Sig: (file-error "Doing fork" "Bad file number") 21.4.0@Sol8

In XEmacs 21.4 (patch 0) "Solid Vapor" [Lucid] (sparc-sun-solaris2.8) of Mon Apr 16 2001 on jena configured using `configure --prefix=/export/home/marcow/usr/local --site-includes=/export/home/marcow/usr/local/include --site-libraries=/export/home/marcow/usr/local/lib --with-gtk=no --error-checking=none --debug=no' I did see the following a couple of times now with the 21.2.beta's and 21.4.0. I don't know what causes it but but it looks like XEmacs has sometimes problems with the fork(). It's not that often, it starts maybe after around 1000 forks (wild guess, vm starts an external program every 2 minutes) and than persists. Has somebody an idea where to start looking?? Thanks, -- Marco --------------------------------------------------------------------------- Signaling: (file-error "Doing fork" "Bad file number") old-call-process-internal("/usr/lib/sendmail" "/tmp/marcow/emacs9cGbhv" 0 nil "-oi" "-f" "marcow" "-oem" "-odb" "-t") apply(old-call-process-internal "/usr/lib/sendmail" "/tmp/marcow/emacs9cGbhv" 0 nil ("-oi" "-f" "marcow" "-oem" "-odb" "-t")) call-process-internal("/usr/lib/sendmail" "/tmp/marcow/emacs9cGbhv" 0 nil "-oi" "-f" "marcow" "-oem" "-odb" "-t") apply(call-process-internal "/usr/lib/sendmail" "/tmp/marcow/emacs9cGbhv" 0 nil ("-oi" "-f" "marcow" "-oem" "-odb" "-t")) call-process("/usr/lib/sendmail" "/tmp/marcow/emacs9cGbhv" 0 nil "-oi" "-f" "marcow" "-oem" "-odb" "-t") apply(call-process "/usr/lib/sendmail" "/tmp/marcow/emacs9cGbhv" 0 nil ("-oi" "-f" "marcow" "-oem" "-odb" "-t")) call-process-region(1 2636 "/usr/lib/sendmail" nil 0 nil "-oi" "-f" "marcow" "-oem" "-odb" "-t") apply(call-process-region (1 2636 "/usr/lib/sendmail" nil 0 nil "-oi" "-f" "marcow" "-oem" "-odb" "-t")) sendmail-send-it() mail-send() vm-mail-send() vm-mail-send-and-exit(nil) call-interactively(vm-mail-send-and-exit) --------------------------------------------------------------------------- --