Idempotency tokens in DynamoDB - Step-by-Step Execution

Concept Flow - Idempotency tokens

Client sends request with token

↓

Check if token exists in DB?

Yes→Return previous result

No↓

Process request

↓

Store token and result

↓

Return new result

The system checks if the idempotency token was used before. If yes, it returns the old result. If no, it processes and stores the result with the token.

Execution Sample

DynamoDB

1. Client sends request with token 'abc123'
2. System checks DB for 'abc123'
3. Token not found, process request
4. Store 'abc123' with result
5. Return result to client

This example shows how a request with an idempotency token is handled to avoid duplicate processing.

Execution Table

Step	Action	Token Exists?	Process Request?	Store Token?	Return Result
1	Receive request with token 'abc123'	Unknown	No	No	No
2	Check DB for token 'abc123'	No	No	No	No
3	Token not found, process request	No	Yes	No	No
4	Store token 'abc123' with result	No	No	Yes	No
5	Return new result to client	No	No	No	Yes
6	Receive duplicate request with token 'abc123'	Unknown	No	No	No
7	Check DB for token 'abc123'	Yes	No	No	No
8	Token found, skip processing	Yes	No	No	No
9	Return stored result to client	Yes	No	No	Yes

💡 Execution stops after returning result to client, either new or stored.

Variable Tracker

Variable	Start	After Step 2	After Step 4	After Step 7	Final
token_exists_in_db	Unknown	No	Yes	Yes	Yes
request_processed	No	No	Yes	No	No
token_stored	No	No	Yes	Yes	Yes
result_returned	No	No	No	No	Yes

Key Moments - 2 Insights

Why does the system check if the token exists before processing?

To avoid doing the same work twice. If the token exists (see step 7 in execution_table), the system returns the old result instead of processing again.

What happens if the token is not found in the database?

The system processes the request (step 3), stores the token with the result (step 4), and then returns the new result (step 5).

Visual Quiz - 3 Questions

Look at the execution_table, at which step does the system decide to process the request?

A Step 3

B Step 2

C Step 7

D Step 9

💡 Hint

Check the 'Process Request?' column in the execution_table.

According to variable_tracker, what is the value of 'token_exists_in_db' after step 4?

A Unknown

B Yes

C No

D Not applicable

💡 Hint

Look at the 'token_exists_in_db' row and the 'After Step 4' column.

If the client sends the same token again, what does the system do according to the execution_table?

A Processes the request again

B Returns an error

C Returns the stored result without processing

D Deletes the token and processes again

💡 Hint

See steps 7 to 9 in the execution_table.

Concept Snapshot

Idempotency tokens prevent duplicate processing.
Client sends a unique token with each request.
System checks if token exists in DB.
If yes, returns stored result.
If no, processes request and stores token with result.
Ensures safe retries without side effects.

Full Transcript

Idempotency tokens are unique identifiers sent with requests to ensure the same operation is not performed multiple times. When a request arrives, the system checks if the token already exists in the database. If it does, the system returns the previously stored result, avoiding duplicate work. If the token is new, the system processes the request, stores the token and result, and returns the new result. This mechanism helps in safely retrying requests without causing unintended repeated actions.

Practice

1. What is the main purpose of using idempotency tokens in DynamoDB operations?

easy

A. To encrypt data stored in the database

B. To prevent duplicate writes when the same request is retried

C. To speed up query performance by caching results

D. To automatically backup data at regular intervals

Solution

1. Step 1: Understand idempotency token role

   Idempotency tokens are unique values sent with requests to ensure the same operation is not applied multiple times.

2. Step 2: Identify the main benefit in DynamoDB

   This prevents duplicate writes if a request is retried due to network issues or timeouts.

3. Final Answer:

   To prevent duplicate writes when the same request is retried -> Option B

4. Quick Check:

   Idempotency tokens prevent duplicates = C [OK]

Hint: Idempotency tokens stop repeated writes on retries [OK]

Common Mistakes:

• Confusing idempotency tokens with encryption keys
• Thinking they improve query speed
• Assuming they trigger backups automatically

2. Which of the following is the correct way to include an idempotency token in a DynamoDB PutItem request using AWS SDK?

easy

A. "IdempotencyToken": "unique-token-value" inside the item attributes

B. Adding IdempotencyToken as a reserved keyword in the table schema

C. Setting IdempotencyToken as a parameter outside the item attributes

D. Using a conditional expression with attribute_not_exists(IdempotencyToken)

Solution

1. Step 1: Understand how idempotency tokens work in DynamoDB

   They are used with conditional expressions to ensure the item is only created if the token does not exist.

