Intro: Linkerd

on YouTube ↗37m

" summary

Linkerd presents a CNCF service mesh with a Rust data plane proxy and Go control plane that is already in production for over 24 months. It ships mTLS on by default, exposes aggregated metrics via Prometheus (temporary retention ~6 hours), and 2.3 adds telemetry, retries, timeouts and auto‑inject while 2.4 targets traffic shifting. Non‑HTTP traffic is proxied as TCP streams, installation is sidecar injection per service for incremental rollout, and development cadence is weekly edge and ~six‑week stable releases. service profiles opt‑in

AI-written summary · about

" chapters (14)

1. 0:00Introduction
2. 1:38What is Linkerd
3. 3:58Why use Linkerd
4. 7:03Linkerd history
5. 8:53Design goals
6. 12:03Data plane
7. 14:01Roadmap
8. 15:51CoopCon EU
9. 20:10QA
10. 21:15Twitter
11. 22:19Co vs Linkerd
12. 23:43Focus on Operational Simplicity
13. 28:31Prometheus
14. 31:44Linkerd

" related talks

• 35mNavigating the Service Mesh EcosystemGeorge Miranda, Diogenes Rittori · KubeCon + CloudNativeCon China 2018
• 04mInterviewDennis Adjei Baah · Scale By The Bay 2018
• 30mService Mesh Is Still Hard. The Things we all Did Wrong & Tales of WoeLin Sun, William Morgan · ServiceMeshCon Europe 2020
• 34mService Mess to Service MeshRob Richardson · DevOpsDays Boston 2020
• 27mService Mess to Service MeshRob Richardson · apidays Paris 2021
• 33mService Meshes: Istio, Linkerd - or No Mesh at All?Eberhard Wolff, Hanna Prinz · GOTOpia Chicago 2021
• 35mUpgrading Your Service Mesh to Linkerd 2Tilen Faganel · KubeCon + CloudNativeCon China 2019
• 28mHow a Service Mesh Helped Us Build Production Cloud-Native ServicesIsrael Sotomayor · KubeCon + CloudNativeCon Europe 2018