You are not logged in.

#1 2021-01-06 14:21:00

regid
Member
Registered: 2016-06-06
Posts: 192

Does http://ix.io/2L3i shows a kernel bug? Traces start at line 178

The log, 483 lines long, is at http://ix.io/2L3i.
The call traces starts at line 178.
The kernel is 5.10.4-arch2-1.


pantum-p1000-p2000-p3000-m5100-m5200-ppd-driver (AUR): PPDs, drivers, for Pantum P1000, P2000, P3000, M5100, M5200 printers.
powerofforreboot.efi (AUR): Utilities to be used from within a UEFI boot manager or shell.

Offline

#2 2021-01-06 14:26:15

V1del
Forum Moderator
Registered: 2012-10-16
Posts: 12,246

Re: Does http://ix.io/2L3i shows a kernel bug? Traces start at line 178

Reads like that usb stick/disk is dead or in the process of dying. Does it work on other systems? Might also just be an issue with the specific USB/SATA bridge.

Last edited by V1del (2021-01-06 14:27:14)

Offline

#3 2021-01-06 14:28:05

schard
Member
From: Hannover
Registered: 2016-05-06
Posts: 1,091
Website

Re: Does http://ix.io/2L3i shows a kernel bug? Traces start at line 178

The log shows that probably either your SATA cable or hard disk (sdc) are defective.
Edit: Ninja'd

Last edited by schard (2021-01-06 14:29:30)

Offline

#4 2021-01-06 14:46:02

regid
Member
Registered: 2016-06-06
Posts: 192

Re: Does http://ix.io/2L3i shows a kernel bug? Traces start at line 178

Both arch, and windows 10, can sometimes read the partition table. As with gdisk -l /dev/sdc. Neither arch, nor windows 10, are able to do something else, such as interactively deleting a partition, with tools like gdisk or diskpart.


pantum-p1000-p2000-p3000-m5100-m5200-ppd-driver (AUR): PPDs, drivers, for Pantum P1000, P2000, P3000, M5100, M5200 printers.
powerofforreboot.efi (AUR): Utilities to be used from within a UEFI boot manager or shell.

Offline

#5 2021-01-06 14:47:48

V1del
Forum Moderator
Registered: 2012-10-16
Posts: 12,246

Re: Does http://ix.io/2L3i shows a kernel bug? Traces start at line 178

Then I'd say it's safe to assume the thing is toast.

FWIW can you run and post the output of

sudo smartctl -a /dev/sdc

?

Offline

#6 2021-01-09 17:03:35

regid
Member
Registered: 2016-06-06
Posts: 192

Re: Does http://ix.io/2L3i shows a kernel bug? Traces start at line 178

lsblk /dev/sdc reports the correct partitions of the hd.
smartctl -i /dev/sdc gets stuck. And so does ctrl-C the stuck smartctl. After a while, the prompt returned. A 2nd attempt after getting the prompt back,

# smartctl -i /dev/sdc
smartctl 7.2 2020-12-30 r5155 [x86_64-linux-5.10.5-arch1-1] (local build)
Copyright (C) 2002-20, Bruce Allen, Christian Franke, www.smartmontools.org

Smartctl open device: /dev/sdc failed: No such device
# lsblk /dev/sdc
lsblk: /dev/sdc: not a block device

A 97 lines long journalctl is at http://ix.io/2LrB.

There were no traces in the journal so far. Aren't traces output only when there is a kernel bug? Also, do note that there is a USB Mass Storage device involved.

Disconnecting and reconnecting the device, smartctl -i is taking a while, yet succeeds.

# smartctl -i /dev/sdc
smartctl 7.2 2020-12-30 r5155 [x86_64-linux-5.10.5-arch1-1] (local build)
Copyright (C) 2002-20, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family:     Fujitsu MHV
Device Model:     FUJITSU MHV2040AT
Serial Number:    NS10T712VL0E
Firmware Version: 000000A0
User Capacity:    40,007,761,920 bytes [40.0 GB]
Sector Size:      512 bytes logical/physical
Device is:        In smartctl database [for details use: -P show]
ATA Version is:   ATA/ATAPI-7 T13/1532D revision 4a
Local Time is:    Sat Jan  9 16:32:01 2021 UTC
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

A 94 lines long output of smartctl -a is at http://ix.io/2LrI.
smartctl -l reports no self-tests have been logged, even though I ran smartctl -t short a few minutes ago. smartctl -H reports

=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

Last edited by regid (2021-01-09 17:12:29)


pantum-p1000-p2000-p3000-m5100-m5200-ppd-driver (AUR): PPDs, drivers, for Pantum P1000, P2000, P3000, M5100, M5200 printers.
powerofforreboot.efi (AUR): Utilities to be used from within a UEFI boot manager or shell.

Offline

#7 2021-01-09 17:30:52

seth
Member
Registered: 2012-09-03
Posts: 18,998

Re: Does http://ix.io/2L3i shows a kernel bug? Traces start at line 178

No reallocations or pending sectors, but "no self-tests have been logged, even though I ran smartctl -t short a few minutes ago" sounds like the firmware took a hit in which case these values cannot be trusted either.
Do the IO errors continue after the re-connection of the device?
Did you try a different cable/slot?

In any event, do not trust the disk for now, that implies to backup all valuable data immediately.

Offline

#8 2021-01-09 19:11:05

regid
Member
Registered: 2016-06-06
Posts: 192

Re: Does http://ix.io/2L3i shows a kernel bug? Traces start at line 178

The device is used for back up. I have another back up, which is why I will not suffer great lost if the problematic device will end its life. Aren't traces imply bugs?

After the events in my previous post, I was able to enlarge one of the partitions with gdisk. Being able to enlarge that partition, I was able to reformat it, and seemingly was able to copy some files. I did not restart the machine, or otherwise notify the kernel explicitly about the change in the partition table of the device. After copying had seemingly finished, running gdisk -l made it hang. Twice. I had to ctrl-C the stuck gdisk. Twice.
The journal now has few traces. A 775 lines long log is at http://ix.io/2Lsu.

Last edited by regid (2021-01-09 19:17:53)


pantum-p1000-p2000-p3000-m5100-m5200-ppd-driver (AUR): PPDs, drivers, for Pantum P1000, P2000, P3000, M5100, M5200 printers.
powerofforreboot.efi (AUR): Utilities to be used from within a UEFI boot manager or shell.

Offline

#9 2021-01-09 19:21:14

V1del
Forum Moderator
Registered: 2012-10-16
Posts: 12,246

Re: Does http://ix.io/2L3i shows a kernel bug? Traces start at line 178

Traces imply that "something" went wrong. As it would be quite a bit of coincidence that both linux and windows were in agreement that the device can't be accessed properly but actually having an issue in their own code it stands to a more likely reason that the HW doesn't respond properly anymore to requests. You can also quite trivially test whether linux's code works correctly by checking whether you see the same traces with another drive, as there's little change on linux or your computers own hardware in the method in which they want to access another disk.

Last edited by V1del (2021-01-09 19:23:17)

Offline

Board footer

Powered by FluxBB