We have installed a new server V240 consisting of 4 HDDs.
All of them are of size 72G and part-number is the same for the fujitsu disks.
however only 2 disks (c1t0d0 and c1t1d0) shows 72G under format and c1t2d0 and c1t3d0 shows the size of 36G. What could it be the reason? pls assist
OS have been installed and mirrored on the 1st 2 disks and firmware is updated to the latest OBP 4.22.33 
Hi.
Did you check the HD firmware?
Should you check the iostat command are the same that the format and verify the hd are 72gb.
# iostat -En
bash-2.05# iostat -En
c0t0d0 Soft Errors: 1 Hard Errors: 0 Transport Errors: 0
Vendor: SEAGATE Product: ST34371W SUN4.2G Revision: 3162 Serial No: 9814Y41471
Size: 4.29GB <4292075520 bytes>
Media Error: 0 Device Not Ready: 0 No Device: 0 Recoverable: 0
Illegal Request: 1 Predictive Failure Analysis: 0
c0t6d0 Soft Errors: 2 Hard Errors: 0 Transport Errors: 0
Vendor: TOSHIBA Product: XM-5401TASUN4XCD Revision: 1036 Serial No: 04/12/95
Size: 0.00GB <0 bytes>
Physically, by looking at the part-number and label on the HDDs, its showing as 72G. But from the OS, it detects them as 36G.
And fyi, the 2 problem disks were used to be cloned from some other server, but not sure if they were of size 36G. We used the dd command to clone it.
Is there a way to recover the 72G back?:o
[quote=quetzalcoatl;302244712]
Hi.
Did you check the HD firmware?
QUOTE]
And btw, how will you check that? And if there's an issue with that, whats the remedy for it? You have come across and resolved such issues before?
Im seeing this for the 1st time:(
I have seen this several times. It happens when you copy a 36 GB disk, vtoc and all, on to a 72 GB disk. The only way I know to go back is to scribble on the vtoc, then run format. format will see that the vtoc is garbage and will make a new one.
To scribble on the disk:
dd if=/bin/sh bs=512 count=4 of=/dev/rdsk/c0t0d0s2 # or whatever device you want to destroy
Note that you lose all data this way.
what do you mean by scribble on the disk?
prtvtoc shows a "typical" cylindrical output of a 36G disk..as what I saw.
Btw, can you explain what this command exactly does before I execute it on my system tomorrow morning.. ?:o
And kindly provide me the steps that I should do in point form, in order for me to get my 72G disks back ..
eg.
1) run this command :dd if=/bin/sh bs=512 count=4 of=/dev/rdsk/c0t0d0s2 # or whatever device you want to destroy
2) type format -> ..... etc etc..
Help is greatly appreciated. Thanks.
Hi
the prtvtoc report information about a disk geometry and partitioning and Solaris use the HDD like a prtvtoc shows, so its very common to paste hdd information when you made a mirror, you could pass it a HD 72G prtvtoc to the HD 36G, but like said Perderabo, this command are destructive..
So you can use the fmthard command to paste the prtvtoc information
for insteance
#root@# prtvtoc /dev/dsk/c1t0d0s0 | fmthard -s - /dev/rdsk/c1t3d0s2
With prtvtoc command it fails with some error and did not work.
However, dd if=/bin/sh bs=512 count=4 of=/dev/rdsk/c1t0d0s2 works!! Im happy to got it working and recover my 72G . Thx Guys!
Typo there:
prtvtoc /dev/rdsk/c1t0d0s0 | fmthard -s - /dev/rdsk/c1t3d0s2
But you need a disk, physically identical to the screwed-up disk, but with a correct vtoc for this to work. My technique, while gross, will work without a correct 2nd disk. I have used it on a 2 disk system where someone copied the first disk to the 2nd, thus reducing the capacity of the second disk. There was no 3rd disk with an intact vtoc to serve as the source for the above command.