Cannot boot after distro upgrade

classic Classic list List threaded Threaded
33 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Cannot boot after distro upgrade

Miroslav Skoric
Hello,

After upgrading the old laptop from jessie to strech, it worked well for
few days (although more slowly than it was with jessie). But after last
proper shutdown, it does not boot anymore. In fact, it starts to boot
until it comes to a point where it says:

"You are in emergency mode. After logging in, type "journalctl -xb" to
view system logs, "systemctl reboot" to reboot, "systemctl default" or
^D to try again to boot into default mode.

Cannot open access to console, the root account is locked.
See sulogin(8) man page for more details.

Press Enter to continue."


Ok, when I press Enter it says:

"Checking in progress on 2 disks (0.0% complete)"

... it takes 1-2 seconds .. while it changes to:

"Checking in progress on 1 disk (11.0% complete)"

... and returns back to the first message:

"You are in emergency mode. After logging in, type "journalctl -xb" to
view system logs, "systemctl reboot" to reboot, "systemctl default" or
^D to try again to boot into default mode.

Cannot open access to console, the root account is locked.
See sulogin(8) man page for more details.

Press Enter to continue."

... and further pressing Enter seems to go nowhere and repeat endlessly.
Ctrl-Alt-F* does not open any new console. This one with error doesn't
accept anything but Enter.

I do not have any rescue/emergency media because it was an old
installation, started several years ago with squeeze, and upgraded over
time to wheezy, jessie, ..

Any idea what to do? Thanks.

Misko

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Reco
        Hi.

On Sat, Aug 31, 2019 at 02:00:18PM +0200, Miroslav Skoric wrote:
> Any idea what to do? Thanks.

Boot with init=/bin/bash kernel commandline parameter, remount root
filesystem read-write, fix your /etc/fstab (systemd is picky about
filesystems it's not able to mount, and no, "noauto" won't fix it),
reboot once more.

Reco

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Étienne Mollier
In reply to this post by Miroslav Skoric
Misko, on 2019-08-31:

> After upgrading the old laptop from jessie to strech, it worked
> well for few days (although more slowly than it was with
> jessie). But after last proper shutdown, it does not boot
> anymore. In fact, it starts to boot until it comes to a point
> where it says:
>
> "You are in emergency mode. After logging in, type "journalctl -xb" to view system logs, "systemctl reboot" to reboot, "systemctl default" or ^D to try again to boot into default mode.
>
> Cannot open access to console, the root account is locked.
> See sulogin(8) man page for more details.
>
> Press Enter to continue."
>
>
> Ok, when I press Enter it says:
>
> "Checking in progress on 2 disks (0.0% complete)"
>
> ... it takes 1-2 seconds .. while it changes to:
>
> "Checking in progress on 1 disk (11.0% complete)"
>
> ... and returns back to the first message:
>
> "You are in emergency mode. After logging in, type "journalctl -xb" to view system logs, "systemctl reboot" to reboot, "systemctl default" or ^D to try again to boot into default mode.
>
> Cannot open access to console, the root account is locked.
> See sulogin(8) man page for more details.
>
> Press Enter to continue."
>
> ... and further pressing Enter seems to go nowhere and repeat
> endlessly. Ctrl-Alt-F* does not open any new console. This one
> with error doesn't accept anything but Enter.
>
> I do not have any rescue/emergency media because it was an old
> installation, started several years ago with squeeze, and
> upgraded over time to wheezy, jessie, ..
>
> Any idea what to do? Thanks.
Good day,

Perhaps you can attempt a boot in "Recovery Mode", see the
"Advanced Boot Options" at the Grub menu stage of the boot.
It could have a positive effect if a faulty kernel module is
loaded and causes this loop in the boot sequence.  Maybe a check
of the memory and SMART data, if those options are available
from your BIOS, could be welcome, especially SMART since some
messages were mentioning checking the disk.

