10 a.m.
Hi,
I tried upgrade , with fedup [1]
I tried a no conventional update, yum localinstall /system-upgrade/*rpm
but now system doesn't boot normally , just boots when I put debug in
kernel boot parameters on grub .
I think is systemd boot order , looking to
journalctl -b | grep -i faile , I see :
Jan 02 14:37:08 segulix systemd[1]: Failed to load configuration for
syslog.service: No such file or directory
Jan 02 14:37:08 segulix systemd[1]: Failed to load configuration for
systemd-readahead-replay.service: No such file or directory
Jan 02 14:37:08 segulix systemd[1]: Failed to load configuration for
systemd-readahead-collect.service: No such file or directory
Jan 02 14:37:20 segulix systemd[1]: Failed to load configuration for
sys-kernel-config.mount: No such file or directory
Jan 02 14:37:20 segulix systemd[1]: Failed to add job
sys-kernel-config.mount/start, ignoring: Invalid argument.
Jan 02 14:37:27 segulix systemd[1]: systemd-udevd-kernel.socket got
notified about service death (failed permanently: no)
Jan 02 14:37:27 segulix systemd[1]: systemd-udevd-control.socket got
notified about service death (failed permanently: no)
Jan 02 14:37:37 segulix systemd[1]: Failed to open private bus
connection: Failed to connect to socket /var/run/dbus/system_bus_socket:
No such file or directory
Jan 02 14:37:37 segulix systemd[1]: Failed to connect to system D-Bus,
retrying later
Jan 02 14:37:37 segulix systemd[1]: Failed to load configuration for
syslog.target: No such file or directory
Jan 02 14:37:37 segulix systemd[1]: Failed to load configuration for
exim.service: No such file or directory
Jan 02 14:37:37 segulix systemd[1]: Failed to load configuration for
postfix.service: No such file or directory
Jan 02 14:37:37 segulix systemd[1]: Failed to load configuration for
sntp.service: No such file or directory
Jan 02 14:37:37 segulix systemd[1]: Failed to load configuration for
lvm2-activation.service: No such file or directory
Jan 02 14:37:37 segulix systemd[1]: Failed to load configuration for
lvm2-activation-early.service: No such file or directory
Jan 02 14:37:37 segulix systemd[1]: Failed to load configuration for
systemd-random-seed-load.service: No such file or directory
Jan 02 14:37:37 segulix systemd[1]: Failed to load configuration for
sysroot.mount: No such file or directory
Jan 02 14:37:37 segulix systemd[1]: systemd-journald.socket got notified
about service death (failed permanently: no)
etc ...
but D-bus seems not start at time ,
and in /etc/systemd/system
syslog.service -> /usr/lib/systemd/system/rsyslog.service
this link should exists ?
but why just boots with kernel with debug boot options , and what I
should do to but with quiet parameter or in default mode ?
[1]
https://bugzilla.redhat.com/show_bug.cgi?id=1045847#c12
TIA,
--
Sérgio M. B.
11:25 a.m.
On Thu, Jan 02, 2014 at 04:00:15PM +0000, Sérgio Basto wrote:
but why just boots with kernel with debug boot options , and what I
should do to but with quiet parameter or in default mode ?
If you can boot at all in some way you are already 95% there as you can
investigate what went wrong, maybe even why, and fix it. Here are
suggestions:
-- run 'package-cleanup --dupes' (package-cleanup is supplied by
yum-utils) and use that with --cleandupes if needed; sometimes
--noscripts too may be well advised
-- run 'package-cleanup --problems' and fix issues if any
-- run 'package-cleanup --orphans' and think what to remove
-- run 'rpm -qV >& /var/tmp/report' and see what is there; reinstall
packages if necessary
-- run 'yum update' as this also may help to smooth some rough edges
If after all of this there are no more reports of any package troubles
and you still cannot boot in a normal way then carefuly comb through
you logs and fix things as required. Re-running corresponding package
scripts can be of help as possibly something failed there the first time
around.
All of the above should not be needed but sometimes things go south;
so document all troublespots you found and attach that information to
your bugzilla report.
Michal
11:49 p.m.
On Qui, 2014-01-02 at 10:25 -0700, Michal Jaegermann wrote:
On Thu, Jan 02, 2014 at 04:00:15PM +0000, Sérgio Basto wrote:
>
> but why just boots with kernel with debug boot options , and what I
> should do to but with quiet parameter or in default mode ?
If you can boot at all in some way you are already 95% there as you
can
investigate what went wrong, maybe even why, and fix it.
Thanks for replying , I had verified all files , done a full prelink
-avmR , I followed some tips on
https://fedoraproject.org/wiki/Common_F20_bugs especially:
3.1 System fails to boot after install using LVM Thin Provisioning
I even try 4.10 suggestions, I changed my disk partitions from MBR to
MPT (works better). In meanwhile also fixed "CUPS hangs on shutdown"
And all options doesn't *worked*, when I almost give up, for today,
occasionally I removed
"systemd.log_level=debug" from kernel boot parameters
and guess what, it boot cleanly ,
systemd.log_level=debug prevents my system to boot !
Conclusion:
GRUB_CMDLINE_LINUX="quiet systemd.log_level=debug" doesn't boot
GRUB_CMDLINE_LINUX="systemd.log_level=debug" doesn't boot
GRUB_CMDLINE_LINUX="debug systemd.log_level=debug" boots
GRUB_CMDLINE_LINUX="quiet" boots
GRUB_CMDLINE_LINUX="" boots
I even think fedup problems was consequence that I had forget
systemd.log_level=debug in default GRUB_CMDLINE_LINUX
Hope that could help something, bye.
--
Sérgio M. B.
3790
days inactive
3791
days old
2 comments
2 participants
participants (2)
-
Michal Jaegermann -
Sérgio Basto