Bought a new 500gb sata samsung and swapped out a 300gb seagate.  Rebuilding went fine.  This morning i get up and see the dirve has 3 errors.  Here's the tail end of the syslog, if someone could tell me what the problem might be.  Did i get a defective drive?  I've never had errors before this.  Thanks!

 

Apr 12 13:22:28 Server kernel: ata2: status=0x50 { DriveReady SeekComplete }

Apr 12 13:22:28 Server kernel: SCSI disk error : host 2 channel 0 id 0 lun 0 return code = 8000002

Apr 12 13:22:28 Server kernel: Current sd08:21: sns = 70  0

Apr 12 13:22:28 Server kernel: Raw sense data:0x70 0x00 0x00 0x00 0x00 0x00 0x00 0x0a 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00

Apr 12 13:22:28 Server kernel:  I/O error: dev 08:21, sector 657347208

Apr 12 13:22:28 Server kernel: md1: read error!

Apr 12 13:22:28 Server kernel: end_read_request 657347208/1, count: 1, uptodate0.

Apr 12 13:46:30 Server kernel: ata2: status=0x50 { DriveReady SeekComplete }

Apr 12 13:46:30 Server kernel: SCSI disk error : host 2 channel 0 id 0 lun 0 return code = 8000002

Apr 12 13:46:30 Server kernel: Current sd08:21: sns = 70  0

Apr 12 13:46:30 Server kernel: Raw sense data:0x70 0x00 0x00 0x00 0x00 0x00 0x00 0x0a 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00

Apr 12 13:46:30 Server kernel:  I/O error: dev 08:21, sector 687083776

Apr 12 13:46:30 Server kernel: md1: read error!

Apr 12 13:46:30 Server kernel: end_read_request 687083776/1, count: 1, uptodate0.

Apr 12 14:38:33 Server kernel: ata2: status=0x50 { DriveReady SeekComplete }

Apr 12 14:38:33 Server kernel: SCSI disk error : host 2 channel 0 id 0 lun 0 return code = 8000002

Apr 12 14:38:33 Server kernel: Current sd08:21: sns = 70  0

Apr 12 14:38:33 Server kernel: Raw sense data:0x70 0x00 0x00 0x00 0x00 0x00 0x00 0x0a 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00

Apr 12 14:38:33 Server kernel:  I/O error: dev 08:21, sector 751486672

Apr 12 14:38:33 Server kernel: md1: read error!

Apr 12 14:38:33 Server kernel: end_read_request 751486672/1, count: 1, uptodate0.

Looks like disk read errors on 3 different sectors.  Suggest you try a parity-check; if more errors pop up, consider replacing the drive.

Parity check completed with no errors.  So what does that mean?

Do you think the errors were reported during the rebuild, or after?

 

Also, have all drives been powered up since then?  Maybe power down, check all cable seating.  Let system sit for an hour or two, then restart and again run parity-sync.

I shut down and reseated cables.  Ran parity sync, still no errors.  Copied 50 gigs or so off it, no errors.  So maybe it was a fluke.  I've got another 500gb seagate on the way, so I'll probably swap that in instead and run some deep diagnostics on the samsung to see if i need to rma it.  Thanks for the input!

Well the day the replacement seagate arrived unraid borked good when writing to the array.  Web interface wouldn't load, couldn't get it shutdown, had to turn it off.  I can't remember all the things that happened or I tried, but none of it was good.  And the new seagate drive exhibited the same problems.  Different cables, different ports.  Finally I hooked up the drive to the motherboard sata instead of the promise card and it has written a few hundred GB so far without issues.  So it would appear either the promise card has issues with new/big drives or the promise linux drivers have some sort of flaw.  The motherboard sata is a VIA chipset.  I'll try 4.0 after I get things stable and see if that remedies things.  I haven't hooked up the samsung 500 i first bought, but i imagine it will work fine attached to the mobo also.

 

Related maybe: http://lime-technology.com/forum/index.php?topic=406.0