If operations here over do not make any difference, then you
really should consider creating a Rescue drive on an USB thumb.
I have had a good experience with SystemRescueCD over the years:

        http://www.system-rescue-cd.org/

But if you have a Debian installation media, you can also
achieve a thing or two by booting on the "Rescue Mode".

If you absolutely positively do not want to build a rescue
media, last chance option would be to edit the Grub menu entry,
and in the linux line, edit (or append if non-existent) the
following "init=" option (hit 'e' to edit the menu entry and F10
to boot):

        linux root=UUID=[...] ro init=/bin/bash

This is a last chance option, do not expect your system to
operate properly without a regular init process.  Boot in this
mode only to proceed to your investigations as of why the
machine fails to spawn the login process.  Quitting this shell
hangs the machine; you have to hard reset.

Kind Regards,
--
Étienne Mollier <[hidden email]>
Fingerprint:  5ab1 4edf 63bb ccff 8b54  2fa9 59da 56fe fff3 882d



signature.asc (673 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Miroslav Skoric
In reply to this post by Reco
On 8/31/19 3:26 PM, Reco wrote:

>
> Boot with init=/bin/bash kernel commandline parameter, remount root
> filesystem read-write, fix your /etc/fstab (systemd is picky about
> filesystems it's not able to mount, and no, "noauto" won't fix it),
> reboot once more.
>
> Reco
>
>

Hi,

Sorry for my ignorance, but where to put that "init=/bin/bash"? Here I
have LILO and it starts booting properly. However, after a page or two,
everything stops at the point I mentioned in my previous mail. And as I
said, I never get root console prompt or anything like that. So how to
remount filesystem, etc ... (Btw, it is encrypted LVM, if it can help.
Seems that decryption part passes well, but somewhere later booting fails.)

Misko

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Miroslav Skoric
In reply to this post by Étienne Mollier
On 8/31/19 3:26 PM, Étienne Mollier wrote:

>
> Perhaps you can attempt a boot in "Recovery Mode", see the
> "Advanced Boot Options" at the Grub menu stage of the boot.
> It could have a positive effect if a faulty kernel module is
> loaded and causes this loop in the boot sequence.  Maybe a check
> of the memory and SMART data, if those options are available
> from your BIOS, could be welcome, especially SMART since some
> messages were mentioning checking the disk.
>

Thanks for idea. However, there is no Grub here, only Lilo. I will check
BIOS for options.

> If operations here over do not make any difference, then you
> really should consider creating a Rescue drive on an USB thumb.
> I have had a good experience with SystemRescueCD over the years:
>
> http://www.system-rescue-cd.org/
>
> But if you have a Debian installation media, you can also
> achieve a thing or two by booting on the "Rescue Mode".
>

Will try some of those. I have Debian installation media for 6.0.1 only
and unsure whether it would work with strech.

> If you absolutely positively do not want to build a rescue
> media, last chance option would be to edit the Grub menu entry,
> and in the linux line, edit (or append if non-existent) the
> following "init=" option (hit 'e' to edit the menu entry and F10
> to boot):
>
> linux root=UUID=[...] ro init=/bin/bash
>
> This is a last chance option, do not expect your system to
> operate properly without a regular init process.  Boot in this
> mode only to proceed to your investigations as of why the
> machine fails to spawn the login process.  Quitting this shell
> hangs the machine; you have to hard reset.
>

As said, no Grub here. Is there similar option for Lilo?

Thanks!

Misko

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Reco
In reply to this post by Miroslav Skoric
On Sat, Aug 31, 2019 at 03:41:12PM +0200, Miroslav Skoric wrote:
> On 8/31/19 3:26 PM, Reco wrote:
>
> >
> > Boot with init=/bin/bash kernel commandline parameter, remount root
> > filesystem read-write, fix your /etc/fstab (systemd is picky about
> > filesystems it's not able to mount, and no, "noauto" won't fix it),
> > reboot once more.
>
> Sorry for my ignorance, but where to put that "init=/bin/bash"? Here I have LILO and it starts booting properly.

http://tldp.org/HOWTO/LILO-2.html , chapter 2.3.

