[Sven Joachim] > Any clue why killall5 did not manage to kill it for good?
I would assume the process is unkillable due to some kernel hang. killall5 simply call the kill(2) system call to signal processes to die. Your analysis make me believe sendsigs is correct, and the problem is between the kernel and the package with the process that fail to die, and this report should be reassigned elsewhere. Happy hacking, -- Petter Reinholdtsen -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org