Azure Database for PostgreSQL Flexible Server stuck in "Restarting" for 4 days while PostgreSQL remains accessible

Sachin Wickramasinghe 20 Reputation points
2026-08-04T05:05:43.26+00:00

I'm experiencing an unusual issue with Azure Database for PostgreSQL Flexible Server and would appreciate any guidance or if anyone has experienced something similar.

Environment

  • Service: Azure Database for PostgreSQL – Flexible Server
  • Region: Canada Central
  • PostgreSQL Version: 14.23
  • SKU: General Purpose (Standard_D2s_v3)
  • High Availability: Disabled
  • Public Network Access: Disabled (VNet integrated)
  • Built-in PgBouncer: Enabled
  • Storage: 32 GB (Auto-grow disabled)

Issue

I manually restarted the PostgreSQL Flexible Server last Friday. Since then, the server has remained in the Restarting state in both the Azure Portal and Azure CLI for 4 days.

Azure Portal:

State: Restarting

Azure CLI:

az postgres flexible-server show  --resource-group <resource-group>  --name <server-name>

Output:

State: Restarting

Current Behavior

Although Azure reports that the server is still Restarting, the PostgreSQL engine appears to be functioning normally.

  • ✅ Applications can successfully connect directly to PostgreSQL on port 5432.
  • ✅ SQL queries execute successfully.
  • ✅ Existing workloads continue to function when connecting directly to PostgreSQL.
  • ❌ The Azure Portal still reports the server as Restarting.
  • ❌ Azure CLI also reports the server state as Restarting.
  • ❌ Built-in PgBouncer (port 6432) is not available.

This makes it appear that the database engine is healthy, but the Azure management plane (or managed PgBouncer service) is stuck in an incorrect state.

Troubleshooting Performed

  • Verified storage utilization is only ~13.7%, so storage exhaustion is not the issue.
  • Verified there are no obvious long-running management operations besides the original restart.
  • Confirmed PostgreSQL is accepting connections and processing queries on port 5432.
  • Confirmed Azure Portal and Azure CLI continue to report the server state as Restarting after 4 days.

Questions

  1. Could this indicate that the Azure Resource Manager/control plane is out of sync with the actual database state?
  2. Has anyone seen the built-in PgBouncer fail while the PostgreSQL engine continues to function?
  3. Is there any customer-side action that can resolve this, or does it require Microsoft backend intervention?
  4. Is there any way to force Azure to refresh the server state or recover the managed PgBouncer service without affecting the running PostgreSQL instance?
Azure Database for PostgreSQL
Answer accepted by question author
Allan Solomon Mejia 8,500 Reputation points
2026-08-04T18:56:06.5766667+00:00

Hello @Sachin Wickramasinghe

Based on the symptoms you've described, this appears more like a platform/control-plane issue than a PostgreSQL engine issue.

If the Flexible Server is stuck in an intermediate state (for example, Starting, Updating, or another long-running operation), customer-initiated operations such as start, stop, restart, or scaling may continue to fail until the backend operation is cleared.

A few things worth checking:

  • Review the Activity Log for any failed or long-running management operations.
  • Verify whether there are any Azure Service Health advisories affecting your region.
  • If management operations consistently fail with internal or "server busy" errors, avoid repeatedly retrying them, as they typically won't resolve a stuck control-plane operation.
  • If the issue persists, open a Microsoft Support request and include the subscription ID, server name, region, timestamps, Activity Log entries, correlation/Tracking IDs, and any CLI or portal error messages. These details allow the engineering team to investigate the backend state.

Similar cases on Microsoft Q&A have required backend intervention from the Azure PostgreSQL service team to clear the stuck operation when the issue originated in the platform rather than the customer configuration.

If you can share the exact error message, the current server state, and any correlation ID, the community may be able to help determine whether this is a configuration issue or one that requires Microsoft support.

Please "Accept the Answer" if this information helped you. This will help us and others in the community as well.

Was this answer helpful?

1 person found this answer helpful.

0 additional answers

Sort by: Newest
Most helpful Newest Oldest

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.