I opened a Sev B support ticket with Microsoft six days ago, but so far I've only received automated responses. Is anyone else having trouble deploying Azure Managed Redis in East US 2?
The portal only reports a generic failure for the Microsoft.Cache/redisEnterprise/write operation:
- Status: Failed
- Error: ResourceOperationFailure
- Message: The resource operation completed with terminal provisioning state 'Failed'.
- Detail: The operation failed.
We already have Managed Redis instances running in East US 2, but new deployments are now failing. We've had to fall back to the older Azure Cache for Redis service. We're aware of capacity problems in West Europe, but this appears to be affecting East US 2 as well. Has anyone found a workaround or confirmed whether this is a regional capacity issue?
3 Answers
We’ve seen the same behavior in West Europe. Deployments fail with only the generic terminal provisioning error, while other regions work. Poland Central has been reliable for our development environments, so switching regions may be the only practical short-term workaround.
Microsoft has acknowledged capacity limitations affecting some Managed Redis Enterprise configurations. They suggested trying a B10 SKU or higher, although that hasn’t worked consistently for everyone. If a larger SKU does deploy successfully, you may be able to scale it down afterward. Using another region is generally more reliable for now, especially since the older Redis service is scheduled to stop accepting new deployments after September 30.
This sounds like a capacity problem rather than a configuration issue. Several teams have reported the same generic provisioning failure across different regions, sometimes succeeding only after many retries. If you contact support, include the deployment timestamp and correlation ID so the service team can trace the failed operation.
The latest failed attempt was at 2026-08-03T14:31:21Z, with correlation ID 3bfef3f2-defa-4989-a485-78f552b0ea64.
The higher-SKU suggestion didn’t work for us either. Our automated tests occasionally succeed in North or West Europe, but only once or twice in a full day, so the capacity issue still appears intermittent.