On Saturday 10 May 2014 09:57:25 Marc Haber wrote: > On Fri, 09 May 2014 20:30:22 +0200, Michael Biebl wrote: > >Am 09.05.2014 19:56, schrieb Steve Langasek: > >> I don't think systemd integration is in a state today that this is ready > >> to become the default. > > > >What are you missing? > > For example, keyscript= in /etc/crypttab. I got systemd just by not > paying attention, and in result an unbootable system since my crypto > setup heavily relies on keyscript=. Thankfully it was only a test VM > that got borked that way before I dared touching production.
I second this. We had several production systems that became unbootable due to this systemd change. It took quite some time to understand the intricate dependency among packages and we were fortunate to have a reliable recovery mechanism. We still do not understand why systemd does not seem to work with the cryptsetup package. The complexity of systemd is substantially higher which makes debugging these types of issues more difficult. -- JP
signature.asc
Description: This is a digitally signed message part.