[BusyBox] init does not kill reparented processes (MYSTERY SOLVED)

[Erik Andersen]

On Sat Aug 28, 2004 at 07:00:42PM +0200, Ignacio Garc?a P?rez wrote:
> > Sigh.  As you no doubt know, as a special case SIG_IGN for
> > SIGCHLD means the signal isn't actually ignored, instead the
> > kernel does automatic child reaping.  We used to do that, and
> > then we received a fairly steady stream of bug reports about init
> > deadlocking, so we reverted the SIGCHLD stuff.
> 
> Now that I realize, if then you set up a ignore handler for SIGCHLD, how the
> heck do you wait for respawn process to terminate?. Shouldn't you instead
> place a true handler that checks if the terminated child is a zombie o a
> process controlled by init ?.

Did you read my patch?  It did not ignore SIGCHLD but instead has
a handler.  As a far as I can see, the code appears correct but
at least for me does not work as expected.

 -Erik

--
Erik B. Andersen             http://codepoet-consulting.com/
--This message was written using 73% post-consumer electrons--