Hello folks, looking for some expertise here. I have a 8TB Seagate parity drive in my system that started throwing a small amount of read errors so I RMA'd the drive. Got the replacement, ran pre-clear without issue. Now today (3 days after installing replacement drive) I get hit with read errors again. Seems a little too unlikely to be another bad drive, but possible I suppose.

 

What's the best way to get to the bottom of this?

 

Is there a way to get a smart report on a running array from a disabled disk?

 

Here is the log for the drive:

Mar 7 21:46:11 UNRAID kernel: sd 1:0:3:0: [sdg] 15628053168 512-byte logical blocks: (8.00 TB/7.28 TiB)
Mar 7 21:46:11 UNRAID kernel: sd 1:0:3:0: [sdg] 4096-byte physical blocks
Mar 7 21:46:11 UNRAID kernel: sd 1:0:3:0: [sdg] Write Protect is off
Mar 7 21:46:11 UNRAID kernel: sd 1:0:3:0: [sdg] Mode Sense: 7f 00 10 08
Mar 7 21:46:11 UNRAID kernel: sd 1:0:3:0: [sdg] Write cache: enabled, read cache: enabled, supports DPO and FUA
Mar 7 21:46:11 UNRAID kernel: sdg: sdg1
Mar 7 21:46:11 UNRAID kernel: sd 1:0:3:0: [sdg] Attached SCSI disk
Mar 7 21:46:15 UNRAID emhttp: ST8000AS0002-1NA17Z_Z840BWJD (sdg) 7814026584
Mar 7 21:46:15 UNRAID kernel: md: import disk0: [8,96] (sdg) ST8000AS0002-1NA17Z_Z840BWJD size: 7814026532

Mar 7 21:46:19 UNRAID emhttp: shcmd (: /usr/local/sbin/set_ncq sdg 1 &> /dev/null
Mar 9 13:04:03 UNRAID kernel: sd 1:0:3:0: [sdg] Synchronizing SCSI cache
Mar 9 13:04:05 UNRAID kernel: sd 1:0:3:0: [sdg] UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
Mar 9 13:04:05 UNRAID kernel: sd 1:0:3:0: [sdg] CDB: opcode=0x8a 8a 00 00 00 00 00 56 f3 54 f0 00 00 02 08 00 00
Mar 9 13:04:05 UNRAID kernel: blk_update_request: I/O error, dev sdg, sector 1458787568

Mar 9 13:04:05 UNRAID kernel: sd 1:0:3:0: [sdg] UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
Mar 9 13:04:05 UNRAID kernel: sd 1:0:3:0: [sdg] CDB: opcode=0x8a 8a 00 00 00 00 02 8e f7 bc 78 00 00 01 00 00 00
Mar 9 13:04:05 UNRAID kernel: blk_update_request: I/O error, dev sdg, sector 10988534904

Mar 9 13:04:05 UNRAID kernel: sd 1:0:3:0: [sdg] Synchronize Cache(10) failed: Result: hostbyte=0x01 driverbyte=0x00

Mar 9 13:04:05 UNRAID kernel: blk_update_request: I/O error, dev sdg, sector 10988544120

 

Thanks everyone!

unraid-diagnostics-20160309-1957.zip

Drive issues are often not the cause of I/O errors. Possibly there was nothing really wrong with the first disk. Did you ask anyone about it?

 

What exactly do you mean the disk is disabled? Does it have a red X next to it?

 

That looks like a very small snippet from the syslog. Seldom useful. Go to Tools - Diagnostics and attach the complete diagnostics zip to your next post.

Drive issues are often not the cause of I/O errors. Possibly there was nothing really wrong with the first disk. Did you ask anyone about it?

 

What exactly do you mean the disk is disabled? Does it have a red X next to it?

 

That looks like a very small snippet from the syslog. Seldom useful. Go to Tools - Diagnostics and attach the complete diagnostics zip to your next post.

 

I've updated the original post with the diagnostic zip attached. When the original disk failed (or so I thought) I didn't do much digging as it had been working well for months without issue. Lesson learned for next time to do a bit more digging before a blind RMA.

 

The disk does have the red X next to it (see attached image). As a result I do not see a corresponding smart report in the diagnostics zip.

Disk dropped offline, power cycle the server and try to get a SMART report.

Disk dropped offline, power cycle the server and try to get a SMART report.

 

Ok, I've tried power cycling the box and the drive never came back. I also tried reseating all the cables and still nothing. Looks to be another bad drive, correct?

 

Thanks!

Ok, I've tried power cycling the box and the drive never came back. I also tried reseating all the cables and still nothing. Looks to be another bad drive, correct?

 

If it’s still offline after that then it’s probably bad.

Please post your experiences in the 8TB SMR drive thread located here https://lime-technology.com/forum/index.php?topic=39526.0 . This is the first I've heard on these forums of a Seagate 8TB dying (and second, lol).

 

Might be a good data point to track, that's all.

In my experience refurbished disks have at best a 50/50 chance of lasting more than a couple of months, I avoid using them, but when if I have to, only on a backup server.