A Journal Node is in Bad Health in Cloudera Manager with "java.io.IOException: Can't scan a pre-transactional edit log"

(Doc ID 2160881.1)

Last updated on AUGUST 30, 2016

Applies to:

Big Data Appliance Integrated Software - Version 4.4.0 and later
Linux x86-64

Symptoms

One of the Journal nodes is in bad health in Cloudera Manager.  Restarting the hdfs service does not help.

The Cloudera Manager console reports:

The health test result for JOURNAL_NODE_WEB_METRIC_COLLECTION has become bad: The Cloudera Manager Agent got an unexpected response from this role's web server.
The health test result for JOURNAL_NODE_FILE_DESCRIPTOR has become good: Open file descriptors: 219. File descriptor limit: 32,768. Percentage in use: 0.67%.
The health test result for JOURNAL_NODE_SWAP_MEMORY_USAGE has become good: 0 B of swap memory is being used by this role's process.
The health test result for JOURNAL_NODE_HOST_HEALTH has become good: The health of this role's host is good.
The health test result for JOURNAL_NODE_FSYNC_LATENCY has become unknown: Not enough data to test: Test of whether the JournalNode fsyncs are completed fast enough.
The health test result for JOURNAL_NODE_GC_DURATION has become unknown: Not enough data to test: Test of whether this role spends too much time in garbage collection.



Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms