I am currently doing a parity sync after replacing a parity disk on a server that has been offline for quite awhile.

It seems one of the disks is throwing errors.

Should I let the parity sync finish, it still has another 16 hours to go, or should I stop it, replace the drive and start again?

 

Diganostics are attached.

Thank you

 

 

 

Sep 17 13:03:23 Backup-Tower kernel: sd 13:0:7:0: [sdae] tag#4 CDB: opcode=0x85 85 06 20 00 00 00 00 00 00 00 00 00 00 40 e0 00 Sep 17 13:03:23 Backup-Tower kernel: md: do_drive_cmd: disk25: ATA_OP e0 ioctl error: -5 Sep 17 13:03:24 Backup-Tower emhttpd: error: mdcmd, 2640: Input/output error (5): write Sep 17 13:03:24 Backup-Tower kernel: mdcmd (1698): spindown 25 Sep 17 13:03:24 Backup-Tower kernel: sd 13:0:7:0: [sdae] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00 Sep 17 13:03:24 Backup-Tower kernel: sd 13:0:7:0: [sdae] tag#0 CDB: opcode=0x85 85 06 20 00 00 00 00 00 00 00 00 00 00 40 e0 00 Sep 17 13:03:24 Backup-Tower kernel: md: do_drive_cmd: disk25: ATA_OP e0 ioctl error: -5

backup_tower-diagnostics-20190917-1605.zip

Syslog doesn't show the problem start, but the disk dropped offline so there's no SMART, check connections and post new diags so we can see SMART.

What would your recommendation be with respect to the parity sync?

No point in continuing with a missing disk.

Thanks