How to install to existing Linux from a LiveDVD, etc...?

[BryanFRitt]

I tried to install software similar to how stated on
http://stackoverflow.com/questions/1086 ... -on-debian
Editing the /etc/apt/sources.list fist with testing, // they did experimental
I did this by exiting out of KDE,
then selecting console login,
then
sudo apt-get install -t testing libc6-amd64 libc6-dev libc6-dbg
I crashed or something in the middle,
I rebooted into 'recovery' mode,
It didn't even leave me a way to type. // rebooted a few times. all times failed, both regular, and recovery
Left with Blinking Alpha and Scroll lock.

I also have Vista 32bit on the this system, and
still have Debian 7.0 64bit DVDs, and have a DVD burner on this system, and 5 blank DVD-R,
I've already commented out the the /etc/apt/sources.list testing line, (using some software to edit Linux files from Windows)

Is there a way I can run Linux commands on this Linux system from Windows, or a Live CD/DVD, etc...?
(since I can't type get to a point where I can type in commands from this Linux system)
sudo apt-get update
sudo apt-get install -t stable libc6-amd64 libc6-dev libc6-dbg
or
sudo apt-get purge libc6-amd64 libc6-dev libc6-dbg
// I hope that's all I have to do to fix this, since I all ready changed that /etc/apt/sources.list line

Can I manually copy the files over? Would that fix it?

[sunrat]

Basically you can't upgrade libc6 version on Wheezy. Are you trying to install Steam or some other software that requires it? Search this forum for "install steam wheezy". There are many threads of people who trashed their system trying to do it.
There is a way to install Steam with GhostSquad57 installer for Wheezy. Again, search the forums.

[BryanFRitt]

Fine I guess, if I can't upgrade libc6;
I'm asking how to recover from this messed up install.
The system crashes before I can even type in a recovery console.

[BryanFRitt]

Trying in recovery mode (or whatever it's called) The computer ends with the 'Alpha' and the 'Scroll' Lock blinking, and
this message...
(typed in manually, this forum is giving me an 'Internal Server Error' when I try to submit the photo.)

Code: Select all

Begin: Running /scripts/init-premount ... done.
Begin: Mounting root file system ... Begin: Running /scripts/local-top ... done.
Begin: Running /scripts/local-premount ... resume: libgcrypt version: 1.5.0
done.
[    2.256017] usb 7-1: new full-speed USB device number 2 using uhci_hcd
[    2.266664] EXT4-fs (sda2): mounted filesystem with ordered data mode. 0pts:
(null)
Begin: Running /scripts/init-bottom ... done.
[    2.405671] usb 7-1: New USB device found, idVendor=0b97, idProduct=7761
[    2.405759] usb 7-1: New USB device strings: Mfr=0 Product=0, SerialNumber=0
[    2.410666] hub 7-1:1.0: 4 ports detected
switch_root: can't execute '/sbin/init': No such file or directory
[    2.552310] Kernel panic - not syncing: Attempted to kill init!
[    2.552405] Pid: 1, comm: busybox Not tainted 3.2.0-4-amd64 #1 Debian 3.2.57-
3+deb7u2
[    2.552504] Call Trace:
[    2.552592] [<ffffffff813492d8>] ? panic+0x95/0x1a2
[    2.552678] [<ffffffff81049da7>] ? do_exit+0x95/0x713
[    2.552678] [<ffffffff8104a6a5>] ? do_group_exit+0x74/0x9e
[    2.552935] [<ffffffff81354c12>] ? system_call_fastpath+0x16/0x1b
_

[dotlj]

If you system has problems as yours clearly does, boot from any Debian image file (CD, DVD or flashdisk)
Select Rescue mode
mount a root file system
It drops you into a root shell where you can check why you get the

switch_root: can't execute '/sbin/init': No such file or directory

man init may be helpful

[BryanFRitt]

I tried chroot from a virtual machine using my Windows partition, but it seams that Windows now prevents raw writes even with admin privileges.*

I ended up formatting the partition, and starting over. Wish I tested this* out from a Live disk disk first.
(I did this impulse I guess; but I backed up the partition first, and even had my home directory on a separate partition, and had made myself directions on how/what to install from my previous installs. So it wasn't too bad, I guess.)

*note: The directions that at least seamed to get me the furthest involved chroot jail, and ssh -ing into my own system.** I tried just using chroot, but wasn't enough. (should have tried from a 'Live' DVD/CD, don't know if it would have worked that way or not, but I could have tried. Too late now...)

**I think from there I would just have to install the Debian 7 stable versions of the files I tried to update to testing. (again, I didn't try this, so I don't know if it would have worked or not)
https://www.debian.org/distrib/packages
search for Debian 7.x stable versions of those software
dpkg -i ...

-

I suppose it would have been less risk if I put the reversal commands afterwards from the command that killed the computer,
` || /usr/bin/sudo /usr/bin/dpkg -i StableVersionsOfSoftware`(maybe add -f, ...) that way I wouldn't have to worry too much about not being able to type in commands afterwards.(again, I didn't try this, so I don't know if it would have worked or not)
If you try this, I suggest doing a full backup of this partition 1st!!!

