[Asterisk-Users] panic() panic() panic() and dma errors

[Jim Gottlieb]

On 2004-06-25 at 22:12, Steve Hanselman (SteveH at brendata.co.uk) wrote:

> If you cat /proc/interrupts is anything else sharing with the TEs?

It doesn't seem to:

          CPU0       CPU1       
  0:       5413       5623    IO-APIC-edge  timer
  1:          0          5    IO-APIC-edge  keyboard
  2:          0          0          XT-PIC  cascade
  8:          1          0    IO-APIC-edge  rtc
 14:       2722       3918    IO-APIC-edge  ide0
 15:         19         19    IO-APIC-edge  ide1
 16:      45283      32056   IO-APIC-level  tor2
 17:      31891      45048   IO-APIC-level  tor2
 18:      31625      44915   IO-APIC-level  tor2
 19:        176          3   IO-APIC-level  eth0
NMI:          0          0 
LOC:      10947      10945 
ERR:          0
MIS:          0

I've turned off everything in the BIOS that I can like all serial
ports, parallel port, APCI.

Interestingly, the system no longer dies in a panic() but with DMA
errors scrolling across the console.

hda: DMA interrupt memory
hda: lost interrupt
hda: dma_timer_expiry: dma status == 0x24

If I pull out the T400P boards, no problems.

I'll leave my original message in below as it's been a while.  I've
been away most of the summer and I leave for Asia on Sunday but I'm
trying to resolve this as best I can.  Thanks.


> -----Original Message-----
> From: Jim Gottlieb [mailto:jimmy-ml at nccom.com] 
> Sent: 25 June 2004 20:31
> To: asterisk-users at lists.digium.com
> Subject: [Asterisk-Users] panic() panic() panic()
> 
> Hi all.  I've been trying to build some new systems, and no matter what
> I do, if I load the zaptel and tor2 drivers, the system panics within
> an hour, even with no traffic.
> 
> These systems are using dual Athlon MP 2800 chips with one, two, or
> three T400P boards and 2 GB of system memory.
> 
> I'm currently using Fedora Core 1, but I also went back to our old
> reliable Red Hat 7.3 and the systems still panic()ed.
> 
> If I don't start the zaptel driver, they don't panic.  If I start the
> zaptel driver, but don't start asterisk, they still panic.  I'm at a
> loss of what to try next.
> 
> A typical Call Trace from the panic message looks like:
> 
> wait_on_irq, [kernel] 0xde
> __global_cli [kernel] 0x62
> flush_to_ldisc [kernel] 0x126
> __run_task_queue [kernel] 0x61
> context_thread [kernel] 0x13b
> context_thread [kernel] 0x0
> context_thread [kernel] 0x0
> kernel_thread_helper 0x5
> 
> Any ideas?  Thanks...