Discussion:
Need Help With Disk Error Messages
(too old to reply)
Jim Beard
2016-08-03 13:46:15 UTC
Permalink
Last night I installed kernel-desktop-4.4.16-1.mga5-1-1.mga5, and this
morning found error messages for /dev/sd7. A look at journalctl turned
up additional messages for /dev/sr.

Neither of these is mounted, and the cd/dvd burner had no disk in it.

Deleting the date and machine name from the left of the journalctl line,
it looks like this:

kernel: sr 6:0:0:0: [sr1] tag#0 FAILED Result: hostbyte=DID_OK
driverbyte=DRIVER_SENSE
kernel: sr 6:0:0:0: [sr1] tag#0 Sense Key : Medium Error [current]
kernel: sr 6:0:0:0: [sr1] tag#0 Add. Sense: Unrecovered read error
kernel: sr 6:0:0:0: [sr1] tag#0 CDB: Read(10) 28 00 00 00 90 48 00 00 02
00
kernel: blk_update_request: critical medium error, dev sr1, sector 147744
kernel: sr 6:0:0:0: [sr1] tag#0 FAILED Result: hostbyte=DID_OK
driverbyte=DRIVER_SENSE
kernel: sr 6:0:0:0: [sr1] tag#0 Sense Key : Medium Error [current]
kernel: sr 6:0:0:0: [sr1] tag#0 Add. Sense: Unrecovered read error
kernel: sr 6:0:0:0: [sr1] tag#0 CDB: Read(10) 28 00 00 00 90 48 00 00 02
00
kernel: blk_update_request: critical medium error, dev sr1, sector 147744
kernel: Buffer I/O error on dev sr1, logical block 18468, async page read

kernel: sd 7:0:0:0: [sdc] tag#0 FAILED Result: hostbyte=DID_ERROR
driverbyte=DRIVER_SENSE
kernel: sd 7:0:0:0: [sdc] tag#0 Sense Key : Hardware Error [current]
[descriptor]
kernel: sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense
information
kernel: sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass through(16) 85 06
20 00 00 00 00 00
kernel: sd 7:0:0:0: [sdc] tag#0 FAILED Result: hostbyte=DID_ERROR
driverbyte=DRIVER_SENSE
kernel: sd 7:0:0:0: [sdc] tag#0 Sense Key : Hardware Error [current]
[descriptor]
kernel: sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense
information
kernel: sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass through(12)/Blank a1
06 20 da 00 00

An attempt to umount sdc7 yields the journalctl message
Aug 03 09:34:19 jdb.home.invalid udisksd[2545]: Cleaning up mount point /
run/media/jim/mblanguage (device 8:39 is not mounted)

The error messages for sdc7 repeat every 10 seconds. The sr1 messages
have not repeated in 50 minutes.

Both devices are old. The cd/dvd burner is on my main machine purchased
fall 2009 and /dev/sdc is an external USB 1TB drive purchased for backup
about that time.

Has anyone else see something like this with the new kernel?
Any ideas why error messages are being generated for a partition and cd
burner that are idle/unmounted?

Suggestions?

Cheers!

