Buffer I/O error problem with drives

[spectromas]

I had a problem with one USB drive where suddenly I was getting a I/O error when trying to delete or do anything with a certain file. I ended up formatting it but the problem kept recurring so I just assumed that the drive was dying. I've just had the same thing happen to a second drive now so I'm suspicious.

 

Initially both drives had ntfs. I suspected perhaps it is just a bug in that so I am now trying with ext4 but I am still having problems. I tried to copy some files from one USB drive to another and the whole system froze and had to be forced off (no tty even). I am seeing this in the logs:

Feb  5 14:23:58 localhost kernel: sd 13:0:0:0: [sdd]
Feb  5 14:23:58 localhost kernel: Result: hostbyte=0x05 driverbyte=0x00
Feb  5 14:23:58 localhost kernel: sd 13:0:0:0: [sdd] CDB:
Feb  5 14:23:58 localhost kernel: cdb[0]=0x2a: 2a 00 00 11 5b b0 00 00 f0 00
Feb  5 14:23:58 localhost kernel: blk_update_request: I/O error, dev sdd, sector 1137584
Feb  5 14:23:58 localhost kernel: EXT4-fs warning (device sdd1): ext4_end_bio:317: I/O error -5 writing to inode 18350086 (offset 75497472 size 8388608 starting block 142228)
Feb  5 14:23:58 localhost kernel: Buffer I/O error on device sdd1, logical block 141942
Feb  5 14:23:58 localhost kernel: Buffer I/O error on device sdd1, logical block 141943
Feb  5 14:23:58 localhost kernel: Buffer I/O error on device sdd1, logical block 141944
Feb  5 14:23:58 localhost kernel: Buffer I/O error on device sdd1, logical block 141945
Feb  5 14:23:58 localhost kernel: Buffer I/O error on device sdd1, logical block 141946
Feb  5 14:23:58 localhost kernel: Buffer I/O error on device sdd1, logical block 141947
Feb  5 14:23:58 localhost kernel: Buffer I/O error on device sdd1, logical block 141948
Feb  5 14:23:58 localhost kernel: Buffer I/O error on device sdd1, logical block 141949
Feb  5 14:23:58 localhost kernel: Buffer I/O error on device sdd1, logical block 141950
Feb  5 14:23:58 localhost kernel: Buffer I/O error on device sdd1, logical block 141951
Feb  5 14:24:32 localhost kernel: usb 8-2: reset high-speed USB device number 5 using ehci-pci
Feb  5 14:25:20 localhost kernel: usb 8-2: reset high-speed USB device number 5 using ehci-pci
Feb  5 14:25:51 localhost kernel: usb 8-2: reset high-speed USB device number 5 using ehci-pci
Feb  5 14:26:22 localhost kernel: usb 8-2: reset high-speed USB device number 5 using ehci-pci
Feb  5 14:26:55 localhost kernel: usb 8-2: reset high-speed USB device number 5 using ehci-pci
Feb  5 14:27:26 localhost kernel: usb 8-2: reset high-speed USB device number 5 using ehci-pci
Feb  5 14:27:26 localhost kernel: sd 13:0:0:0: [sdd]
Feb  5 14:27:26 localhost kernel: Result: hostbyte=0x05 driverbyte=0x00
Feb  5 14:27:26 localhost kernel: sd 13:0:0:0: [sdd] CDB:
Feb  5 14:27:26 localhost kernel: cdb[0]=0x2a: 2a 00 00 13 5e 80 00 00 f0 00
Feb  5 14:27:26 localhost kernel: blk_update_request: I/O error, dev sdd, sector 1269376
Feb  5 14:27:26 localhost kernel: EXT4-fs warning (device sdd1): ext4_end_bio:317: I/O error -5 writing to inode 18350086 (offset 109051904 size 4194304 starting block 158702)
Feb  5 14:27:26 localhost kernel: buffer_io_error: 20 callbacks suppressed
Feb  5 14:27:26 localhost kernel: Buffer I/O error on device sdd1, logical block 158416
Feb  5 14:27:26 localhost kernel: Buffer I/O error on device sdd1, logical block 158417
Feb  5 14:27:26 localhost kernel: Buffer I/O error on device sdd1, logical block 158418
Feb  5 14:27:26 localhost kernel: Buffer I/O error on device sdd1, logical block 158419
Feb  5 14:27:26 localhost kernel: Buffer I/O error on device sdd1, logical block 158420
Feb  5 14:27:26 localhost kernel: Buffer I/O error on device sdd1, logical block 158421
Feb  5 14:27:26 localhost kernel: Buffer I/O error on device sdd1, logical block 158422
Feb  5 14:27:26 localhost kernel: Buffer I/O error on device sdd1, logical block 158423
Feb  5 14:27:26 localhost kernel: Buffer I/O error on device sdd1, logical block 158424
Feb  5 14:27:26 localhost kernel: Buffer I/O error on device sdd1, logical block 158425
Feb  5 14:28:47 localhost kernel: usb 8-2: reset high-speed USB device number 5 using ehci-pci

It is hadn't happened with a second drive I wouldn't even be questioning this but assuming this is not just a dying drive and it is not a bug in ntfs, can it be something else? I'm starting to wonder if there is something in my kernel that might be causing this?

[spectromas]

Thought I would post a follow up to this and mark it as solved. I started to have a look in to the message I was getting when I got a kernel panic, it turned out to be related to scheduling so I tried disabling BFS and sure enough there have been no more issues.

[digifuzzy]

idle thoughts...

1) If these are USB devices - I would lean towards the USB hub being buggy.

2) ehci? ahci driver instead?

[spectromas]

idle thoughts...

1) If these are USB devices - I would lean towards the USB hub being buggy.

 

This is a laptop so would it be all the ports if this were the case?

 

2) ehci? ahci driver instead?

 

I tried disabling ehci in the usb section for the kernel but I would only achieve 1 mbps transfer rate like that, unless that's not what you mean?

 

I'm trying now with a different port and not copying usb>usb, rather usb>internal drive then internal drive>usb. Don't know if it will make a difference and it's a much longer process considering how much I have to transfer.

[digifuzzy]

Then I would look at the USB cable. I've heard of flakey/bizarre errors from bad or unshielded cables.

Beyond that, I got nothing. Sorry

[spectromas]

I suppose that could account for one of the drives (the original one that keeps giving me this problem) but the other was a stick I plugged directly in to the port. Unless there are any other suggestions I guess I'll just have to bite the bullet and buy a new drive.

 

Thanks again for the help.

[spectromas]

Just ran badblocks on it, it got to a certain stage and the whole computer froze again. Think it must be dead a dead drive. Hopefully the other was just a coincidence.