SUMMARY: Parity_Error_Register_92<ERROR,CHECK,ERR16>

From: when will I finish the decorating ? (pallan@clare.risley.aeat.co.uk)
Date: Fri Jul 14 1995 - 04:29:33 CDT


Dear All,

While musing on the summary deficit (perhaps indicating that not all
problems get solved) I remembered I owe a summary.

Here it, well, sort of is. As an intermittent fault it hasn't a hard-and-fast
miracle cure.

> Our 4-year-old IPC (SunOS 4.1.1) has recently started crashing
> and reporting other errors as seen below. Some crashes have been
> out of working hours, when there was no activity.
>
> May 23 15:10:48 Safety vmunix: DVMA Parity Error, ctx = 0x0, virt addr = 0xfff12ea8
> May 23 15:10:48 Safety vmunix: pme = e20020ec, phys addr = 20ecea8
> May 23 15:10:48 Safety vmunix: Parity Error Register 92<ERROR,CHECK,ERR16>
> May 23 15:10:48 Safety vmunix: bad module/chip at: ?
> May 23 15:10:48 Safety vmunix: System operation cannot continue, will test location anyway.
> May 23 15:10:48 Safety vmunix: parity error at 20ecea8 is transient.
> May 23 15:10:48 Safety vmunix: panic: dvma parity error

That was between 0 and 20 reboots per week. Power-on memory testing didn't report faults.

I got one reply From: gunn@lardav.com (David Gunn) suggesting:

  o Run diagnostics

  o Remove SIMMs one at a time

  o Switching SIMMs around inside the IPC

  o Swap SIMMs with other IPCs.

Sundiag found no faults.
We had the memory replaced (and increased) by a hardware bod.
No problem since then.

 Peter Allan peter.allan@aeat.co.uk.



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:10:29 CDT