ALM-14012 JournalNode Is Out of Synchronization Is Generated in the Cluster

Symptom

Alarm "ALM-14012 JournalNode Is Out of Synchronization" is generated in the MRS cluster.

Cause Analysis

1. Log in to the node for which the alarm is generated and search for the startDetail.log file of the JournalNode instance in the /var/log/Bigdata/hdfs/nn directory. The log shows that the JournalNode instance has stopped.
2. Check the latest edits log file in the /srv/BigData/journalnode/hacluster/current directory on the node for which the alarm is generated and other JournalNode nodes. It is found that data on the faulty node is not synchronized with other nodes.

Procedure

1. Log in to FusionInsight Manager and choose Cluster > Services > HDFS > Instance. On the displayed page, select the JournalNode instance of the node for which the alarm is generated, click More, and select Stop Instance.
2. Log in to the node for which the alarm is generated, and move all files in the /srv/BigData/journalnode/hacluster/current directory to another new directory (for example, /opt/test) to clear the directory.
3. Log in to FusionInsight Manager and choose Cluster > Services > HDFS > Instance. On the displayed page, select the stopped JournalNode instance and click Start Instance.
4. Wait for a while and check whether the alarm is cleared.