How to debug Seagate SAS drive failure to spin-up?

Hi everyone, wondering if any of you might be able to help me debug some odd drive behavior.

I bought 6x3TB Seagate Constellation ES.2 drives from Bitdeals back in Feb and they just arrived yesterday (package lost in transit, and found). I plugged the 6 drives into my case (with some help from the discord since I had some mechanical alignment issues), and powered on to find that only 4 of the drives seem to be working. The 4 that work are working as expected, smart looks clean, I’m checking them more thoroughly now.

I can see that all drives are recognized by the system at plug, which implies that SAS communication seems to work since I can see the WWN and general data about the drives at that time. But then the 2 fishy ones try to spin up…and nothing. dmesg just stops reporting anything after attempting to spin the drive up (just a bunch of “.” lines that I assume are the system waiting). smartctl can’t reach them. I do hear them when they try to spin up, but I don’t get anything after that.

All 6 drives are connected to the same LSI SAS9210-8i HBA and they all use the same style of SAS cable (recommended here). I’ve tried swapping ports on the HBA, as well as swapping plugs that are physically connected to the SAS drives, all the same result.

I’ve tried unplugging other drives to lower power demands on the system just to be safe and no change either. Since all drives are the same model and use the same set of SAS cables/power supply/HBA I’m inclined to think that this isn’t a simple 3.3V power disable issue or a cabling issue.

Any ideas on what to try next?

I’d start with the 3.3v mod and go from there.

Would that really apply here though? The drives are all same make and model (and some from same date code too), and are using same cables, same HBA, same power cables. I can’t seem to find a difference and the working drives are working without any 3.3v mod.

The 3.3v wire is not necessary in any HDD/SSD application nowadays, it’s part of an older spec. Give it a shot, then we’ll see if anything changes after that.

I’ll try to find some kapton tape lying around and report back. I did check the Constellation ES.2 spec though and these drives don’t use the pin 3 power disable feature (they mark pins 1/2/3 all as 3.3V reserved) in their pin descriptions.

If the drives were newer I’d be inclined to agree that this is the issue, but these seem old enough where this 3.3v “feature” wouldn’t apply.

Just snip the wire closest to the L, it’s not needed! Kapton tape is the worst way to 3.3v mod.

It has nothing to do with the age of the drives, btw.