I'm planning to upgrade one of my disks from 6TB to 10TB. Unfortunately, it is super slow and would take 200 days to complete...

 

Attached the log. Any thoughts?

tower-diagnostics-20190529-1101.zip

There appears to be a connection problem:

 

May 29 19:00:07 Tower kernel: ata1: hard resetting link
May 29 19:00:13 Tower kernel: ata1: link is slow to respond, please be patient (ready=0)
May 29 19:00:17 Tower kernel: ata1: COMRESET failed (errno=-16)
May 29 19:00:17 Tower kernel: ata1: hard resetting link
May 29 19:00:19 Tower kernel: ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
May 29 19:00:20 Tower kernel: ata1.00: configured for UDMA/33
May 29 19:00:20 Tower kernel: ata1: EH complete
May 29 19:00:21 Tower kernel: ata1.00: exception Emask 0x50 SAct 0x0 SErr 0x4090800 action 0xe frozen
May 29 19:00:21 Tower kernel: ata1.00: irq_stat 0x00400040, connection status changed
May 29 19:00:21 Tower kernel: ata1: SError: { HostInt PHYRdyChg 10B8B DevExch }
May 29 19:00:21 Tower kernel: ata1.00: failed command: READ DMA EXT
May 29 19:00:21 Tower kernel: ata1.00: cmd 25/00:40:18:79:11/00:05:00:00:00/e0 tag 0 dma 688128 in
May 29 19:00:21 Tower kernel:         res 50/00:00:5f:4c:6a/00:00:00:00:00/41 Emask 0x50 (ATA bus error)

 

Replace cables.

Can you see from the log, which cable is it?

 

The cable should be fine though as it has been perfectly working with the replaced disk (which was working well).

 

I can plug the cable better in again and try again. Question though is whether I can shut down the server and then restart the rebuild?

1 minute ago, steve1977 said:

which cable is it?

Disk13

 

1 minute ago, steve1977 said:

Question though is whether I can shut down the server and then restart the rebuild?

Yes, cancel rebuild and shutdown, it will restart from the beginning on next array start.

Thanks. Disk 13 is not the new disk. Let me look into it.

 

I am still a bit worried that I am adding too many TB to one SATA controller. Had issue before when upgrading disks though I remember from forum conversation that this is not seen as a potential issue.

Shut down and put all cables back in.

 

Now it seems to be even worse as the file system cannot be read anymore. See log attached. Any more thoughts?

tower-diagnostics-20190529-1141.zip

Check filesystem on disk13:

https://wiki.unraid.net/Check_Disk_Filesystems#Checking_and_fixing_drives_in_the_webGui

or

https://wiki.unraid.net/Check_Disk_Filesystems#Drives_formatted_with_XFS

Done, see log. Thanks!

tower-diagnostics-20190529-1230.zip

from the smart log on disk13 the diskready and notready counts might be concerning, Also i cannot locate those erors on any of my smarttests

0x0009  2            5  Transition from drive PhyRdy to drive PhyNRdy
0x000a  2            5  Device-to-host register FISes sent due to a COMRESET

this article speculates that may be a failing drive or error on drive

https://superuser.com/questions/1027912/many-transition-from-drive-phyrdy-to-phynrdy-causing-latencies-in-windows-7

 

Edited May 29, 20197 yr by Fiservedpi

Anything you'd suggest me to do?

You can run the run xfs_repair in maintenance mode from the link above or maybe test a different drive in slot 13 to rule out cables/controller if you have a spare one, if that works then RMA the drive. What are the speeds when moving data to the drive? are they normal or crap? just moving data not a rebuild 

Edited May 29, 20197 yr by Fiservedpi

1 hour ago, steve1977 said:

Done, see log. Thanks!

No need for new diags, jun run xfs_repair (without -n) and it should fix the filesystem, if it doesn't post the xfs_repair output.

 

Note, there were still ATA errors on the previous diags, though only 1, but there's still a connection issue.

Thanks. The disk is visible again. Let's see whether the data rebuild completes.

 

Understand from you that there may be some cabling issue, but no evidence of any harddisk failure, right?

26 minutes ago, steve1977 said:

but no evidence of any harddisk failure, right?

Correct, at least nothing apparent on the SMART data.