CE590 unmount disks due to read/write errors

[noc]

I have CE590 together with Cisco's Storage array with 6 HDD.

Software Release: CE ver 2.51 (Build: FCS 05/24/01)

It unmount some disks after 10-20 hours of working:

ce-rtk#sh cfs volumes

/c0t0d0s3: mounted

/c0t1d0s3: mounted

/c0t2d0s3: not mounted

/c0t3d0s3: mounted

/c0t4d0s3: not mounted

/c0t5d0s3: mounted

/c0t6d0s3: not mounted

/c0t7d0s3: mounted

sometime "cfs mount xxxx" working, sometime not, and i must do "disk prepare", sometime "disk prepare" doesn't work with message something about "No_Mem"

show cfs stat for unmounted disk shows:

Filesystem Statistics for volume /c0t4d0s3 Status: unmounted

Disk Wraps 0

Inode Hits 18152

Inode Misses 591856

CFS Read error 46

CFS Write error 219

Inode Load error 2

Attribute Load error 0

CFS Object Truncations 8014

Truncated CFS Object Flushes 8014

Volume Unmount Reason CFS TOC Write Failure

Volume Clears 0

Format time Wed Apr 9 08:51:39 2003

Mount time Wed Apr 9 11:11:46 2003

Unmount time Thu Apr 10 04:26:18 2003

Can it be hardware problem?

[owillins · Accepted solution]

If the read/write errors exceeds a threshold, then the cfs volume will be automatically unmounted. I guess the disk manufacture or the disk prepare would not work due to the lack of memory. If you do a reload, the memory would be cleaned and hence you would be able to do a disk manufacture after that.

You could use this document about disk recovery. Might help you sometime.

http://www.cisco.com/warp/public/117/cache_engine/ce_diskrec.html