FYI failed boot with 3.18.0-0.rc1.git1.2

[Bruno Wolff III]

On Mon, Oct 27, 2014 at 23:42:59 -0500,
  Bruno Wolff III <bruno at wolff.to> wrote:
>On Thu, Oct 23, 2014 at 10:04:26 -0500,
> Bruno Wolff III <bruno at wolff.to> wrote:
>>This one is more likely due to something in the initrd than the kernel.
>>The boot got to where dbus was activated and then services started failing.
>>Again I was in a hurrry and just went back to 3.18.0-0.rc0.git9.5 
>>and the system booted normally.
>
>I suspect this is some interaction between systemd and some unusual 
>configuration on a couple of my machines. I have at least one machine 
>that boots rc2 and one that still has the problem with rc2 (and one I 
>don't want to reboot right now, but was having the problem when I last 
>checked).

I did a bit of testing and the problem does seem to be dependent on 
the kernel version as the version that worked previously kept working 
after I rebuilt the initramfs. The problem is still happening as of 
3.18.0-0.rc2.git3.1.fc22.

The logs don't get written to disk for the failed boots even though they 
appear to get pretty far and I was hoping they would. So I need to go 
back and looking at things during the failed boot and see if I can find 
where things start going wrong. Given that other people aren't screaming 
about this, I suspect it is related to old hardware or maybe software raid.

I'm pretty busy this weekend so I may not make much progress, but we're 
still early in the 3.18 series so there isn't a rush yet.