Reco

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Étienne Mollier
In reply to this post by Miroslav Skoric
Misko, on 2019-08-31:

> On 8/31/19 3:26 PM, Étienne Mollier wrote:
> > If operations here over do not make any difference, then you
> > really should consider creating a Rescue drive on an USB thumb.
> > I have had a good experience with SystemRescueCD over the
> > years:
> >
> >     http://www.system-rescue-cd.org/
> >
> > But if you have a Debian installation media, you can also
> > achieve a thing or two by booting on the "Rescue Mode".
> >
>
> Will try some of those. I have Debian installation media for
> 6.0.1 only and unsure whether it would work with strech.
Since your system probably inherited stuff from Squeeze (like
partition format version), it might work, but I wouldn't bet on
it either.

> > If you absolutely positively do not want to build a rescue
> > media, last chance option would be to edit the Grub menu entry,
> > and in the linux line, edit (or append if non-existent) the
> > following "init=" option (hit 'e' to edit the menu entry and F10
> > to boot):
> >
> >     linux root=UUID=[...] ro init=/bin/bash
> >
> > This is a last chance option, do not expect your system to
> > operate properly without a regular init process.  Boot in this
> > mode only to proceed to your investigations as of why the
> > machine fails to spawn the login process.  Quitting this shell
> > hangs the machine; you have to hard reset.
> >
>
> As said, no Grub here. Is there similar option for Lilo?
Using lilo, on Debian 10 but it shouldn't have much changed, I
could boot as such by retyping a command line looking like the
following on my virtual machine, at the "boot:" prompt:

        linux /vmlinuz root=/dev/vda1 ro init=/bin/bash

Your command line will be surely different since you are running
on real hardware.  Please, per Reco's recommendation, consider
to have a look at the document to properly use this mode:
> http://tldp.org/HOWTO/LILO-2.html , chapter 2.3.
>
> Reco


Kind Regards,
--
Étienne Mollier <[hidden email]>
Fingerprint:  5ab1 4edf 63bb ccff 8b54  2fa9 59da 56fe fff3 882d



signature.asc (673 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Miroslav Skoric
In reply to this post by Étienne Mollier
On 8/31/19 3:26 PM, Étienne Mollier wrote:

>
> Maybe a check
> of the memory and SMART data, if those options are available
> from your BIOS, could be welcome, especially SMART since some
> messages were mentioning checking the disk.
>

I checked the system memory and hard disk self test (quick test and
SMART check), and the tests passed OK.

> If operations here over do not make any difference, then you
> really should consider creating a Rescue drive on an USB thumb.
> I have had a good experience with SystemRescueCD over the years:
>
> http://www.system-rescue-cd.org/
>
> But if you have a Debian installation media, you can also
> achieve a thing or two by booting on the "Rescue Mode".
>

Ok, I borrowed from a friend the first installation CD for Debian 8.11.1
i386 and tried to use it for rescue. (I still have the CDs for Debian
6.0.1a i386 but supposed that the newer version is better.)

So, when entered rescue mode, I tried fsck at first. It refused because
/dev/mapper/localhost-root was mounted as root file system. So I
unmounted it and tried fsck again. It reported several issues in
/dev/mapper/localhost-home and /dev/mapper/localhost-tmp, and offered to
truncate and/or to fix them. When finished, the subsequent fsck reported
all clean (/dev/mapper/localhost-home, /dev/mapper/localhost-tmp,
/dev/mapper/localhost-usr, /dev/mapper/localhost-var).

