Ramping Down the CF API V2 Usage: Embracing Improvements and Managin...

on YouTube ↗25m

" summary

A dedicated V2 rate limiter is deployed (rolled out from December with biweekly halving) that returns V2-specific headers and 429 when exceeded; this cuts V2 traffic from 60% to 5%, reduces Cloud Controller VMs by more than 50%, and lands at 1,000 V2 requests per user per hour. An exemption scope via UAA grants temporary relief (applied by a Concourse pipeline, expires after six months). Demo shows V3 remains unaffected and V2 clients must re-authenticate to pick up exemptions.

AI-written summary · about

" chapters (8)

1. 0:00Introduction to V2 API ramp-down
2. 1:25Motivation for V2 API deprecation
3. 4:38Challenges in migration to V3
4. 6:43Rate limiting V2 API strategy
5. 9:48Exemption process for V2 API
6. 12:19Live demo of rate limiting
7. 16:41Summary and future outlook
8. 19:53Q&A session

" related talks

• 15mVelocity CultureJon Jenkins · Velocity 2011
• 06mVelocity 09: Alistair CrollAlistair Croll · Velocity 2009
• 12mCF Application Runtime Demo: Weighted Routing (Istio/Envoy), Multiple Custom App Ports...Dieu Cao · Cloud Foundry Summit North America 2019
• 05mRunning Cloud Foundry at ScaleMark Kropf · Cloud Foundry Summit 2014
• 47mScaling: from 0 to 25 million usersJosip Stuhli · Open Source Monitoring Conference 2024
• 37m73,000 Pods a Day, Lessons From Misadventures In Multi-TenantShane Corbett, Wil Reed · KubeCon + CloudNativeCon North America 2022
• 20mUnlocking the power API Management for developersAly Ibrahim · Commit Your Code 2024
• 55mCVE Overload: Scaling Container Vulnerability ScansLucas Ferreira · OWASP Global AppSec Singapore 2023