I'm looking at implementing my first LVM system. It looks pretty straight-forward. (Very similar to the LVM stuff I saw in AIX v3.x and Veritas management Sun used to manage SPARCStorage Arrays.)
The one nagging question in my mind is:
How resilient is LVM to a sudden renumbering of physical hard disks?
Assume no disks have actually failed, just that a new disk
gets added and suddenly all the LVM drives get notched up
a drive letter.
For example,
Before After SCSI ID* description ------ ----- -------- -------------------------------- sda sda 0,0,0,0 operating system disk (HW RAID) -- sdb 0,0,1,0 new added disk (e.g. a CD drive)
sdb sdc 1,0,0,0 LVM PV#1 sdc sdd 1,0,0,1 LVM PV#2 sdd sde 1,0,0,2 LVM PV#3 ---------------------------------------------------------- * The four numbers are: (1) controller, (2) controller channel, (3) SCSI target, (4) SCSI LUN.
Can I just assume LVM will do the right thing, and my VG created from the three physical volumes will be OK?
Thanks, - Don Buchholz
_______________________________________________ linux-lvm mailing list linux-lvm@redhat.com https://www.redhat.com/mailman/listinfo/linux-lvm read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
