🎨

go/base/context.go

@@ -119,7 +119,7 @@ type MigrationContext struct {
 	PostponeCutOverFlagFile              string
 	CutOverLockTimeoutSeconds            int64
 	CutOverExponentialBackoff            bool
-	CutOverExponentialBackoffMaxInterval int
+	CutOverExponentialBackoffMaxInterval int64
 	ForceNamedCutOverCommand             bool
 	PanicFlagFile                        string
 	HooksPath                            string
@@ -343,7 +343,7 @@ func (this *MigrationContext) SetCutOverLockTimeoutSeconds(timeoutSeconds int64)
 	return nil
 }
 
-func (this *MigrationContext) SetCutOverExponentialBackoffMaxInterval(intervalSeconds int) error {
+func (this *MigrationContext) SetCutOverExponentialBackoffMaxInterval(intervalSeconds int64) error {
 	if intervalSeconds < 2 {
 		return fmt.Errorf("Minimal maximum interval is 2sec. Timeout remains at %d", this.CutOverExponentialBackoffMaxInterval)
 	}

go/cmd/gh-ost/main.go

@@ -83,7 +83,7 @@ func main() {
 	flag.BoolVar(&migrationContext.SwitchToRowBinlogFormat, "switch-to-rbr", false, "let this tool automatically switch binary log format to 'ROW' on the replica, if needed. The format will NOT be switched back. I'm too scared to do that, and wish to protect you if you happen to execute another migration while this one is running")
 	flag.BoolVar(&migrationContext.AssumeRBR, "assume-rbr", false, "set to 'true' when you know for certain your server uses 'ROW' binlog_format. gh-ost is unable to tell, event after reading binlog_format, whether the replication process does indeed use 'ROW', and restarts replication to be certain RBR setting is applied. Such operation requires SUPER privileges which you might not have. Setting this flag avoids restarting replication and you can proceed to use gh-ost without SUPER privileges")
 	flag.BoolVar(&migrationContext.CutOverExponentialBackoff, "cut-over-exponential-backoff", false, "Wait exponentially longer times between failed cut-over attempts (obeys a maximum interval configurable with 'cut-over-exponential-backoff-max-interval'). Ignores 'default-retries.'")
-	cutOverExponentialBackoffMaxInterval := flag.Int("cut-over-exponential-backoff-max-interval", 64, "Maximum number of seconds to wait between failed cut-over attempts. Ignored unless 'cut-over-exponential-backoff' is 'true.' When the maximum is reached, attempts will stop, regardless of whether the last was successful.")
+	cutOverExponentialBackoffMaxInterval := flag.Int64("cut-over-exponential-backoff-max-interval", 64, "Maximum number of seconds to wait between failed cut-over attempts. Ignored unless 'cut-over-exponential-backoff' is 'true.' When the maximum is reached, attempts will stop, regardless of whether the last was successful.")
 	chunkSize := flag.Int64("chunk-size", 1000, "amount of rows to handle in each iteration (allowed range: 100-100,000)")
 	dmlBatchSize := flag.Int64("dml-batch-size", 10, "batch size for DML events to apply in a single transaction (range 1-100)")
 	defaultRetries := flag.Int64("default-retries", 60, "Default number of retries for various operations before panicking")

go/logic/migrator.go

@@ -149,20 +149,22 @@ func (this *Migrator) retryOperation(operation func() error, notFatalHint ...boo
 	return err
 }
 
+// retryOperation attempts running given function, waiting 2^(n-1) seconds
+// between each attempt, where n is the running number of attempts. exits
+// as soon as the function returns with non-error, or as soon as the next
+// wait interval exceeds `CutOverExponentialBackoffMaxInterval`.
 func (this *Migrator) retryOperationWithExponentialBackoff(operation func() error, notFatalHint ...bool) (err error) {
-	var numAttempts int
-	for {
+	numAttempts := 0
+	var interval int64
+	maxInterval := this.migrationContext.CutOverExponentialBackoffMaxInterval
+	for interval < maxInterval {
+		time.Sleep(time.Duration(interval) * time.Second)
 		err = operation()
 		if err == nil {
 			return nil
 		}
-
-		interval := int(math.Exp2(float64(numAttempts)))
-		if interval > this.migrationContext.CutOverExponentialBackoffMaxInterval {
-			break
-		} else {
-			time.Sleep(time.Duration(interval) * time.Second)
-		}
+		interval = int64(math.Exp2(float64(numAttempts)))
+		numAttempts++
 	}
 	if len(notFatalHint) == 0 {
 		this.migrationContext.PanicAbort <- err