From: Urie, Todd (TUrie@trueposition.com)
Date: Tue May 28 2002 - 15:38:14 EDT
It was pointed out to me that my initial question was not very clear. This
is what I get for just writing out a quick question as it popped into my
head and not thinking about what I was really asking. Also, some things
that make sense in a face to face conversation don't always work in an email
exchange. I know this, but apparently don't always follow my own advice.
So, here's what I should have said in my first attempt:
The use of 'fmthard -s new_vtoc /dev/rdsk/c0t0d0s0' on a disk that is
currently mounted has the potential for wreaking havoc upon a system. I
want to ensure that I understand the situations where I'm safe and those
where I am at risk.
If I construct a new VTOC based upon an existing one and the new VTOC makes
use of 'extra' space that was on the drive by allocating the extra space to
newly created slices, then the 'fmthard -s new_vtoc /dev/rdsk/c0t0d0s0'
construct should not be destructive of existing data on the drive.
Therefore, changes of this nature would be OK.
If I construct a new VTOC where the new VTOC has changes to slice
definitions (e.g. changes to any of starting, ending, or cylinder count),
then I am at risk of losing data contained either in the modified slice or
any slices that may be affected by virtue of the fact that cylinders are
reallocated from one to another slice.
Is there any case where I can redefine the VTOC on a drive that has all
space allocated to existing slices and not be at risk for corrupting data in
one or more slices? My current understanding is that the answer is "No".
In other words, anytime that I make changes where the definition of existing
slices remain unchanged, I should be safe. When I make changes that change
existing slices, then I'm at risk. Is this correct? Or, are there other
cases where I might be safe or not? Such as, if I merely expand a slice by
adding more cylinders on to the end? What are the special cases, if any,
when dealing with the disk that contains the root files system?
Thanks for any help and sorry for adding to the general noise level on my
first attempt.
Todd
> -----Original Message-----
> From: Urie, Todd
> Sent: Tuesday, May 28, 2002 1:39 PM
> To: 'sunmanagers@sunmanagers.org'
> Subject: VTOC question
>
> My previous question regarding testing disk failures when
> using DiskSuite lead me to another question. I reviewed a
> few scripts that are intended to configure DiskSuite
> mirroring on a system. One of them used 'fmthard' to rewrite
> the VTOC on a disk that was in use as the root disk. I think
> I understand how 'fmthard' works to rewrite the VTOC. Based
> upon my understanding, it would seem that using the 'fmthard
> -s new_vtoc /dev/rdsk/c0t0d0s0' construct on a mounted disk,
> especially one that contains the root filesystem, would be bad.
>
> Am I correct in my understanding? Or is there something that
> I don't understand about how this works?
>
> Todd Urie
> Unix Support Specialist
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:24:23 EDT