Originally posted by: robint
Hey all
I have a production server (p52A) that was powered off for a scheduled power outage but on wake up it failed to boot.
Initially, it crashed with an 888-102-300-0c5 (remember them), DSI. I suspected a disk problem as the dump also failed to write.
I removed the disks (there are two and each LV is mirrored) and placed them on another identical server. Rebooted and the same 888.
I booted from media to maint shell and attempted to mount / over /mnt. This also resulted in 888 (DSI)
I ran fsck on /dev/hd4 and it has detected and cleared a bad block as well as sorted the inode map.
I then attempted to mount / again and it just errors with "invalid argument" on "mount /dev/hd4 /mnt"
As the media I booted from was 5304(32 bit) and the server was now running 53 TL12(64 bit), I made mksysb on DAT and booted from that, just to see if the level difference accounted for the mount problem, but same outcome for mount. ( I can now access the contents of /usr/bin which previously were failing with the message "killed")
The mount command refuses to mount the / partition.
I ran "logform /dev/hd8", not really sure if this was relevant, and it succeeded, althoug, I could still not mount /
When booting from maint, but using option #1, boot and mount. The following error is given:
Importing Volume group...
PV Status: hdisk0 0000bef841441482 PVREMOVED
hdisk1 0000bef841462324 PVACTIVE
varyonvg: Volume group rootvg is varied on.
0516-1281 synclvodm: WARNING lv control block of hd7
has been overwritten
rootvg
Checking the /usr filesystem.
/dev/hd4 (/): ** Unmounted cleanly - Check suppressed
Checking the /usr filesystem.
/dev/rhd2 (/usr): ** Unmounted cleanly - Check suppressed
Replaying log for /dev/hd4.
Data Storage Interrupt - PROC
.test_and_set+000020 stwcx. r9,0,r5 r9=F6250800,r=...
KDB(2)>
A normal reboot from disk just hangs on 557.
Is there a way I can repair hd4? Or is there something I can do to recover it?
All help appreciated!
Rob
#AIX-Forum