Equal speeds during certify?

That is very normal, drives can vary widely.

2% is a very modest difference.

What about 68:56 remaining on 2 volumes, 26:57 - 27:15 remaining on the other 3 volumes. 2 are writing pattern at 2.99T, 3 are verifying, offset 2.5-2.7T.

There seems to be a clear difference between 2 of the 5 and the rest. I'm using an Oyen Digital Mobius 5-bay Thunderbolt 2 case, all have passed the SMART, all are on firmware 01.01A01.

First 2 are OS X disk6 and disk7, the others are disk1, 2, and 5
Thunderbolt reports SATA bus 0, ID 0 and 1 for the slow 2, SATA bus 0, ID 2-4 for the faster drives

Is there an explanation for this big a discrepancy?

This is just the disks. They are not identical, and apparently two of them are somewhat slower.

I could try to guess at causes, but unless the differences are dramatic, not worry about it.

Hey, here I am again 15 mo later, same problem: just started the certify 3-pass, same Oyen Digital Mobius 5-bay Thunderbolt 2 case, new drives. Ran out of space on my 3TB WD Blacks, replaced with 5 6TB HGST's.

Once again, 3 disks show 52:56-59 remaining
2 disks show 92:48 remaining

Same buses are slow, same are fast: SATA bus 0 ID 0-1 are slow again, ID 2-4 are fast. I think this may be a controller board issue? But I've had the controller board swapped out because the first one died.

You say unless the differences are dramatic, not to worry. But the faster drives are 1.75X faster than the slower drives. I would say nearly twice the speed IS dramatic. Grrrrrr.

Try swapping the drives to the other slots.

Cancel the rebuild.
Swap drives 1-2 and 3-4
Certify (SoftRAID will offer to pick up)
See what happens.

Try swapping the drives to the other slots.

Cancel the rebuild.
Swap drives 1-2 and 3-4
Certify (SoftRAID will offer to pick up)
See what happens.

When you say “pick up”, you mean where it left off, so I don’t have to start all over?

Yes, in most cases, SoftRAID can resume where it left off.

Yes, in most cases, SoftRAID can resume where it left off.

Interesting results from shuffling the drives around.

1. So I removed drives from the 1 - 4 slots, left #5 in place.

2. I put drives from slots 1 & 2 into slots 3 & 4 and vice versa. I did not replace them in any particular order.

3. Initially a few of them showed drive errors, made it sound like bad drives. I popped them out and reinserted, they worked just fine. Zero errors since.

4. The certify did start up where they left off, but for some reason one drive did not, it started over from offset 0.

5. Now it appears they're all on par speed wise. 63-64 hrs remaining, 81 hours on the one that started over from scratch. This is about 18 hours in since I restarted the certify process.

So I don't know what to think about the performance difference between slots, as there doesn't appear to be any difference at the moment.

Perhaps there was bad connectivity, that is always a possibility.

You can always check "read" performance on any disk by running activity Monitor and seeing what the throughput is.

Write performance would require a certify