On Wednesday, April 21, 2004, at 02:09 PM, Glen Harris wrote:
I have a similar system - a fibre channel RAID box. I added a disk to it
and expanded the RAID5 device, then used the scsiadd command to rescan
the device. If your card emulates a SCSI disk, this may work. Otherwise,
try unloading/loading the kernel driver module(if possible), and failing
that a reboot should work.
Anyway, the new size was reported correctly as 1.25Tb, so I then used pvresize to grow the pv to the new size of the device.
Just curious, but what kind of rebuild time do you see on that 1.25TB RAID5 volume?
I know you have a hardware controller, but I've found that with Linux's software RAID5, anything over 400GB or so will take over 3 hours on a decent disk system to rebuild. Thats why I use LVM to aggregate several RAID5 volumes, rather than one giant RAID5. I generally expect that it will take a couple of hours to see the failure, find a replacement, and get it installed in the machine (thats if I'm onsite, and have spares on hand ) With a 2 hour rebuild time, I'm already looking at 4 hours of exposure where another disk failure means data loss. Thats already pretty high to me, especially if there's something systemic causing drives to fail.
Even with hot spares, you still have to be careful during the array rebuild. :-P
_______________________________________________ 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/
