Reallocated sector count continues to climb

zamboni wrote:
What did you do if SeaTools did not give you an "error code" - what did you put on the RMA?

Drive 2 is really about to fail. Pick something from this list, probably 659DC7AE:

http://www.seagate.com/staticfiles/supp ... codes.html

Honestly Zamboni, I've done so many RMAs I've lost track of how I reached my current process. I do not bother with Seatool but just make a note of the reallocated sector count. As I said, Seagate has always replaced them without comment. If I was you, I would ship the drive out ASAP.

Replying to this thread because it contained the most valuable info for my (probably VERY TYPICAL) situation: I am starting to see "bad sectors" after ~16months of service on one of my drives. In considering my situation and anticipating eventual drive replacement (I already have a backup on hand) I looked at the User Manual (!) for my NV+ and it says only "don't even power-off, just pull the questionable drive and replace it later" implying I can replace the original drive assuming it tests OK, or the new blank drive, doesn't matter (running XRAID).

Is it really this simple, and that's why there's no "sticky" about this anywhere? Any advantage or safety in Powering-Down first before pulling the questionable drive? My NV+ serves only to backup my PCs once-a-day so I can take it offline for as long as I want.

If you are going to remove a drive, test it and return it to the unit, that is best done with the unit powered off. If on the other hand are going to pull a drive and then replace it with another drive that is best done with the power on, the "hot-pull" + "hot-add" routine. If you hot pull a drive and immediately plug it back in, most often there will be not impact. When you hot pull a drive, the system will initially report the drive as dead, then report it as missing. At that point, is when the replacement should be inserted. The drive will be recognized, verified, tested, formatted and then resynched to the array. While it is best to add a drive with no partitions, it is not necessary, but you may get a message that the drive contains data.

I always use Windows Disk Management to wipe the partitions on a drive that has been used before I add it to an array.

For all of the reading I've done on this, I don't recall before seeing the info in your post.

Very clear PapaBear--thank you very much! Adding to my "emergency procedures" docs! :)

PapaBear wrote:

If you are going to remove a drive, test it and return it to the unit, that is best done with the unit powered off. If on the other hand are going to pull a drive and then replace it with another drive that is best done with the power on, the "hot-pull" + "hot-add" routine. If you hot pull a drive and immediately plug it back in, most often there will be not impact. When you hot pull a drive, the system will initially report the drive as dead, then report it as missing. At that point, is when the replacement should be inserted. The drive will be recognized, verified, tested, formatted and then resynched to the array. While it is best to add a drive with no partitions, it is not necessary, but you may get a message that the drive contains data. I always use Windows Disk Management to wipe the partitions on a drive that has been used before I add it to an array.

PapaBear wrote:
If you hot pull a drive and immediately plug it back in, most often there will be not impact.

I would think it should be more like this;

If you COLD pull a drive and immediately plug it back in (after testing), there will be no impact (unless the testing has modified the disk).

In my experience, if you HOT pull any drive, there will always be an impact and it will go through a resync cycle.

When you are replacing a disk, this is fine and expected.

If you are just testing and reusing a disk, you can avoid the resync by shutting down the nas and COLD pulling the drive, then returning it before powering up.

PapaBear wrote:
I always use Windows Disk Management to wipe the partitions on a drive that has been used before I add it to an array.

Is this really necessary? I thought the entire drive will be overwritten when it goes through the sync process.

dminches wrote:
Is this really necessary?

no, but doing it does not hurt anything.

I thought the entire drive will be overwritten when it goes through the sync process.

it does.