How To Prevent DynamoDb exception ProvisionedThroughputExceededException in Laravel - laravel

Question

I am updating a certain number of records in a long period of time, and I have no certainty about the timing in which the records will be produced. Sometimes, when many records are produced at the same time, I get an Error Log Entry saying that I hit the ProvisionedThroughputExceededException.
I'd like to prevent this exception to happen, or at least be able to catch the exception (and then re-throw it so that I don't alterate the logic) but all I get is the error log below:
[2019-02-12 15:50:48] local.ERROR: Error executing "UpdateItem" on "https://dynamodb.eu-central-1.amazonaws.com"; AWS HTTP error: Client error: `POST https://dynamodb.eu-central-1.amazonaws.com` resulted in a `400 Bad Request` response:
The Log continues and we can find a little more detail:
ProvisionedThroughputExceededException (client): The level of configured provisioned throughput for the table was exceeded. Consider increasing your provisioning level with the UpdateTable API. -
{
"__type": "com.amazonaws.dynamodb.v20120810#ProvisionedThroughputExceededException",
"message": "The level of configured provisioned throughput for the table was exceeded. Consider increasing your provisioning level with the UpdateTable API."
}
{"exception":"[object] (Aws\\DynamoDb\\Exception\\DynamoDbException(code: 0): Error executing \"UpdateItem\" on \"https://dynamodb.eu-central-1.amazonaws.com\"; AWS HTTP error: Client error: `POST https://dynamodb.eu-central-1.amazonaws.com` resulted in a `400 Bad Request` response:
So, the exception was thrown, but it looks like it's already caught, while I'd love to catch it myself, even only to keep track of it, and possibly to avoid the exception at all.
Is there a way to do so ?

Answer 1

To prevent the Exception, the obvious answer would be "Use Autoscaling on DynamoDb capacity". And that's what I did, with a certain degree of luck: when a spike in the request arose, I still had the exception, but in average, autoscaling worked pretty well. Here is the CloudFormation snipped for autoscaling:
MyTableWriteScaling:
Type: AWS::ApplicationAutoScaling::ScalableTarget
Properties:
MaxCapacity: 250
MinCapacity: 5
ResourceId: !Join ["/", ["table", !Ref myTable ]]
ScalableDimension: "dynamodb:table:WriteCapacityUnits"
ServiceNamespace: "dynamodb"
RoleARN: {"Fn::GetAtt": ["DynamoDbScalingRole", "Arn"]}
WriteScalingPolicy:
Type: AWS::ApplicationAutoScaling::ScalingPolicy
Properties:
PolicyName: !Join ['-', [!Ref 'AWS::StackName', 'MyTable', 'Write', 'Scaling', 'Policy']]
PolicyType: TargetTrackingScaling
ScalingTargetId: !Ref MyTableWriteScaling
ScalableDimension: dynamodb:table:WriteCapacityUnits
ServiceNamespace: dynamodb
TargetTrackingScalingPolicyConfiguration:
PredefinedMetricSpecification:
PredefinedMetricType: DynamoDBWriteCapacityUtilization
ScaleInCooldown: 1
ScaleOutCooldown: 1
TargetValue: 60
DependsOn : MyTableWriteScaling
That said, I still had the Exception. I knew that the throttled requests would eventually be written, but I was looking for a way to prevent the exception, since I could not catch it.
The way to do it was introduced by Amazon on November 28 and it is DynamoDB on demand.
Quite usefully, in the announcement we read:
DynamoDB on-demand is useful if your application traffic is difficult to predict and control, your workload has large spikes of short duration, or if your average table utilization is well below the peak.
Configuring on-demand in CloudFormation couldn't be easier:
HotelStay:
Type: AWS::DynamoDB::Table
Properties:
BillingMode: PAY_PER_REQUEST
...
Changing the BillingMode and removing the ProvisionedThroughput prevented this kind of exception to be thrown, they're just gone forever.