Shipping / 4 / 4
Feature flags
A shop ships a new checkout flow. When deploy and release are the same act, a bug reaches every user and the way back is a 15-minute redeploy. Feature flags separate the two: the code ships dark, a flag service decides who sees it, the SDK evaluates locally and gets changes over a stream, a kill switch turns it off in two seconds, and owners with expiry dates keep old flags from piling up.
- 13 min
- Intermediate
- 12 steps
- Checkout errors
- 0.4 %
- Time to turn off
- 900 s
- Checkout p99
- 180 ms
- Stale flags
- 0
- request
- reply
- redelivery
- duplicate
- lost
Numbers are a model scenario, not measurements
The real request
CI/CD → Checkout APIRequest
14:00:00 deploy checkout v412 -> 24/24 pods1
14:00:00 includes: new checkout flow (PR 1183)2
14:03:40 rollout complete: 100% of traffic on v4123
-- merging the PR was the release decision4
v412— One build: the new flow and every other change of the week together.new checkout flow— The feature itself is just part of the code.100% of traffic— No step between nobody and everybody.release decision— Who sees the feature is decided by a merge button.
Build it yourselfCheat sheetCompare lessons
Course: 74 lessons
01 Foundations
02 Scaling
03 Data & storage
04 Communication
05 Reliability
06 Architecture
07 Security
08 Shipping
09 Whole systems
10 Business processes
11 Case studies