ado@elsie.UUCP (Arthur David Olson) (09/10/86)
I've been discovering core dumps in / from /etc/getty on our 4.1BSD system of late. Always caused by SIGQUIT. My first look at "init.c" and "getty.c" leads me to believe that there's a window of vulnerability between the time a line is opened and the time the line is put into raw mode where a quit character received on the line would indeed cause getty (which doesn't do anything with QUIT signals) to dump core. A look at System V also seems to show the same window (though on System V the window is entirely in getty itself, which has the responsibility for opening the line). While this is not a major problem--init just starts getty up again-- I'm tempted to have getty ignore QUIT signals to avoid getting the core dumps (to avoid the possibility of overwriting some more useful core dump). Are there any second opinions on doing this? If so, please mail them to me. Thanks. -- UNIX is a registered trademark of AT&T. -- UUCP: ..decvax!seismo!elsie!ado ARPA: elsie!ado@seismo.ARPA DEC, VAX, Elsie & Ado are Digital, Borden & Ampex trademarks.