Several times a single LXC guest became unreachable, the host's "load" started to rise slowly but continously up to ~1000 (!) and more, fortunately with the host still allowing to login via ssh but unfortunately with only one way out: Reset button. Ouch. WTF.
The details and further solution attempts are now described and tracked in Ever-rising load on Debian jessie + DRBD8 + LXC host pairs.
Initial suspicion and upstream DRBD 8 kernel modules
In 3 of the first 4 cases of the problem described above the first unusual syslog entry hinted at DRBD.
Therefore we packaged (Debianized) the most-recent upstream kernel modul.
We are currently unsure wether DRBD has anything to do with the problem, but are still maintaining those packages, see Debian jessie builds of DKMSed upstream DRBD8 Kernel Module.
Solution: Patching DRBD 8 kernel module
See bottom of Ever-rising load on Debian jessie + DRBD8 + LXC host pairs.