Merge pull request #325 from hubert-s/feature_324

Fixes #324 by disabling health-check when splunk.hec.lb.poll.interval…

Author: kashyap-splunk

README.md

@@ -163,7 +163,7 @@ Use the below schema to configure Splunk Connect for Kafka
 | `splunk.hec.max.outstanding.events` | Maximum amount of un-acknowledged events kept in memory by connector. Will trigger back-pressure event to slow down collection if reached. | `1000000` |
 | `splunk.hec.max.retries` | Amount of times a failed batch will attempt to resend before dropping events completely. Warning: This will result in data loss, default is `-1` which will retry indefinitely  | `-1` |
 | `splunk.hec.backoff.threshhold.seconds` | The amount of time Splunk Connect for Kafka waits to attempt resending after errors from a HEC endpoint." | `60` |
-| `splunk.hec.lb.poll.interval`  |  Specify this parameter(in seconds) to control the polling interval(increase to do less polling, decrease to do more frequent polling) |  `120` |
+| `splunk.hec.lb.poll.interval`  |  Specify this parameter(in seconds) to control the polling interval(increase to do less polling, decrease to do more frequent polling, set `-1` to disable polling) |  `120` |
 | `splunk.hec.enable.compression` | Valid settings are true or false. Used for enable or disable gzip-compression. |`false`|
 ### Acknowledgement Parameters
 #### Use Ack

src/main/java/com/splunk/hecclient/LoadBalancer.java

@@ -63,7 +63,15 @@ public LoadBalancer(HecConfig hecConfig, CloseableHttpClient client) {
         Runnable r = () -> {
             run();
         };
-        scheduledExecutorService.scheduleWithFixedDelay(r, 0, this.hecConfig.getlbPollInterval(), TimeUnit.MILLISECONDS);
+
+	if (this.hecConfig.getlbPollInterval() > 0)
+	{
+            scheduledExecutorService.scheduleWithFixedDelay(r, 0, this.hecConfig.getlbPollInterval(), TimeUnit.MILLISECONDS);
+	}
+	else
+	{
+            log.info("Disabling out-of-band health-check");
+	}
     }
 
     public void setHttpClient(CloseableHttpClient httpClient) {