Hi there 

 

I would be grateful for advice on my Unraid 5.0-rc12a system with 8 drive array - main page capture attached.

No problems with system for 5 months. I checked parity a few days ago with no errors and then after loading 80GB I noticed a redball and 270 write errors on the parity drive (Seagate ST3000DM001).

 

I checked the smart, and ran a long disk test on all drives, then checked the smart again. This shows current pending sectors and reallocated sector count both zero but runtime bad block = 2.

 

google suggests Runtime bad blocks are the result of an unexpected voltage shift of a cell during a read, write or erase operation however I'm running a Corsair psu on the unraid approved list and have had no problems for months, and the server is on a UPS.

 

The system rebooted with a blue ball next to the parity drive so I reassigned the Seagate as the parity drive and rewrote parity to the parity drive - again without error.  I've since written another 80GB with no errors, so is the parity drive ok now?

 

So what's going on?? All I can see is the syslog showing BadCRC handshaking. What does this mean? Should I try moving the SATA cable, rebooting and checking the syslog?

 

Here's the syslog extract (full syslog and smart for the Seagate attached):

Dec  7 21:20:13 Tardis kernel: ata7.00: exception Emask 0x10 SAct 0x0 SErr 0x780100 action 0x6

Dec  7 21:20:13 Tardis kernel: ata7.00: irq_stat 0x08000000

Dec  7 21:20:13 Tardis kernel: ata7: SError: { UnrecovData 10B8B Dispar BadCRC Handshk }

Dec  7 21:20:13 Tardis kernel: ata7.00: failed command: READ DMA EXT

Dec  7 21:20:13 Tardis kernel: ata7.00: cmd 25/00:00:c0:a2:03/00:04:15:00:00/e0 tag 0 dma 524288 in

Dec  7 21:20:13 Tardis kernel:          res 50/00:00:7f:36:48/00:00:15:00:00/e0 Emask 0x10 (ATA bus error)

Dec  7 21:20:13 Tardis kernel: ata7.00: status: { DRDY }

Dec  7 21:20:13 Tardis kernel: ata7: hard resetting link

Dec  7 21:20:13 Tardis kernel: ata7: SATA link up 3.0 Gbps (SStatus 123 SControl 300)

Dec  7 21:20:13 Tardis kernel: ata7.00: configured for UDMA/133

Dec  7 21:20:13 Tardis kernel: ata7: EH complete

Dec  7 22:21:00 Tardis kernel: ata7.00: exception Emask 0x10 SAct 0x0 SErr 0x780100 action 0x6

Dec  7 22:21:00 Tardis kernel: ata7.00: irq_stat 0x08000000

Dec  7 22:21:00 Tardis kernel: ata7: SError: { UnrecovData 10B8B Dispar BadCRC Handshk }

Dec  7 22:21:00 Tardis kernel: ata7.00: failed command: READ DMA EXT

Dec  7 22:21:00 Tardis kernel: ata7.00: cmd 25/00:00:40:90:3e/00:04:f5:00:00/e0 tag 0 dma 524288 in

Dec  7 22:21:00 Tardis kernel:          res 50/00:00:ff:23:83/00:00:f5:00:00/e0 Emask 0x10 (ATA bus error)

Dec  7 22:21:00 Tardis kernel: ata7.00: status: { DRDY }

Dec  7 22:21:00 Tardis kernel: ata7: hard resetting link

Dec  7 22:21:01 Tardis kernel: ata7: SATA link up 3.0 Gbps (SStatus 123 SControl 300)

Dec  7 22:21:01 Tardis kernel: ata7.00: configured for UDMA/133

Dec  7 22:21:01 Tardis kernel: ata7: EH complete

 

Many thanks, Judy

syslog071213.txt

smartsdg061213.txt

Try a new SATA cable.

Thanks, will advise if a new cable fixed it in a month or two

DO NOT tie-wrap all the SATA cables together.  That will cause crosstalk from one cable to another and result in CRC errors unless you use shielded cables (and normally shielded cables are only available for ESATA use)

 

DO NOT run SATA cables parallel to power supply cables (keep a bit of distance between them)

 

These two tips to not be OCD compulsive to make it look "neat" will minimize CRC errors.

 

Joe L.