SSSergei Solodinsergeisolod.hashnode.dev·16h ago · 16 min readMy FDCServers VPS Stayed Online While Disk Reads Took 18.7 SecondsI actually liked FDCServers. For quite a while, my VPS worked exactly the way I wanted it to. It handled real production traffic. Nginx behaved normally. The backend behaved normally. I was not watchi00
SSSergei Solodinsergeisolod.hashnode.dev·Sep 1 · 9 min readAVA Hosting Said “No Sharing.” My VPS Reported 32.73% CPU StealA production debugging story about one KVM VPS, Linux steal time, and why a running server can still be starved of CPU. This is not a negative review of AVA Hosting. It is simply a record of what happ00
SSSergei Solodinsergeisolod.hashnode.dev·Aug 31 · 11 min readMy REGXA VPS Had 94% CPU Steal With Zero TrafficHow I separated an application bottleneck from hypervisor contention using mpstat, PSI, localhost tests, and one simple control experiment. I am not writing this as a negative review of REGXA, and I a00
SSSergei Solodinsergeisolod.hashnode.dev·Aug 28 · 22 min readI Replaced Bunny Storage With a Private Media Origin. The Biggest Win Was Not BandwidthFor almost two months, I served a large collection of images, video, and audio through a setup that looked reasonable on paper: Browser ↓ Working server ↓ Nginx ↓ local proxy_cache ↓ cache00
SSSergei Solodinsergeisolod.hashnode.dev·Aug 27 · 20 min readDesigning a Safer WebP→AVIF Pipeline with SSIMULACRA2I thought I had a threshold problem. I actually had a reference problem, followed by an acceptance-logic problem. The image pipeline was already doing something fairly sophisticated: instead of assign00
