Seagate 8TB SED causes VERY slow OS load

ethanf

Dabbler
Joined
Jul 1, 2017
Messages
26
TL;DR: I added some Seagate 8TB SED (self encrypting drive) and it seems to have added MASSIVE delays in recognizing the drives and booting, along with some errors thrown here and there. Anyone else have issues or any suggestions? Possible bad drives?

Full Story:
I'm in the process of trying to retire my old HP server and recently spun up a new server (Supermicro X10SRH-CF) with a built in LSI SAS3008 onboard HBA. It did have an older firmware (P14 IT) but I left it as is and installed lastest TrueNAS without issue. I added 12x 10TB Seagate drives and created a new pool--awesome, no problems.

I needed more space to get all the data migrated from the old server since it has a massive single pool. I ordered 12x 8TB Seagate Enterprise SED drives (ST8000NM0105) and it has caused some weird stuff to happen. I am new to SED drives and from the research I've done I should be able to just use them without managing the encryption key. It should just use a default key. I don't want to manage the keys.

I added the drives in while powered on (hot swap) and created a new pool from them. I started transferring data and everything seemed fine. Then, the pool went completely offline and none of the drives were visible. I rebooted the server and it took FOREVER to POST. The lights on the backplane for the new drives seem to 'get stuck'. I decided to update the firmware on the SAS3008 to P16 since that's what the latest firmware shows on Broadcom's website. I updated the firmware and skipped the BIOS so that it would boot faster (my OS disk is on another controller). It booted quicker but then TrueNAS took nearly an hour to load up. I kept seeing "request for sata passthrough page completed with error 60" and "root mount waiting for CAM". Again, the lights on the drive bays would sit for a long time on the newly added drives. I ended up shutting the server down. Removing all the newly added drives, and started it back up. Booted up quickly with the existing 10TB drives and running without any issues.

I'm in the process of testing each drive individually on another machine, but it's going to take a long time. I saw another post that someone got some 'waiting for CAM' messages on known bad drives. These drives are supposed to be new even though they were manufactured in 2016.

Any ideas as to what is going on? Possible multiple drive failures?

Edit: The reason I decided to update the firmware on the SAS3008 was in case the HBA had some weird issue with self-encrypting drives.
 
Last edited:

ethanf

Dabbler
Joined
Jul 1, 2017
Messages
26
I tested the drives and several failed short generic test and one was totally dead. I suspect the bad drives caused all the issues--crazy how there isn't a smooth recovery when the HBA runs into issues connecting to a drive.
 
Top