Oystein Viggen <[EMAIL PROTECTED]> writes: > Experiment: > > $ cp /hurd/null somedir/ > $ settrans somedir somedir/null > $ cat somedir > > If translators within directories within translators worked, cat somedir > would return nothing, just the same as cat /dev/null. On my system, cat > would hang, waiting for input, until killed. I did not investigate this > further.
Ah yes, this is indeed the case. The parent filesystem tries to run somedir/null, blocking your "cat" command until it returns. And, that "tries to run" includes lookup up the same somedir/null. So I was wrong in what I wrote earlier. This doesn't bother me, as long as C-c works. Can you verify that it does? > If you set up an active translator instead (settrans -a), it all works > just beautifully. Right, because here the program is run by settrans *before* it gets attached to the "somedir" node. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]