[connector/failover] Ability to skip max_retries for levels

[sinkingpoint]

Is your feature request related to a problem? Please describe.

The failover connector currently has max_retries setting that allows setting a maximum number of retries before completely removing an exporter from the pool. In our case, our high priority receiver might be out for hours, or even days, so there's no real setting that makes sense for us.

Describe the solution you'd like

We should have a way to disable the max_retries setting so that the connector will continue to poll the higher priority receivers, regardless of how long they have been out. Other components use a magic value of 0 for this purpose, but in the failover connector a max_retries of 0 automatically rejects every pipeline and errors.

Describe alternatives you've considered

The alternative is just to set this to a very large value, potentially calculated based off of the retry time, and some maximum acceptable length of time, but this just sets a time bomb

[github-actions]

Pinging code owners:

• connector/failover: @akats7 @djaglowski @fatsheep9146

See Adding Labels via Comments if you do not have permissions to add labels yourself.

[github-actions]

Pinging code owners for connector/failover: @akats7 @djaglowski @fatsheep9146. See Adding Labels via Comments if you do not have permissions to add labels yourself.