protin@pica.army.mil (Arthur W. Protin Jr.) (07/18/90)
In article <1277@tuewsd.win.tue.nl> wsinpdb@lso.win.tue.nl (Paul de Bra) writes: Sender: wsinpdb@win.tue.nl (Paul de Bra) Organization: Eindhoven University of Technology, The Netherlands Lines: 15 In article <960004@teecs.UUCP> belkin@teecs.UUCP (Hershel Belkin) writes: >... >Every so often I fins a shell process (sh or ksh) which has somehow >become dis-associated with its logon session. By this I mean that >the shell's PPID is "1", and the user is no longer logged on. >... This is the infamous trap/signal/eof bug, which I don't know exactly, but some combination of trapping and sending signals and having end-of-file on standard input causes an infinite loop in the Bourne and Korn shell, at least in some Unix versions which haven't fixed the bug. Anyone know the full scoop? Paul. Probably not, but here is another piece of the puzzle that may be of some help. A couple of years ago, while at another company, we ported the Korn shell to the SGI Personal Iris 4D/20. We found many instances of the shell running detached and soaking up so many cycles as to alarm us all. The problem was traced to a conflict of features. The trigger was a remote login session being disconnected. We tuned in with the debugger in the middle of the infinite loop. Ksh would detect an error and "long jump" to clean up. In its clean up it tried to display a message, which would fail. Ksh would detect the error ... The problem was that when errors were detected on input/output operations the errno was checked for recovery and/or to be ignored. The errno on the 4D/20 for connection loss was not a number that was special cased as meaning the loss of controling terminal. Thus, when I added that errno to the "if", the problem was cured. Arthur Protin protin@pica.army.mil