Cosmos DB Multi-Region Sync Enabled – Azure Search Indexer Failures Observed, Requesting Sync Disablement or Resource Scale-Up

Nivetha J 40 Reputation points • Microsoft External Staff

We scaled up Cosmos DB to 1.2 million RU/s and enabled Multi-Region Read/Write sync. However, the synchronization operation has been running for almost 18 hours and is still not completed.

Due to this ongoing background synchronization process, a few services connected to Cosmos DB are currently experiencing downtime/performance impact. In addition, we are unable to perform further scale-up operations on the Cosmos account while this process is in progress.

At this point, we need immediate assistance on one of the following options:

  1. Increase/scale up the Cosmos DB resources further from the backend to help the synchronization complete faster.

OR

  1. Stop/disable the ongoing multi-region synchronization process so that we can stabilize the environment and proceed with the required scaling operations manually. We scaled up Cosmos DB to 1.2 million RU/s and enabled Multi-Region Read/Write sync. However, the synchronization operation has been running for almost 18 hours and is still not completed. Due to this ongoing background synchronization process, a few services connected to Cosmos DB are currently experiencing downtime/performance impact. In addition, we are unable to perform further scale-up operations on the Cosmos account while this process is in progress. At this point, we need immediate assistance on one of the following options:
    1. Increase/scale up the Cosmos DB resources further from the backend to help the synchronization complete faster.
    OR
    1. Stop/disable the ongoing multi-region synchronization process so that we can stabilize the environment and proceed with the required scaling operations manually.
0 comments No comments
Sign in to comment

Answer accepted by question author

Amira Bedhiafi 42,941 Reputation points • MVP • Volunteer Moderator

Hello Nivetha !

Thank you for posting on MS Learn Q&A.

This needs to be handled as an urgent Azure support case.

When a Cosmos DB account is being updated for global replication or multi-region config, Azure may need to fully replicate and commit the data before the region or write config is considered complete. During this type of operation, other management operations such as throughput scaling can be blocked or paused until the region operation finishes.

Since the op has already been running for around 18 hours and is now causing service downtime or performance impact, I would recommend opening or escalating a severity A support case immediately with Azure Cosmos DB support.

0 comments No comments
Sign in to comment

Answer accepted by question author

Pilladi Padma Sai Manisha 10,190 Reputation points • Microsoft External Staff • Moderator

Hi Nivetha J,
Thankyou for reaching microsoft Q&A!
Thank you for your patience while we investigated this issue.

The behavior you experienced was due to the internal multi-region synchronization process triggered after enabling multi-region read/write (active-active) on the Cosmos DB account. During this process, temporary backend metadata throttling led to Azure Search indexer failures and service impact. [https://ou...essageItem | Outlook]

We engaged the Azure Cosmos DB Product Group (PG) team, and they have now successfully completed the synchronization and applied the necessary backend interventions.

At this point:

The synchronization process has been completed successfully.

Backend throttling has been mitigated.

Dependent services, including Azure Search indexers, should now function as expected.

You should be able to proceed with normal operations, including any required scaling activities.

Please monitor your workloads and let us know if you observe any further issues—we’re happy to assist

0 comments No comments
Sign in to comment

0 additional answers

Sign in to answer

Your answer