How to repair system if kernel panic?

Question

I've just updated packages:

2013-05-23 11:15:34 startup archives unpack
2013-05-23 11:15:35 upgrade locales:all 2.13-38 2.17-3
2013-05-23 11:15:35 status half-configured locales:all 2.13-38
2013-05-23 11:15:35 status unpacked locales:all 2.13-38
2013-05-23 11:15:35 status half-installed locales:all 2.13-38
2013-05-23 11:15:35 status triggers-pending man-db:amd64 2.6.3-3
2013-05-23 11:15:35 status half-installed locales:all 2.13-38
2013-05-23 11:15:38 status half-installed locales:all 2.13-38
2013-05-23 11:15:38 status unpacked locales:all 2.17-3
2013-05-23 11:15:39 status unpacked locales:all 2.17-3
2013-05-23 11:15:39 upgrade libc6:i386 2.13-38 2.17-3
2013-05-23 11:15:39 status half-configured libc6:i386 2.13-38
2013-05-23 11:15:40 status unpacked libc6:i386 2.13-38
2013-05-23 11:15:40 status half-configured libc6:amd64 2.13-38
2013-05-23 11:15:40 status half-installed libc6:i386 2.13-38
2013-05-23 11:15:47 status half-installed libc6:i386 2.13-38
2013-05-23 11:15:47 status unpacked libc6:i386 2.17-3
2013-05-23 11:15:48 status unpacked libc6:i386 2.17-3
2013-05-23 11:15:48 upgrade libc6:amd64 2.13-38 2.17-3
2013-05-23 11:15:48 status half-configured libc6:amd64 2.13-38
2013-05-23 11:15:48 status unpacked libc6:amd64 2.13-38
2013-05-23 11:15:48 status half-installed libc6:amd64 2.13-38
2013-05-23 11:15:53 status half-installed libc6:amd64 2.13-38
2013-05-23 11:15:53 status unpacked libc6:amd64 2.17-3
2013-05-23 11:15:54 status unpacked libc6:amd64 2.17-3
2013-05-23 11:15:54 trigproc man-db:amd64 2.6.3-3 2.6.3-3
2013-05-23 11:15:54 status half-configured man-db:amd64 2.6.3-3
2013-05-23 11:16:00 status installed man-db:amd64 2.6.3-3
2013-05-23 11:16:01 startup packages configure
2013-05-23 11:16:01 configure libc6:amd64 2.17-3 <none>
2013-05-23 11:16:01 status unpacked libc6:amd64 2.17-3
2013-05-23 11:16:02 status unpacked libc6:amd64 2.17-3
2013-05-23 11:16:02 status half-configured libc6:amd64 2.17-3
2013-05-23 11:16:57 status installed libc6:amd64 2.17-3
2013-05-23 11:16:57 configure libc6:i386 2.17-3 <none>
2013-05-23 11:16:57 status unpacked libc6:i386 2.17-3
2013-05-23 11:16:57 status unpacked libc6:i386 2.17-3
2013-05-23 11:16:57 status half-configured libc6:i386 2.17-3
2013-05-23 11:16:57 configure locales:all 2.17-3 <none>
2013-05-23 11:16:57 status unpacked locales:all 2.17-3
2013-05-23 11:16:57 status unpacked locales:all 2.17-3
2013-05-23 11:16:57 status half-configured locales:all 2.17-3

Debian GNU/Linux jessie/sid \n \l

and since that my computer gives me kernel panic. :/

To be more exact something like this:

init1 segfault at 3f ip .. [i don't remeber rest of message]

kernel panic not syncing. Attempted to kill init.

Init not taited..

How can I deal with it?

I've found something like this: http://forums.debian.net/viewtopic.php?f=10&t=96879 As far as I remeber (or maybe I'm wrong) some kind of libc was updated.. I tried to solved according to comment 21 but I have no:

 /lib/libc.so.6

file.

Added pic when booting stops.

added logs from /var/log/apt/term.log:

Log started: 2013-05-23  09:57:05
Selecting previously unselected package less.
(Reading database ... 
(Reading database ... 5%
(Reading database ... 10%
(Reading database ... 15%
(Reading database ... 20%
(Reading database ... 25%
(Reading database ... 30%
(Reading database ... 35%
(Reading database ... 40%
(Reading database ... 45%
(Reading database ... 50%
(Reading database ... 55%
(Reading database ... 60%
(Reading database ... 65%
(Reading database ... 70%
(Reading database ... 75%
(Reading database ... 80%
(Reading database ... 85%
(Reading database ... 90%
(Reading database ... 95%
(Reading database ... 100%
(Reading database ... 320906 files and directories currently installed.)
Unpacking less (from .../archives/less_458-1_amd64.deb) ...
Processing triggers for mime-support ...
Processing triggers for man-db ...
Setting up less (458-1) ...
Log ended: 2013-05-23  09:57:16

