https://bugzilla.novell.com/show_bug.cgi?id=299267#c22
Thomas Fehr changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |cthiel@novell.com, kukuk@novell.com
--- Comment #22 from Thomas Fehr 2007-08-29 06:20:42 MST ---
Part of the problem is that it is not easily detectable if the same device was
bigger with older releases in the general case.
There is no guarantee that the disk even has a partition table (e.g. it is
not uncommon to use /dev/hda unpartitioned as a PV in a LVM or EVMS Volume
group or the disk could be part of a dmraid array). Additionally I am not sure
what happens on other disk labels (e.g. I assume a GPT disks will appear
completely inconsistent since the backup of GPT at end of disk is missing
and new disk size does not match size in GPT metadata).
What I can do is check for the error message "Can't have a partition outside
the disk!" by parted. Unfortunately parted will refuse to print any further
data
about the partition table in such a case and so the execution of the binary you
propose will be somewhat speculative (e.g. the partition tables contained on
the first disk of a raid0 managed by dmraid also regulary has partitions
outside
of the disk). So if it does not harm to sometime call the binary on a disk
where HPA does not even exists I can do the needed changes in YaST2.
Nevertheless this covers not the cases where a partition table does not exist
or is more complex as msdos (e.g. GPT). Another problem is that we will have
beta#3 on Monday so there is not too much time left for try-and-error.
In the end, this needs to be decided by the people responsible for SL 10.3 and
SLES11 anyway. Adding Christoph Thiel and Thorsten Kukuk to CC. AFAIK they are
responsible for 10.3 and SLES11.
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.