lkml.org 
[lkml]   [2000]   [Mar]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: memory tester and BP6 problems
Date
Quoting John Newman:
:Can anyone recommend a good app to stress test the memory in my
:Linux box? I'm getting these strange lockups in a BP6 where
:the box locks hard, usually emitting a constant beep when it's
:locked. When I reboot it it usually doesn't come up for about
:five minutes but instead makes a beep pattern that is alternating
:a high and lower pitch beep continuously.

Funny you mention this, because it happened to me this morning for the
first time in 6 months, with 2.2.14.

:usually stays up for anywhere from a day to a week. Not
:overclocked, happens with any kernel config and I've got my ram
:settings at the most conservative. Also, the temperature stays
:anywhere from 29C - 34C on the CPU's and in the system as reported
:by lm_sensors and looking at logs of the temps it is never overly
:hot when it dies.

Dual Celeron config, not overclocked. I don't have lm_sensors installed,
but when I rebooted I went to the BIOS and looked at the temperature.
It was 50C, as usual (I have 2 fans on the Celerons, but no thermal
glue, so I guess the heat does not dissipate properly--it's stable though).

I have put mem=288M in my loadlin paramerters, which is correct. I rebooted
with mem=287M, and we'll see if I get new lockups.

It's not the first time I see those lockups, and they always happen when my
machine is having many windows opened for a while (4,5 Netscape windows,
VMWare running with 96 Mb configured, around 20 xterms).
Usually nothing in the logs, but this morning, I saw this (not on my
SMP machine, but on a remote syslog) -- the SMP machine is called "nice",
after the French town Nice, not the English adjective.

------
nice kernel: Unable to handle kernel NULL pointer dereference at
nice kernel: current->tss.cr3 = 0d7e1000, %cr3 = 0d7e1000
nice kernel: *pde = 00000000
nice kernel: Oops: 0000
nice kernel: CPU: 1
nice kernel: EIP: 0010:[smp_call_function_interrupt+8/64]
nice kernel: EFLAGS: 00013046
nice kernel: eax: 00000000 ebx: 0000006a ecx: 08259384 edx: 00000018
nice kernel: esi: 00000000 edi: 082fcdf0 ebp: bffff614 esp: cd793fbc
nice kernel: ds: 0018 es: 0018 ss: 0018
nice kernel: Process vmware (pid: 22828, process nr: 68, stackpage=cd793000)
nice kernel: Stack: 082fcdf0 c010ad5e 0000006a 08259384 08259384 00000000 082fcdf0
bffff614
nice kernel: 63c4be29 0000002b 0000002b ffffffff 080d10d1 00000023 00003246
bffff568
nice kernel: 0000002b
nice kernel: Call Trace: [call_function_interrupt+26/32]
nice kernel: Code: 8b 30 8b 58 04 8b 78 10 e8 ab fa ff ff 8b 15 cc a0 20 c0 8d
nice kernel: PI wait (CPU#0)
nice kernel: stuck on TLB IPI wait (CPU#0)
------
And that's it. The machine locked up, because I don't see any new indication
of periodic cron jobs ("rmmod -a") being run on the machine past that date,
on the syslog server.

It looks like a spinlock deadlock problem happened, after the initial oops.

Raphael

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

\
 
 \ /
  Last update: 2005-03-22 13:56    [from the cache]
©2003-2011 Jasper Spaans