Bug 781644 - systemd-38-3.fc17 prevents machine from booting "multiuser"
Summary: systemd-38-3.fc17 prevents machine from booting "multiuser"
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Fedora
Classification: Fedora
Component: systemd
Version: 17
Hardware: Unspecified
OS: Unspecified
unspecified
high
Target Milestone: ---
Assignee: systemd-maint
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2012-01-14 00:34 UTC by Michal Jaegermann
Modified: 2012-03-13 12:35 UTC (History)
9 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2012-03-13 12:35:39 UTC
Type: ---


Attachments (Terms of Use)

Description Michal Jaegermann 2012-01-14 00:34:15 UTC
Description of problem:

After installting systemd-38-3.fc17 a machine becomes unsable.  It does not boot
into multiuser or graphics session.  An attempt to boot to "level 3" becomes soon painfully slow and it eventually stops after printing on a screen

Started Permit User Sessions

and it just sits there without any further progress.  Nothing is logged anywhere, with an exception of /var/log/boot.log which does not really say very much, so even a postmortem attempts to check what went wrong come to naught.

It is possible to boot to level 1, misnamed "rescue", but that is of not much help as even attempts to bring network interfaces up do not succeed.

After downgrading, with difficulty, all this mess to systemd-37-4.fc17 a machine boots as it used to.

Version-Release number of selected component (if applicable):
systemd-38-3.fc17

How reproducible:
always

Steps to Reproduce:
1. just try to boot

Comment 1 Michal Schmidt 2012-01-14 00:44:22 UTC
What's the SELinux mode? Try with 'enforcing=0'.

Comment 2 Michal Jaegermann 2012-01-14 01:19:54 UTC
(In reply to comment #1)
> What's the SELinux mode? Try with 'enforcing=0'.

It is off. Forgot to mention that. On the system in question selinux is never in use.

I can see one more option.  Due to a bug 768778 a dracut is not updated to the current one.  Maybe this has something to do with it?  I will check that later.  Only if this is the case then the whole concotion is getting waaaay to fragile for a sane init.

Comment 3 Michal Schmidt 2012-01-14 01:55:54 UTC
When in the rescue shell, does "systemctl --failed" list any units?
Does "systemctl status systemd-journald.{socket,service}" show both units as running?
What syslog implementation do you use? rsyslog?

Comment 4 Michal Jaegermann 2012-01-14 02:50:28 UTC
(In reply to comment #2)
> 
> I can see one more option.  Due to a bug 768778 a dracut is not updated to the
> current one.

Yes, that was behind it.  After hacking broken dracut-014-9.git20111215.fc17.noarch to produce a usable initrd I was able to boot using systemd-38-3.fc17.  Only I immediately run into then next
bug 781657 (no reboot/shutdown).

> ... does "systemctl --failed" list any units?

Booting that way an older kernel with initrtd produces by the previous dracut
and running "systemctl --failed" gets me:

systemd-journald.service loaded failed

and nothing else (even with "--all").

I am really glad that this is not a remote machine or I would be totally screwed.

Comment 5 Mamoru TASAKA 2012-02-05 05:32:30 UTC
I see this issue on both current rawhide and also on F-16
(from perhaps systemd-37-6.fc16)

Comment 6 Michal Schmidt 2012-02-06 11:38:08 UTC
(In reply to comment #5)
> I see this issue on both current rawhide and also on F-16
> (from perhaps systemd-37-6.fc16)

In F-16 you may be seeing bug 786182 or a similar bug in a different service.

Comment 7 waltersheridan 2012-02-13 11:31:15 UTC
With Alt + SysRq + I
it boots into graphic mode even if it hangs during "permit user sessions"

Comment 8 Lennart Poettering 2012-03-13 01:53:06 UTC
Hmm, so is there anything left to fix here that bug 768778 or bug 786182 can't explain?

Comment 9 Michal Jaegermann 2012-03-13 03:23:13 UTC
(In reply to comment #8)
> Hmm, so is there anything left to fix here that bug 768778 or bug 786182 can't
> explain?

AFAICT I already replied to this "No" in a comment #4 from 2012-01-13.  Some other people added later comments indicating that they still have issues.  What is a status with these I have no idea.

Comment 10 Lennart Poettering 2012-03-13 12:35:39 UTC
OK, people should probably report new bugs if there's something to fix here from anybody but the original reporter. Closing.


Note You need to log in before you can comment on or make changes to this bug.