Azure Storage Queues messages should be processed in a timely manner. When messages get stuck in a queue for too long, it might highlight issues with other parts of the system (e.g. service that should pick up the messages and process them). Observing queues is a good way to ensure that other parts of the system are functioning correctly, even if they can't be monitored directly.
In order to ensure that Azure Storage Queue is depleted in a timely manner:
- Open the Azure Queue Storage configuration dialog.
- Define a new metric called OldestMessageAge (category AzureStorageQueueOldestMessageAgeInMinutes).
- Define a new alert with severity Error that will be raised whenever the age of the oldest message exceeds 10 min.
Other metrics that you might find useful are:
- AzureStorageQueueLength - raise an alert whenever length exceeds a specific value;
-
AzureStorageQueueOldestMessageId - raise an alert whenever the id doesn't change in subsequent monitoring cycles.
All three metrics can be used in combination for defining a single alert or individually.