have ssh access to the server