Log started: 2013-05-23  11:15:33
(Reading database ... 
(Reading database ... 5%
(Reading database ... 10%
(Reading database ... 15%
(Reading database ... 20%
(Reading database ... 25%
(Reading database ... 30%
(Reading database ... 35%
(Reading database ... 40%
(Reading database ... 45%
(Reading database ... 50%
(Reading database ... 55%
(Reading database ... 60%
(Reading database ... 65%
(Reading database ... 70%
(Reading database ... 75%
(Reading database ... 80%
(Reading database ... 85%
(Reading database ... 90%
(Reading database ... 95%
(Reading database ... 100%
(Reading database ... 320928 files and directories currently installed.)
Preparing to replace locales 2.13-38 (using .../locales_2.17-3_all.deb) ...
Unpacking replacement locales ...
Preparing to replace libc6:i386 2.13-38 (using .../archives/libc6_2.17-3_i386.deb) ...
De-configuring libc6:amd64 ...
Checking for services that may need to be restarted...
Checking init scripts...
Unpacking replacement libc6:i386 ...
Preparing to replace libc6:amd64 2.13-38 (using .../libc6_2.17-3_amd64.deb) ...
Checking for services that may need to be restarted...
Checking init scripts...
Unpacking replacement libc6:amd64 ...
Processing triggers for man-db ...
Setting up libc6:amd64 (2.17-3) ...
Checking for services that may need to be restarted...
Checking init scripts...

Restarting services possibly affected by the upgrade:
  samba: restarting...done.
  mysql: restarting...done.
  cron: restarting...done.
  atd: restarting...done.
  apache2: restarting...done.

Services restarted successfully.
Setting up libc6:i386 (2.17-3) ...
dpkg: error processing libc6:i386 (--configure):
 subprocess installed post-installation script was killed by signal (Segmentation fault)
Setting up locales (2.17-3) ...
Installing new version of config file /etc/locale.alias ...
dpkg: error processing locales (--configure):
 subprocess installed post-installation script was killed by signal (Segmentation fault)
Errors were encountered while processing:
 libc6:i386
 locales

Log ended: 2013-05-23 11:16:59

Solution

I did as schaiba said, so:

1. backup your sensitive stuff (for me only /home dir)
2. reinstall operating system

Comments

[user]

Try booting from rescue media, mount and chroot into your root partition, mount anything else needed (including /usr) and then reinstall glibc (or whichever C library you are using; most likely it is glibc). apt-get --reinstall --only-upgrade install libc-bin libc6 might be a good place to start. When you get your system working again, apt-get --reinstall --only-upgrade it all, in case something is broken.

[xliiv]

chroot gives me "segmentation fault" :/

[user]

Does chroot segfault even if you use that on known good media such as your installation CD? If so, it's likely your hardware is broken somehow.

[user]

ld is the dynamic linker, so I would focus on the very basic parts of the system: binutils, glibc, ...

[xliiv]

hardware seems to be fine, because on the same computer i also have a Linux Mint and I can use it. When I boot from live USB i can chroot Mint but fails with my Debian

[user]

But which chroot are you using to do it?

[user]

I also just now saw that this is about Debian sid. Sid is known to see regular breakage and even the official Debian web site says Use it at your own risk!. If you want something reasonably stable, at least go with testing.

[Jeff Hewitt]

@MichaelKjörling This doesn't necessarily point to faulty hardware. I faced a similar issue recently (though with Squeeze) after a forced reinstall of libc. chroot failed to execute bash.

[schaiba]

I'd suggest you reinstall, as it's the simplest and shortest way, since it's a libc issue, an essential component of the system, then restore from backups.

[Jeff Hewitt]

@xliiv I'm sorry to report that I couldn't solve my libc issue and had to reinstall as @schaiba suggests.

[xliiv]

Ok, i reinstalled it. So what should i do with this topic? Thanks to you all for effort.

[Jeff Hewitt]

@xliiv Add this as an answer to your question maybe.

[clerksx]

"All distributions"? Arch Linux doesn't.

[vonbrand]

@ChrisDown, then dump it. This is an elementary fallback measure.

[clerksx]

I see no reason to. It has already been discussed -- there are plenty of far better ways. I use the (maintained) LTS kernel as fallback, and not some unsupported previous version which could potentially be incompatible with the rest of the system. That is a far better method.