Disked droped from array after LOADSSSSSSSS of writes

[KoNeko]

i have a disk that is pushed off the array and the write count said 18,446,744,073,709,545,472

i didnt get any error /notice on it. i saw it by chance when i logged in.

i did a restart of the server and than i got a notice.

Unraid Disk 5 error
Alert [THANEKOS] - Disk 5 in error state (disk dsbl)
1593026701
WDC_WD30EFRX-68AX9N0_WD-WMC1T0944207 (sdg)
alert

Jun 24 21:23:37 ThaNekos kernel: sd 7:0:1:0: [sdg] 5860533168 512-byte logical blocks: (3.00 TB/2.73 TiB)
Jun 24 21:23:37 ThaNekos kernel: sd 7:0:1:0: [sdg] 4096-byte physical blocks
Jun 24 21:23:37 ThaNekos kernel: sd 7:0:1:0: [sdg] Write Protect is off
Jun 24 21:23:37 ThaNekos kernel: sd 7:0:1:0: [sdg] Mode Sense: 9b 00 10 08
Jun 24 21:23:37 ThaNekos kernel: sd 7:0:1:0: [sdg] Write cache: enabled, read cache: enabled, supports DPO and FUA
Jun 24 21:23:37 ThaNekos kernel: sdg: sdg1
Jun 24 21:23:37 ThaNekos kernel: sd 7:0:1:0: [sdg] Attached SCSI disk
Jun 24 21:24:24 ThaNekos emhttpd: WDC_WD30EFRX-68AX9N0_WD-WMC1T0944207 (sdg) 512 5860533168
Jun 24 21:24:24 ThaNekos kernel: mdcmd (6): import 5 sdg 64 2930266532 0 WDC_WD30EFRX-68AX9N0_WD-WMC1T0944207
Jun 24 21:24:24 ThaNekos kernel: md: import disk5: (sdg) WDC_WD30EFRX-68AX9N0_WD-WMC1T0944207 size: 2930266532

 

i can still check the Smart values of the disk in unraid but those are all good.

how is it possible it will do so many writes?

 

i had it do a health check monday and it was all ok.

 

it says drive is disabled. but no warnings in the logs further.

Edited June 24, 2020 by KoNeko

[itimpi]

Did you gather diagnostics zip file (via tools -> Diagnostics) BEFORE you rebooted?   We would need those to see what lead up to the event as the logs get reset on reboot unless you have enabled the syslog server (via Settings -> Syslog).    If not then the current diagnostics might still give a clue.

 

Note that a disk gets disabled when a write to it fails, and it takes manual intervention to get it back as discussed in this section of the online documentation.    Do you have notifications enabled?    If so you you should have been informed when the disk got disabled.

[KoNeko]

8 hours ago, itimpi said:

Did you gather diagnostics zip file (via tools -> Diagnostics) BEFORE you rebooted?   We would need those to see what lead up to the event as the logs get reset on reboot unless you have enabled the syslog server (via Settings -> Syslog).    If not then the current diagnostics might still give a clue.

 

Note that a disk gets disabled when a write to it fails, and it takes manual intervention to get it back as discussed in this section of the online documentation.    Do you have notifications enabled?    If so you you should have been informed when the disk got disabled.

I have syslog enabled.

i did reenable the disk again. and its done its party check and rebuild thing and all is fine again now. i added the diagnostics.

 

i did get a message the disk got disabled. but only after the restart.

 

thanekos-diagnostics-20200625-0450.zip