# mount -a  reported (among the others):

  "wrong fs type, bad option, bad superblock on
/dev/mapper/localhost-home, missing codepage or helper program, or other
error"

and

  "wrong fs type, bad option, bad superblock on
/dev/mapper/localhost-tmp, missing codepage or helper program, or other
error"


# dmesg | tail reported (among the other):

EXT4-fs (dm-6): couldn't mount as ext3 due to feature incompatibilities
EXT4-fs (dm-5): couldn't mount as ext3 due to feature incompatibilities


# mount reported that /, /usr, and /var were there, but not /home and /tmp.


My /etc/fstab includes:

/dev/mapper/localhost-home /home ext3 defaults 0 2
/dev/mapper/localhost-tmp /tmp ext3 defaults 0 2


When I reboot, those two fail.

Any idea?


PS: I can also try with that SystemRescueCD but just wanted to try this
quickest way ...

Misko

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Pascal Hambourg-2
Le 01/09/2019 à 01:04, Miroslav Skoric a écrit :

>
> # dmesg | tail reported (among the other):
>
> EXT4-fs (dm-6): couldn't mount as ext3 due to feature incompatibilities
> EXT4-fs (dm-5): couldn't mount as ext3 due to feature incompatibilities
>
>
> # mount reported that /, /usr, and /var were there, but not /home and /tmp.
>
>
> My /etc/fstab includes:
>
> /dev/mapper/localhost-home /home ext3 defaults 0 2
> /dev/mapper/localhost-tmp /tmp ext3 defaults 0 2

Mount them as ext4.

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Étienne Mollier
On 01/09/2019 01.20, Pascal Hambourg wrote:

> Le 01/09/2019 à 01:04, Miroslav Skoric a écrit :
>>
>> # dmesg | tail reported (among the other):
>>
>> EXT4-fs (dm-6): couldn't mount as ext3 due to feature incompatibilities
>> EXT4-fs (dm-5): couldn't mount as ext3 due to feature incompatibilities
>>
>>
>> # mount reported that /, /usr, and /var were there, but not /home and /tmp.
>>
>>
>> My /etc/fstab includes:
>>
>> /dev/mapper/localhost-home /home ext3 defaults 0 2
>> /dev/mapper/localhost-tmp /tmp ext3 defaults 0 2
>
> Mount them as ext4.
>
Hi Miroslav,

Pascal is probably right.  If you manage to have access to the
command "dumpe2fs" in your rescue environment, what is the
output of:

        # dumpe2fs /dev/mapper/localhost-home | grep '^Filesystem features:'

Actual Ext3 should probably not have more features than:

        has_journal ext_attr resize_inode
        dir_index filetype sparse_super
        large_file

If you have more, you most likely want to edit your /etc/fstab
entries to use ext4.  Out of curiosity, how are defined your /,
/usr and /var in fstab ?  It could be interesting for us to see
the differences, since you mention those are mounting properly.

Cheers,
--
Étienne Mollier <[hidden email]>
              5ab1 4edf 63bb ccff 8b54  2fa9 59da 56fe fff3 882d


signature.asc (673 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Miroslav Skoric
In reply to this post by Reco
On 8/31/19 3:48 PM, Reco wrote:

> On Sat, Aug 31, 2019 at 03:41:12PM +0200, Miroslav Skoric wrote:
>> On 8/31/19 3:26 PM, Reco wrote:
>>
>>>
>>> Boot with init=/bin/bash kernel commandline parameter, remount root
>>> filesystem read-write, fix your /etc/fstab (systemd is picky about
>>> filesystems it's not able to mount, and no, "noauto" won't fix it),
>>> reboot once more.
>>
>> Sorry for my ignorance, but where to put that "init=/bin/bash"? Here I have LILO and it starts booting properly.
>
> http://tldp.org/HOWTO/LILO-2.html , chapter 2.3.
>
> Reco
>
>

