How to recover a HeyPuter instance

Replace 1 with the instance number (1–4). Port = 410N.

docker ps -a | grep puter1
cd ~/puter/instance1 && docker compose ps
cd ~/puter/instance1 && docker compose logs puter --tail 80
cd ~/puter/instance1 && docker compose up -d puter
# Required after puter recreate — nginx may cache old upstream IP
cd ~/puter/instance1 && docker compose restart nginx
curl -sI http://127.0.0.1:4101/ | head -10
curl -sI https://puter1.cadenceworks.tech | head -10

Common causes

Symptom	Likely cause
DB access denied	.env / config.json credential mismatch
HTTPS 502 after upgrade	nginx not restarted after puter recreate
S3 / asset failures	publicEndpoint or Caddy s3.* hostname mismatch
Crash loop on pull	Unpinned :main / bad tag — pin 26.05.2

Reference: HeyPuter fleet. Tutorial: Getting started.