Azure app service resiliency - How to handle appservice when site is down

Question

Azure app service resiliency - How to handle appservice when site is down

madu243 0

I created App service and connect Azure front door to it. I have some doubts regarding when application is down. How to handle resiliency of app when application is down. Also how to handle disaster recovery in when app is down

Locked Question. You can vote on whether it's helpful, but you can't add comments or replies or follow the question.

Christos Panagiotidis 3,546 Reputation points

2026-07-29T16:10:58.14+00:00

Front Door can route around an unhealthy origin, but it cannot recover an App Service application when that app is the only origin. For instance-level resiliency, run at least two App Service instances, enable Health Check against a meaningful endpoint, and use zone redundancy where the selected region and plan support it. For regional disaster recovery, deploy the same application to a second region and place both apps in one Front Door origin group. Configure health probes, use different priorities for active-passive failover or equal priorities for active-active routing, and validate the behavior during a planned outage. Replicate every dependency separately, including the database, storage, Key Vault secrets, configuration, and deployment artifacts; otherwise the web tier can fail over while the application still fails. Automate both deployments and document the required recovery time and data-loss tolerance. A single App Service origin behind Front Door does not provide regional disaster recovery

Christos Panagiotidis 3,546 Reputation points

2026-07-29T16:10:58.14+00:00

Front Door can route around an unhealthy origin, but it cannot recover an App Service application when that app is the only origin. For instance-level resiliency, run at least two App Service instances, enable Health Check against a meaningful endpoint, and use zone redundancy where the selected region and plan support it. For regional disaster recovery, deploy the same application to a second region and place both apps in one Front Door origin group. Configure health probes, use different priorities for active-passive failover or equal priorities for active-active routing, and validate the behavior during a planned outage. Replicate every dependency separately, including the database, storage, Key Vault secrets, configuration, and deployment artifacts; otherwise the web tier can fail over while the application still fails. Automate both deployments and document the required recovery time and data-loss tolerance. A single App Service origin behind Front Door does not provide regional disaster recovery