Ok Reco, I managed to crawl through this a little bit (I am an old
person), so when I reached prompt, did that:

root@(none):/# mount
sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
udev on /dev type devtmpfs
(rw,nosuid,relatime,size=1514900k,nr_inodes=215931,mode=755)
devpts on /dev/pts type devpts
(rw,nosuid,noexec,relatime,gid=5,mode=620,ptmxmode=000)
tmpfs on /run type tmpfs (rw,nosuid,noexec,relatime,size=304560k,mode=755)
/dev/mapper/localhost-root on / type ext3
(rw,relatime,errors=continue,barrier=1,data=ordered)
/dev/mapper/localhost-usr on /usr type ext3
(rw,relatime,errors=continue,barrier=1,data=ordered)
root@(none):/#

root@(none):/# mount -a
[ 1024.810706] EXT3-fs (dm-6): error: couldn't mount because of
unsupported optional features (8000)
mount: wrong fs type, bad option, bad superblock on
/dev/mapper/localhost-home, missing codepage or helper program, or other
error

In some cases useful info is found in syslog - try dmesg | tail or so.
[ 1024.839252] EXT3-fs (dm-5): error: couldn't mount because of
unsupported optional features (8000)
mount: wrong fs type, bad option, bad superblock on
/dev/mapper/localhost-tmp, missing codepage or helper program, or other
error

In some cases useful info is found in syslog - try dmesg | tail or so.
[ 1024.874388] kjournald starting. Commit interval 5 seconds
[ 1024.874702] EXT3-fs (dm-3): using internal journal
[ 1024.874761] EXT3-fs (dm-3): mounted filesystem with ordered data mode
root@(none):/#

root@(none):/# dmesg | tail
[ 19.828971] kjournald starting. Commit interval 5 seconds
[ 19.836188] EXT3-fs (dm-2): using internal journal
[ 19.836246] EXT3-fs (dm-2): mounted filesystem with ordered data mode
[ 19.838924] aufs: module is from the staging directory, the quality is
unknown, you have been warned.
[ 19.840477] aufs 3.2.x+setfl-debian
[ 1024.810706] EXT3-fs (dm-6): error: couldn't mount because of
unsupported optional features (8000)
[ 1024.839252] EXT3-fs (dm-5): error: couldn't mount because of
unsupported optional features (8000)
[ 1024.874388] kjournald starting. Commit interval 5 seconds
[ 1024.874702] EXT3-fs (dm-3): using internal journal
[ 1024.874761] EXT3-fs (dm-3): mounted filesystem with ordered data mode

root@(none):/# nano /etc/fstab
#
...
proc /proc proc defaults 0 0
/dev/mapper/localhost-root ext3 errors=remount-ro 0 1
# /boot was on /dev/sda1 during installation
UUID=...... /boot ext2 defaults 0 2
/dev/mapper/localhost-home /home ext3 defaults 0 2
/dev/mapper/localhost-tmp /tmp ext3 defaults 0 2
/dev/mapper/localhost-usr /usr ext3 defaults 0 2
/dev/mapper/localhost-var /var ext3 defaults 0 2
/dev/mapper/localhost-swap_1 none swap sw 0 0
/dev/scd0 /media/cdrom0 udf,iso9660 user,noauto 0 0


Whatever I tried to fix in fstab, did not bring improvement. As Pascal
Hambourg suggested, tried also to mount /home and /tmp as ext4  but no
avail. At next reboot:

...
...
[FAILED] Failed to mount /home.
See 'systemctl status home.mount' for details.
...
...
...

and similar for /tmp


By the way, there appeared some other errors (I hope not in connection
with the main issue):

...
...
[FAILED] Failed to start Console System Startup Logging.
See 'systemctl status console-kit-log-system-start.service' for details.
[FAILED] Failed to start Enable support for additional executable binary
formats.
See 'systemctl status binfmt-support.service' for details.
...
...

Any idea?

Misko

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Reco
        Hi.

