[erlang-questions] catch supervisor failure

[Bengt Kleberg]

Would it be possible to have a supervisor between the master supervisor
and the one that dies?
This new supervisor could handle the behavior you described before, but
has gotten lost from this email. It was something about 1 minute, I
think.


bengt

On Fri, 2011-08-26 at 07:42 +0200, Max Lapshin wrote:
> On Fri, Aug 26, 2011 at 9:25 AM, Bengt Kleberg
> <bengt.kleberg@REDACTED> wrote:
> > Greetings,
> >
> > Supervisor tree. Ie, your supervisor has a supervisor.
> >
> > See http://www.erlang.org/doc/design_principles/des_princ.html
> >
> 
> And what next?
> 
> worker has failed for 100 times during 3 seconds. It's supervisor is
> dead and is restarted by parent. This happens for 5 times during 15
> seconds and then master supervisor dies with all application.
> 
> I don't see any callback: "handle supervisor death"