This is the only info i can find on the drive. There use to be a section in unRAID that would show me if the drive was prefailure, old, etc.

You should use an HBA that gives unRAID direct access to the disks, not a RAID controller, though there are some SMART config options to get the info from some RAID controllers.

I have a 24 disk array in a real server. I have to use a raid controller. There is no place on the MBD to plug in the array.

An HBA is a controller without RAID functions, e.g., LSI 9201-16i, 9207-8i, etc, or a controller in IT mode, e.g., 9211-8i.

I don't have any plans to upgrade the server at this time, but thank you for the suggestion. My only question is the availability of log files indicating the cause of the drive failure. Do they exist?

syslog may give a clue, together with the SMART report.

Tools Tab --> System Log is the first place i looked. Searching for the disk /dev/sdt specifically yielded some results.

Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 Sense Key : 0x5 [current] 
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 ASC=0x20 ASCQ=0x0 
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 CDB: opcode=0x85 85 06 20 00 00 00 00 00 00 00 00 00 00 40 e5 00
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 Sense Key : 0x5 [current] 
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 ASC=0x20 ASCQ=0x0 
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 CDB: opcode=0x85 85 06 20 00 00 00 00 00 00 00 00 00 00 40 98 00
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 Sense Key : 0x5 [current] 
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 ASC=0x20 ASCQ=0x0 
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 CDB: opcode=0x85 85 06 20 00 00 00 00 00 00 00 00 00 00 40 e5 00
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 Sense Key : 0x5 [current] 
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 ASC=0x20 ASCQ=0x0 
Sep 16 10:53:49 Tower01 kernel: sd 1:0:16:0: [sdt] tag#0 CDB: opcode=0x85 85 06 20 00 00 00 00 00 00 00 00 00 00 40 98 00

The last line of the SMART report offered the answer..."Device does not support Self Test logging".  Good thing you mentioned the SMART report, because i missed this line the first time.

 

smartctl 6.5 2016-05-07 r4318 [x86_64-linux-4.9.30-unRAID] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Vendor:               LSI
Product:              9750-4i    DISK
Revision:             5.12
User Capacity:        319,988,695,040 bytes [319 GB]
Logical block size:   512 bytes
Logical Unit id:      0x600050e05571c000dc5b00006cf10000
Serial number:        5VH36CLG5571C000DC5B
Device type:          disk
Local Time is:        Sat Sep 16 11:02:56 2017 EDT
SMART support is:     Available - device has SMART capability.
SMART support is:     Disabled
Temperature Warning:  Disabled or Not Supported

=== START OF READ SMART DATA SECTION ===
SMART Health Status: OK
Current Drive Temperature:     0 C
Drive Trip Temperature:        0 C

Error Counter logging not supported

Device does not support Self Test logging

 

This is the screen i was looking for that is available on my 2nd unRAID server. Which does not use a raid card and plugs directly into the MBD.

 

 

The reason i wanted a log file is to confirm that there was an actual error and this is not some bug with my raid card. Like in my post here. When i remove these "bad" drives from the array and test them in other machines, they are working properly. Re-partitioning and formating these drives works too. If i add them back to the array they work again for about a week before reporting the same error.

 

So these failures could be false positives. On the other hand every drive i replaced continued to work without issue. As you originally suggested, i may need a HBA card to find a definitive answer. Thanks for the help.

Edited September 16, 20178 yr by chris1259

Like I said already you're are using a RAID controller, and that's why you can't get a SMART report, hence why HBAs are recommended.