Back again with more disk issues. ESA10000 with dual HSZ70's attached to an
ES40 runn V4.0F. Had some horrific controller problems over the past few days
and after all is said and done, I'm having one final problem with a 4 disk,
RAID-5 set of 18.2G drives.
disks in the set are DISK11100, DISK21100, DISK31100 and DISK41100. Had
problems with channel one over the past week and had all the disks on that
channel fail. 11100 failed out of the set and the spare (DISK60900) failed in.
The RAID rebuild completes and I try to fsck the filesystem (rz16c, in this
case). The fsck fails with cannot read block number ....... Console on
controllers spews error messages targeting DISK60900. Fine, I say DISK60900 is
bad. Had DISK11100 set up as a spare and failed it in. Let the rebuild finish
and then tried the fsck. Same error SAME BLOCK number. Console messages logged
targeted DISK11100 as being bad! As a final test, I failed out DISK11100 so the
the set was running reduced and tried the fsck. same result, same block number,
messages on console identfied the PTL of the bad disk as 255 255 255?!?!
So... can i get the data back? mount says the filesystem is dirty... no
duh... can't mount it. fsck yaks that there are hoards on unreadble blocks.
is my last result to go newfs the raidset?
Thanks
Jim Anderson
Received on Tue Nov 09 1999 - 13:58:10 NZDT