Take V100R019C10 as an example.
A single instance is deployed for the MapReduce service. If the node of this instance fails, the MapReduce service status will become abnormal accordingly. Since the Hive and Spark2x services are dependent on the MapReduce service, the status of these two services will become abnormal as well. According to the alarm cause displayed in the alarm list, the node failure is due to the abnormal MapReduce service. Although the status of the Hive and Spark2x services is displayed as abnormal, the two services are running properly.
iMaster NCE-Campus services are not affected because they do not use the MapReduce function.
Rectify the node fault first. After the fault is rectified, the alarm is automatically cleared.
