Can't remove drive due to I/O error

[Starbuckz8]

I'm finally getting around to removing the initial 1TB drives with larger.  It's been on my to-do list for a while but as there isn't much free space, I had to start out replacing one by one. 

 

My current problem, is when I go to remove a 1TB drive, drivepool errors out with "The request could not be performed because of an I/O device error". 

 

I currently have 8 drives in the server.  At the time of the removal, I can tell it was reading from J:\ - drive to be removed and writing to E:\ - just replaced larger drive.

 

I've since added the cleaned drives back off and am duplicating the entire pool as there is now enough space to handle it.  But what I'd like to know; and didn't see logs for, are if the I/O error is from the read or the write.

 

I've set the load balancer on J to nothing to prepare it for removal, but can't tell if I should actually have it set on E

[Starbuckz8]

No need to reply.  Out of impatience waiting for it to complete, I decided I'd just copy the last few remaining files to desktop and then copy them onto a different drive within the pool.  Even though I wasn't sure how well storage pool would handle direct copy into it's hidden folders.  That's when I realized that every time it hit one of Microsoft's backup 4GB blocks it would slow down to 0 and then error out as well.

 

As this drive wasn't one directly connected to the motherboard but an Intel SAS expanded I moved ports around and it told me what I already knew - the news is the gravest...

 

The irony, and incredibly funny part is when I added the drive that went bad I said to myself "if I add in more than is need now, I won't get around to this until it's too late". 

 

Gots nobody to blame but meself.  I just hope it's one of those 4GB files that takes an hour to recover from, instead of 2 days.  With MSFT you really, never know...  But hey!  If we all followed best practices; all the time, half of us probably would have best-practiced ourselves out of a job, right?

 

After it's done repairing the backup database I'll set everything to duplicate.  I promise.  Yeah - I've also been saying I'm going to CrashPlan Free to the office.  I haven't gotten around to that either.

[Christopher (Drashna)]

Well, I'm sorry to hear about the drive issues.

 

As for removing the drive, there is a "Force Damaged Disk Removal" option that will skip problem files like this. That way, it at least allows you to remove the drive from the pool.

And when it finishes, there will be a non-hidden PoolPart.xxxxx folder with the remaining contents of the disk, that it couldn't move. 

 

And yes, the software is fine with moving files around in the hidden PoolPart folders. Though, we generally recommend against it, there isn't a technical reason that you cannot. However, we do recommend remeasuring the pool after doing so (as well as stopping the StableBit DrivePool Service temporarily, if you're moving a lot of files around. 

 

 

 

 

As for the 4GB dat files for the Client Backup Database (at least, that's what I'm assuming you're referring to), that ... may cause issues depending on the EXACT data that is affected.  

 

If the data was duplicated, just skip it, and you'll be fine. If it wasn't .... that's a problem.  

At least, it's something to consider for the future. But if you're running StableBit Scanner, it may be able to help recover the file in question. Depending.