topic SERVER-HANG in Operating System - Linux

SERVER-HANG

prash_1 — Tue, 20 Mar 2007 09:40:44 GMT

Dear gurus,

I am using suse linux 2.6.16.

machine having Xeon-dual Processor.
oracle 10g & sap installed in the system.

After finishing installation it worked @10hr
then it get hang. when I restarted server & checked message file it contain following error-----------
sapinstexe[5358]: segfault at 0000000032833664 rip 00002aaaabbb2d10 rsp 00007fffac0d0b88 error 6
kernel: klogd 1.4.1, ---------- state change ----------

what would be the Probably thing to hang.

Thanks in adv.

Re: SERVER-HANG

Steven E. Protter — Tue, 20 Mar 2007 09:52:48 GMT

Shalom,

A number of interesting possibilities.

segfault may indicate a memory problem. Perhaps a bad chip. It also can indicate the presence of software generating the sgmentation fault.

Take a peak.
http://gcc.gnu.org/ml/gcc-bugs/2000-10/msg00259.html

Is there more in /var/log/messages ??

SEP

Re: SERVER-HANG

prash_1 — Tue, 20 Mar 2007 10:09:34 GMT

Thanks,

I am giving some lines in message I cant get what is that. & that is before 4 hr when server hang.
rcpowersaved: CPU frequency scaling is not supported by your processor.
rcpowersaved: enter 'CPUFREQ_ENABLED=no' in /etc/powersave/cpufreq to avoid this warning.
rcpowersaved: Cannot load cpufreq governors - No cpufreq driver available

rcpowersaved: s2ram does not know your machine. See 's2ram -i' for details. (127)
rcpowersaved: Use SUSPEND2RAM_FORCE=yes to override this detection

Re: SERVER-HANG

Steven E. Protter — Tue, 20 Mar 2007 10:14:48 GMT

Shalom,

I'd continue this line of investigation.

Check that the machine is current on bios and firmware. The cpu scaling issue could really whack any Linux operating system.

Also, if possible start the system with less memory to try and detect a bad chip.

Also, bad software can cause this. Segmentation faults can occur when one program tries to grab and use memory already allocated to another program.

SEP