Hi, I am having a nightmare and hope that someone has a few words of advice. We have two unix boxes which are (but shouldn't) be looking at the same volume group. Net result is corrupt data and we have been advised to restore from mksysb and then restore data from backups. This is new territory and apologise if the question has been answered many times before. How do I start doing this? Do I stop the system, put key in service mode to restore from mksysb? What impact will that have on the corrupted volume groups I have eg datavg? Will I need to redefine all the filesystems within the volume groups? Any advice on what to do, what to watch out for and certainly what not to do would be very much appreciated.

Go to www.redbooks.ibm.com and download SG24-5432 and SG24-5432. These are extensive redbooks on the AIX Logical Volume Manager. If you have 2 systems accessing the same volumes, perhaps you once had HACMP configured. One volume group would be varied on the the other varied off. However, HACMP would most likely be accessing your eg datavg, not your rootvg. If these are SSA disks, perhaps you have them looped improperly. If they are twin tailed (differential) SCSI, then their is a good chance that these system were once part of an HACMP cluster. Bottom line, your mksysb should not interfer with the other volume group. The mksysb has information regarding which disks comprise the volume group and will create the logical volumes and restore the filesystems. Hopefully, your mksysb is current. Putting the system into maintenance mode will allow you to restore from the mksysb image. Just follow the menu prompts.