Mailinglist Archive: opensuse-factory (1324 mails)

< Previous Next >
Re: [opensuse-factory] i/o error on sdb shown in strg+alt+f10
  • From: Thomas Langkamp <thomas.lassdiesonnerein@xxxxxx>
  • Date: Sat, 10 Oct 2015 23:04:36 +0200
  • Message-id: <56197D64.6060804@gmx.de>
Thanks Carlos and the others
good information :)
reply inline

Am 10.10.2015 um 14:08 schrieb Carlos E. R.:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On 2015-10-10 09:24, Thomas Langkamp wrote:

Could this mean that my SSD dies?

Might be, but first try replacing the cable. I really do hope SSDs
last longer than four years, unless it runs full time.

uptime 6000 hours.
I did a smartctl -t short which showed no errors

then I saw that smartctl tells me that tomshardware/crucial suggests a firmware update because some of those drives are known to hang every hour after 5000 hours uptime. However, then my ssd would be 1000 hours late with this known bug. I updated the firmware nevertheless AND also changed the cable - why not. Now I will have to wait if the error comes back.

And where is the output of Strg+Alt+F10 stored?

In the system log, /if/ the system doesn't crash before it can write
to the disk.

my system log seems to be non-persistend. however I grabbed pen and paper when it showed up some minutes ago. There are many repeating lines with this:

"date+time kernel: [long-number] blk_update_request: I/O error /dev/sdb2 sector $changing-number"

sdb2 is root, ext4, SSD, latest TW x64

I canĀ“t find the i/o error messages in /var/log/messages, dmesg,
xsession-errors or boot.log. Are there other important log-files I
missed?

dmesg would show everything, if the system is still running after the
crash. boot.log would show nothing. "/var/log/messages" should have
the same as dmesg shows, /if/ it is enabled, because in 13.2 it is
not. Instead, you have to query "journalctl".

Which diag program do you use to check your SSD with mounted root?

smartctl :-)
See the man page, it has examples.

The test runs while the system runs, in the disk hardware, without
involving the system cpu. When it finishes, you just use "smartctl -a
/dev/sdX" to query the result.

I don't use SSD, but rotating disks, so there will be differences in
the output. There should be a life estimation.

- --
Cheers / Saludos,

Carlos E. R.

(from 13.1 x86_64 "Bottle" (Minas Tirith))
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)

iF4EAREIAAYFAlYY/7AACgkQja8UbcUWM1xGVwD/caEIr2btRSv5IL9wm3cdi69Z
KK+dTBl7NFOjRZwycqIA+gIukqKABqWgJXsRW+PyAve/JJn+Np4zXF/vrGoDKFbN
=iD8k
-----END PGP SIGNATURE-----



--
To unsubscribe, e-mail: opensuse-factory+unsubscribe@xxxxxxxxxxxx
To contact the owner, e-mail: opensuse-factory+owner@xxxxxxxxxxxx

< Previous Next >