extrem slow Parity-Sync/Data-Rebuild.

[MrYoshii]

Hi i got a new parity drive. Now its realy slow.

 

I checkt with glances no io wait.

im not sure why its so slow.

im thankfull for every tipp

 

greetings MrYoshii

[JorgeB]

Please post the diagnostics.

[MrYoshii]

sry i forget that one

homeserver01-diagnostics-20210930-1643.zip

[JorgeB]

Check/replace cables on disk6 and try again.

[MrYoshii]

why do you think ther is a problem?

[MrYoshii]

i checkt the cabels still same thing i generated a new diagnostics

homeserver01-diagnostics-20210930-1800.zip

[JorgeB]

Still constant ATA errors on disk6:

 

Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: cmd 60/08:70:78:9a:05/05:00:00:00:00/40 tag 14 ncq dma 659456 in
Sep 30 17:59:33 HOMESERVER01 kernel:         res 40/00:88:f8:9f:05/00:00:00:00:00/40 Emask 0x50 (ATA bus error)
Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: status: { DRDY }
Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: failed command: READ FPDMA QUEUED
Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: cmd 60/38:78:80:9f:05/00:00:00:00:00/40 tag 15 ncq dma 28672 in
Sep 30 17:59:33 HOMESERVER01 kernel:         res 40/00:88:f8:9f:05/00:00:00:00:00/40 Emask 0x50 (ATA bus error)
Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: status: { DRDY }
Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: failed command: READ FPDMA QUEUED
Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: cmd 60/40:80:b8:9f:05/00:00:00:00:00/40 tag 16 ncq dma 32768 in
Sep 30 17:59:33 HOMESERVER01 kernel:         res 40/00:88:f8:9f:05/00:00:00:00:00/40 Emask 0x50 (ATA bus error)
Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: status: { DRDY }
Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: failed command: READ FPDMA QUEUED
Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: cmd 60/40:88:f8:9f:05/05:00:00:00:00/40 tag 17 ncq dma 688128 in
Sep 30 17:59:33 HOMESERVER01 kernel:         res 40/00:88:f8:9f:05/00:00:00:00:00/40 Emask 0x50 (ATA bus error)
Sep 30 17:59:33 HOMESERVER01 kernel: ata4.00: status: { DRDY }
Sep 30 17:59:33 HOMESERVER01 kernel: ata4: hard resetting link
Sep 30 17:59:38 HOMESERVER01 kernel: ata4: SATA link up 3.0 Gbps (SStatus 123 SControl 320)

 

Replace/swap both cables

[MrYoshii]

i have changed cables on that wd red drive an switch sata ports but still getting the same error. The error on the cache drive is now gone.

[JorgeB]

If errors persist with new cables/port it's likely a bad disk.

[MrYoshii]

but why is unraid not reporting a bad disk? the smart locks also good.

[ChatNoir]

18 minutes ago, MrYoshii said:

but why is unraid not reporting a bad disk? the smart locks also good.

SMART is only one part (problematic sectors, etc).

From your logs some commands do not get to the disk. Unraid tries again, reset the link, try again, etc. This is what slows down activities.

If it only happens to a single drive, it is probably not a controller issue.

If it is not a bad cable or individual port issue, it is probably the drive.

[MrYoshii]

ok

no i tryed it in a other pc with ubuntu. no errors in syslog i used testdisk to restore the partition and xfs_repair to restore the drive now i got the data back ^^ but i dont understand why on my unraid server it is not working.. i tryed diverent sata controllers and cabels

 

edit: i redid the wireing on all drives. now its working again

Edited October 1, 2021 by MrYoshii