I am looking for a way to minimize downtime while replacing 4GB disks
in systems with 9GB (or larger) disks. In particular, I'd like to come
up with a scenario that would allow me to restore the OS partitions to
new disks, then reboot the server, coming up on the new disks, and then
pulling old OS disks and replacing them with larger disks for other uses.
Here's my scheme - what am I missing?
* Take system down, add 2 new 9GB disks to open slots, reboot.
* partition new disks as desired, with OS partitions and user
space.
* use LSM to create a mirror out of the pair of new disks.
* Create ADVFS domains/fsets and mount at some temporary mount point
* Restore content from most recent backup onto OS partitions
onto newly mounted mirror.
* Modify /etc/fstab to boot from new disk.
* Take system down, modify console parameters to boot from new disk.
* Reboot, make sure it comes up correctly.
* Take system down, remove old small disks, replace with more new
disks, reboot.
Any reason why this wouldn't work?
TIA!
Judith Reed
jreed_at_appliedtheory.com
Received on Tue Mar 26 2002 - 18:23:06 NZST