On Sun, Sep 01, 2019 at 05:01:39PM +0200, Miroslav Skoric wrote:

> On 8/31/19 3:48 PM, Reco wrote:
>
> > On Sat, Aug 31, 2019 at 03:41:12PM +0200, Miroslav Skoric wrote:
> > > On 8/31/19 3:26 PM, Reco wrote:
> > >
> > > >
> > > > Boot with init=/bin/bash kernel commandline parameter, remount root
> > > > filesystem read-write, fix your /etc/fstab (systemd is picky about
> > > > filesystems it's not able to mount, and no, "noauto" won't fix it),
> > > > reboot once more.
> > >
> > > Sorry for my ignorance, but where to put that "init=/bin/bash"? Here I have LILO and it starts booting properly.
> >
> > http://tldp.org/HOWTO/LILO-2.html , chapter 2.3.
> >
>
> Ok Reco, I managed to crawl through this a little bit (I am an old person), so when I reached prompt, did that:
>
...
> root@(none):/# mount -a
> [ 1024.810706] EXT3-fs (dm-6): error: couldn't mount because of unsupported optional features (8000)

This. Everything else is secondary.
8000 (it's in hex) means EXT4_FEATURE_INCOMPAT_INLINE_DATA (had to
search kernel source for this), which is "inline_data" ext4-only
feature.

So, either the superblock of this filesystem is damaged, or you're
trying to mount ext4 filesystem as ext3 (won't going to work).

So, let's do something easy and non-destructive first (I assume that
/tmp does not contain anything useful):

tune2fs -l /dev/localhost/tmp

fsck.ext3 -f -n /dev/localhost/tmp

fsck.ext4 -f -n /dev/localhost/tmp

fsck.ext3 -b 8193 -f -n /dev/localhost/tmp

fsck.ext4 -b 8193 -f -n /dev/localhost/tmp

And, for the good measure,

uname -a

Reco

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Miroslav Skoric
In reply to this post by Étienne Mollier
On 9/1/19 1:20 PM, Étienne Mollier wrote:

>
> Hi Miroslav,
>
> Pascal is probably right.  If you manage to have access to the
> command "dumpe2fs" in your rescue environment, what is the
> output of:
>
> # dumpe2fs /dev/mapper/localhost-home | grep '^Filesystem features:'
>
> Actual Ext3 should probably not have more features than:
>
> has_journal ext_attr resize_inode
> dir_index filetype sparse_super
> large_file
>
> If you have more, you most likely want to edit your /etc/fstab
> entries to use ext4.  Out of curiosity, how are defined your /,
> /usr and /var in fstab ?  It could be interesting for us to see
> the differences, since you mention those are mounting properly.
>

Hi Étienne,

The output of the above command (version dumpe2fs 1.43.4 31-Jan-2017) is
the same as yours, with *one addition*: inline_data

root@(none):/# nano /etc/fstab
#
...
proc    /proc    proc    defaults    0    0
/dev/mapper/localhost-root /   ext3    errors=remount-ro 0    1
# /boot was on /dev/sda1 during installation
UUID=......    /boot    ext2    defaults    0    2
/dev/mapper/localhost-home    /home    ext3    defaults 0    2
/dev/mapper/localhost-tmp    /tmp    ext3    defaults 0    2
/dev/mapper/localhost-usr    /usr    ext3    defaults 0    2
/dev/mapper/localhost-var    /var    ext3    defaults 0    2
/dev/mapper/localhost-swap_1    none    swap    sw    0    0
/dev/scd0    /media/cdrom0    udf,iso9660 user,noauto    0    0


Regards,

Misko

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Étienne Mollier
Misko, on 2019-09-01:

> On 9/1/19 1:20 PM, Étienne Mollier wrote:
> >
> > Hi Miroslav,
> >
> > Pascal is probably right.  If you manage to have access to the
> > command "dumpe2fs" in your rescue environment, what is the
> > output of:
> >
> >     # dumpe2fs /dev/mapper/localhost-home | grep '^Filesystem
> >     features:'
> >
> > Actual Ext3 should probably not have more features than:
> >
> >     has_journal    ext_attr    resize_inode
> >     dir_index    filetype    sparse_super
> >     large_file
> >
> > If you have more, you most likely want to edit your /etc/fstab
> > entries to use ext4.  Out of curiosity, how are defined your /,
> > /usr and /var in fstab ?  It could be interesting for us to see
> > the differences, since you mention those are mounting
> > properly.
> >
>
> Hi Étienne,
>
> The output of the above command (version dumpe2fs 1.43.4
> 31-Jan-2017) is the same as yours, with *one addition*:
> inline_data
Good news, this is consistent with Reco's observation in the
other thread.  Follows his recommendations and see what happens.

