topic manager should retry

[frairon]

in rare cases, topic creation can get delayed, so the topic manager should retry when verifying a topic, like this:

func (m *topicManager) getTopicConfigMap(topic string) (map[string]sarama.ConfigEntry, error) {
	var (
		err error
		cfg []sarama.ConfigEntry
	)

	err = withRetry(func() error {
		cfg, err = m.admin.DescribeConfig(sarama.ConfigResource{
			Type: sarama.TopicResource,
			Name: topic,
		})

		return err
	}, "get topic config", 3, NewSimpleBackoff(1*time.Second, 1*time.Second))

	// now it does not exist anymore -- this means the cluster is somehow unstable
	if err != nil {
		return nil, fmt.Errorf("Error getting config for topic %s: %w", topic, err)
	}
	// remap the config values to a map
	cfgMap := make(map[string]sarama.ConfigEntry, len(cfg))
	for _, cfgEntry := range cfg {
		cfgMap[cfgEntry.Name] = cfgEntry
	}
	return cfgMap, nil
}

func withRetry(actor func() error, message string, retries int, backoff Backoff) error {
	var errors *multierror.Error
	for retries > 0 {
		err := actor()
		if err != nil {
			errors = multierror.Append(errors, err)
		} else {
			return nil
		}
		time.Sleep(backoff.Duration())
		retries--
	}
	return multierror.Append(errors, fmt.Errorf("%s failed with too many retries", message))
}

see if we can use a library to do the retry instead of a custom solution