[Linux-HA] Failing over on resource failure - is it possible ?

[Dmitry Korsun]

Hi,
   The application that I am considering using linux-ha for requires a 
failover to occur when a locally hosted resource fails.  From 
documentation and brief testing of heartbeat it does not seem like the 
resources configured in haresources file are monitored by heartbeat.
   So, if a master is hosting an FTP server and that FTP server dies 
(while the node itself remains operational in all other respects), is it 
possible to have hearbeat failover - or essentially commit suicide to 
force a failover - to another node ?  Ideally, the node that detected a 
hosted resource failure should then attempt a re-start of the resources 
and the node itself and on success become the master again.

Thanks for any help.

-- 
::::::::::::...... .  .   .
:: Dmitry Korsun
:: x4095
:: dmitry.korsun at corp.idt.net
:::::::::::::::::....... .   .       .