Monitor Azure Storage with Datadog
Microsoft Azure Storage is a cloud-based service that is designed to deliver highly available, geographically redundant storage for on-premise and cloud applications. Azure Storage offers client libraries for various languages, including PHP, Java, Python, and Ruby.
If you’re using Azure Storage, you will want to track its usage and get notified about any issues before the services and applications that rely on Azure Storage start to fail or degrade. That’s why we are pleased to announce that we have added Azure Storage to the growing list of Microsoft Azure cloud services that you can monitor with Datadog.
Azure Storage overview
General-purpose Azure Storage accounts have access to four services that address different needs:
- Blob storage, or Object storage, is designed for unstructured data. Blobs are stored in containers, each of which can hold any number of blobs. There are three types of blobs:
- block blobs: objects like image files and documents
- page blobs: designed to support random read/write access, like a virtual hard disk
- append blobs: designed to be append-only, not overwritten/deleted, making it suitable for log data, for example
- Table storage is meant for structured NoSQL data
- Queue storage is designed to buffer message delivery within and across various services
- File storage enables legacy applications to gain access to shared storage using the standard Server Message Block (SMB) protocol
Azure Storage offers several data redundancy options: locally redundant storage (multiple copies of your data in the same facility), zone-redundant storage (redundancy across multiple facilities), and geo-redundant storage (replication across more than one region).
Unpack your Azure Storage metrics
Datadog’s Azure Storage integration surfaces dozens of metrics that you can graph, alert on, or visualize in an out-of-the-box dashboard, including:
- end-to-end latency of successful requests
- percentage of requests that were successful
- request throughput
- error rates (with individual metrics for throttling errors, authorization errors, timeouts, etc.)
- bytes received/sent (ingress/egress)
In addition to Azure Storage, Datadog also integrates with other Microsoft cloud services—including Azure Redis Cache, Azure App Service, and Azure SQL Databases—to give you a comprehensive view of your entire Azure infrastructure.
Storage resource tags
If you are using Azure Resource Manager (ARM) to organize and manage your services, Datadog will automatically pull in resource metadata so that you can aggregate and filter your metrics using those same logical groups.
The integration automatically pulls in the following tags for all of your Azure Storage resources:
storage_type (blob, table, etc.),
subscription_name (e.g. pay as you go). It will also pick up any custom tags you may have added to your Azure resources through the Azure portal.
You can use these tags to filter and customize your dashboard to zoom in on specific subcategories of your Azure Storage resources. For example, in the screenshot below, we filtered the dashboard to only display
table storage metrics from the
test-vms resource group.
Tags are also instrumental for creating targeted alerts that can help you surface insights and identify trends across specific subsets of your Azure resources.
For example, you could create an alert to get notified when an important resource group starts to experience an increased rate of throttling errors. If this issue is not temporary (e.g. someone is conducting load tests), you may want to look into optimizing your queries and/or repartitioning data so that you can distribute transactions across more than one partition. The service documentation contains more guidelines for troubleshooting issues like this once you’re alerted to them.
Monitor Azure Storage (and more)
If you’re already using Datadog and you’d like to start monitoring Azure Storage, first set up the main Azure integration, if you haven’t done so already. Then turn on the integration for Azure Storage and enter your account name and the read-only Shared Access Signature (SAS) token that was generated by this script.
If you’re not currently using Datadog, but would like to bring more observability to your Azure cloud infrastructure and services, get started with a free 14-day trial.