System Metrics Not Updating After Scaling Azure DocumentDB (MongoDB vCore) from M10 to M20

Question

Problem Description: We are experiencing an issue with system-level metrics (CPU percent, Memory percent, Storage percent, IOPS, and overall Resource Utilization metrics) for our Azure DocumentDB (with MongoDB compatibility) cluster, deployed in Central India.

All metrics were working correctly when the cluster was running on M10 tier. After scaling the cluster from M10 to M20 on Sunday, the system metrics stopped updating entirely. Insights and Azure Monitor now show a flat line starting from the moment of the scale operation. Mongo request duration is still updating properly, which confirms that the cluster is running and processing workload.

Based on Microsoft documentation and our observations, this appears to be a backend issue where the Metrics Publisher / Metrics Agent failed to re-register with Azure Monitor after the scale-up operation. This is a known issue that can occur after tier upgrades in DocumentDB (MongoDB vCore) environments.
Can you help us resolve this issue

Answer 1

Hi Avinash Dhanukonda,

Thank you for reaching on Microsoft Q&A!

it looks like you're facing a frustrating issue with your Azure DocumentDB (MongoDB vCore) metrics not updating after scaling from M10 to M20. This does seem to align with known problems where the Metrics Publisher or Metrics Agent doesn't re-register with Azure Monitor following a scale-up operation.

• Here’s a step-by-step approach you can take to potentially resolve the issue:
  1. Check Diagnostic Settings: Ensure that your diagnostic settings are properly configured to collect metrics. You can do this by navigating to your Azure Cosmos DB resource and reviewing the diagnostics options.
  2. Restart the Instance: Sometimes, a simple restart of the DocumentDB instance can resolve temporary issues with metrics not displaying correctly.
  3. Monitor Azure Service Health: Check if there are any ongoing issues with Azure services in the Central India region that might be impacting your metrics. Look for service outages or issues in the Azure Status page.
  4. Validate Permissions: Ensure that there are no permission-related issues that could prevent metrics from being collected or displayed.
  5. Wait for Recovery: Occasionally, it takes some time for metrics to start updating post-scale. If it's just been a short while since the scaling, monitor the situation for a bit longer.
  6. Review Resource Utilization: Since the Mongo request duration is still updating, your cluster is likely processing workloads. Keep an eye on various resource utilization metrics, even if they're not currently visible.
  In the meantime, here are some follow-up questions that might help narrow down the troubleshooting:
  • When exactly did you scale from M10 to M20?
  • How long after the scale-up did you start noticing the metrics not updating?
  • Have you checked the diagnostic settings for any potential misconfigurations?
  • Are there any alerts configured for the metrics that might notify you of underlying issues?
  • Can you verify if other metrics (apart from Mongo request duration) are still functioning or updating?

Reference documents:

• Monitor Azure Cosmos DB for MongoDB vCore Metrics with Azure Monitor
• How to manage alerts for Azure Cosmos DB for MongoDB vCore
• Azure Cosmos DB Cluster Scaling Documentation
• Azure Monitor workbooks overview