Hello, I was checking this bug during the Paris BSP which is taking place right now. What about throwing in another check like the following one in addition to the tty and SSH_CONNECTION ones?
is_child_of_sshd() { pid=$$ ppid=$PPID # Walk up to init. while [ $pid -ne 1 ]; do grep -q sshd /proc/$ppid/cmdline && return 0 pid=$ppid ppid=$(grep ^PPid: /proc/$pid/status | tr -dc 0-9) done return 1 } [...] if ! pgrep -f "^sshd.+${PTS#/dev/}\>" >/dev/null \ && [ -z "${SSH_CONNECTION:-}" ] \ && ! is_child_of_sshd; then [...] It seems reasonable to me to try to climb up the process tree until we meet a process with "sshd" in the command line (or, maybe better (?), with "sshd" in the proc/<pid>/exe symlink). Still not the cleanest of the solutions, but should be quite portable. Ludovico -- <l...@dovi.co> IRC: garden@freenode OpenPGP: 1024D/63D2D5D907F89BB8 Jabber/gtalk: garde...@gmail.com -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org