Is Azure down right now?

1. 1

   Read the live status above.

   If the portal or Resource Manager fail for us too, it is an Azure-side incident.

2. 2

   Open Service Health in the portal.

   Azure posts most incidents only to affected subscriptions: Azure portal, Service Health, Service issues. It names regions, services and tracking IDs.

3. 3

   Check Resource Health for the failing resource.

   Each resource has a Resource Health blade that says whether Azure sees a platform problem with that specific VM, database or app.

4. 4

   Try the CLI if the portal is the problem.

   Run az account show. If it works, Resource Manager is up and only the portal UI is struggling.

5. 5

   Rule out your own changes.

   A new Azure Policy, RBAC change, firewall rule, private endpoint or expired secret can fail deployments and look exactly like an outage.

• Use the tracking ID. Every Service Health incident has one. Use it in support cases and in your own incident notes.

• Set up Service Health alerts. Azure Monitor can email, text or call you when an incident or planned maintenance affects your subscriptions and regions. Set it up before you need it.

• Hold deployments. If Resource Manager or a regional service is impaired, deployments can fail half-way. Pause pipelines until the incident is mitigated.

• Fail over where you designed for it. Paired regions, availability zones and geo-replicated storage are what keep you up when one region is not.

Global configuration changes

The October 2025 Azure Front Door outage came from a configuration change that safety checks failed to stop before it went global.

Storage and compute dependencies

In July 2024 a configuration change cut Central US compute off from its storage, restarting virtual machines across the region.

Shared edge and identity services

Front Door and Entra ID sit in front of many services, so their incidents spread to Microsoft 365, Xbox and customer apps.

Regional power and network events

Data-centre or network problems usually stay within one region, which is why multi-region designs matter.

The biggest Azure incidents and what caused them. Durations are approximate and come from public reporting and official reviews.

• OCT 29, 2025

  Azure Front Door: latency, timeouts and errors for customers and Microsoft 365 for about 8 hours

  An inadvertent configuration change deployed globally because of a defect in deployment safety systems. Source: Parametrix

• JUL 18, 2024

  Central US: virtual machines and dependent services down overnight

  A configuration change blocked access between Azure Storage clusters and compute resources. Source: Futurum

Every minute our servers request Azure's main endpoints, resolve management.azure.com through Cloudflare, Google and microsoft's own nameservers, and read azure.status.microsoft. We combine that with one-tap user reports, filtered for duplicates and abuse: a spike needs at least 12 reports in 15 minutes from 8 or more different networks.

Built to avoid false alarms: a failed request is retried before it counts, a single bad check never says "down" on its own, and a check that fails for reasons on our side is shown as inconclusive rather than as an outage. Limits: we check from one location, so a regional problem may first show up in user reports. Reports store only the problem type, a country code and anonymous codes that change daily; never your IP address. Protego is independent and not affiliated with Azure. Last reviewed October 26, 2026.