Trying the Spot + Ondemand in Managed Autoscaling configuration, but there is ParamValidation Exception #3963
Labels
emr
response-requested
Waiting on additional information or feedback.
service-api
This issue is caused by the service API, not the SDK implementation.
Describe the bug
Hi, recently i am testing the functionality where i was thinking of using few cores as on-demand and few core nodes as spot, but there seems to be validation exception, when i manually tried the same config in EMR, it is allowing those values.
{
"ComputeLimits": {
"UnitType": "Instances",
"MinimumCapacityUnits": 2,
"MaximumCapacityUnits": 26,
"MaximumOnDemandCapacityUnits": 12,
"MaximumCoreCapacityUnits": 15
}
}
my expectation was:
12 core on-demand, 3 core spot
11 spot task nodes
1 master node
But my boto version gives exception.
Expected Behavior
it shouldn't throw validation exception
Current Behavior
it is throwing
botocore.exceptions.ClientError: An error occurred (ValidationException) when calling the RunJobFlow operation: Invalid configuration for managed scaling. If you select On-Demand market type for core nodes in a cluster with instance groups, MaximumCoreCapacityUnits must be less than or equal to MaximumOnDemandCapacityUnits.
Reproduction Steps
same as description
Possible Solution
same as description
Additional Information/Context
No response
SDK version used
1.18.59
Environment details (OS name and version, etc.)
amazon linux
The text was updated successfully, but these errors were encountered: