Hi Listers. yesterday I shot myself in the foot by increasing the number of blocks on the SWAPGEN command. When I rebooted, I could no longer find my LVM. When the e2fsck ran it came up with the message "Incorrect metadata area header checksum". I ran the vgdisplay, pvdisplay, and lvdisplay and while I continued to get the "incorrect metadata" messages, the displays looked correct. I tried running e2fsck with an alternate superblock, but that failed. I tried to run vgcfgrestore and it failed. I was eventually able to correct the problem by issuing the pvcreate with the --restorefile option.
I found a statement in the LVM HOWTO that read "The above command (pvcreate with the --restorefile option) has been known to recover a PV overwritten with mkswap." Does anyone know if there is a size limitation to the number of blocks using SWAPGEN? Does anyone know how increasing the number of blocks corrupted the VGDA? If the VGDA is strored at the beginning of the physical volume I'm not sure how this was corrupted. Thanks in advance for your help... Susan Zimmerman Senior Systems Software Programmer ISD Tech Support (703) 206-4509 ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