2. Step 2: Identify correct syntax for conditional check

   The expression attribute_not_exists(IdempotencyToken) prevents overwriting if the token is already present.

3. Final Answer:

   Using a conditional expression with attribute_not_exists(IdempotencyToken) -> Option D

4. Quick Check:

   Conditional check with attribute_not_exists = B [OK]

Hint: Use conditional expression attribute_not_exists for idempotency [OK]

Common Mistakes:

• Placing token inside item without condition causes duplicates
• Trying to set token outside request parameters incorrectly
• Assuming token is a reserved schema keyword

3. Given the following DynamoDB PutItem request with idempotency token condition:

{
  "TableName": "Orders",
  "Item": {"OrderId": {"S": "123"}, "IdempotencyToken": {"S": "abc-123"}},
  "ConditionExpression": "attribute_not_exists(IdempotencyToken)"
}

What happens if this request is sent twice with the same token?

medium

A. The first request succeeds; the second fails with a conditional check error

B. Both requests succeed and create duplicate items

C. Both requests fail due to token conflict

D. The second request overwrites the first item silently

Solution

1. Step 1: Analyze the condition expression effect

   The condition attribute_not_exists(IdempotencyToken) allows the write only if the token is not present.

2. Step 2: Predict behavior on repeated requests with same token

   The first request writes the item and token. The second request fails because the token now exists, triggering a conditional check failure.

3. Final Answer:

   The first request succeeds; the second fails with a conditional check error -> Option A

4. Quick Check:

   First write OK, second fails condition = D [OK]

Hint: Same token blocks second write with condition failure [OK]

Common Mistakes:

• Assuming both writes succeed causing duplicates
• Thinking second write silently overwrites
• Believing both writes fail

4. You wrote this DynamoDB PutItem request with idempotency token:

{
  "TableName": "Payments",
  "Item": {"PaymentId": {"S": "p001"}, "IdempotencyToken": {"S": "token-xyz"}},
  "ConditionExpression": "attribute_not_exists(IdempotencyToken)"
}

But you get a ConditionalCheckFailedException on the first request. What is the likely cause?

medium

A. An item with the same IdempotencyToken already exists in the table

B. The IdempotencyToken attribute is missing from the item

C. The table does not have a primary key defined

D. The ConditionExpression syntax is invalid

Solution

1. Step 1: Understand the meaning of ConditionalCheckFailedException

   This error means the condition attribute_not_exists(IdempotencyToken) was false, so the token attribute already exists.

2. Step 2: Identify why the first request fails

   If the token already exists in an item, the condition prevents overwriting, causing the failure.

3. Final Answer:

   An item with the same IdempotencyToken already exists in the table -> Option A

4. Quick Check:

   Token exists causes condition failure = A [OK]

Hint: Condition fails if token already exists in table [OK]

Common Mistakes:

• Assuming missing token causes this error
• Thinking primary key absence triggers this error
• Believing condition syntax is wrong

5. You want to design a DynamoDB table to support idempotent order creation using tokens. Which approach best ensures no duplicate orders are created when clients retry requests?

hard

A. Use a composite primary key with OrderId and IdempotencyToken and allow overwrites

B. Use OrderId as the primary key and store IdempotencyToken as a unique attribute with a conditional write checking attribute_not_exists(IdempotencyToken)

C. Use IdempotencyToken as the primary key and reject writes if the token exists

D. Do not store tokens; rely on client-side checks to avoid duplicates

Solution

1. Step 1: Consider uniqueness enforcement for idempotency

   Using IdempotencyToken as the primary key ensures each token is unique and prevents duplicates at the database level.

2. Step 2: Evaluate other options

   Use OrderId as the primary key and store IdempotencyToken as a unique attribute with a conditional write checking attribute_not_exists(IdempotencyToken) risks duplicates if tokens are not primary keys. Use a composite primary key with OrderId and IdempotencyToken and allow overwrites allows overwrites, breaking idempotency. Do not store tokens; rely on client-side checks to avoid duplicates relies on unreliable client checks.

3. Final Answer:

   Use IdempotencyToken as the primary key and reject writes if the token exists -> Option C

4. Quick Check:

   Primary key on token ensures uniqueness = A [OK]

Hint: Make token primary key to enforce uniqueness [OK]

Common Mistakes:

• Storing token as non-key attribute without condition
• Allowing overwrites with composite keys
• Relying only on client-side checks