[sunrat]

If you try this, I suggest doing a full backup of this partition 1st!!!

Always do this before doing something potentially destructive.

[BryanFRitt]

!@#$%^&*()
I reinstalled a fresh a system, except for /home
It was working, rebooting, everything since my last post.
Now it's back this same kind of boot error:
(the '________' in [<ffffffff________>] are different from last time, and after trying several times, I can see that the ________ are consistent)

Code: Select all

switch_root: can't execute '/sbin/init': No such file or directory
kernel panic - not syncing: Attempted to kill init!
Pid: 1, comm: busybox Not tainted 3.2.0-4-amd64 #1 Debian 3.2.60-1+deb7u1
Call Trace:
[<ffffffff81349c26>] ? panic+0x95/0x1a2
[<ffffffff81049da7>] ? do_exit+0x95/0x713
[<ffffffff8104a6a5>] ? do_group_exit+0x74/0x9e
[<ffffffff81355592>] ? system_call_fastpath+0x16/0x1b

This time I don't remember doing anything risky, and an attempt to backup my computer using windows, caused all files on my backup drive to be deleted.
After this backup error, I tried to log back in to Linux, but then got the init error.
Even though I've been loyal to Linux for quite a while, If this init issue isn't fixed, I might end up having to switch to Windows

[BryanFRitt]

I got far enough to see that it seams to be the same problem as before. libc6 issue, but I'm pretty sure never tried to upgrade it this time! Not sure how to fix it. This is as far as I got in a 'live-disk'. Debian 7.5 77575f9a-0983-457c-8a62-7512f0989de5_ is were I have my Debian installation mounted. I can't run a `sudo apt-get -f install` like most websites are suggesting because that affects the 'live-disk' and not the one on the hard drive. (One way I can tell which is which, is by if busybox is installed, on Debian KDE 7.5 Live-Disk it's installed, on my system it's not... although based on the messages near the init kernel panic, it might still have parts left behind?)

Code: Select all

sudo dpkg --root=/media/77575f9a-0983-457c-8a62-7512f0989de5_ --remove --force-all libc6:amd64 && sudo dpkg --root=/media/77575f9a-0983-457c-8a62-7512f0989de5_ --force-all -i libc6_2.13-38+deb7u3_amd64.deb
dpkg: libc6:amd64: dependency problems, but removing anyway as you requested:
 kde-runtime depends on libc6 (>= 2.11).
 libxau6:amd64 depends on libc6 (>= 2.2.5).
 libssh2-1:amd64 depends on libc6 (>= 2.4).
 liblockfile1:amd64 depends on libc6 (>= 2.4).
 lsof depends on libc6 (>= 2.11).
 pipelight-multi depends on libc6 (>= 2.2.5).
 libmono-system-drawing4.0-cil depends on libc6 (>= 2.13) | libc6.1 (>= 2.13) | libc0.1 (>= 2.13); however:
  Package libc6:amd64 is to be removed.
  Package libc6.1 is not installed.
  Package libc0.1 is not installed.
 libkjsembed4 depends on libc6 (>= 2.2.5).
 libkrb5-3:amd64 depends on libc6 (>= 2.9).
 cups depends on libc6 (>= 2.7).
 libwrap0:amd64 depends on libc6 (>= 2.11).
 kdocker depends on libc6 (>= 2.2.5).
 usbmuxd depends on libc6 (>= 2.4).
 growisofs depends on libc6 (>= 2.2.5).
 protobuf-compiler depends on libc6 (>= 2.2.5).
 p7zip depends on libc6 (>= 2.3.2).
 gnotski depends on libc6 (>= 2.7).
 libcupsimage2:amd64 depends on libc6 (>= 2.7).
 kapptemplate depends on libc
dpkg: warning: overriding problem because --force enabled:
 Package is in a very bad inconsistent state - you should
 reinstall it before attempting a removal.
(Reading database ... 288864 files and directories currently installed.)
Removing libc6:amd64 ...
dpkg (subprocess): unable to execute installed pre-removal script (/var/lib/dpkg/info/libc6:amd64.prerm): No such file or directory
dpkg: error processing libc6:amd64 (--remove):
 subprocess installed pre-removal script returned error exit status 2
dpkg (subprocess): unable to execute installed post-installation script (/var/lib/dpkg/info/libc6:amd64.postinst): No such file or directory
dpkg: error while cleaning up:
 subprocess installed post-installation script returned error exit status 2
Errors were encountered while processing:
 libc6:amd64

