2.3.3 ESIntegTestCase

I am currently trying to implement integration tests in Java for some sample data I have come up with to test a feature I just finished. Everything is going well and I am able to start a node, create the mapping, seed the data and make my assertions. The problem is occurring during the @after in ESIntegTestCase where there are several calls that look something like this: client().admin().cluster().someOtherCall()... which will randomly cause this error:

NoNodeAvailableException[None of the configured nodes are available: [{#transport#-1}{local}{local[5]}]]
at __randomizedtesting.SeedInfo.seed([2C373C1B13097282:742FFB2F7E484278]:0)
at org.elasticsearch.client.transport.TransportClientNodesService.ensureNodesAreAvailable(TransportClientNodesService.java:290)
at org.elasticsearch.client.transport.TransportClientNodesService.execute(TransportClientNodesService.java:207)
at org.elasticsearch.client.transport.support.TransportProxyClient.execute(TransportProxyClient.java:55)

Where the '{local[X]}' (5 in this case) in the error changes from run to run.

My java to create the index/mapping/data looks roughly like this:

protected void initializeData() throws JsonParseException, JsonMappingException, IOException {
   createIndex(INDEX);
   ensureGreen(INDEX);
   refresh();
   CLIENT = client();
	
   String mappingOne = readFile(MAPPING_RESOURCE);
   String mappingTwo = readFile(MAPPING_RESOURCE);
	
   CLIENT.admin().indices()
	        .putMapping(new PutMappingRequest(INDEX).type(TYPE_1).source(mappingOne))
	        .actionGet();
	
   CLIENT.admin().indices()
	        .putMapping(new PutMappingRequest(INDEX).type(TYPE_2).source(mappingTwo)).actionGet();
	 
   CLIENT.prepareIndex(INDEX, TYPE_1, getId().toString())
	          .setSource(my_test_json_data).execute().actionGet();
			  
   CLIENT.prepareIndex(INDEX, TYPE_2, getId().toString())
	          .setSource(my_test_json_data).execute().actionGet();
			  
   CLIENT.admin().indices().refresh(new RefreshRequest(INDEX)).actionGet();
}

Has anyone experienced anything like this or have any insight into the error I am seeing?

Solved my own issue. The problem turned out to be that the implementation that was building and executing the query was also closing the transport client. After removing this 'mistake' (query builder should not be responsible for closing the client connection) my tests are all passing.