Hi Ethan, Just trying to help narrow down where the corruption is occurring. :-) Lemme see if I understand the setup:
1.) Production MySQL DB server operating off of LUN A sourcing the data/zpool to MySQL. 2.) LUN B gets added from another SAN device. 3.) zpool gets imported off of LUN B. 4.) ibbackup of data on LUN A targetting LUN B. 5.) LUN B gets exported and detached. 6.) LUN B gets attached and imported on a backup server where the ibbackup file is loaded into a MySQL instance. 7.) MySQL is shut-down and data files are then scurried off to tape on the backup server. Essentially, the goal is a LAN-less backup? -J On Thu, Jun 5, 2008 at 3:56 PM, Ethan Erchinger <[EMAIL PROTECTED]> wrote: > > > Jason J. W. Williams wrote: >> >> Do the filesystems pass a ZFS scrub? If so, the corruption is >> somewhere inside MySQL or ibbackup flying the bits around. This is >> MySQL 5 right? >> >> > > We don't issue zfs scrub because, like resilvering (at least in u4/u5), > there is no way to limit the impact of that scrub/resilver on normal IO. > > MySQL 4.1 and 5. I would bet on the corruption happening inside MySQL. I > never blamed ZFS, but ibbackup is a nice safety net. > > Ethan > _______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
