Gubernator Brings Autoscaling to Docker Without Kubernetes Overhead

·2 views

A new open-source container orchestrator called Gubernator (gbnt) has been developed to provide horizontal autoscaling for Docker-based workloads without requiring Kubernetes. The tool lets teams use standard docker-compose.yml files while gaining automatic scaling based on real-time CPU and NVIDIA GPU utilization metrics. Every 10–15 seconds, Gubernator's autoscaler evaluates container performance via Prometheus and cAdvisor, then calculates desired replica counts using a threshold-based formula with cooldown guards to prevent erratic scaling. Autoscaling policies are declared directly through gbnt.autoscaling.* labels inside existing Compose files, eliminating the need for proprietary manifests. The platform positions itself as a middle ground between the simplicity of Docker Swarm and the governance features of enterprise-grade orchestration systems.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

GoSumo Launches AI-Powered Smart Routing for India's Small Businesses

GoSumo, a customer messaging platform, has introduced an AI-driven smart routing feature aimed at helping small and medium businesses in India manage conversations across WhatsApp, Instagram, SMS, web chat, and email from a single interface. The system uses OpenRouter AI to detect the intent of each incoming message and automatically assigns it to the most suitable available agent based on skill mappings stored in a PostgreSQL database. BullMQ and Redis are used to queue routing jobs and track agent availability in real time, ensuring busy or offline agents are skipped. If no suitable agent is available, the system sends customers an automated reply with an expected response time to maintain service quality. Recent security updates to the platform have also addressed privilege escalation vulnerabilities and password-hashing performance issues, strengthening data protection for business users.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Developer Rebuilds Agentic RAG App on AWS Serverless, Ditching EC2 and a Database

A developer has documented migrating an agentic retrieval-augmented generation (RAG) application from a single AWS EC2 instance to a fully serverless AWS architecture. The original setup ran six Docker Compose containers on a t4g.small instance, loading a 24,775-fragment corpus into RAM on every run and costing around $15 per month even when idle. The new architecture replaces this with a Lambda-based API, S3 Vectors for storage, DynamoDB for conversations and feedback, and Amazon Bedrock for inference using Nova and Titan models. A key decision in the migration was deleting a Postgres database rather than migrating it, treating the corpus as a build artifact instead of a runtime dependency. The project also adopts proper infrastructure as code, eliminating the earlier reliance on bash scripts and manual SSH deployments.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Developer Finds Safari MCP Server Bug That Silently Mints Invalid Tab Tokens

A developer discovered a critical bug in their Safari MCP server where browser tab ownership tokens became invalid after a daemon restart. The flaw stemmed from a cached browser-session epoch in the extension's background worker that was never refreshed from storage, causing newly minted tokens to carry a stale identifier. When the worker eventually cold-started, it discarded all stored records tied to the outdated epoch, instantly invalidating every token created during the restart window. Affected tabs became permanently stranded — commands were refused and the tokens could not be rotated or recovered from the client side. The bug went undetected because six separate safety checks all passed silently, with only the final rehydration path actually comparing the epoch against storage.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

CSS Flexbox vs Grid: A Beginner's Guide to Choosing the Right Layout Tool

CSS offers two primary layout tools — Flexbox and Grid — each designed to solve different design challenges. Flexbox is best suited for one-dimensional layouts, arranging items along a single row or column, making it ideal for navigation bars or button groups. Grid, on the other hand, handles two-dimensional layouts involving both rows and columns, making it a better fit for image galleries, dashboards, or product pages. Understanding the core difference between the two helps developers pick the right tool for each use case. The guidance is aimed at beginners looking to build a stronger foundation in front-end web development.

0 comments Read more at DEV Community