Troubleshooting

Enterprise

Bacula Enterprise Only

This solution is only available for Bacula Enterprise. For subscription inquiries, please reach out to sales@baculasystems.com.

This article lists common issues and the first checks to perform when using the HDFS Plugin.

Connection Problems

Symptoms:

  • The job fails before the plugin starts the backup or restore workflow.

  • The joblog shows a connection error or a missing URL warning.

Likely causes and checks:

  • Confirm that url is set

  • Verify that the File Daemon host can reach the HDFS endpoint.

  • If Kerberos is used, confirm that user_principal, keytab, and service_principal are set and valid.

  • If XML-based Hadoop configuration is used, confirm that the paths provided in core_site_path, hdfs_site_path, and ssl_client_path are readable.

Snapshot Problems

Symptoms:

  • Backup jobs fail when trying to create or access snapshots.

  • Query or list operations return no snapshot data.

Likely causes and checks:

  • Confirm that the target directory in base_snap_dir exists.

  • Confirm that the HDFS directory is snapshot-enabled.

  • Verify that the current HDFS user has permission to create and inspect snapshots.

  • If incremental or differential jobs unexpectedly behave like full backups, check that the previous Bacula snapshot still exists.

Restore Problems

Symptoms:

  • A restore writes files to an unexpected location.

  • Existing files are not replaced the way you expected.

  • The restore finishes but some files are skipped.

Likely causes and checks:

  • Remember that the current implementation uses where to decide whether the restore is local.

  • Check the Bacula Replace setting in the restore job.

  • Confirm that the destination path on the File Daemon host exists and is writable.

  • If abort_on_error is enabled, the restore stops as soon as a write fails.

Operational Checks

  • Review the plugin log path when a job fails early.

  • Check whether the selected snapshot directory still exists.

  • Validate the chosen fstype and ensure it matches the intended workflow.

  • Reduce the scope of include or exclude filters if a large namespace job is taking too long to complete.

See also

Previous articles:

Next articles:

Go back to: HDFS Plugin.