Thanks for all of the comments and suggestions, though I'm still stuck. To provide a little background, this actually comes from the lock.t test for IO::All.
Maybe my issues come from possibly incorrect assumptions, which are: - processes should block when requesting a shared lock on a filehandle that is already exclusively locked (unless specifically asking for non-blocking). - the parent process in the test should do the testing (&is). - in the test, the child process requests an exclusive lock, writes a few lines and sleeps, then unlocks. - in the test, the parent process requests a shared lock and blocks while the child still has the exclusive lock. After the child has finished writing, the parent reads in the lines and ensures that they're what the client wrote. Or, maybe I need a brain adjustment? Any clues are appreciated :) -- Ian Langworth Project Guerrilla Northeastern University College of Computer and Information Science _______________________________________________ Boston-pm mailing list [EMAIL PROTECTED] http://mail.pm.org/mailman/listinfo/boston-pm
