Comment # 28 on bug 1197192 from Franck Bui

(In reply to Thomas Blume from comment #23)
> 
> That's a good point Franck. I still have doubts whether the systemd upstream
> approach of waiting for ever for the resume device is correct.

IIRC that was done to prevent the timeout from expiring when booting requires
manual interventions (for example when the rootfs or the swap partition is
encrypted).

> I mean, one can easily hang the machine, by just modifying or adding a wrong
> resume parameter at the bootloader stage.
> I deem that quite risky.

Well one could argue that this can be easily fixed by removing resume= option
from the kernel command line.