chris gareau wrote:
> I turned on my computer today, to find that I couldn't login at all, I
> was able to ssh into the system from another system and login to change
> the dm reboot and try agane, that got me in as my regular user but
> didn't fix the hole problem.
>
> when I try to log in as root or do anything that needs root access I get
> errors, could not open session when I try to su root.
>
> kicked back to the login prompt when I try to log in as any user from
> the console.
>
> what could be wrong??????
ok I booted from the trk and scaned the file system for errors, it found
and fixed them, then I rebooted no help, I tried reseting the root and
diehard passwords, using chroot and passwd from the trk, no help.

I can log in using kdm but not gdm, and not at the console.

HELP!!!!!!!!!!!!!!!!!!!!!!!!!

I am out of ideas!!!!!!!!!!!!

On Thu, 12 Apr 2007 16:28:24 -0400, chris gareau wrote:

>> what could be wrong??????
> ok I booted from the trk and scaned the file system for errors, it found

If this is an ext2 or ext3 filesystem, check the directory /lost+found
for any files you recognize.

If your system has filesystem corruption, due to a hard reset,
it may be best to to a reinstall, and then turn off hard drive
write caching.

> and fixed them, then I rebooted no help, I tried reseting the root and
> diehard passwords, using chroot and passwd from the trk, no help.
> I can log in using kdm but not gdm, and not at the console.
> HELP!!!!!!!!!!!!!!!!!!!!!!!!!
> I am out of ideas!!!!!!!!!!!!

Is mingetty running? It should be started by /etc/inittab having lines like
1:2345:respawn:/sbin/mingetty tty1

Also check cat /etc/securetty shows ...
tty1
vc/1
tty2
vc/2
tty3
vc/3
tty4
vc/4
tty5
vc/5
tty6
vc/6

More info needed, such as exact actions/error messages.

Regards, Dave Hodgins

--
Change nomail.afraid.org to ody.ca to reply by email.
(nomail.afraid.org has been set up specifically for
use in usenet. Feel free to use it yourself.)

David W. Hodgins wrote:
> On Thu, 12 Apr 2007 16:28:24 -0400, chris gareau wrote:
>
>>> what could be wrong??????
>> ok I booted from the trk and scaned the file system for errors, it found
>
> If this is an ext2 or ext3 filesystem, check the directory /lost+found
> for any files you recognize.
>
> If your system has filesystem corruption, due to a hard reset,
> it may be best to to a reinstall, and then turn off hard drive
> write caching.
>
>> and fixed them, then I rebooted no help, I tried reseting the root and
>> diehard passwords, using chroot and passwd from the trk, no help.
>> I can log in using kdm but not gdm, and not at the console.
>> HELP!!!!!!!!!!!!!!!!!!!!!!!!!
>> I am out of ideas!!!!!!!!!!!!
>
> Is mingetty running? It should be started by /etc/inittab having lines like
> 1:2345:respawn:/sbin/mingetty tty1
>
> Also check cat /etc/securetty shows ...
> tty1
> vc/1
> tty2
> vc/2
> tty3
> vc/3
> tty4
> vc/4
> tty5
> vc/5
> tty6
> vc/6
>
> More info needed, such as exact actions/error messages.
>
> Regards, Dave Hodgins
>
nothing I recognise

yes mingetty is running, I can cs but I just get kicked back to the
login prompt when I try to log in as anubody.

I can't tell what is saies to me when I log in but I think it is the
same as when I try to su

[diehard@diehardp4 ~]$ su root
Password:
could not open session
[diehard@diehardp4 ~]$

I am backing up everything I can to my laptop in prep for a reinstall
but would like to avoide a reinstall if posible.

On Thu, 12 Apr 2007 18:07:13 -0400, chris gareau wrote:

Check # cat /etc/securetty shows ...
tty1
vc/1
tty2
vc/2
tty3
vc/3
tty4
vc/4
tty5
vc/5
tty6
vc/6

Check # grep root /etc/passwd
root:x:0:0:root:/root:/bin/bash

Check # grep root /etc/group
root:x:0:

Run passwd to update the roots password (again, if already done).

Run pwconv and grpconv to update the shadow files.

Check # cat /etc/pam.d/login shows ...
#%PAM-1.0
auth required pam_securetty.so
auth required pam_nologin.so
auth include system-auth
account include system-auth
password include system-auth
session optional pam_console.so
session include system-auth

Regards, Dave Hodgins

--
Change nomail.afraid.org to ody.ca to reply by email.
(nomail.afraid.org has been set up specifically for
use in usenet. Feel free to use it yourself.)

