I am using Microsoft.Azure.Cosmos .NET SDK, v3.13.0 to automate management of CosmosDB Containers and am running into an issue migrating some existing CosmosDB containers management using .NET rather than Powershell.
Many of our containers are large and/or high traffic requiring large Autoscale Max RU/s to be provisioned:
When I try and provision over 100k RU/s using the .NET SDK an exception is thrown:
Response status code does not indicate success: BadRequest (400); Substatus: 0; ActivityId: XXXX; Reason: (Message: {"Errors":["The offer should have valid throughput values between 10000 and 100000 inclusive in increments of 1000."]}
ActivityId: XXXX, Request URI: /apps/XXXX/services/XXXX/partitions/85af63d2-5f5c-4aff-9eaf-dd67b8116b0c/replicas/XXXX, RequestStats:
RequestStartTime: 2020-10-01T22:47:46.9903602Z, RequestEndTime: 2020-10-01T22:47:46.9903602Z, Number of regions attempted:1
ResponseTime: 2020-10-01T22:47:46.9903602Z, StoreResult: StorePhysicalAddress: rntbd://XXXX/apps/XXXX/services/XXXX/partitions/XXXX/replicas/XXXX, LSN: 60923, GlobalCommittedLsn: 60923, PartitionKeyRangeId: , IsValid: True, StatusCode: 400, SubStatusCode: 0, RequestCharge: 1.24, ItemLSN: -1, SessionToken: XXXX, UsingLocalLSN: False, TransportException: null, ResourceType: Offer, OperationType: Replace
, SDK: Microsoft.Azure.Documents.Common/2.11.0, Please see CosmosDiagnostics, Windows/10.0.19041 cosmos-netstandard-sdk/3.13.2);
This is clearly an issue in the .NET SDK since Azure portal and Powershell cmdlets allow Autoscale provisioning of RU/s over 100k?
Is there another approach to provision over 100k RU/s with the .NET SDK avoiding this error?