RAID-5 expansion / harddisk replacement (with bigger disks) without total data-loss

Jay Kline list at
Tue Aug 19 18:40:01 CDT 2003

On Tue, 19 Aug 2003 12:49:15 +0200
>I think it should be possible, by first replacing all the 9GB harddisks,
>one by one (like if they were broken and needed to be replaced) and after
>that by extending the container's size to the full extent that should be
>possible by then (found hints about that, using afacli commands "container
>reconfigure"/"container extend").

Well, maybe.  Before you even venture down that road, what filesystem is it?
ext2 is realizable, but I have seen more failures from people trying than
anything else. ext3 is in the same boat, but because of the journal it does
seem a bit more reliable on resize.  reiserfs handles resizes very well. I
dont know about other filesystems, so you might want to double check how
well the filesystem itself will handle.

As for the raid drive-swap, Im not so sure that will work, either. I cant
speak for your raid controller, but its not uncommon for raids to require all
identical geometry drives (keeping in mind the newest raid control I have
worked with is now 7 years old). If that is the case, the raid controller
wont mirror properly to a different drive size. However, if your raid
controller only has the requirement that the drive must be >= 9G, you should
be ok.  I would want to test it on another machine first, if at all possible.
And at the very least, have good backups of everything, because whenever you
mess with the filesystem, no matter how confident you are, something can go
wrong and mess it all up.


More information about the Linux-PowerEdge mailing list