On Thu, Dec 02, 2004 at 01:51:24PM +0200, Anton Zinoviev wrote: > On Wed, Dec 01, 2004 at 10:51:42PM +0100, Sven Luther wrote: > > > > I still don't get why the kernel should complain though ? > > Because it can not reread the partition table if it uses some of the > partitions for something - swap (partman automaticaly activates any > existing swap to allow small memory installs), LVM or RAID phisical > volume or mounted partition (used as installation medium). Consider > this:
You are meaning that if the kernel uses one of the partitions of the partition table, it can then not modify the other ones ? > 1. The user creates the necessary partitions and goes to configure LVM Ok. > 2. Partman writes the partition table to the disks and invokes the LVM > configurator Ok. > 3. The configurator creates logical volumes. At that stage the kernel > is no more able to reread the partition table Why ? There is really no reason this should happen. > 4. The user specifies file systems and mount points for the logical > volumes but decides / to be in a regular partition rather than in a > logical volume I would go for /boot only, but well. > 5. Partman sets_system_type for the partition for / Ok. > 6. When the user finishes partman writes the partition tables to the > disks. The kernel complains. Mmm. Then the kernel is buggy. Friendly, Sven Luther > > Anton Zinoviev > > > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
