Glenn, here is the output of both the trap command run at the end of the script and when run separately in the interactive shell after the script:
$ ./ttt.ksh 0 0 0 0 From main program: trying to trap SIGINT ...signal INT trapped trying to trap SIGQUIT ... trying to trap SIGTERM ...signal TERM trapped trying to trap SIGHUP ...signal HUP trapped trap -- trap_TERM TERM trap -- '' QUIT trap -- trap_INT INT trap -- trap_HUP HUP end $ trap $ To debug further I also created a completely new user with ksh as the default shell, and interesting enough, the SIGQUIT trap then started working. So I guess this means the problem must indeed be somewhere in my user environment. There is nothing obvious to me in the .* files at the moment, but I'll take it from here and sum up when I find the culprit. Thanks for your help, --Marcus Glenn Fowler wrote on 03/ 1/10 05:22 PM: > On Mon, 01 Mar 2010 17:16:51 +0000 Marcus Heine wrote: > >>> at the same prompt you run the script (that fails) from >>> what does this command output >>> trap >>> >>> >> Output below, also enhanced with trap command exit stati. >> > >> So it looks like in my environment it does not seem to accept the trap >> definition (trap trap_QUIT QUIT), however the exit status of that command >> is 0. >> > >> $ ./ttt.ksh >> > ... > >> trap -- '' QUIT >> > this shows that QUIT is ignored inside the script > at the same prompt that you ran ./ttt.ksh from, run the following command > trap > and send the output -- I (and dgk) expect that QUIT is also ignored in your > interactive shell -- and that would implicate your .profile or $ENV file > (and possibly ~/.kshrc depending on how your local administration) > > _______________________________________________ > ksh93-integration-discuss mailing list > ksh93-integration-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/ksh93-integration-discuss
