Aha: after more searching I found the solution in another thread:
Solution: create /usr/local/bin/rsync
containing
#!/bin/sh
exec /usr/bin/rsync --xattrs --delete --compress --compress-level=2 "$@"
(and this even works with the stock rsync-3.1.1 xenial package)
Phew! Glad to be getting rid of this legacy server
Edit: the filesystem copied successfully, but the container’s init
process (symlink to /lib/systemd/systemd
) doesn’t launch any processes. At least the filesystem copied OK, so it’s a different problem to investigate…
root@ix-dns1-new:~# lxc exec ix-dc1 bash
root@ix-dc1:~# ps auxwww
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 1 0.0 0.0 36612 2020 ? Ss 16:20 0:00 /sbin/init
root 22 0.0 0.0 21804 3748 pts/1 Ss 16:21 0:00 bash
root 34 0.0 0.0 37764 3060 pts/1 R+ 16:21 0:00 ps auxwww
root@ix-dc1:~#
… and it seems to be this problem (Ubuntu 16.04 won’t work in a cgroup2 environment). Fortunately, this is a container I can retire (I tested it first because it was the least important), and the important ones are 20.04.