> root@(none):/# nano /etc/fstab
> #
> ...
> proc    /proc    proc    defaults    0    0
> /dev/mapper/localhost-root    /   ext3    errors=remount-ro 0
> 1
> # /boot was on /dev/sda1 during installation
> UUID=......    /boot    ext2    defaults    0    2
> /dev/mapper/localhost-home    /home    ext3    defaults 0    2
> /dev/mapper/localhost-tmp    /tmp    ext3    defaults 0    2
> /dev/mapper/localhost-usr    /usr    ext3    defaults 0    2
> /dev/mapper/localhost-var    /var    ext3    defaults 0    2
> /dev/mapper/localhost-swap_1    none    swap    sw    0    0
> /dev/scd0    /media/cdrom0    udf,iso9660 user,noauto    0    0
I don't suppose these /, /usr and /var have this "inline_data"
flag in "dumpe2fs" output, do they?

Kind Regards,
--
Étienne Mollier <[hidden email]>
Fingerprint:  5ab1 4edf 63bb ccff 8b54  2fa9 59da 56fe fff3 882d



signature.asc (673 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Pascal Hambourg-2
In reply to this post by Miroslav Skoric
Le 01/09/2019 à 17:01, Miroslav Skoric a écrit :
>
> EXT3-fs (dm-6): error: couldn't mount because of
> unsupported optional features (8000)

This is not the same as the previous error message you showed while
using the Debian Jessie 8.11 installer in rescue mode :

> EXT4-fs (dm-6): couldn't mount as ext3 due to feature incompatibilities

See the differences ?

EXT3-fs vs EXT4-fs : "recent" versions of the kernel use the ext4 driver
for all ext* versions. Seeing "EXT3-fs" indicates that you run a very
old kernel, older that the 3.16 one included in Jessie.

"unsupported feature" vs "feature incompatibilities" : the former means
that the kernel does not support the feature (too old, again) whereas
the latter means that the feature is incompatible with mounting as ext3.

man ext4 states that "inline_data" feature is supported since version
3.8, so the 3.16 kernel from Jessie would support it.

what kind of outdated kernel version are you running ? Is it still the
3.2 version from Wheezy ? Looks like the dist upgrade was not complete.
If so, install a 3.16 kernel, replace ext3 with ext4 or auto for the two
filesystems, reboot and see if they mount fine.

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Miroslav Skoric
In reply to this post by Reco
On 9/1/19 5:33 PM, Reco wrote:

>
> So, let's do something easy and non-destructive first (I assume that
> /tmp does not contain anything useful):
>
> tune2fs -l /dev/localhost/tmp
>
> fsck.ext3 -f -n /dev/localhost/tmp
>
> fsck.ext4 -f -n /dev/localhost/tmp
>
> fsck.ext3 -b 8193 -f -n /dev/localhost/tmp
>
> fsck.ext4 -b 8193 -f -n /dev/localhost/tmp
>
> And, for the good measure,
>
> uname -a
>
> Reco
>
>

Reco,

There is a lot in the output of those commands. I can put the outputs
into txt files, but cannot mount the usb stick to transfer the content
to the other comp to send it. Syntax ..?

