Fixing possible race condition when passing backoff around as a pointer.

realis.go

@@ -90,7 +90,7 @@ type RealisConfig struct {
 	timeoutms                   int
 	binTransport, jsonTransport bool
 	cluster                     *Cluster
-	backoff                     *Backoff
+	backoff                     Backoff
 	transport                   thrift.TTransport
 	protoFactory                thrift.TProtocolFactory
 	logger                      Logger
@@ -141,7 +141,7 @@ func ZKUrl(url string) ClientOption {
 	}
 }
 
-func Retries(backoff *Backoff) ClientOption {
+func Retries(backoff Backoff) ClientOption {
 	return func(config *RealisConfig) {
 		config.backoff = backoff
 	}
@@ -159,7 +159,7 @@ func ThriftBinary() ClientOption {
 	}
 }
 
-func BackOff(b *Backoff) ClientOption {
+func BackOff(b Backoff) ClientOption {
 	return func(config *RealisConfig) {
 		config.backoff = b
 	}
@@ -220,7 +220,7 @@ func NewRealisClient(options ...ClientOption) (Realis, error) {
 
 	// Default configs
 	config.timeoutms = 10000
-	config.backoff = &defaultBackoff
+	config.backoff = defaultBackoff
 	config.logger = NoopLogger{}
 	config.options = options
 
@@ -570,14 +570,14 @@ func (r *realisClient) CreateService(auroraJob Job, settings *aurora.JobUpdateSe
 
 	resp, err := r.StartJobUpdate(update, "")
 	if err != nil {
-		return resp, nil, errors.Wrap(err, "unable to create service")
+		return nil, nil, errors.Wrap(err, "unable to create service")
 	}
 
 	if resp != nil && resp.GetResult_() != nil {
 		return resp, resp.GetResult_().GetStartJobUpdateResult_(), nil
 	}
 
-	return resp, nil, errors.New("results object is nil")
+	return nil, nil, errors.New("results object is nil")
 }
 
 func (r *realisClient) ScheduleCronJob(auroraJob Job) (*aurora.Response, error) {

realis_e2e_test.go

@@ -61,7 +61,7 @@ func TestMain(m *testing.M) {
 }
 
 func TestNonExistentEndpoint(t *testing.T) {
-	backoff := &realis.Backoff{ // Reduce penalties for this test to make it quick
+	backoff := realis.Backoff{ // Reduce penalties for this test to make it quick
 		Steps:    5,
 		Duration: 1 * time.Second,
 		Factor:   1.0,

retry.go

@@ -134,7 +134,7 @@ func (r *realisClient) thriftCallWithRetries(thriftCall auroraThriftCall) (*auro
 				adjusted = Jitter(duration, backoff.Jitter)
 			}
 
-			r.logger.Printf("A retriable error occurred during thrift call, backing off for %v before retrying\n", adjusted)
+			r.logger.Printf("A retriable error occurred during thrift call, backing off for %v before retry %v\n", adjusted, curStep+1)
 
 			time.Sleep(adjusted)
 			duration = time.Duration(float64(duration) * backoff.Factor)
@@ -146,6 +146,7 @@ func (r *realisClient) thriftCallWithRetries(thriftCall auroraThriftCall) (*auro
 		func() {
 			r.lock.Lock()
 			defer r.lock.Unlock()
+			r.logger.Println("Beginning Aurora Thrift Call")
 			resp, clientErr = thriftCall()
 		}()
 
@@ -169,6 +170,9 @@ func (r *realisClient) thriftCallWithRetries(thriftCall auroraThriftCall) (*auro
 			return nil, errors.New("Response from aurora is nil")
 		}
 
+		// Log response code for debugging
+		r.logger.Printf("Aurora replied with code: %v\n", resp.GetResponseCode().String())
+
 		// Check Response Code from thrift and make a decision to continue retrying or not.
 		switch responseCode := resp.GetResponseCode(); responseCode {
 
@@ -191,6 +195,7 @@ func (r *realisClient) thriftCallWithRetries(thriftCall auroraThriftCall) (*auro
 		// The only case that should fall down to here is a WARNING response code.
 		// It is currently not used as a response in the scheduler so it is unknown how to handle it.
 		default:
+			r.logger.Println("unhandled response code received from Aurora")
 			return nil, errors.Errorf("unhandled response code from Aurora %v", responseCode.String())
 		}