Kafka & events · Design under changing requirements
Checkout fails when the order service is down
Checkout calls the order service synchronously and fails with it. How do you redesign the hand-off with Kafka so checkout keeps working?
Monday 10:00 · design review, checkout and order-serviceIllustrative scenario, not a real incident
Checkout returned 503 to 3,100 customers on Saturday evening
When a customer pays, checkout calls POST /orders on the order service and waits for the answer before it shows the confirmation page.
On Saturday the order service was unavailable from 20:10 to 20:35. Every checkout in that time failed, although payment and the rest of the shop were healthy.
What the dashboards and logs show
- Failed checkouts
- 3,100 between 20:10 and 20:35
- Order service
- unavailable for 25 minutes
- Normal load
- about 20 orders a second in the evening
- Kafka
- already in use by other teams, three brokers
Clarify · optional · scored under "Hypothesis before change"
Ask the interviewer
0 of 3 asked
Each question reveals a fact. What you ask before you act is part of your score.
First move · scored under "Mitigate before root cause"
The redesign will take weeks, and the order service could go down again tonight. What do you do first?
Your rubric score, the model answer, the follow-up questions and the related lessons appear once you have answered the last step.