Oh, and fordem: Why would I need to recreate the RAID when I put in the disk I took out? They'd still have both the same contents, it shouldn't have to do nothing to the disk but accept it as its long lost brother, right? Otherwise, my spider sense is detecting suboptimal implementation xD
gunrunnerjohn has already answered this, I just want to add one thing - ALL RAID arrays work this way, lose a disk - flag the array as degraded, add a disk to a degraded array - do a rebuild - the only way around this is to not power up the unit with a disk removed.
Interestingly enough, you
should be able to remove both disks, power the unit up, upgrade the firmware power the unit down, replace the disks (in the same slots please) and on power up it should recognize them as a valid RAID1 array. The ability to do a firmware flash without disks was added in 1.03.
Please note that
should, I haven't tested this, but I do know that you can put a RAID1 pair in a new enclosure and it will recognize them with no difficulty.
Thanks everyone for your replies... I'll wait for the new HDD to arrive before I do anything. But if rebuilding a RAID is really so shaky with the DNS-323, is there any point to use RAID1 at all in it? Wouldn't I be better off having twice the amount of storage, then? Of course that'd require me to buy one more disk to have a complete backup, but if I can't rely on the RAID for anything, I'd rather not waste the disk space on it.
I need to make one thing VERY clear, although seeing as I have already stated it and you apparently still haven't gotten it -
RAID is not a backup.
What RAID is about is continuing to have your data available even if a disk were to fail - a RAID array is not
required to provide a mechanism to rebuild, even though that is a very desirable feature and although most manufacturers do provide it - there are some that don't.
Consider these scenarios ...
#1 - you are running a small business and your point of sale system does not have redundant disks, but you do backup to tape every night. At 9:00 am just as you're opening up you have a disk failure, so you call a tech who comes in and replaces the disk and restores the backup, but it techs the tech an hour to get there, and the process takes another hour, so you're without your PoS (and losing business) for two hours.
#2 - you are running a small business and your point of sale system has redundant disks which do not rebuild. At 9:00 am just as you're opening up you have a disk failure, but the system continues to function because the data remains accessible. You call a tech who comes in at 5:00 pm, backs up the system, replaces the disk and restores the backup - in this case you had the use of the PoS all day and lost no business.
That is what RAID is about - minimizing the impact of downtime - in this case by deferring the required downtime to a convenient period outside of business hours - it's not about backup, and so, if you don't need your data to be constantly available then you don't need RAID and if you can make better use of the disk space then by all means do so.