Technical article 110823

I get the error message "Not enough memory for change table" or "Segmentation Fault - core dumped" or "Bus Error - core dumped" when building my kernel.

OpenServer 6.0.0UnixWare 7.1.0UnixWare 7.1.1UnixWare 7.1.3UnixWare 7.1.4Open UNIX 8

Written · Last revised

Problem

When trying to rebuild the kernel, one of the following error messages is displayed:

UX:idmkunix: ERROR: Not enough memory for change table
OR
Segmentation Fault - core dumped
OR
Bus Error - core dumped
and then the idbuild fails.

Cause

It is possible that the file /etc/conf/cf.d/changedb has become corrupted.

Solution

First, rename the file to changedb.bak and then relink the kernel using /etc/conf/bin/idbuild with both the -B or -# options.

This removal of the changedb file should resolve the problem.

The "-#" option will generate verbose debug output, which may be of use if the relink still fails.

You can also run:
# /etc/conf/bin/idcpunix
to copy the unix kernel to /stand
and:
# /etc/conf/bin/idmkenv

to remake the kernel environment, usually done upon reboot.

NOTES:    One of the /etc/conf/pack.d/<driver>/space.c files may include
          a header file which does not exist on the system.

Try running a "sar -r 1 1" to find out how much memory is actually available on the system should the problem be a true lack of memory.

In the case of a core dump, it is likely that the process which
dumped core is an instance of /etc/conf/bin/idmkunix, and that the
core file itself will be located in /etc/conf/cf.d.  So if removing
the changedb file does not resolve the issue, try to verify whether
idmkunix was indeed the binary which caused the core dump.  To do
this, you can use the debug(1) debugger; for example:
# debug -ic core.23631
Error: Could not access idmkunix
debug>

Here, "core.23631" is the core file, found in /etc/conf/cf.d; 23631 is the process ID of the process which dumped core. The error message immediately shows that it was "idmkunix" which dumped core. Thus in the case where removing changedb does not fix the problem, you can note what binary _did_ lead to the core dump, along with output generated by the "-#" option to idbuild, to hopefully obtain enough useful information to resolve the issue.

From the Xinuos technical article database, record 110823. Reproduced with permission. The original is at www.sco.com/ta/110823, 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.