KB-1211 "Inconsistent DBID on GW1" error thrown in checkengine script

Symptoms

One or more Appian engines fail to start. The output of checkengine.(bat|sh) shows the following:

Inconsistent DBID on GW1. Server Database ID: XXXX, Gateway Database ID: ON

Cause

If this is a migration, this error indicates the environment was not fully stopped before copying the engine database files *.kdb to the new Appian version.

For other scenarios where there’s no migration involved, the root cause is usually that Appian was shut down incorrectly or suddenly without completing the checkpoint.

Action

If this is not the result of a migration, perform the following:

1. Stop Appian per Starting and Stopping Appian.
2. REQUIRED: If running Appian 16.2 or 16.3, apply the latest hotfix.
3. Run the repair script located in <APPIAN_HOME>/server/_scripts/tools to fix the .kdb files for those engines with issues. Note: If the repair script successfully identifies that a KDB is in need of repair, it will create a .repaired file, which is the original file with some transactions dropped.
4. Rename the .repaired file to a higher numbered KDB. For instance, if the file is named pe445.kdb.repaired, you can rename it to pe446.kdb or pe500.kdb. Just ensure that the repaired KDB is the highest numbered KDB in the list for that engine. In most instances, the .repaired file will be missing data as expected (since the repair script truncates invalid transactions), so the new KDB will need to be tested for integrity.
5. Start Appian. Run checkengine.(bat|sh) to ensure the engine starts up successfully.

For migrations, the solution is to fully stop the source environment with stop-suite.(bat|sh) and perform the migration again.

Warning: Renaming .kdb files is not supported beyond what is documented in Appian documentation, the Appian Playbook, and KB articles.

Affected Versions

This article applies to versions 17.2 and earlier.