Technical article 104882

Kernel check failed, kernel = (null), code = Unknown

OpenServer 5.0.0

Written · Last revised

Problem

I occasionally receive this mail message from cron:

From cron Fri Oct 27 13:40:04 1995
From: Cron daemon <cron@sco.com>
X-Mailer: SCO OpenServer Mail Release 5.0
To: sys@sco.com
Date: Fri, 27 Oct 95 13:40:02 PDT
Message-ID:  <9510271340.aa01952@sulawesi.sco.com>
Status: RO

kernel check failed, kernel = (null), code = Unknown

*************************************************
Cron: The previous message is the standard output
and standard error of one of your cron commands.
What does this message mean?

Cause

There are two known causes of this problem:

Situation #1:

There is a cron(C) job that runs in the background, every 20 minutes, Monday through Friday from 8:00 A.M. to 5:00 P.M. (and every hour during non-working hours), to gather system performance data using sar(ADM).

The operating system contains several counters that are
incremented as various system actions occur.  These include
counters for CPU utilization, buffer usage, disk and tape
I/O activity, TTY device activity, switching and system-call
activity, file access, queue activity, inter-process
communications and paging.  The counters are reset to 0 every
time the machine is booted into multiuser mode.
At some point, a new kernel was relinked but the system was
not rebooted.  The cron(C) job that runs the data collection jobs
no longer has access to the valid kernel counters and sends
the warning message.
Situation #2:
In this situation, the system was upgraded from a previous release
 of SCO Unix and therefore has a combined boot/root filesystem,
 with all copies of the kernel stored in the / directory.  Most
 OpenServer Release 5 systems are fresh installs, and they have
 separate boot and root filesystems, with copies of the kernel
 stored in /stand.
If the inode number of the kernel is greater than 32767, it is
 incorrectly sign-extended while being stored into a 32-bit field
 in a structure passed from /boot to the booted kernel.  That number
 is later extracted and used to identify the file from which the
 kernel was loaded, allowing programs like sar(ADM) to be
 independent of the name of kernel.  In this situation, however,
 sar(ADM) references the wrong file and the data collection
 routine fails.
Other utilities have already been changed to use different
mechanisms to access kernel state, so that they don't need to
open the kernel image at all.  The problem with sar(ADM) has
been reported to SCO Engineering and will be fixed in a later
release of the operating system.

Solution

Situation #1: Reboot the machine, using the new kernel.

Situation #2: Try this patch to /boot, which should correct the problem.

# cp -p /boot /boot.orig
# what /boot
/boot:
         SCO boot 23 Apr 95
 # sum -r /boot
 00357   134 /boot       <-- if these numbers don't match,
                                 don't proceed
 # echo 3f:feb?h 90 | _fst -w /boot -
* 3f:feb?h 90
 0x3f:0xfeb:     0x99    =       0x90
 # sum -r /boot
 09082   134 boot

Once patched and rebooted, the system should function normally.

See also

See also the sar(ADM) man page.

From the Xinuos technical article database, record 104882. Reproduced with permission. The original is at www.sco.com/ta/104882, which is reachable but blocked from search engines. Xinuos marks this article against a retired product line. That refers to the product’s lifecycle, not to whether the advice still works — plenty of it still applies to running systems.

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.