Fix metric name for e2e test (#5985)

* Add log pattern condition and fix metric name for e2e

* remove incorrect metric from comments

* fix style

kafka/datadog_checks/kafka/data/conf.yaml.example

@@ -616,7 +616,7 @@ init_config:
         attribute:
           Value:
             metric_type: gauge
-            alias: kafka.net.handler.avg.idle.pct.rate
+            alias: kafka.net.processor.avg.idle.pct.rate
     - include:
         domain: 'kafka.server'
         bean: 'kafka.server:type=KafkaRequestHandlerPool,name=RequestHandlerAvgIdlePercent'

kafka/tests/common.py

@@ -35,7 +35,7 @@
 """
 
 KAFKA_E2E_METRICS = [
-    "kafka.net.handler.avg.idle.pct.rate",
+    "kafka.net.processor.avg.idle.pct.rate",
     # Request metrics:
     "kafka.request.channel.queue.size",
     "kafka.request.fetch_consumer.time.99percentile",

kafka/tests/conftest.py

@@ -6,12 +6,17 @@
 import pytest
 
 from datadog_checks.dev import docker_run
+from datadog_checks.dev.conditions import CheckDockerLogs
 from datadog_checks.dev.utils import load_jmx_config
 
 from .common import HERE
 
 
 @pytest.fixture(scope='session')
 def dd_environment():
-    with docker_run(os.path.join(HERE, 'compose', 'docker-compose.yml')):
+    compose_file = os.path.join(HERE, 'compose', 'docker-compose.yml')
+
+    with docker_run(
+        compose_file, conditions=[CheckDockerLogs(compose_file, [r'\[KafkaServer id=\d+\] started'], matches="all")]
+    ):
         yield load_jmx_config(), {'use_jmx': True}