Workspace Offline (503) Error: How Can I Check Memory Usage and Plan Limits?

Below is some guidance based on the official documentation:

Memory Limitations of Pro Plans

Depending on your subscription tier, n8n Cloud has set memory limits:

• Pro-1 (10k executions): 640 MiB RAM
• Pro-2 (50k executions): 1280 MiB RAM

Since you mentioned you’re on a Pro plan, your workspace is governed by one of the above limits. If your workflow’s data processing pushes usage near these limits, it can cause the instance to crash or become temporarily offline (resulting in 503 errors).

Monitoring Memory Usage

n8n Cloud doesn’t offer a direct memory usage dashboard. However, you can take the following steps to help verify if high memory consumption is the culprit:

1. Review Logs and Error Messages:
   Look for warnings like “n8n may have run out of memory while executing it” in your execution logs. Also, check for messages such as "JavaScript heap out of memory" if you have self-hosting–these are strong indicators of memory exhaustion.

2. Audit Workflow Design:
   Ensure that your workflows are built with memory efficiency in mind. Large JSON data, binary data, and memory-heavy nodes like the Code node can quickly increase memory consumption. See the Cloud Data Management guidelines for best practices.

3. Adjust Execution Data Saving:
   If you don’t need to save every successful production execution, consider turning off execution logging. This can reduce overall data (and memory usage) and is configurable from the workflow settings.

What to Do Now

• Optimize Workflow:
  If possible, simplify or break up the workflow to reduce the memory demand. For example, you might:

  • Avoid storing unnecessary execution data.
  • Process large data in smaller batches.
  • Use nodes that help reduce the memory footprint.

• Review Usage Soon After Workflow Runs:
  Sometimes the automatic data pruning (which runs based on storage and time limits) may take a little while. If your instance was temporarily overwhelmed, it might restart quickly. There isn’t a fixed “Monday” wait time, so checking logs and trying a modified (more memory-efficient) workflow may help.

• Contact Support:
  If the issue persists and you suspect it’s due to a memory overload, it could be useful to reach out to n8n Cloud support. They may be able to confirm the memory state and advise if an upgrade or further tuning is necessary.

By reviewing the logs, confirming your workflow’s memory footprint, and optimizing where possible, you should be able to determine whether the 503 error is indeed a memory-related issue and take steps to mitigate it.