David W. Hodgins wrote:
> On Thu, 12 Apr 2007 18:07:13 -0400, chris gareau wrote:
>
> Check # cat /etc/securetty shows ...
> tty1
> vc/1
> tty2
> vc/2
> tty3
> vc/3
> tty4
> vc/4
> tty5
> vc/5
> tty6
> vc/6
>
> Check # grep root /etc/passwd
> root:x:0:0:root:/root:/bin/bash
>
> Check # grep root /etc/group
> root:x:0:
>
> Run passwd to update the roots password (again, if already done).
>
> Run pwconv and grpconv to update the shadow files.
>
> Check # cat /etc/pam.d/login shows ...
> #%PAM-1.0
> auth required pam_securetty.so
> auth required pam_nologin.so
> auth include system-auth
> account include system-auth
> password include system-auth
> session optional pam_console.so
> session include system-auth
>
> Regards, Dave Hodgins
>
ok did all that everything normal but still can't su to root or login
ans anybody at the console

chris gareau writes:



>chris gareau wrote:
>> I turned on my computer today, to find that I couldn't login at all, I
>> was able to ssh into the system from another system and login to change
>> the dm reboot and try agane, that got me in as my regular user but
>> didn't fix the hole problem.
>>
>> when I try to log in as root or do anything that needs root access I get
>> errors, could not open session when I try to su root.
>>
>> kicked back to the login prompt when I try to log in as any user from
>> the console.
>>
>> what could be wrong??????
>ok I booted from the trk and scaned the file system for errors, it found
>and fixed them, then I rebooted no help, I tried reseting the root and
>diehard passwords, using chroot and passwd from the trk, no help.

>I can log in using kdm but not gdm, and not at the console.

>HELP!!!!!!!!!!!!!!!!!!!!!!!!!

>I am out of ideas!!!!!!!!!!!!

Full filesystem.
/etc/password is corrupted ( eg a blank line near the start of the file)
/etc/shadow corrupted.
partitions not mounted that should be.

chris gareau wrote:
>
>
> David W. Hodgins wrote:
>> On Thu, 12 Apr 2007 18:07:13 -0400, chris gareau wrote:
>>
>> Check # cat /etc/securetty shows ...
>> tty1
>> vc/1
>> tty2
>> vc/2
>> tty3
>> vc/3
>> tty4
>> vc/4
>> tty5
>> vc/5
>> tty6
>> vc/6
>>
>> Check # grep root /etc/passwd
>> root:x:0:0:root:/root:/bin/bash
>>
>> Check # grep root /etc/group
>> root:x:0:
>>
>> Run passwd to update the roots password (again, if already done).
>>
>> Run pwconv and grpconv to update the shadow files.
>>
>> Check # cat /etc/pam.d/login shows ...
>> #%PAM-1.0
>> auth required pam_securetty.so
>> auth required pam_nologin.so
>> auth include system-auth
>> account include system-auth
>> password include system-auth
>> session optional pam_console.so
>> session include system-auth
>>
>> Regards, Dave Hodgins
>>
> ok did all that everything normal but still can't su to root or login
> ans anybody at the console

well thanx for all the help, but a reinstall was unavoidable.

this time I am pouting the /home in a hole other partition so I don't
have to go threw all the work of copying off all the stuff from the
computer to an other 1.

thanx anyway

On Thu, 12 Apr 2007 20:38:01 -0400, chris gareau wrote:

> well thanx for all the help, but a reinstall was unavoidable.
> this time I am pouting the /home in a hole other partition so I don't
> have to go threw all the work of copying off all the stuff from the
> computer to an other 1.

I expected that was going to be the end result. When you get filesystem
corruption, there are just too many files to check/restore.

I strongly suggest turning off hard drive write caching, to reduce
the likelyhood of corruption from a hard reset.
Add
hdparm -W 0 /dev/hda
to /etc/rc.local, for each of your hard drives.

Running hdparm -I /dev/hda will then show you the status of the drive.
Under the Commands/features, there should not be an asterisk beside
the Write Cache.

Regards, Dave Hodgins

--
Change nomail.afraid.org to ody.ca to reply by email.
(nomail.afraid.org has been set up specifically for
use in usenet. Feel free to use it yourself.)

David W. Hodgins wrote:
> On Thu, 12 Apr 2007 20:38:01 -0400, chris gareau wrote:
>
>> well thanx for all the help, but a reinstall was unavoidable.
>> this time I am pouting the /home in a hole other partition so I don't
>> have to go threw all the work of copying off all the stuff from the
>> computer to an other 1.
>
> I expected that was going to be the end result. When you get filesystem
> corruption, there are just too many files to check/restore.
>
> I strongly suggest turning off hard drive write caching, to reduce
> the likelyhood of corruption from a hard reset.
> Add
> hdparm -W 0 /dev/hda
> to /etc/rc.local, for each of your hard drives.
>
> Running hdparm -I /dev/hda will then show you the status of the drive.
> Under the Commands/features, there should not be an asterisk beside
> the Write Cache.
>
> Regards, Dave Hodgins
>
thanx I will look into it, alteast when I am doing crazy things.