This is a clone of issue RHEL-84910 to use for version rhel-10.1
–
Original description:
Cause: When disabling linger for non-root users, we did not wait long
enough for the user state to be "closing". Then we restart
systemd-logind to force linger to be cancelled.

Consequence: On some systems, this starts a timer which will kill
the session for root, causing the sshd session to be terminated,
and the Ansible play to fail with `UNREACHABLE`.

Fix: Wait much longer for the user linger to be cancelled, and only
restart logind if absolutely necessary.

Result: The role does not fail with `UNREACHABLE` errors when
removing resources.

Signed-off-by: Rich Megginson <rmeggins@redhat.com>