Parity Disk - BLK_UPDATE_REQUEST: I/O ERROR

[deadsix]

My parity disk tonight suffered a BLK_UPDATE_REQUEST: I/O ERROR which caused it to be disabled by UnRAID automatically.During the initial parity build on the drive I encountered issues that caused the SATA link speed to be negotiated down to 3.0Gbps. Though I am not certain if the two issues are related. 

I am wondering if anyone can scour my logs to see if I should replace the drive or if I should start a parity rebuild with the same drive? Thanks!

Edit: Attempting to run a SMART test leads to the SMART test never completing with the following warnings appearing on the disk log

 

Oct 10 22:39:38 Normandy emhttpd: read SMART /dev/sdf
Oct 10 22:39:43 Normandy emhttpd: shcmd (20): /usr/local/sbin/set_ncq sdf 1
Oct 10 22:39:43 Normandy root: set_ncq: setting sdf queue_depth to 1
Oct 10 22:39:43 Normandy emhttpd: shcmd (21): echo 128 > /sys/block/sdf/queue/nr_requests
Oct 10 22:56:17 Normandy kernel: sd 5:0:0:0: [sdf] Synchronizing SCSI cache
Oct 10 22:56:17 Normandy kernel: sd 5:0:0:0: [sdf] Synchronize Cache(10) failed: Result: hostbyte=0x04 driverbyte=0x00
Oct 10 22:56:17 Normandy kernel: sd 5:0:0:0: [sdf] Stopping disk
Oct 10 22:56:17 Normandy kernel: sd 5:0:0:0: [sdf] Start/Stop Unit failed: Result: hostbyte=0x04 driverbyte=0x00

 

normandy-diagnostics-20211010-2222.zip

Edited October 11, 2021 by deadsix

[JorgeB]

Replace cables on parity disk and try again.

[deadsix]

8 hours ago, JorgeB said:

Replace cables on parity disk and try again.

 

Started a parity rebuild after replacing the SATA (data) cable and my disk log is clean I am no longer seeing the SATA link speed being negotiated to 3.0 Gbps. Thanks!

[deadsix]

 

On 10/11/2021 at 11:16 AM, deadsix said:

 

Started a parity rebuild after replacing the SATA (data) cable and my disk log is clean I am no longer seeing the SATA link speed being negotiated to 3.0 Gbps. Thanks!

Bumping this threaded because when I added a new hard to gain more space the data rebuild spawned the same SATA link speed negotiated 3.0Gbps.

Any ideas on a fix, thanks!

Edited October 13, 2021 by deadsix

[JorgeB]

Diags please.

[deadsix]

5 hours ago, JorgeB said:

Diags please.

 

Attached, thank you!

 

normandy-diagnostics-20211013-0833.zip

[JorgeB]

I'm not seeing any issues, what drive is linking @ SATA2?

[deadsix]

1 hour ago, JorgeB said:

I'm not seeing any issues, what drive is linking @ SATA2?

ST16000NM001G which is sdf in my array and also my parity drive.

[JorgeB]

Quote

Model Family:     Seagate Exos X16
Device Model:     ST16000NM001G-2KK103

SATA Version is:  SATA 3.3, 6.0 Gb/s (current: 6.0 Gb/s)

 

[deadsix]

1 hour ago, JorgeB said:

Model Family:     Seagate Exos X16
Device Model:     ST16000NM001G-2KK103

SATA Version is:  SATA 3.3, 6.0 Gb/s (current: 6.0 Gb/s

That can be explained by me turning off the server when I saw the errors return, I was very concerned that my parity could become invalid due to a read error and that I would lose my entire array. When I dumped the diagnostics the HDD wasn't reporting any negotiating errors and was running normally.  Thanks!

[JorgeB]

Syslog starts over after every reboot, if it happens again save the diags before rebooting.

[deadsix]

5 hours ago, JorgeB said:

Syslog starts over after every reboot, if it happens again save the diags before rebooting.

 

Understood, I fired up my server again and ran into a similar issue on just my parity drive, but this time the SATA link speed was not being negotiated down to 3.0Gbps. I've included the diagnostic file for review. 

Thanks!

normandy-diagnostics-20211013-1926.zip

[JorgeB]

That is usually a power/connection problem, replace cables on that disk.

[deadsix]

On 10/14/2021 at 2:56 AM, JorgeB said:

That is usually a power/connection problem, replace cables on that disk.

I rearranged my power cable application and the issues haven't resurfaced. I might need a more powerful PSU for these new high-density drives. Thanks JorgeB for your continued assistance.