Azure Service Health

Azure Service Health is a suite of experiences that provide a personalized view of the health of Azure services, regions, and resources you rely on. The business problem it solves is simple: in a cloud environment, you depend on Microsoft to keep the underlying infrastructure running. But outages happen—network failures, regional disasters, software bugs. Without a monitoring tool, you might learn about an outage from your users or customers, leading to downtime, lost revenue, and reputational damage. Service Health gives you proactive notifications, detailed incident reports, and root cause analyses, enabling you to respond faster and communicate effectively with stakeholders.

Azure Service Health is actually three integrated services:

- Service Health (personalized view): Available in the Azure portal, it shows only the services and regions you are actually using. It includes: - Service Issues: Problems that are affecting you now. - Planned Maintenance: Upcoming events that may impact your resources. - Health Advisories: Changes that require your action (e.g., service retirements, feature deprecations).

Azure Service Health is free for all Azure customers. There is no pricing tier—every subscription gets access to the same features. However, to receive alerts, you need to configure Service Health alerts using Azure Monitor (which may incur costs for certain actions like SMS or voice calls).

Key components: - Service Health Dashboard: The central place in the portal to view active events. - Health Alerts: You can create alerts based on service health events (service issues, planned maintenance, health advisories, security advisories). These alerts can trigger email, SMS, webhook, ITSM, or Azure Functions. - Health History: A 90-day history of events for auditing and compliance. - Resource Health History: Individual resource health history for the last 90 days.

In an on-premises data center, you would typically have a monitoring solution like System Center Operations Manager (SCOM) to monitor servers, network, and applications. You would set up alerts for hardware failures, disk space, etc. Azure Service Health is the cloud equivalent, but with a twist: you can't monitor Microsoft's infrastructure directly—you rely on Microsoft's communication. Instead of monitoring the physical hardware, you monitor the service's health from your perspective. The on-premises equivalent would be a combination of a vendor's health portal (like Dell's hardware status) and your own monitoring tools. Service Health replaces the need to manually check status pages or rely on third-party monitoring for Azure infrastructure.

You can access Service Health from the Azure portal under Help + support > Service Health. From there, you can view the dashboard, create alerts, and review history.

Using Azure CLI, you can list service health events:

az account health check

But more commonly, you would use Azure Monitor CLI to create alerts:

az monitor activity-log alert create --resource-group myResourceGroup --name "ServiceHealthAlert" --condition category=ServiceHealth --action-groups /subscriptions/.../actiongroups/myActionGroup

PowerShell cmdlets:

Get-AzActivityLog -StartTime (Get-Date).AddDays(-7) | Where-Object {$_.Category -eq 'ServiceHealth'}

ARM/Bicep templates can deploy Service Health alert rules. For example, a Bicep snippet:

resource serviceHealthAlert 'Microsoft.Insights/activityLogAlerts@2020-10-01' = {
  name: 'ServiceHealthAlert'
  location: 'Global'
  properties: {
    scopes: [
      subscription().id
    ]
    condition: {
      allOf: [
        {
          field: 'category'
          equals: 'ServiceHealth'
        }
      ]
    }
    actions: {
      actionGroups: [
        {
          actionGroupId: actionGroup.id
        }
      ]
    }
    enabled: true
  }
}

Scenario 1: Retail company during Black Friday

A retail company running an e-commerce platform on Azure needs to ensure high availability. They configure Service Health alerts for any service issues affecting their region and services (App Service, SQL Database). During Black Friday, a planned maintenance event for SQL Database is announced. The team gets a notification 7 days in advance, allowing them to schedule a maintenance window during low traffic hours. Without Service Health, they might have been caught off guard.

Scenario 2: Healthcare provider with compliance requirements

A healthcare provider must document all incidents that could affect patient data availability. They use Service Health's 90-day history to produce reports for auditors. When a regional outage occurs, they receive detailed incident reports from Service Health, including root cause and mitigation steps, which they include in their compliance documentation.

Scenario 3: Startup with limited operations team

A startup with a small team uses Resource Health to monitor their critical VMs. They set up Resource Health alerts to email the team if a VM becomes unavailable. When a VM fails due to underlying host issues, they get an alert within minutes, and the Resource Health dashboard shows 'Unavailable' with a reason. They can quickly redeploy the VM or open a support ticket with the incident ID.

An e-commerce company runs its website on Azure App Service and uses Azure SQL Database for transactions. During the holiday shopping season, they cannot afford downtime. The operations team configures Service Health alerts for all services in their primary region (East US) and a secondary region (West US) for disaster recovery. They set up an action group that sends email to the on-call engineer and posts a message to a Microsoft Teams channel. When a planned maintenance event for SQL Database is announced, the team gets a notification 7 days in advance. They schedule a maintenance window during the lowest traffic period (3 AM) and update their customers via a banner. Without this alert, they might have been surprised during peak hours, causing transaction failures.

A financial services firm must comply with regulatory requirements that mandate documentation of all infrastructure incidents. They use Azure Service Health's health history to automatically download monthly reports of all service issues and planned maintenance events. When a regional outage affects their Azure Kubernetes Service (AKS) cluster, they receive a detailed incident report from Service Health, including root cause analysis and mitigation steps. This report is attached to their incident management system for audit evidence. The team also sets up Resource Health alerts for their critical VMs to get immediate notifications if a VM becomes unavailable, ensuring they can quickly failover to a standby VM.

A small startup with a single administrator uses Azure for all its infrastructure. They have a few VMs and a SQL database. They configure Service Health alerts to send SMS to the admin's phone for any service issues affecting their region. One day, a network issue causes a VM to become unavailable. The admin receives an SMS from Azure Service Health within minutes. They open the Resource Health blade for the VM and see 'Unavailable' with a message that the underlying host has a hardware issue. They quickly redeploy the VM from a snapshot and open a support ticket with the incident ID. Without Service Health, the admin might not have known about the issue until users reported it.