Fix flaky ml test by increasing timeout (elastic#9523)

The ml tests sometimes fail on Travis or Jenkins but it seems when looking at the output of the files normally all entries are there. As they pass most of the time I assume it is an issue related to the timeout. Increasing the timeout to 60s.

(cherry picked from commit 220cdee)

filebeat/tests/system/test_ml.py

@@ -133,6 +133,6 @@ def _run_ml_test(self, setup_flag, modules_flag):
         for obj in ["Datafeed", "Job", "Dashboard", "Search", "Visualization"]:
             self.wait_log_contains("{obj} already exists".format(obj=obj),
                                    logfile=output_path,
-                                   max_timeout=30)
+                                   max_timeout=60)
 
         beat.kill()

x-pack/libbeat/docker-compose.yml

@@ -22,15 +22,15 @@ services:
       kibana: { condition: service_healthy }
     healthcheck:
       interval: 1s
-      retries: 1200
+      retries: 2400
 
   elasticsearch:
     extends:
       file: ${ES_BEATS}/testing/environments/${TESTING_ENVIRONMENT}.yml
       service: elasticsearch
     healthcheck:
       test: ["CMD", "curl", "-u", "elastic:changeme", "-f", "http://localhost:9200"]
-      retries: 300
+      retries: 1200
       interval: 1s
     environment:
       - "ES_JAVA_OPTS=-Xms512m -Xmx512m"
@@ -47,6 +47,6 @@ services:
       service: kibana
     healthcheck:
       test: ["CMD-SHELL", 'python -c ''import urllib, json; response = urllib.urlopen("http://elastic:changeme@localhost:5601/api/status"); data = json.loads(response.read()); exit(1) if data["status"]["overall"]["state"] != "green" else exit(0);''']
-      retries: 300
+      retries: 1200
       interval: 1s
-    command: /usr/local/bin/kibana-docker --xpack.security.enabled=true --elasticsearch.username=elastic --elasticsearch.password=changeme
+    command: /usr/local/bin/kibana-docker --xpack.security.enabled=true --elasticsearch.username=elastic --elasticsearch.password=changeme