Comment # 20 on bug 927776 from Richard Weinberger

(In reply to Dr. Werner Fink from comment #19)
> (In reply to Richard Weinberger from comment #18)
> 
> > I never said, that it is a systemd bug.
> 
> Currently most errors reported by systemd will be assigned to systemd. That
> is that the bug dispatcher or reporter isn't able to distinguish between
> reporting and causing en error. 
> 
> > Let's sum up the facts we know.
> > 1. The passphrase was correct as my root filesystem was mounted correctly.(In 
> > reply to Richard Weinberger from comment #18)
> 
> Belive me: In rare case it is possible that a passpharse error does not
> always protect that a file system can be mounted.  AFAICR I had seen this in
> past on a virtual test system.  In this case the file system check had
> recovered the afterward broken ext3 file system.

Hmm, your idea is that instead of passpharse P I've entered passpharse P'.
P' was able to open the LUKS disk but as it was not identical to P the blocks
got decrypted in a wrong manner and caused bad files?

I don't think that this can happen with LUKS as the hash would not match.
A false positive hash _and_ mountable (bad) filesystems sounds very unlikely
to me. 

> > 2. For reasons I don't know dbus did not start. 
> >   The top most errors on the log are:
> 
> Adding dbus-1 maintainers to carbon copy.

Thx!