Re: partman problems (sparc and elsewhere)
On Wed, Dec 01, 2004 at 07:21:04PM +0100, Sven Luther wrote:
> On Wed, Dec 01, 2004 at 09:13:47PM +0200, Anton Zinoviev wrote:
> > On Wed, Dec 01, 2004 at 03:36:14PM +0100, Sven Luther wrote:
> > >
> > > > Or partition we are mounting without formating if its type doesn't
> > > > correspond to the file system it contains.
> > >
> > > Well, i don't think so, not automatically at least.
> >
> > (Lib)parted repairs the wrong partition types in all cases. In some
> > cases partman will not write the partition table but if it has to for
> > some reason then the partition type will be changed.
> >
> > > > At the time when some partition is being formatted (or simply sceduled
> > > > for mounting) the partition table may already contain physical volumes
> > > > of activated volume groups.
> > >
> > > So what ? a raid or LVM partition is simply a partition with a given flag. The
> > > fact that you modify the other partition on the partition table should in no
> > > way affect any raid or lvm partitions you have,
> >
> > The kernel will complain that it can not reread the partition table.
> > This complain is not dangerous and partman can probably safely hide it
> > from the user but this already is not nice.
> >
> > Moreover the confirmation dialog contains a list of the changed
> > partition tables. If we decide to write all partition tables to the
> > disk just because they contain some formatted partition this means that
> > they have to be included in this list. In most cases this list will
> > contain all partition tables and because of that it will be less
> > informative.
>
> I don't follow this. You read the partition table, modify the type for one
> partition type, write it back.
If the file system type is unchanged partman can not know whether the
partition type is modified or not. Thats why you proposed that partman
writes the partition table if it contains formatted partition (or
partition with phisical volume for RAID or LVM). In many cases this
means that partman will have to write unchanged partition tables.
In order to fix #238388 (i.e. partman allows user to make changes in
partition tables that are used by the kernel) partman has to know which
operations change the partition table and disallow them. However even
in this case it will have to write the unchanged partition tables
because some some partition type may have changed without its knowledge. :(
> This should not change the rest of the
> partitions, or it is a RC parted bug, and your duty is to fill a bug report
> about this.
Ah, I see. It changes only the type of the partitions that have
set_system_type.
Anton Zinoviev
Reply to: