Quoth Martin Turon on Mon, Oct 20, 2008 at 03:57:00AM -0700:
> Is there a way to control the # of retries based on start method failures
> before going into maintenance mode?
No. This change has been requested as
6450661 start method failures should be configurable
> Is there a way to control the delay between retries of start caused by
> method failures?
There's no way to tell SMF to do this for you. You could build this
logic into your start method, though. Perhaps by making the start
method wait until the target is ping'able before starting the tunnel.
You might even include a process in the service which watches your
tunnel process, and when it dies, sleeps for a certain amount of time,
or waits until the target is pingable, before exiting itself. I suspect
this can be done with the shell, but am not sure.
David