Or to wait that I copy/paste the output by hand :-)

Misko

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Reco
On Sun, Sep 01, 2019 at 09:19:43PM +0200, Miroslav Skoric wrote:

> On 9/1/19 5:33 PM, Reco wrote:
>
> >
> > So, let's do something easy and non-destructive first (I assume that
> > /tmp does not contain anything useful):
> >
> > tune2fs -l /dev/localhost/tmp
> >
> > fsck.ext3 -f -n /dev/localhost/tmp
> >
> > fsck.ext4 -f -n /dev/localhost/tmp
> >
> > fsck.ext3 -b 8193 -f -n /dev/localhost/tmp
> >
> > fsck.ext4 -b 8193 -f -n /dev/localhost/tmp
> >
> > And, for the good measure,
> >
> > uname -a
>
> There is a lot in the output of those commands. I can put the outputs into txt files, but cannot mount the usb stick to transfer the content to the other comp
> to send it. Syntax ..?

lsblk to get the device name of your USB stick.
mount /dev/<devname> /mnt

Don't forget to "umount /mnt" afterwards.

> Or to wait that I copy/paste the output by hand :-)

Might get a photo of it as well ;)

Reco

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Miroslav Skoric
In reply to this post by Pascal Hambourg-2
On 9/1/19 8:40 PM, Pascal Hambourg wrote:

> Le 01/09/2019 à 17:01, Miroslav Skoric a écrit :
>>
>> EXT3-fs (dm-6): error: couldn't mount because of unsupported optional
>> features (8000)
>
> This is not the same as the previous error message you showed while
> using the Debian Jessie 8.11 installer in rescue mode :
>
>> EXT4-fs (dm-6): couldn't mount as ext3 due to feature incompatibilities
>
> See the differences ?
>
> EXT3-fs vs EXT4-fs : "recent" versions of the kernel use the ext4 driver
> for all ext* versions. Seeing "EXT3-fs" indicates that you run a very
> old kernel, older that the 3.16 one included in Jessie.
>
> "unsupported feature" vs "feature incompatibilities" : the former means
> that the kernel does not support the feature (too old, again) whereas
> the latter means that the feature is incompatible with mounting as ext3.
>
> man ext4 states that "inline_data" feature is supported since version
> 3.8, so the 3.16 kernel from Jessie would support it.
>
> what kind of outdated kernel version are you running ? Is it still the
> 3.2 version from Wheezy ? Looks like the dist upgrade was not complete.
> If so, install a 3.16 kernel, replace ext3 with ext4 or auto for the two
> filesystems, reboot and see if they mount fine.
>
>

Hi Pascal,

root@(none):/# uname -a
Linux (none) 3.2.0-4-486 #1 Debian 3.2.96-2 i686 GNU/Linux
root@(none):/#


If the dist upgrade was not complete (and if the system not functional
anymore), how to change/upgrade the kernel now?

Tnx.

Misko

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Pascal Hambourg-2
Le 01/09/2019 à 22:59, Miroslav Skoric a écrit :
>
> root@(none):/# uname -a
> Linux (none) 3.2.0-4-486 #1 Debian 3.2.96-2 i686 GNU/Linux

So you upgraded from Jessie to Stretch but still ran the old kernel from
Wheezy all this time ? Wow.

> If the dist upgrade was not complete (and if the system not functional
> anymore), how to change/upgrade the kernel now?

apt-get install linux-image-586

Reply | Threaded
Open this post in threaded view
|

Re: Cannot boot after distro upgrade

Miroslav Skoric
On 9/2/19 12:26 AM, Pascal Hambourg wrote:

> Le 01/09/2019 à 22:59, Miroslav Skoric a écrit :
>>
>> root@(none):/# uname -a
>> Linux (none) 3.2.0-4-486 #1 Debian 3.2.96-2 i686 GNU/Linux
>
> So you upgraded from Jessie to Stretch but still ran the old kernel from
> Wheezy all this time ? Wow.
>

Sure. I upgraded Jessie to Stretch last week. And it worked well for me
until Friday eve. (And before that I upgraded Wheezy to Jessie cca year
ago. It worked well for me too.) It's almost a ten year old laptop now.

>> If the dist upgrade was not complete (and if the system not functional
>> anymore), how to change/upgrade the kernel now?
>
> apt-get install linux-image-586
>

Hmmm, without network? Ok, I will see if I could enable networking ...

12