Code: Select all

sudo dpkg --root=/media/77575f9a-0983-457c-8a62-7512f0989de5_ --force-all -i libc6_2.13-38+deb7u3_amd64.deb
Selecting previously unselected package libc6:amd64.
(Reading database ... 288865 files and directories currently installed.)
Preparing to replace libc6:amd64 2.13-38+deb7u3 (using libc6_2.13-38+deb7u3_amd64.deb) ...
dpkg (subprocess): unable to execute old pre-removal script (/var/lib/dpkg/info/libc6:amd64.prerm): No such file or directory
dpkg: warning: subprocess old pre-removal script returned error exit status 2
dpkg: trying script from the new package instead ...
dpkg (subprocess): unable to execute new pre-removal script (/var/lib/dpkg/tmp.ci/prerm): No such file or directory
dpkg: error processing libc6_2.13-38+deb7u3_amd64.deb (--install):
 subprocess new pre-removal script returned error exit status 2
dpkg (subprocess): unable to execute installed post-installation script (/var/lib/dpkg/info/libc6:amd64.postinst): No such file or directory
dpkg: error while cleaning up:
 subprocess installed post-installation script returned error exit status 2
Errors were encountered while processing:
 libc6_2.13-38+deb7u3_amd64.deb

Code: Select all

sudo dpkg --root=/media/77575f9a-0983-457c-8a62-7512f0989de5_ --configure -a
dpkg: error processing libc6:amd64 (--configure):
 Package is in a very bad inconsistent state - you should
 reinstall it before attempting configuration.
Errors were encountered while processing:
 libc6:amd64

Update: I also found some links had the '/' in the wrong direction. Fixing this allowed me to chroot into the correct directory! (How did they get reversed?, and is there an easy fix if it turns out there's a lot of them?)

Code: Select all

sudo apt-get -f install
Reading package lists... Done
Building dependency tree       
Reading state information... Done
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
2 not fully installed or removed.
Need to get 0 B/4,221 kB of archives.
After this operation, 0 B of additional disk space will be used.
dpkg: error processing libc6:amd64 (--configure):
 Package is in a very bad inconsistent state - you should
 reinstall it before attempting configuration.
dpkg: dependency problems prevent configuration of man-db:
 man-db depends on libc6 (>= 2.11); however:
  Package libc6:amd64 is not configured yet.

dpkg: error processing man-db (--configure):
 dependency problems - leaving unconfigured
Errors were encountered while processing:
 libc6:amd64
 man-db
E: Sub-process /usr/bin/dpkg returned an error code (1)

Ok! this may have fixed this part of the problem! `sudo apt-get purge 'man-db'` followed by a `sudo apt-get install --reinstall libc6:amd64` `sudo dpkg --configure -a` in the correct chroot! restarting now... wish me luck! <--Well, that may have fixed this problem, but there were file now found errors(wpasupplicant, console-screen.sh, ...) unknown filesystem errors (ext3, NTFS...), only recovery mode would let me do anything, regular mode kept giving me invalid password, no gui... I added in a /forcefsck file, and rebooted, it checked the disk, and rebooted again, but same errors.

note: in uninstalling man-db
" The following packages will be REMOVED:
alien* debhelper* equivs* googleearth* kde-sc-dev-latest* lintian* lsb-core* man-db* "

Side note:
There's a '_' at the end of it, I had done a
`sudo debootstrap --arch amd64 wheezy /media/77575f9a-0983-457c-8a62-7512f0989de5 http://http.debian.org/debian`
which uses the spot without '_' for a chroot so ...

[edbarx]

I tried to install software similar to how stated on
http://stackoverflow.com/questions/1086 ... -on-debian
Editing the /etc/apt/sources.list fist with testing, // they did experimental

Code: Select all

http://stackoverflow.com/questions/10863613/how-to-upgrade-glibc-from-version-2-13-to-2-15-on-debian

Oh dear, what a glaringly bad advice and it has got 19 points! Simply dangerous, horrible and verging on the malicious.

With that step you ruined your installation. Sorry, reinstall.

[BryanFRitt]

I also found some links had the '/' in the wrong direction. Fixing this allowed me to chroot into the correct directory! (How did they get reversed?, and is there an easy fix if it turns out there's a lot of them?)

Now I've made a script to switch all the '\' to '/' in links, and it's links to links, if switching the '\' to '/' would change the link from invalid to valid. (also kept owner/group of the files the same) Turns out there was around 13,000 bad links. [2.4K 'absolute' links, and 8.6K 'relative' links] (I also backed up the bad links to a separate folder, so I can find out which ones changed.) But only doing what I had done up to this point only got me to where I can type in recovery mode.
[p.s. Still don't know how all the '/' in links could have gotten switched to '\'; any ideas?]
[p.s. Found out if you edit/save a script while its running, the changes take place while the script is running <-- could be very dangerous!!! ]

Back in a chroot'd environment I did a '--download-only' apt-get everything already installed, then in the recovery not chroot, a simular script to one found on this link, and the one posted to in it's comments
http://ubuntuforums.org/showthread.php?t=735693
Then went back and reinstalled whatever was skiped by this script, and rebooted.
In doing so everything went back to working!

p.s. Another thing I found wrong and fixed along the way was the my username:groupname got changed to user:user (or was it 'users'?) instead of bryan:bryan. Files were still 'owned' by 1000, but 1000 had changed to user:user.