Join us

ContentUpdates and recent posts about Fleet..
Link
@varbear shared a link, 3 weeks, 2 days ago
FAUN.dev()

Rust unit testing: file writing

To test file writes without hitting the disk, the author swaps in a closure that takes a file handle. That handle’s a test double, so after the code runs, you can crack it open and inspect what got written... read more  

Link
@varbear shared a link, 3 weeks, 2 days ago
FAUN.dev()

Full Unicode Search at 50× ICU Speed with AVX‑512

StringZilla v4.5drops a major speed bomb on Unicode text processing. Think10× faster tokenization and case folding. Up to150× faster for case-insensitive substring search. It leaves ICU and PCRE2 wheezing in the dust. Under the hood: SIMD all the way, AVX-512 on newer chips, plus script-aware SIMD k.. read more  

Full Unicode Search at 50× ICU Speed with AVX‑512
Link
@varbear shared a link, 3 weeks, 2 days ago
FAUN.dev()

pqr.sql: Generate QR Codes with Pure SQL in PostgreSQL

A developer jammed out aQR code generator in pure SQL, just PostgreSQL, no extensions or libraries. One gnarly single-statement query. It even runs faster onPostgreSQL 17than on 16, thanks to engine tweaks... read more  

pqr.sql: Generate QR Codes with Pure SQL in PostgreSQL
Link
@varbear shared a link, 3 weeks, 2 days ago
FAUN.dev()

5 engineering dogmas it's time to retire

Dependencies are risky, especially in smaller companies - avoid unnecessary packages to prevent security incidents and maintain code simplicity. Feature flags can become overwhelming if abused, leading to complex codebases and false sense of security - use them wisely. Commenting code is a balance -.. read more  

Link
@kaptain shared a link, 3 weeks, 2 days ago
FAUN.dev()

Dapr Deployment Models

Daprstarted as a humble Kubernetes sidecar. Now? It's a full-blownmulti-mode runtimethat runs wherever you need it,edge,VM, orserverless APIs. Diagrid’sCatalysttakes that further. It wraps Dapr in a fully managed API layer that’s detached from your app’s lifecycle. No infra lock-in, just token-based.. read more  

Dapr Deployment Models
Link
@kaptain shared a link, 3 weeks, 2 days ago
FAUN.dev()

v1.35: Job Managed By Goes GA

In Kubernetes v1.35,spec.jobControllerManagedByhits GA. That means full handoff of Job reconciliation to external controllers is now official. It unlocks tricks likeMultiKueue, where a single management cluster fires off Jobs to multiple worker clusters, without losing sight of what’s running where... read more  

Link
@kaptain shared a link, 3 weeks, 2 days ago
FAUN.dev()

Troubleshooting Cilium network policies: Four common pitfalls

Cilium’s Day 2 playbook covers the real work: dialing inL7 policy controls, tuningHubble observability, and wringing performance fromBPF. It's how you keep big Kubernetes clusters sane. The focus?Multi-tenant isolation,node-to-node encryption, and scaling cleanly withexternal etcdso the network does.. read more  

Link
@kaptain shared a link, 3 weeks, 2 days ago
FAUN.dev()

93% Faster Next.js in (your) Kubernetes

Next.js brings advanced capabilities to developers out-of-the-box, but scaling it in your own environment can be challenging due to uneven load distribution and high latency. Watt addresses these issues by leveragingSO_REUSEPORTin the Linux kernel, resulting in significantly improved performance met.. read more  

Link
@kaptain shared a link, 3 weeks, 2 days ago
FAUN.dev()

1.35: In-Place Pod Resize Graduates to Stable

In-Place Pod Resizehits GA in Kubernetes 1.35. You can now tweak CPU and memory on live pods without restarts. This is finally production-ready! What’s new since beta? It now handlesmemory limit decreases, doesprioritized resizes, and gives you betterobservabilitywith fresh Kubelet metrics and Pod e.. read more  

Link
@kaptain shared a link, 3 weeks, 2 days ago
FAUN.dev()

Kubernetes OptimizationInPlace Pod Resizing,ZoneAware Routin

Halodoc cut EC2 costs and shaved latency by leaning into two Kubernetes tricks: In-place pod resizing(v1.33) lets them dial pod resources up or down on the fly, especially handy during off-peak hours. Zone-aware routingviatopology-aware hintskeeps inter-service traffic close to home (same AZ), skipp.. read more  

Kubernetes OptimizationInPlace Pod Resizing,ZoneAware Routin
Fleet is a high-scale GitOps system built to support the realities of multi-cluster operations. Instead of pushing YAMLs or relying on brittle scripts, Fleet treats Git as the authoritative state and continuously reconciles that state across every cluster under management.

Its architecture uses lightweight agents, bundling, and content distribution to propagate changes efficiently - whether you’re managing five clusters or five thousand. Policies, Helm charts, CRDs, and raw manifests all become versioned, reviewable, and auditable through Git.

Fleet integrates cleanly with Rancher, enabling teams to automate cluster bootstrapping, enforce standards, roll out updates safely, and instantly detect drift. It excels in environments that demand consistency: edge fleets, hybrid cloud estates, regulated sectors, and platform teams building opinionated Kubernetes platforms.