Re: init scripts problems ?

"Also sprach denis bonnenfant:"
> > think your test is something like "kill clients while under raid; start
> > new client", but I am not sure).
> 
> Just stop server on the other side.

Oh, OK! In that case the clients should die off, or at least get pretty
darn sick. 

So that explains why they were dying ... but ONE of them was on the way
in. It appears that as one old client died, a new one was born (that's
expected), but that the second client to die couldn't die all the way.

I would assume that the second client to die was stuck in the dying,
which explained why the new replacement for its sibling couldn't start.

So I expect this should be reproducible. 


Peter