>Number:         4511
>Category:       port-sun3
>Synopsis:       savecore will save old dumps repeatedly...
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    gnats-admin (GNATS administrator)
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Nov 16 16:20:00 1997
>Last-Modified:
>Originator:     Greg A. Woods
>Organization:
Planix, Inc.; Toronto, Ontario; Canada
>Release:        1997-11-12
>Environment:

System: NetBSD sometimes 1.3_ALPHA NetBSD 1.3_ALPHA (GENERIC) #188: Thu Nov 13 02:51:07 PST 1997 root@chronos.clock.org:/usr/src/sys/arch/sun3/compile/GENERIC sun3
Machine: Sun-3/260 w/32MB RAM, SCSI

>Description:

After a panic it is possible that savecore will save all or part of the
dump after every subsequent re-boot.

This problem may or may not be port specific.  I'd like to ask anyone looking
at this PR to verify this fact and expand it's category if necessary.

>How-To-Repeat:

Enable savecore.
Cause a system panic.
Reboot without cycling system power.
Notice normal savecore operation.
Reboot again without cycling system power.
Notice savecore trying to retrieve previous dump.

>Fix:

A possible fix would be to have savecore wipe out the magic number of the
dump after retreiving it to prevent subsequent retrieval.

A normal and successful system halt should also probably wipe out any dump
magic number to prevent a savecore upon reboot.
>Audit-Trail:
>Unformatted: