Mageia forum

This is weird.
I have tried booting several times different elder kernels, and safe boot, and during further reboots to author this forum topic it suddenly booted OK!
Anyhow here is at least one bug (boot getting stuck because no timeout) https://bugs.mageia.org/show_bug.cgi?id=14452

I ran updates on a thinkpad T43 yesterday, and then it could not boot.

For more than ten tries it got stuck waiting forever. Last lines of log on screen (I have it to write messages on display during boot, no startup picture)

Code: Select all

[  OK  ] Started Save boot dmesg content.
The default is the current profile.
                                                           [  OK  ]
Looking for new hardware                       [  OK  ]
[  OK  ] Started Reconfigure the system on administrator request.
[  **   ] A start job is running for Trigger Flushing of Journal to Persistent Storage (14 min 7s / no limit)

Can someone give an idea how to get past this if it happens again, to some prompt to possibly debug, and even better to get rid of it...?
I found no way to get to a prompt. Ctrl Alt Fx do not work at this stage.
Of the earlier lines scrolling through i did not catch a problem, and the only [FAILED] line was the normal Failed to load legacy modules ... something like that.
I get the same problem if i choose earlier kernels.
If I select safe boot it stops with a blinking cursor after two lines of systemd-fsck having checked the /boot and /home on top of a black screen.
If i select safe boot without splash and quiet i also see other lines but my eyes did not see any of interest except two "mounted filesystem with ordered data mode"

EDIT: Tried updates of today (kernel 3.17.2-3) , rebooted, and am stuck again... /EDIT
EDIT2: selecting safe boot option on latest kernel worked - at least this time /EDIT2

morgano wrote:I ran updates on a thinkpad T43 yesterday, and then it could not boot.
...
EDIT2: selecting safe boot option on latest kernel worked - at least this time /EDIT2

Try the following pattern

Code: Select all

journalctl -b -p err --since='2014-05-17 16:50' --until='2014-05-17 17:10'

with your date / time.

Maybe you should have a look at the size of the journal files.

Code: Select all

du -sm /var/log/journal/

Maybe you should try to configure it to volatile storage, meaning you lose the journal on every reboot. Or at least set a size limit.

But without further log excerpts as shown by unklar nobody will know why it failed.

Thanks for the hints.
Progressing at https://bugs.mageia.org/show_bug.cgi?id=14452

Changing swap entry in /etc/fstab from

Code: Select all

UUID=xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx swap swap defaults 0 0

to

Code: Select all

UUID=xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx none            swap    sw              0       0

will help.

ps: I'm back (as promised) in Mageia 5. Nice. grub2 is much improved though grub-legacy is also installed
and udevadm hwdb --update needs "--root="

Code: Select all

# udevadm hwdb --update --root=

gohlip wrote:and udevadm hwdb --update needs "--root="

Code: Select all

# udevadm hwdb --update --root=

How does that fit with the topic of this thread, and in what context?

gohlip, are you talking about https://bugs.mageia.org/show_bug.cgi?id=14457 ?

@doktor5000, changing fstab for swap will solve the problem per topic.
As for udevadm, it is in "ps:", not main body text. It might help as a separate issue, yes.
But again, you don't have to be so pedantic, right lan? I did not miss you that much.

@morgano, I did not see this bug report before, but a similar one on redhat. Yes, it's related.

Cheers everybody.

gohlip wrote:But again, you don't have to be so pedantic, right lan? I did not miss you that much.

That was just a simple question out of curiosity ...

doktor5000 wrote:

gohlip wrote:But again, you don't have to be so pedantic, right lan? I did not miss you that much.

That was just a simple question out of curiosity ...

It's good to be back and nice that some things don't change.
Grubbe.