If using fallback, should stop respecting the interval period

[nicholaschiasson]

The fallback becomes the peer to use in the case of DNS resolution failure. In many cases, this can be due to a sporadic failure, thus it becomes undesirable to use the fallback address for the full interval duration. Allowing one or a few errors is acceptable, but allowing errors for several seconds, possibly minutes, due to an interval configured to match a DNS TTL, this can be very bad.

In the case where the fallback address is being used, we should attempt the DNS resolution until it succeeds.

[nicholaschiasson]

Perhaps also this could be configurable. We could add another boolean attribute to the directive like fallback_recover, which when true (present), we use the behaviour I described above, otherwise if not present, then we continue with the current behaviour. I think this would be the best solution.

[nicholaschiasson]

Superseded by #10