jim b.
--
UNIX is not user-unfriendly; it merely
expects users to be computer-friendly.
Jim Beard
2016-08-03 13:54:18 UTC
Permalink
Post by Jim Beard
Last night I installed kernel-desktop-4.4.16-1.mga5-1-1.mga5, and this
morning found error messages for /dev/sd7. A look at journalctl turned
up additional messages for /dev/sr.
Neither of these is mounted, and the cd/dvd burner had no disk in it.
Deleting the date and machine name from the left of the journalctl line,
kernel: sr 6:0:0:0: [sr1] tag#0 FAILED Result: hostbyte=DID_OK
Unrecovered read error kernel: sr 6:0:0:0: [sr1] tag#0 CDB: Read(10) 28
00 00 00 90 48 00 00 02 00 kernel: blk_update_request: critical medium
error, dev sr1, sector 147744
kernel: sr 6:0:0:0: [sr1] tag#0 FAILED Result: hostbyte=DID_OK
Unrecovered read error kernel: sr 6:0:0:0: [sr1] tag#0 CDB: Read(10) 28
00 00 00 90 48 00 00 02 00
kernel: blk_update_request: critical medium error, dev sr1, sector 147744
kernel: Buffer I/O error on dev sr1, logical block 18468, async page read
kernel: sd 7:0:0:0: [sdc] tag#0 FAILED Result: hostbyte=DID_ERROR
Hardware Error [current] [descriptor]
kernel: sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense
information kernel: sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass
through(16) 85 06 20 00 00 00 00 00 kernel: sd 7:0:0:0: [sdc] tag#0
FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_SENSE kernel: sd
7:0:0:0: [sdc] tag#0 Sense Key : Hardware Error [current] [descriptor]
kernel: sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense
information kernel: sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass
through(12)/Blank a1 06 20 da 00 00
An attempt to umount sdc7 yields the journalctl message Aug 03 09:34:19
jdb.home.invalid udisksd[2545]: Cleaning up mount point /
run/media/jim/mblanguage (device 8:39 is not mounted)
The error messages for sdc7 repeat every 10 seconds. The sr1 messages
have not repeated in 50 minutes.
Correction, the set of errors on sd7 will repeat almost immediately, but
then the next set will be about 10 minutes later.

Cheers!

jim b.
--
UNIX is not user-unfriendly; it merely
expects users to be computer-friendly.
Bit Twister
2016-08-03 15:20:58 UTC
Permalink
Post by Jim Beard
Last night I installed kernel-desktop-4.4.16-1.mga5-1-1.mga5, and this
morning found error messages for /dev/sd7. A look at journalctl turned
up additional messages for /dev/sr.
snip
Post by Jim Beard
Has anyone else see something like this with the new kernel?
Nope and I installed the kernel when it came out.
Post by Jim Beard
Any ideas why error messages are being generated for a partition and cd
burner that are idle/unmounted?
Item in /etc/fstab maybe
Post by Jim Beard
Suggestions?
Cold start of box, get a good known to be readable backup.
plan on new disk install in the near future.

Running with two hard drives and burner. and here is a
trimed/snipped snippet of my log.

$ uname -srvi
Linux 4.4.16-desktop-1.mga5 #1 SMP Tue Jul 26 09:23:40 UTC 2016 x86_64


$ journalctl -b | grep ' sd '
[sda] 1953525168 512-byte logical blocks: (1.00 TB/932 GiB)
[sda] 4096-byte physical blocks
[sda] Write Protect is off
[sda] Mode Sense: 00 3a 00 00
[sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[sda] Attached SCSI disk
[sdb] 1953525168 512-byte logical blocks: (1.00 TB/932 GiB)
[sdb] 4096-byte physical blocks
[sdb] Write Protect is off
[sdb] Mode Sense: 00 3a 00 00
[sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[sdb] Attached SCSI disk
Jim Beard
2016-08-03 19:02:23 UTC
Permalink
Post by Bit Twister
Post by Jim Beard
Last night I installed kernel-desktop-4.4.16-1.mga5-1-1.mga5, and this
morning found error messages for /dev/sd7. A look at journalctl turned
up additional messages for /dev/sr.
Item in /etc/fstab maybe
Commented out, with # at the start of the line.
Post by Bit Twister
Post by Jim Beard
Suggestions?
Cold start of box, get a good known to be readable backup.
plan on new disk install in the near future.
I have done the cold start of the external usb drive. The set of error
messages recurs each 10 minutes.

The cd drive no longer returns an error.

I will likely reboot the machine, but will leave that for later. I did
have to reboot my backup machine yesterday, as it went incommunicado.
Perhaps a power surge or something affecting both.

