Kernel panic -supervisor read, page not present

[Doug Hardie]

On Sunday, May 18, 2003, at 09:34 US/Pacific, Doug Hardie wrote:

> I have a system that is almost a year old.  It started rebooting last 
> week several times each day with the message "supervisor read, page 
> not present".   A search of the archives indicates this is most likely 
> a hardware failure, but there is no indication of what might have 
> failed.  The only messages preceeding the panic have been a gzip that 
> failed, several crons and atruns that failed.  Most of the panics show 
> nothing unusual preceeding them in console.log or messages.  The 
> machine is an archive server.  Its role in life is to do an rsync 
> every 5 minutes for off-site archival storage.  Nothing else is 
> running on it.  Any ideas how to determine the failure as its still 
> under warrantee for a few more weeks.

The situation is deteriating.  Now rsync is core dumping.  ls works, 
but ls -l does not.  It also core dumps.  The problem occurs when 
reading /etc/spwd.db.  Something must be corrupt in it.  I can't find 
any info on that file.