2.5.60 problems with dbench and unkillable processes.

From: Steven Cole (elenstev@mesatop.com)
Date: Tue Feb 11 2003 - 17:40:22 EST


With 2.5.60, dbench starts, prints "1 clients started", but never prints
any "..." and doesn't finish.

With 2.5.59-mm10, dbench runs normally.

I ran this:
ps -ewo user,pid,priority,%cpu,stat,command,wchan
with the 2.5.60 System.map copied to /boot, and got this:

root 1103 15 0.1 S /usr/sbin/sshd schedule_timeout
steven 1104 15 0.0 S -bash wait4
steven 1142 15 0.0 S ./dbench 1 wait4
steven 1143 16 0.0 S ./dbench 1 pause
steven 1153 16 0.0 S /bin/sh /home/st wait4
steven 1154 16 0.0 R ps -ewo user,pid -

I control-c'ed the dbench 1, and started ./dbench 4, and got this:

root 1103 15 0.0 S /usr/sbin/sshd schedule_timeout
steven 1104 15 0.0 S -bash wait4
steven 1143 16 0.0 S ./dbench 1 pause
steven 1160 15 0.0 S ./dbench 4 wait4
steven 1161 16 0.0 S ./dbench 4 pause
steven 1162 16 0.0 S ./dbench 4 pause
steven 1163 16 0.0 S ./dbench 4 pause
steven 1164 16 0.0 S ./dbench 4 pause
steven 1166 16 0.0 S /bin/sh /home/st wait4
steven 1167 17 0.0 R ps -ewo user,pid -

That process 1143 is now unkillable with kill -9.

After control-c-ing the ./dbench 4, I now have 4
dbench 4 processes which are unkillable with kill -9.

The system is 2-way PIII, kernel is SMP, PREEMPT.

Steven

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sat Feb 15 2003 - 22:00:36 EST