The readable backup and a new disk seem to be a good idea. The sdc drive
itself is a backup, so I am wondering if a Passport or some such might be
a good way to go. I like a hot backup running, but it has been months
since I used it, so do I really need or want it?

Any recommendations on manufacturer or model/spec in maybe 2TB size?
I have a bias for Western Digital, but not everything they make is what
one might hope for.

Cheers!

jim b.
--
UNIX is not user-unfriendly; it merely
expects users to be computer-friendly.
Doug Laidlaw
2016-08-04 02:41:39 UTC
Permalink
Post by Jim Beard
Any recommendations on manufacturer or model/spec in maybe 2TB size?
I have a bias for Western Digital, but not everything they make is what
one might hope for.
I use a 1 TB Seagate, the brand my shop pushes, model SRD00F1. Backups
were constantly going unreadable until I changed the filesystem to XFS.
No problems since. One Seagate HD failed early, repaired under guarantee,
but that was exceptional.

Doug.
--
Mageia release 5 (Official) for x86_64
running 4.4.13-desktop-1.mga5 with DE=MATE
faeychild
2016-08-04 03:53:07 UTC
Permalink
Post by Bit Twister
journalctl -b | grep ' sd '
Odd! I get no report, with three disks and a burner
--
faeychild
Running kde on 3.14.43-desktop-1.mga4 kernel.
Mageia release 4 (Official) for x86_64
Bit Twister
2016-08-04 06:10:30 UTC
Permalink
Post by faeychild
Post by Bit Twister
journalctl -b | grep ' sd '
Odd! I get no report, with three disks and a burner
could be a byproduct of your setup. I picked that pattern
by looking in my journal for something that matched the OP's
log. Looking in my journal showd that I could use the six'th
word as unique. See

$ journalctl -b | grep cache:
Aug 02 11:58:58 wb.home.test kernel: sd 0:0:0:0: [sda] Write cache:
Aug 02 11:58:58 wb.home.test kernel: sd 1:0:0:0: [sdb] Write cache:

PS: for any lurkers:

Reason the above works from my user account is because I have
added my GUID to the systemd-journal group.

Saves me having to get into root for journal research.

Also allows me to have a 5 line window pinned to top of my
desktop, watching the journal. My KDE autostart script
automagically launches it everytime I bring KDE up.

nohup xterm -fn 7x13 -fg black -bg SpringGreen3 \
-geom 175x5+20+0 -display :0 \
-e journalctl -fa --no-pager &
Jim Beard
2016-08-04 13:32:18 UTC
Permalink
Post by Jim Beard
Last night I installed kernel-desktop-4.4.16-1.mga5-1-1.mga5, and this
morning found error messages for /dev/sd7. A look at journalctl turned
up additional messages for /dev/sr.
Neither of these is mounted, and the cd/dvd burner had no disk in it.
Deleting the date and machine name from the left of the journalctl line,
kernel: sr 6:0:0:0: [sr1] tag#0 FAILED Result: hostbyte=DID_OK
Unrecovered read error kernel: sr 6:0:0:0: [sr1] tag#0 CDB: Read(10) 28
00 00 00 90 48 00 00 02 00 kernel: blk_update_request: critical medium
error, dev sr1, sector 147744
Grabbed a can of air, opened up the desktop case, and cleaned inside the
main machine and then cleaned the external usb drive of dust. That was
needed, so worthwhile in itself.

Moved the /dev/sdc usb external 1TB drive to another machine, and the
same problem recurs. The external drive is bad. It is still readable,
despite the error messages, so I have disconnected it completely and put
it to one side.

Next task is to decide what the replacement will be, another usb drive in
an external drive case for hot backup or a Passport-type backup to
connect for updating and then disconnect and store. Whatever I get, I
will attempt to transfer everything from the failing drive to the new
one, update from the machines backed up, and carry on.

Cheers!

jim b.
--
UNIX is not user-unfriendly; it merely
expects users to be computer-friendly.
Loading...