Elasticube stuck in "saving files" phase and unable to build cube [Linux]
Step-by-step guide
When building an Elasticube in Sisense Linux, the process may halt at the “Initializing,” “Saving files,” or “Attaching cube” stages, especially after manually cancelling a build. Standard service restarts do not resolve these stuck states.
1. Check RabbitMQ queues for stuck messages
Sometimes, message queues hold up the completion of the build.
Command (bash):
Review the output for queues with unexpected message counts.
Screenshot: Terminal with rabbitmqctl command output highlighting nonzero queues.
(Description: Output showing a queue with many pending messages that may indicate blockage.)
If stuck messages are found, consider purging the queue.
2. Identify and remove stuck cube build entries in MongoDB
Persistent build entries can block new builds.
Find your MongoDB pod:
Access the MongoDB pod:
Enter MongoDB shell:
Switch to Sisense's 'be' database:
Find running build entries:
(Description: Output should show records with your cube’s name and running status.)
Delete stuck builds for the specific cube (replace 'cube name' accordingly):
3. Retry the cube build
Return to the Sisense UI and try building the cube again.
Troubleshooting Tips
- Double-check the cube name before using the MongoDB delete command.
- Always back up your database before performing direct deletions.
- For multiple stuck cubes, repeat the deletion process for each one.
- If problems persist, review Sisense system logs and reach out to Sisense Support.
Conclusion
Sisense cube builds can get stuck in “Saving files” due to residual process states, especially after manual cancellation. Clearing RabbitMQ queues and removing “RUNNING” build entries from MongoDB resolves this for most users. Always exercise caution when making database changes.
References/related content
Build Log Walkthrough
How To Troubleshoot Build Failures (Linux OS)
Disclaimer: This post outlines a potential custom workaround for a specific use case or provides instructions regarding a specific task. The solution may not work in all scenarios or Sisense versions, so we strongly recommend testing it in your environment before deployment. If you need further assistance with this, please let us know.
