On January 10, I posted the following request:
> For the past couple of months, my system has been experiencing
> strange panics of various flavors. Here's a sample:
>
>
>
> Jan 9 11:44:52 vt vmunix: pid 116, `':
> Jan 9 11:44:52 vt vmunix: vector 0x38, pc 0xe06f86e, sr 0x2700, stkfmt 0x0, context 0x0
> Jan 9 11:44:52 vt vmunix: D0-D7 0 f0c13fc f110034 2 2000 11a 2100 39
> Jan 9 11:44:52 vt vmunix: A0-A7 e147800 e0a7498 e0cd962 f708f0c f110000 e0cdc32 e15f29c e090470
> Jan 9 11:44:52 vt vmunix: Begin traceback...fp = e15f29c, sp = e090470
> Jan 9 11:44:52 vt vmunix: End traceback...
> Jan 9 11:44:52 vt vmunix: panic: Stack format error
> Jan 9 11:44:52 vt vmunix: zs3: silo overflow
>
>
> Jan 8 10:33:52 vt vmunix: panic: pagefault
> Jan 8 10:33:52 vt vmunix: syncing file systems... [4] [3] [1] done
> Jan 8 10:33:52 vt vmunix: zs3: silo overflow
>
>
> Jan 2 09:36:27 vt vmunix: pid 124, `':
> Jan 2 09:36:27 vt vmunix: vector 0x8, pc 0xe01b4b4, sr 0x2504, stkfmt 0xb, context 0x0
> Jan 2 09:36:27 vt vmunix: Bus Error Reg 80<INVALID>
> Jan 2 09:36:27 vt vmunix: data fault address 10038 faultc 0 faultb 0 dfault 1 rw 1 size 0 fcode 5
> Jan 2 09:36:27 vt vmunix: KERNEL MODE
> Jan 2 09:36:27 vt vmunix: page map 23000000 pmgrp ff
> Jan 2 09:36:27 vt vmunix: D0-D7 4e20 f0d9674 0 0 0 2000 0 0
> Jan 2 09:36:27 vt vmunix: A0-A7 10000 e0041be ffffff88 0 0 e153d1c e090460 e090450
> Jan 2 09:36:27 vt vmunix: Begin traceback...fp = e090460, sp = e090450
> Jan 2 09:36:27 vt vmunix: Called from e004700, fp=0, args=e0041be 2000 0 0
> Jan 2 09:36:27 vt vmunix: End traceback...
> Jan 2 09:36:27 vt vmunix: panic: Bus error
> Jan 2 09:36:27 vt vmunix: zs3: silo overflow
>
>
> The only pattern I can discern is the "zs3: silo overflow" which
> appears as a part of every panic log. I have no idea what
> zs3 is ... zs0 and zs1 are config'd into the kernel, but there's
> not a mention of zs3.
>
> Any similar problems? Fixes?
>
> My system is a 3/50 under SunOS 4.1. I have a Wren-V SCSI on which
> the a partition is my root, b is my swap, and g is my home directory
> partition. /usr and /usr/kvm are NFS imports. Crashes seem to be
> independent of application: I've been nailed while running under X,
> SunView, and directly on the big white screen (no windows).
The general consensus is the following:
- zs2 is the keyboard
- zs3 is the mouse
- The silo overflows take place as a side effect of the panic. While
the kernel is logging the trace, keyboard/mouse events overflow
relatively small silos for those devices. Those overflows are also
logged.
- I'm hosed with a probable bad disk (the only component I haven't
replaced, and something is corrupting program images in memory).
Thanks to all of those who sent me detailed and informative stuff on
this one:
proton!baumann@ucrmath.ucr.edu (Michael Baumann)
mrwallen@UCSD.EDU (Mark R. Wallen)
bisc@zellweger.ch (Thomas K. Bischoff)
drl@vuse.vanderbilt.edu
jimh@nsd.fmc.com. (Jim Hendrickson)
smb@ulysses.att.com
auspex!guy@uunet.UU.NET (Guy Harris)
Phil Kearns
kearns@CS.WM.EDU
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:09 CDT