iv had almost all drives showing errors, as well as 3 drives disabled and had to rebuild twice in 48 hrs! also dockers disappearing. now im afraid of losing data, always happens over night but everything is fine during the day

tower-diagnostics-20181222-0900.zip

Edited December 22, 20187 yr by vixfix

All the disks with errors are connected to the same expander, there are 6 disks on a different expander with no errors, also no errors for the single disk on the onboard SATA ports, so start there, either the expander or the cables connecting it to the HBA, also possibly power for those disks if it's separate from the other ones.

Forgot to mention, make sure you disable spindown for disk27 and parity2, Unraid can't spin down SAS disks and it's spamming your log with the errors below causing several syslog hours to be missing:

 

Dec 21 16:11:53 Tower kernel: mdcmd (3485): spindown 27
Dec 21 16:11:53 Tower kernel: md: do_drive_cmd: disk27: ATA_OP e0 ioctl error: -5
Dec 21 16:11:53 Tower kernel: mdcmd (3486): spindown 29
Dec 21 16:11:53 Tower kernel: md: do_drive_cmd: disk29: ATA_OP e0 ioctl error: -5
Dec 21 16:11:54 Tower emhttpd: error: mdcmd, 2639: Input/output error (5): write
Dec 21 16:11:54 Tower kernel: mdcmd (3487): spindown 27
Dec 21 16:11:54 Tower kernel: md: do_drive_cmd: disk27: ATA_OP e0 ioctl error: -5
Dec 21 16:11:54 Tower kernel: mdcmd (3488): spindown 29
Dec 21 16:11:54 Tower emhttpd: error: mdcmd, 2639: Input/output error (5): write

 

thank you, trying to learn how to read the diag output. understand about 5% of the files. what file has the output i should read?

 

edit: found file

Edited December 22, 20187 yr by vixfix