SUMMARY: Spurious VME interrupt

From: Jeff Pack (jfp@inel.gov)
Date: Fri Jun 12 1992 - 14:48:21 CDT


Once again, the list delivers. Here's a brief summary of the responses.
Thanks to:

From: Steve Elliott <se@computing.lancaster.ac.uk>
From: kevin@ukw.uucp (Kevin Sheehan {Consulting Poster Child})
From: gdmr@dcs.edinburgh.ac.uk (George D M Ross)
From: Brent Alan Wiese <brent@crick.ssctr.bcm.tmc.edu>
From: Dave Classen <classen@crick.ssctr.bcm.tmc.edu>
From: dan@breeze.bellcore.com (Daniel Strick)
From: david@elroy.Jpl.Nasa.Gov (David Robinson)
From: alastair@eucad.co.uk (Alastair Young)
From: sls@cs.duke.edu (Shelley L. Shostak)
From: kiva!risto@snowbird.Central.Sun.COM (Risto M. Tolonen)

The original message:

> I have a Sun 4/370 running SunOS 4.1 that is giving me some VME interrupts
> that are troubling. Here's the configuration:
>
> 4/370
> SunOS 4.1
> 56 MB memory, 32 on CPU bord, 24 in Slot 2
> One Xylogics 753 SMD controller in Slot 6 with one Fujitsu MK2382 SMD disk
> One Wangtek 60 MB 1/4" SCSI tape
> One Fujitsu M2263SA SCSI disk
> One additional Sun Ethernet controller (ie1) in Slot 5
>
> The messages are the following:
>
> .
> .
> .
> Jun 10 14:15:58 mica vmunix: spurious VME interrupt at processor level 5
> Jun 10 14:15:58 mica vmunix: VME level 3, VME vector 0xffffffff
> Jun 10 14:39:20 mica vmunix: spurious VME interrupt at processor level 5
> Jun 10 14:39:20 mica vmunix: VME level 3, VME vector 0xffffffff
> Jun 10 14:48:17 mica vmunix: spurious VME interrupt at processor level 5
> Jun 10 14:48:17 mica vmunix: VME level 3, VME vector 0xffffffff
> .
> .
> .
>
>
> I have tried to find a time pattern or process associated with the
> messages, but there is none that I can see. As you can see by the times,
> the interrupts seem to be random. The interrupts don't seem to be causing
> a problem, other than reducing space in var :-). Sun replaced the CPU card
> with no results. So, again I turn to the list for my answers. Anyone seen
> this before? If you need more info, just email. I'll summarize.

Steve (and a couple others) mentioned a patch for this that's available;
here's a few lines from the README:

Patch-ID# 100116-01
Keywords: spurious level 3 interupt ie1
Synopsis: spurious level 3 interupt for ie1
Date: 06/Sept/90
 
SunOS release: 4.0.3, 4.1 4.1_PSR_A
 
Unbundled Product:
 
Unbundled Release:
 
Topic:
 
BugId's fixed with this patch: 1009297

Architectures for which this patch is available: sun3 running 4.0.3
                                                 sun3 running 4.1
                                                 sun4 running 4.1
                                                 sun4/390 running 4.1_PSR_A
ONLY these architectures show the symptoms.

Obsoleted by: 5.0

Problem Description:

        This patch fixes a problem which occurs in SunOS 4.0.3 and 4.1
        on sun-3 and sun4 gateway machines. When a large number of packets
        are being routed through the gateway(ie1 interface), the following
        error message will appear in the console window:

                vmunix: spurious level 3 interrupt
        
        These messages are displayed as a result of the way the
        Intel 82586 ethernet controller works and this patch is
        designed to suppress the message.

This really sounds like the fix I need. Why couldn't Sun have told me
this? Grrr.

Several others stated that ie1 was causing the message, that it was
harmless, and could be ignored. That's what I had been doing, but thought
I should at least be brave and try to figure out what was causing it.

Reseating boards and backplane configuration was also suggested by a few
folks and I had tried that earlier. Same thing.

David Classen suggested a Rev. A Xylogics board may be causing trouble. I have
Rev. B which is what cured his trouble. Thanks anyway.

Upgrading ie1 to the new VME based Ethernet board available with the 4/400
series was also mentioned. This may be something to plan for in FY-93.

Anyway, I'm trying the patch first and then talking to Sun about a new
Ethernet board (hopefully under my maintenance contract). Thanks again to
all.

-- 
---------------------------------------------------------------
Jeff Pack                         UUCP:  ...!uunet!inel.gov!jfp
Idaho National Engineering Lab    Internet:  jfp@inel.gov
P. O. Box 1625 M.S. 2603          Phone:  (208) 526-0007
Idaho Falls, ID 83415             FAX:  (208) 526-9936



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:43 CDT