Re: 4gl Segmentation Fault

In article <44ciba$1b1@urania.cs.du.edu> aburt@urania.cs.du.edu "Andrew Burt" writes: > > >} We have ported a 4gl application from a MIPS to DEC box. > >} The application has 4ge's run from other 4ge's (ie run "prog.4ge"). > >} In certain places the message "Segmentation Fault" flashes up and > >} it returns to the calling program. For a given terminal it happens > >} in exactly the same place each, time but varies between terminals. > >} (We've tried xterms, wy120 and vt220 emulators) > >} > >} If the 4ge where the problem occurs is run separately from the > >} command line everything is OK. > >} (OSF V3.0, 4gl 4.13 UD1, Online 5.05 UC1) > >} > >} Any ideas? > >} > >} Thanks, William > > You might check the termcaps verrrry carefully. If you have it wrong, > e.g., #rows or #cols funny, it can core dump. I can't remember exactly > how I got it to do it before, but I know I had that once and found > the terminal type and/or termcap weren't "normal". I have found this to happen on background programs ('without waiting') when they use *something* which tries to access $TERMCAP when there isn't a valid value for $TERM in it - examples are the various 'SET OPTIONS' stuff. -- ============================================================================ Sally Woolrich | This mail contains my personal sally@excelsis.demon.co.uk | views not those of my employer! ============================================================================