Question regarding <defunct> processes

[Doug Gwyn]

In article <413 at servio.UUCP> penneyj at servio.UUCP (D. Jason Penney) writes:
>Oh, the astonishing things that happen in pUnyx.  What you're struggling with
>is the infamous "SIGCHLD" problem.  To allow a process to complete its
>death, you must trap the SIGCHLD signal and then do a wait() to clear it.

Wrong.
It is not necessary to trap SIGCHLD to reap zombies.
In fact I don't recommend relying on weird SIGCLD/SIGCHLD semantics
for any purpose whatsoever.
I emailed the original requestor a proper explanation and solutions.