Troubleshoot common issues in Azure Cosmos DB's API for MongoDB
APPLIES TO:
MongoDB
The following article describes common errors and solutions for deployments using the Azure Cosmos DB for MongoDB.
Note
Azure Cosmos DB does not host the MongoDB engine. It provides an implementation of the MongoDB wire protocol version 4.0, 3.6, and legacy support for wire protocol version 3.2. Therefore, some of these errors are only found in Azure Cosmos DB's API for MongoDB.
Common errors and solutions
| Code | Error | Description | Solution |
|---|---|---|---|
| 2 | BadValue | One common cause is that an index path corresponding to the specified order-by item is excluded or the order by query doesn't have a corresponding composite index that it can be served from. The query requests a sort on a field that isn't indexed. | Create a matching index (or composite index) for the sort query being attempted. |
| 2 | Transaction isn't active | The multi-document transaction surpassed the fixed 5-second time limit. | Retry the multi-document transaction or limit the scope of operations within the multi-document transaction to make it complete within the 5-second time limit. |
| 9 | FailedToParse | Indicates that the Cosmos DB server was unable to interpret or process a parameter because the provided input did not conform to the expected or supported format. | Ensure only valid and supported parameters are included in your queries. |
| 13 | Unauthorized | The request lacks the permissions to complete. | Ensure you're using the correct keys. |
| 26 | NamespaceNotFound | The database or collection being referenced in the query can't be found. | Ensure your database/collection name precisely matches the name in your query. |
| 50 | ExceededTimeLimit | The request has exceeded the timeout of 60 seconds of execution. | There can be many causes for this error. One of the causes is when the currently allocated request units capacity isn't sufficient to complete the request. This can be solved by increasing the request units of that collection or database. In other cases, this error can be worked-around by splitting a large request into smaller ones. Retrying a write operation that has received this error may result in a duplicate write. If you're trying to delete large amounts of data without impacting RUs: - Consider using TTL (Based on Timestamp): Expire data with Azure Cosmos DB's API for MongoDB - Use Cursor/Batch size to perform the delete. You can fetch a single document at a time and delete it through a loop. This will help you slowly delete data without impacting your production application. |
| 61 | ShardKeyNotFound | The document in your request didn't contain the collection's shard key (Azure Cosmos DB partition key). | Ensure the collection's shard key is being used in the request. |
| 66 | ImmutableField | The request is attempting to change an immutable field | "_id" fields are immutable. Ensure that your request doesn't attempt to update that field or the shard key field. |
| 67 | CannotCreateIndex | The request to create an index can't be completed. | Up to 500 single field indexes can be created in a container. Up to eight fields can be included in a compound index (compound indexes are supported in version 3.6+). |
| 112 | WriteConflict | The multi-document transaction failed due to a conflicting multi-document transaction | Retry the multi-document transaction until it succeeds. |
| 115 | CommandNotSupported | The request attempted isn't supported. | Other details should be provided in the error. If this functionality is important for your deployments, create a support ticket in the Azure portal, and the Azure Cosmos DB team will get back to you. |
| 11000 | DuplicateKey | The shard key (Azure Cosmos DB partition key) of the document you're inserting already exists in the collection or a unique index field constraint has been violated. | Use the update() function to update an existing document. If the unique index field constraint has been violated, insert or update the document with a field value that doesn't exist in the shard/partition yet. Another option would be to use a field containing a combination of the ID and shard key fields. |
| 16500 | TooManyRequests | The total number of request units consumed is more than the provisioned request-unit rate for the collection and has been throttled. | Consider scaling the throughput assigned to a container or a set of containers from the Azure portal or you can retry the operation. If you enable SSR (server-side retry), Azure Cosmos DB automatically retries the requests that fail due to this error. |
| 16501 | ExceededMemoryLimit | As a multi-tenant service, the operation has gone over the client's memory allotment. This is only applicable to Azure Cosmos DB for MongoDB version 3.2. | Reduce the scope of the operation through more restrictive query criteria or contact support from the Azure portal. Example: db.getCollection('users').aggregate([{$match: {name: "Andy"}}, {$sort: {age: -1}}])) |
| 40324 | Unrecognized pipeline stage name. | The stage name in your aggregation pipeline request wasn't recognized. | Ensure that all aggregation pipeline names are valid in your request. |
| - | MongoDB wire version issues | The older versions of MongoDB drivers are unable to detect the Azure Cosmos DB account's name in the connection strings. | Append appName=@accountName@ at the end of your connection string, where accountName is your Azure Cosmos DB account name. |
| - | MongoDB client networking issues (such as socket or endOfStream exceptions) | The network request has failed. This is often caused by an inactive TCP connection that the MongoDB client is attempting to use. MongoDB drivers often utilize connection pooling, which results in a random connection chosen from the pool being used for a request. Inactive connections typically time out on the Azure Cosmos DB end after four minutes. | You can either retry these failed requests in your application code, change your MongoDB client (driver) settings to teardown inactive TCP connections before the four-minute timeout window, or configure your OS keepalive settings to maintain the TCP connections in an active state.To avoid connectivity messages, you may want to change the connection string to set maxConnectionIdleTime to 1-2 minutes.- Mongo driver: configure maxIdleTimeMS=120000 - Node.JS: configure socketTimeoutMS=120000, autoReconnect = true, keepAlive = true, keepAliveInitialDelay = 3 minutes |
| - | Mongo Shell not working in the Azure portal | When user is trying to open a Mongo shell, nothing happens and the tab stays blank. | Check Firewall. Firewall isn't supported with the Mongo shell in the Azure portal. - Install Mongo shell on the local computer within the firewall rules - Use legacy Mongo shell |
| - | Unable to connect with connection string | The connection string has changed when upgrading from 3.2 -> 3.6 | When using Azure Cosmos DB's API for MongoDB accounts, the 3.6 version of accounts has the endpoint in the format *.mongo.cosmos.azure.com whereas the 3.2 version of accounts has the endpoint in the format *.documents.azure.com. |
Next steps
- Learn how to use Studio 3T with Azure Cosmos DB's API for MongoDB.
- Learn how to use Robo 3T with Azure Cosmos DB's API for MongoDB.
- Explore MongoDB samples with Azure Cosmos DB's API for MongoDB.