The failsafe example in the docs doesn’t really wo...

# http4k

[snowe]

The failsafe example in the docs doesn’t really work. The circuitbreaker never triggers, it always continues to retry. You can see it if you add an onRetry block

[snowe]

val failsafeExecutor = Failsafe.with(
            Fallback.of(Response(OK).body("Fallback")),
            RetryPolicy.builder<Response>()
                .withMaxAttempts(5)
                .handleResultIf { !it.status.successful }
                .onRetry { result ->
                    <http://logger.info|logger.info> { "${result.attemptCount} Retry is occurring ${result.isRetry}" }
                    <http://logger.info|logger.info> { "${result.lastResult} " }
                }
                .build(),
            CircuitBreaker.builder<Response>()
                .withFailureThreshold(1)
                .handleResultIf { it.status.serverError }
                .build(),
            Timeout.of(Duration.ofMillis(100))
        )

I’m pretty sure I’m not doing anything wrong here. I checked the failsafe docs and it says this is how it should work, but that the circuit breaker should stop the retries.

[dave]

This is all down to failsafe - the filter is exceptionally simple 🙂 .

Still - I've update the example to hopefully be a little easier to grok. If you comment out the last 2 policies you can see that the retry is working as intended (ie. only on >299). I suspect you were seeing weird interplay between the different policies.

[snowe]

Yes I know the retry is working properly, but according to failsafe’s docs, the circuit breaker should kill the retry (I mean, that’s the whole point of the circuit breaker right?)

[snowe]

https://github.com/failsafe-lib/failsafe/issues/90