On Fri, Nov 12, 2004 at 11:31:45AM -0000, Ranveer Attalia wrote:
>  
> Hi 
> 
> Was wonderig if someone could help. I got a failure on one of my
> partitions yesterday night complaining about the error in the subject
> heading. I thought it may be a problem with the actual partition on the
> server Telsun4 (see info. below). 
> Therefore I umounted the /ccmbackup partition and then fsck'd it and it
> was fine.
> 
> 
> Any ideas would be helpful

In addition to Paul's comments about backing up during heavy activity,
let me add a concern that perhaps you, or someone with expertise in
disk layout, look at the partitioning of that drive.

> ?   DUMP: bread: dev_seek error: Error 0
> ?   DUMP: bread: dev_seek error: Error 0
> ?   DUMP: bread: dev_seek error: Error 0
> ?   DUMP: bread: dev_seek error: Error 0
> |   DUMP: Warning - block 1108709828 is beyond the end of
> `/dev/rdsk/c2t2d0s3'
> |   DUMP: Warning - block 120731398 is beyond the end of
> `/dev/rdsk/c2t2d0s3'
> |   DUMP: Warning - block 81522582 is beyond the end of
> `/dev/rdsk/c2t2d0s3'
> |   DUMP: Warning - block 791677494 is beyond the end of
> `/dev/rdsk/c2t2d0s3'
> |   DUMP: Warning - block 705708460 is beyond the end of
> `/dev/rdsk/c2t2d0s3'

One possibility is that your disk slices and/or partitions
have small overlaps (which they should not :)).  While it
can happen on both Sparc and x86 versions of Solaris, it
is particularly easy to do on the latter (x86) versions.
This is because different fdisk programs from different
sources, may view the disk geometry slightly differently.

Use a tool like Solaris' format or prtvtoc command to view
the slice geometry by sector number to ensure there is no
overlap (other than slice 2 which is the entire disk by
convention).

-- 
Jon H. LaBadie                  jon AT jgcomp DOT com
 JG Computing
 4455 Province Line Road        (609) 252-0159
 Princeton, NJ  08540-4322      (609) 683-7220 (fax)