Drive Errors - Can't nail down culprit - What am I missing?

[rukiddin]

For whatever reason, my ATA4 drive gives me lots of errors (see attached log).

 

It's easy enough to identify the drive that's having trouble, but not so easy to ascertain why.

 

I have replaced cables, drives, even added a SATA controller, and still I'm having this extremely frustrating problem.  I MUST be missing something so simple.

 

I'm running on a CS2EE Mother Board (the original Official UNRAID MB), it has 6 SATA ports, when I use just 4 of them, everything works great.  When I add a 5th drive, either to a SATA controller, or to one of the remaining SATA ports on the motherboard, I get the errors in the log.

 

Sep 24 12:26:38 Jensen kernel: ata4: hard resetting link

Sep 24 12:26:44 Jensen kernel: ata4: SATA link up 1.5 Gbps (SStatus 113 SControl 310)

Sep 24 12:26:44 Jensen kernel: ata4.00: configured for UDMA/100

Sep 24 12:26:44 Jensen kernel: ata4: EH complete

Sep 24 12:26:44 Jensen kernel: ata4.00: limiting speed to UDMA/33:PIO4

Sep 24 12:26:44 Jensen kernel: ata4.00: exception Emask 0x10 SAct 0x0 SErr 0x4810000 action 0xe frozen

Sep 24 12:26:44 Jensen kernel: ata4.00: irq_stat 0x08400040, interface fatal error, connection status changed

Sep 24 12:26:44 Jensen kernel: ata4: SError: { PHYRdyChg LinkSeq DevExch }

Sep 24 12:26:44 Jensen kernel: ata4.00: failed command: READ DMA EXT

Sep 24 12:26:44 Jensen kernel: ata4.00: cmd 25/00:00:c0:a5:48/00:04:00:00:00/e0 tag 7 dma 524288 in

Sep 24 12:26:44 Jensen kernel:        res 50/00:00:bf:a5:48/00:00:00:00:00/e0 Emask 0x10 (ATA bus error)

Sep 24 12:26:44 Jensen kernel: ata4.00: status: { DRDY }

Sep 24 12:26:44 Jensen kernel: ata4: hard resetting link

Sep 24 12:26:50 Jensen kernel: ata4: SATA link up 1.5 Gbps (SStatus 113 SControl 310)

Sep 24 12:26:50 Jensen kernel: ata4.00: configured for UDMA/33

Sep 24 12:26:50 Jensen kernel: ata4: EH complete

 

Anybody got any ideas of anything I have tried, or is there a setting I should be changing?

 

Thanks

 

Mike

log_dump.txt

[dgaschk]

Try a different drive.

[rukiddin]

I have replaced the drive.. with 2 different drives..

[RobJ]

The errors indicate a temporary loss of the drive, as if it is jiggling loose very briefly.  Now that could be a loose SATA or power cable, either end, or a loose backframe connection, or it could be poor power, insufficient power due to too many devices on that power rail?  Because you said it works fine with 4 drives, but has problems when a fifth drive is added, also implicates a power issue.  I think you may need a better power supply, one with a single rail.  Or it could be that fifth cable is defective, loose leads.

[rukiddin]

I think it has to be the power supply.  I was hoping it was something else.  I'm going to run some tests with that in mind.  I don't think it's the power connectors, as I've run it with the drive in a NORCO drive bay, and outside the drive bay using 2 different power connectors.  I can copy data to the drive, but usually in that case the drive is the only one powered  up.  It primarily fails when all the drives are going at the same time.

 

I have an extra (smaller) power supply, I'll try hooking it up only to this one drive, and see what happens, that should confirm or eliminate the power supply as the problem.

 

Thanks

 

Mike