Azure Managed Redis stuck in Failed/ScalingFailed after B3 to B5 scale fails with AllocationFailed (East US), how to clear state and get B5 capacity?

Question

Azure Managed Redis stuck in Failed/ScalingFailed after B3 to B5 scale fails with AllocationFailed (East US), how to clear state and get B5 capacity?

Alican 5

Hoping for some help with an Azure Managed Redis (Redis Enterprise) instance - Balanced tier, HA + zone-redundant (2 nodes), running in East US.

I'm trying to scale it up from Balanced_B3 to Balanced_B5 with no downtime (HA is on). Every attempt goes into Updating/Scaling for a couple of minutes and then fails with:

AllocationFailed: Request failed due to insufficient capacity. Retry using a different Azure Managed Redis size, region or contact Azure support for assistance.

I've tried it both from the CLI and the Portal Scale blade, and retried many times over the last 4+ days, including weekend and off-peak hours, hoping some capacity would free up. Same result every time.

The bigger problem now is that after these failed attempts the resource is stuck showing provisioningState = Failed / resourceState = ScalingFailed. The cache itself is completely fine - clients read and write with no issues - but the Scale option in the Portal is greyed out, so I can't retry scaling from there at all.

It's been sitting like this for hours (7+ untouched) and hasn't cleared on its own.

Why it's time-sensitive: this is a production backend for a live game, and the instance holds live-season player state. Memory is around 91% and climbing ~0.11 GB/day, so it's on track to hit the eviction ceiling in about 4 days. I need headroom before then.

Two questions:

How do I clear the Failed / ScalingFailed provisioning state so the Portal scale option works again? Is that something a support engineer has to reset on the backend, or can I do it myself?
Is there a known Balanced_B5 capacity shortage in East US right now, and any rough ETA? If it won't clear soon, what's the closest region with B5 available — or is there a recommended way to add capacity without downtime mid-event?

I can share the exact subscription and resource IDs privately if needed. Thanks!

SAI JAGADEESH KUDIPUDI 3,640 Reputation points Microsoft External Staff Moderator

2026-07-27T14:11:46.76+00:00

Hi @Alican ,
Please carefully follow my instructions below to create a new Service and subscription limits (quotas) request to allow access to deploy Azure Database for MySQL Flexible Server in your preferred region.
1. Navigate to this link to start the process:

https://portal.azure.com/#create/Microsoft.Support
Alican 5 Reputation points

2026-07-27T15:16:02.23+00:00

Thanks. I think this was pasted from another thread; it references Azure Database for MySQL Flexible Server and stops at step 1. I'm not deploying anything new and this isn't a quota cap, so a quota request doesn't look like the right vehicle - please correct me if it is. Additionally, I do have a Developer support plan for this subscription.

My case is Azure Managed Redis (Redis Enterprise),

Balanced_B3 → Balanced_B5, East US, failing with AllocationFailed: insufficient capacity.

Used memory is at 93% of a measured ~4.80 GB ceiling and rising, so I'm days from eviction.
SAI JAGADEESH KUDIPUDI 3,640 Reputation points Microsoft External Staff Moderator

2026-07-29T14:41:39.7566667+00:00

Hi @Alican ,
Could you please share the requested details in a private message so we can take a closer look and assist you further?

2 answers

Your answer

SAI JAGADEESH KUDIPUDI 3,640 Reputation points Microsoft External Staff Moderator

2026-07-27T14:11:46.76+00:00

Hi @Alican ,
Please carefully follow my instructions below to create a new Service and subscription limits (quotas) request to allow access to deploy Azure Database for MySQL Flexible Server in your preferred region.
1. Navigate to this link to start the process:

https://portal.azure.com/#create/Microsoft.Support
Alican 5 Reputation points

2026-07-27T15:16:02.23+00:00

Thanks. I think this was pasted from another thread; it references Azure Database for MySQL Flexible Server and stops at step 1. I'm not deploying anything new and this isn't a quota cap, so a quota request doesn't look like the right vehicle - please correct me if it is. Additionally, I do have a Developer support plan for this subscription.

My case is Azure Managed Redis (Redis Enterprise),

Balanced_B3 → Balanced_B5, East US, failing with AllocationFailed: insufficient capacity.

Used memory is at 93% of a measured ~4.80 GB ceiling and rising, so I'm days from eviction.
SAI JAGADEESH KUDIPUDI 3,640 Reputation points Microsoft External Staff Moderator

2026-07-29T14:41:39.7566667+00:00

Hi @Alican ,
Could you please share the requested details in a private message so we can take a closer look and assist you further?

Answer 1

Hello Alican,

Welcome to the Microsoft Q&A and thank you for posting your questions here.

I understand that your Azure Managed Redis stuck in Failed/ScalingFailed after B3 to B5 scale fails with AllocationFailed (East US), how to clear state and get B5 capacity.

The scale request from Balanced_B3 to Balanced_B5 failed because Azure could not allocate the required Managed Redis infrastructure in East US, resulting in an AllocationFailed provisioning error. The existing B3 cache continues to serve traffic normally because the scale operation never completed. - https://learn.microsoft.com/en-us/azure/redis/how-to-scale

Your questions require different answers:

There is currently no documented customer-facing operation in Azure Portal, Azure CLI, PowerShell, or ARM that manually clears a Managed Redis resource stuck in: provisioningState = Failed and resourceState = ScalingFailed following an AllocationFailed event. If the Portal Scale blade remains disabled and no further management operations can be initiated, Microsoft Support must investigate the backend provisioning record and resource provider state. This is a platform-level condition and not something that can be reset from the customer side. - https://docs.azure.cn/en-us/redis/monitor-troubleshoot-faq, https://learn.microsoft.com/en-us/azure/redis/how-to-scale
To obtain additional capacity before memory exhaustion. Waiting for capacity is not a reliable mitigation, the immediate production-safe actions are followings:
- Attempt scaling to another supported SKU that provides more capacity and is currently available.
- Evaluate a move to Memory Optimized tier if it satisfies the workload requirements.
- Reduce memory pressure by removing stale keys and shortening TTL values.
- Deploy a larger Azure Managed Redis instance in another region and perform a controlled migration if regional capacity remains unavailable.
- Open an Azure Support case immediately to determine whether East US capacity is blocked for the target SKU and whether backend intervention is required.

I hope this is helpful. Please! Do not hesitate to let me know if you have any other questions, steps or clarifications.

Please do not close the thread by upvoting and accepting the answer if any part of it is helpful.

Answer 2

Your B3 cache is serving traffic, but the B5 resize cannot allocate capacity, and the failed control-plane operation has left Scale disabled. Microsoft documents that HA caches remain available during successful scaling, with possible brief connection blips; AllocationFailed means the target infrastructure was not allocated.

Check Activity log and az redisenterprise show for the latest error and correlation ID, then stop retries. There is no documented customer command to clear ScalingFailed. Open an Azure Managed Redis support request with the resource ID, timestamp, target SKU, and correlation ID, requesting state recovery and B5 capacity confirmation.

Microsoft does not publish live regional SKU capacity or ETAs, so support must confirm East US or alternatives. Do not delete the working cache. Meanwhile, expire or delete nonessential keys and verify eviction and TTL behavior. Moving regions requires a planned migration to another cache; it is not an in-place, zero-downtime switch.