Long AEMaaCS deployment times caused by custom indexing jobs and Elastic App Search errors
Deployment times in Adobe Experience Manager as a Cloud Service can be longer than expected when custom indexing jobs reference deleted or unavailable Elastic App Search engines. Update the jobs to reference only existing engines, then redeploy.
Description description
During deployments using Cloud Manager in Adobe Experience Manager as a Cloud Service, deployment times may be significantly longer than expected. This can happen when custom indexing jobs are configured to connect to Elastic App Search engines that have been deleted or are unavailable. In these cases, deployment logs typically show repeated errors with HTTP 410 Gone responses, indicating that the requested resource has been deleted. These errors do not cause the deployment to fail but increase processing load during environment startup, resulting in longer deployment times.
Environment
Adobe Experience Manager as a Cloud Service (all versions)
Symptoms
- Deployment duration is longer than expected.
- Deployment logs contain repeated errors related to custom indexing jobs and Elastic App Search engine connectivity.
- Example log entry:
Error performing request URI [ /api/as/v1/engines/<engine-name>...]
Cause
Custom indexing jobs are configured to connect to Elastic App Search engines that have been deleted or are unavailable, resulting in repeated HTTP 410 Gone errors during environment startup. This increases processing load and delays the deployment process.
Resolution resolution
Follow the steps below to resolve the issue.
- Review the deployment logs for errors related to custom indexing jobs attempting to connect to Elastic App Search engines. Look for HTTP 410 Gone responses indicating deleted resources.
- Identify the custom indexing jobs or services responsible for these requests. These may be custom schedulers or services implemented in the AEM project.
- Verify the configuration of the Elastic App Search engines referenced by these jobs. Ensure that the engine names are valid and that the engines exist and are accessible.
- Update the configuration or code of the custom indexing jobs to reference only existing and available Elastic App Search engines.
- Redeploy the build using Cloud Manager.
- Monitor the deployment logs to confirm that HTTP 410 Gone errors no longer appear and that deployment times return to expected levels.