Summary: excessive zs? interrupt count solved.

From: kevin d bonifield (kevnb@world.std.com)
Date: Mon Nov 11 1991 - 20:47:08 CST


 This is a summary posting that discusses the high zs? interrupt
count that suddenly appeared on our machine several days ago and
used about 20% CPU time to service. The zs? count displayed by
vmstat was averaging 4100 over 5 seconds. It turns out that
someone had "borrowed" the mouse from the workstation keyboard.
When the mouse was returned the zs? count dropped back to normal.

 I still don't understand what causes this. The Sparc that the
mouse was borrowed for did not show zs? increasing when the
mouse was disconnected. Any ideas Sun-Wizards?

 My thanks to mills@ccu.UManitoba.CA for pointing me in the direction
of serial hardware. His message (with headers edited) follows:
----------------------------------------------------------------------------
>From mills@ccu.UManitoba.CA Sat Nov 9 10:30:28 1991
Date: Sat, 9 Nov 91 09:25:50 CST
From: Gary Mills <mills@ccu.UManitoba.CA>
Subject: Re: SUN 3/260 excessive zs? count
Newsgroups: alt.sys.sun

I had this happen once on a 4/280, and it turned out that one of the
control lines in the RS-232 cable was connected on the Sun end, but
was open on the far end of the cable. It acted like an antenna, picking
up noise, and causing thousands of interrupts.

-- 
-Gary Mills-         -Networking Group-          -U of M Computer Services-
----------------------------------------------------------------------------

Kevin Bonifield Inet:kevnb@world.std.com FlightSafety International



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:17 CDT