Ignore drive marked as Damaged with virtually no damage

[Parallax Abstraction]

So I have a pile of drives in my pool and one of them is an older 4TB Red. Recently, 8 of sectors totaling a whopping 4kb of data were flagged as bad and the drive was evacuated. I reformatted the drive and did multiple re-checks of it to see if this was the start of a spreading problem, but it isn't and I'm confident the drive is OK for the next while at least. However, if I re-add it to the pool, it won't balance to it because it reports as Damaged by Scanner. I'd like to ignore the Damaged warning on this drive for now unless it gets worse (similar to how you can tell Scanner to ignore SMART warnings), but I don't see a way to do that without telling the balancers to ignore all damaged drives, which I don't want to do. But I don't want to take this drive out of service over 4kb of bad sectors that aren't getting worse.

Is there any way I can have Scanner treat the drive as OK while still evacuating others that go Damaged? Thanks.

[Shane]

Hmm. Scanner checks by disk, not by volume, so I suspect an old trick where you split the drive into multiple partitions with the bad sectors in unallocated space won't work:

Disk: |<- volume 1 -> <- unallocated space containing 8 bad sectors -> <- volume 2 ->|

Only other option I can think of is telling Scanner to NOT scan the surface of that disk automatically (in Scanner GUI, click that disk, click "Disk settings" (has the three green horizontal bars next to it), tick "Never scan surface automatically", click Save) then expand the disk's entry so you can click on Edit blocks and select "Mark all Unreadable blocks Unchecked".

Scanner should then still alert on any new SMART errors but it'll be up to you to remember to manually scan that disk from time to time.

[Shane]

If you manually put a test file into that disk's poolpart and do a Re-measure, does it evacuate the file?

[Parallax Abstraction]

Yeah, I think that could work! I'll give that a shot and see what happens. Thanks!

[Parallax Abstraction]

Hmm, so this is odd. I did what you suggested and removed the drive from scanning and marked it as unchecked. It no longer shows as Damaged in StableBit Scanner. However, I added it back to my pool and it refuses to write anything to it. 🤔

[Parallax Abstraction]

2 hours ago, Shane said:

If you manually put a test file into that disk's poolpart and do a Re-measure, does it evacuate the file?

Tried that out. It's not evacuating it, but it's also not writing new data to it.

[Shane]

@Christopher (Drashna) Any ideas?

[Parallax Abstraction]

So, the plot thickens.

I did a lengthy re-format of the drive and the bad sector cleared. I've seen this happen before when a single block gets marked as bad. Kind of like doing an old-school low level format. I actually had Scanner do a full surface re-analysis of the drive and it now reports as healthy. But when I added it back to the pool, it's still not being written to. Now this got real odd. 🤔

[Parallax Abstraction]

I ended up opening a ticket with support and provided some info they requested but to my surprise this morning, the problem seems to have solved itself. One of my systems ran a new large backup last night and the formally empty drive now has 225GB on it. I recently added 2 12TB drives to the pool so maybe the balancing algorithm just didn't see a need to use the 4TB until recently? Not sure, but it seems to be working now so I guess all's well. I'll take it.

Thanks!