I/O error on device sdb
 

I have an old IBM X345 server that I am using and I suddenly started getting drive I/O problems. Below are the exact messages that are showing up in /var/log/messages related to this.

Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] Unhandled error code
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] CDB: Read(10): 28 00 05 0f b4 c3 00 00 80 00
Apr 10 17:15:21 localhost kernel: end_request: I/O error, dev sdb, sector 84915395
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] Unhandled error code
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] CDB: Read(10): 28 00 05 0f b5 43 00 00 80 00
Apr 10 17:15:21 localhost kernel: end_request: I/O error, dev sdb, sector 84915523
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] Unhandled error code
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] CDB: Read(10): 28 00 05 0f b4 fb 00 00 08 00
Apr 10 17:15:21 localhost kernel: end_request: I/O error, dev sdb, sector 84915451
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] Unhandled error code
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] CDB: Read(10): 28 00 05 0f b4 fb 00 00 08 00
Apr 10 17:15:21 localhost kernel: end_request: I/O error, dev sdb, sector 84915451
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] Unhandled error code
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
Apr 10 17:15:21 localhost kernel: sd 5:0:0:0: [sdb] CDB: Read(10): 28 00 05 0f b4 fb 00 00 08 00
Apr 10 17:15:21 localhost kernel: end_request: I/O error, dev sdb, sector 84915451

The server is running Fedora 12, everything else seems to be working fine. It has all 6 drive bays filled, 5 drive RAID 5 with 1 hot spare. I used the IBM tools to check the controller and according to that there are no problems with any of the drives or the controller, with the exception that the controller shows 16 misc messages.

Does anyone have any idea what is going on here? or how to fix it? I am moving everything to a new system, and reading seems to be working without any problems right now, but I am not ready with the new system yet so I need the old server that is giving the errors to continue working for now.

It report that particulair sector on your disk B has problem. Some application can cause the issue. It doesn't mean that hard disk has problem.

I was able to read the data without any issue and the problem sectors seemed to be changing depending on the user. I swapped out the SCSI controller and re-copied some of the data and the issue was fixed. I am not sure which one of those fixed it but it is working now.