Subject: port-i386/13748: fdisk(8) always uses in-core boot record
To: None <gnats-bugs@gnats.netbsd.org>
From: Don Yuniskis <auryn@gci-net.com>
List: netbsd-bugs
Date: 08/18/2001 17:23:38
>Number: 13748
>Category: port-i386
>Synopsis: fdisk(8) fails to reread updated on-disk boot record
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: port-i386-maintainer
>State: open
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Sat Aug 18 17:25:00 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator: Don Yuniskis
>Release: NetBSD 1.5.1
>Organization:
Don Yuniskis
>Environment:
System: NetBSD Some.Machine.Name 1.5.1 NetBSD 1.5.1 (GENERIC) #56: Mon
Jul 2 15:54:23 CEST 2001
he@nsa.uninett.no:/usr/src/sys/arch/i386/compile/GENERIC i386
>Description:
When the "master boot record" is altered with fdisk(8)
(e.g., 'fdisk -b CC/HH/SS sd0', 'fdisk -u sd0', etc.),
the changes are written back to disk (assuming the user
indicates his desire to do so!). However, subsequent
invocations of fdisk(8) present the original parameters
as existed prior to the alteration. I.e. the disk is
never reexamined.
>How-To-Repeat:
fdisk -u sd0
alter the "BIOS geometry"
save the changes to disk
quit
fdisk -u sd0
note BIOS geometry APPEARS to not have been changed
>Fix:
What's the downside to forcing a read of the physical
media (instead of relying on in-core information)?
>Release-Note:
>Audit-Trail:
>Unformatted:
X-send-pr-version: 3.95