Technical article 104628

My SCO Unix panics with "Too many queued CPU# interrupts" or "Unable to freeze processor"

OpenServer 5.0.0OpenServer 5.0.2OpenServer 5.0.4OpenServer 5.0.5OpenServer 5.0.6OpenServer 5.0.7

Written · Last revised

Problem

My SCO Unix system panics with the error message "Too many queued CPU# interrupts" (where # is some number) or "Unable to freeze processor"

Cause

The kernel has a finite interrupt buffer queue which has been
exceeded.  This is a fatal error.

Solution

The size of this interrupt buffer queue is affected by the kernel tunable parameter MAXACPUS.

This value can be increased to a higher value.  SCO Support
suggests 9 as a starting point.  The current maximum recommended
value is 28.

To change the value please do the following :

1. Log in as the root user (if you have not already done so).

2. cd  /etc/conf/cf.d
3. Save a copy of 'stune' for backup  (for example,
   cp stune /tmp/stune)
3. vi  stune
4. Search for the parameter MAXACPUS

a. If it is found, then change the number following the parameter to a desired value.

b. If it is not found, then edit the file to include the following line:

MAXACPUS  9
5. You will need to relink the kernel and reboot before this
   change takes effect.  It is recommended that you change to
   single user mode to do this.  The steps are as follows:

a. Notify all users that the system will be unavailable for a few minutes.

b. Issue the command:  shutdown -y -g0 -i1

c. At the Ctrl-D prompt, enter the root password.

d. Once in Maintenance mode, type:  cd /etc/conf/cf.d
e. Issue the command:  ./link_unix -y
f. Reboot the computer.
Note:  If the relink fails, replace the file /etc/conf/cf.d/stune
       with the backup file from /tmp.  Try the relink (step 5)
       again.  If that succeeds then re-edit the stune file.

If the relink fails, then you will need to correct the Link Kit before you continue.

If this does not resolve the issue then check to ensure the server:

a.  Is running the latest BIOS provided by the OEM.
b.  Is running the latest Maintenance provided by SCO.
c.  Run "hwconfig -hc" to identify any devices which might share
    IRQ's and seperate them to try to ensure, for example, the hard
    disk controller is unique.

This issue has been seen running the SMP MultiProcessor Supplement.

From the Xinuos technical article database, record 104628. Reproduced with permission. The original is at www.sco.com/ta/104628, which is reachable but blocked from search engines.

Next step

Need this fixed rather than explained?

Tell us the release, what the system runs and what it is doing. We will reply by email with what we would do next.

  • You deal with the people who do the work
  • No obligation, and no upgrade required before we will help
  • We tell you plainly if something is not worth doing

Send an enquiry

We reply by email. If a server is down, say so in the first line. You can also write to help@scoexpert.com.