A developer wants the ability to roll back to a previous version of an AWS Lambda function in the event of errors caused by a new deployment. How can the developer achieve this with MINIMAL impact on

Question

Options

• AChange the application to use an alias that points to the current version. Deploy the new version
• BChange the application to use an alias that points to the current version. Deploy the new version
• CDo not make any changes to the application. Deploy the new version of the code. If too many
• DCreate three aliases: new, existing, and router. Point the existing alias to the current version.

• A
  80% (12)
• C
  13% (2)
• D
  7% (1)

Explanation

Using a Lambda alias as an indirection layer is the correct approach because the application always invokes the alias ARN (not a version ARN directly), so a rollback is as simple as updating the alias to point back to the previous version - no application code changes, no redeployment, near-instant recovery for users.

Why the distractors are wrong:

• B appears identical to A as shown (likely a formatting artifact in this question - both describe the alias pattern correctly).
• C is wrong because deploying without an alias means the application is coupled to $LATEST or a hardcoded version ARN. Rolling back requires a full redeployment of old code rather than a quick pointer swap, increasing user impact time.
• D is wrong because a three-alias "router" setup is not a standard Lambda pattern and introduces unnecessary complexity. Lambda does support weighted alias routing natively (shifting a percentage of traffic to a new version), but you don't need three separate aliases to accomplish this.

Memory tip: Think of a Lambda alias like a DNS record - it's a named pointer you control independently of the underlying version. Just as you can update a DNS entry to redirect traffic, you can update an alias to instantly redirect invocations to any published version. "Alias = rollback handle."

No community discussion yet for this question.