ReadMeFeed / Community knowledge
Real questions. Useful conversations. Find the people who know your stack.
A decision guide for PM2 zero-downtime deployments: compare restart, reload, and gracefulReload, understand cluster-mode constraints, and validate connection draining under load.
Learn how to use Fiber's ShutdownWithTimeout to drain in-flight requests during deployments, avoid dropped connections, and handle common pitfalls like deadlocks and context leaks.
Stop locking production tables for schema changes. Learn how to use Oracle DBMS_REDEFINITION to add columns or change compression online while keeping your application active.
Stop using slow DELETE/INSERT cycles for large partitions. Learn how Oracle Partition Exchange allows sub‑second data swaps using metadata‑only operations.
Learn how PM2's master process orchestrates graceful worker replacement, the role of SIGINT/SIGKILL, and why wait_ready matters for reliable deployments.
Discover how Scalingo’s rolling‑update deployment keeps services online, how to verify it, and the trade‑offs you should watch for.
PM2 cluster mode can keep your Node.js API online while you reload workers one at a time. Learn how to enable it, test zero‑downtime reloads, and understand the trade‑offs for long‑lived connections.
Changing a mapping often means building a new index. Point your application at an alias, backfill with the reindex API, then flip the alias atomically—without dropping traffic.
Zero‑downtime deployments on Discloud let you update Discord bots without a blink. Learn how the feature works, step‑by‑step deployment, a concrete example, trade‑offs, and how to verify the deployment in practice.
pm2
fiber
nodejs
scalingo
elasticsearch
docker