Executing a Lambda Python Query on a DynamoDB table

Question

I am trying to do something simple.  I need to make sure at the time of registration in my app, that the Username they request is not used in my AWS backend.

To complete this, I have a small table in DynamoDB named: D1UsersLookup.  It has the following attributes:
- string Username - the PK
- string Role - the Sortkey
- string AccountID
Currently, I have 2 records (One day expect 100ks records)

Username	Role	AccountID
runghof		Fan	        1
Elvis		Artist	2

My Lambda Python function attempts to query the table to make sure the table does not have a Username a new user would like to have during a registration process.  If there is no match for my query, then I can add it to this table. Conversely, if there is a match, then I need to inform my client app (.Net-Maui/C# code) that they need to choose a different Username.
Sounds simple, but it hasn't been.

Here's the base of the code, Logging and timing stripped out for now:

import boto3
from boto3 import resource
from boto3.dynamodb.conditions import Key, Attr

def lambda_handler(event, context):
    myUsername = 'Elvis'
    dynamodb = boto3.resource("dynamodb")
    table= dynamodb.Table("D1UsersLookup")
    response = table.query(KeyConditionExpression=Key('Username').eq(myUsername))
    print('Searching for myUsernane ' + myUsername)
    print(response)
    IItems = response ['Items']
    print(IItems)

The complete response string is:
{'Items': [], 'Count': 0, 'ScannedCount': 0, 'ResponseMetadata': {'RequestId': 'QO9KDI66TK3T3LQMC9D8P93PFJVV4KQNSO5AEMVJF66Q9ASUAAJG', 'HTTPStatusCode': 200, 'HTTPHeaders': {'server': 'Server', 'date': 'Wed, 08 May 2024 01:14:11 GMT', 'content-type': 'application/x-amz-json-1.0', 'content-length': '39', 'connection': 'keep-alive', 'x-amzn-requestid': 'QO9KDI66TK3T3LQMC9D8P93PFJVV4KQNSO5AEMVJF66Q9ASUAAJG', 'x-amz-crc32': '3413411624'}, 'RetryAttempts': 0}}

Note: Since there's a record with the Username set to: Elvis, I would expect the Count to be
'Count': 1
Continueing

myUsername = 'bob'
    response = table.query(KeyConditionExpression=Key('Username').eq(myUsername))
    print('Searching for myUsernane ' + myUsername)
    print(response)
    IItems = response ['Items']
    print(IItems)

The response is:
{'Items': [], 'Count': 0, 'ScannedCount': 0, 'ResponseMetadata': {'RequestId': 'IJ7V182G2AIGPQOQV54RGT8VMVVV4KQNSO5AEMVJF66Q9ASUAAJG', 'HTTPStatusCode': 200, 'HTTPHeaders': {'server': 'Server', 'date': 'Wed, 08 May 2024 01:14:11 GMT', 'content-type': 'application/x-amz-json-1.0', 'content-length': '39', 'connection': 'keep-alive', 'x-amzn-requestid': 'IJ7V182G2AIGPQOQV54RGT8VMVVV4KQNSO5AEMVJF66Q9ASUAAJG', 'x-amz-crc32': '3413411624'}, 'RetryAttempts': 0}}

Note: In this case, I expected the count to be 0-zero

Finally, I've chosen query because I've read a scan can get expensive.

I feel I'm missing something basic.

Any help would be appreciated.

Petrus

Steve Kinsman · Accepted Answer

Eventual consistency means consistency is usually achieved within a second.  So if the records have been there a while then no, you won't need to do a strongly-consistent read.

Didier_Durand · Answer

Hi,

Which consistency rule did you apply to your DDB table? By default it is "eventually consistent" which means:

```
When issuing eventually consistent reads to a DynamoDB table or an index, 
the responses may not reflect the results of a recently completed write operation.
 If you repeat your read request after a short time, the response should eventually
 return the more recent item. 
```
as per documentation at https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.ReadConsistency.html

In your use case, you may want to use Strongly Consistent Read.

Best,

Didier

Executing a Lambda Python Query on a DynamoDB table

Relevant content