Files
Juan DenisandClaude Opus 5 da2b77f3d2 feat(templates): tell people whether a template actually fits the server
Deploying checked three things — the name is free, the server row exists, the
directory isn't taken. Nothing asked whether the box could carry the thing, so
a 2 GB app landed on a 1 GB VPS at full speed and the first sign of trouble was
the OOM killer.

Every piece was already here and unconnected: templates describe what they cost
(`requirements: memory / storage`), servers report what they have
(Server.total_memory / total_disk) and what they are using (the newest
ServerMetrics sample). This is the join, answering in the shape a person asks
it — does it fit, will it be tight, or can't we tell?

Advisory on purpose. It reports; it never refuses. An estimate is an estimate,
real usage depends on configuration and traffic, so the operator keeps the
final say and "tight" is a sentence to read rather than a wall to argue with.
A full server still returns valid:true, and a test asserts that so this cannot
quietly become a gate.

Three judgements worth stating:

- The reserve is a floor-and-fraction pair (10% of the box, at least 256 MB RAM
  / 2 GB disk). A bare percentage is wrong at both ends: 10% of 1 GB is
  nothing, 10% of 64 GB is more than most apps need.
- `unknown` outranks `ok`. No recent sample, or a template that declares
  nothing, says so — a confident wrong answer is worse than no answer.
- Free memory comes from psutil's `available`, not `free`: cache and buffers
  are reclaimable, and counting them as used calls every healthy Linux box
  full.

Catalog coverage goes from 52/106 to 106/106. The 54 estimates are each app's
typical idle-to-light-use footprint, taken from the project's own documented
minimum where it publishes one and rounded to the values the existing 52
already use — sonarqube 4 GB because that is its documented heap floor, immich
4 GB for the ML models, prometheus 20 GB because retention dominates, traefik
256 MB.

Surfaced two ways: a capacity block alongside (never inside) `errors` on
validate-install, and a read-only GET the deploy drawer can poll while the
operator is still choosing a target, so the answer updates as they switch
servers instead of waiting for a submit.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-07-25 15:48:36 -04:00

74 lines
1.4 KiB
YAML

id: caddy
name: Caddy
version: "2.7"
description: Fast, multi-platform web server with automatic HTTPS
icon: https://serverkit.ai/imgs/template-icons/caddy.svg
website: https://caddyserver.com
documentation: https://caddyserver.com/docs/
categories:
- networking
- proxy
- ssl
variables:
- name: HTTP_PORT
description: HTTP port
type: port
default: "80"
hidden: true
- name: HTTPS_PORT
description: HTTPS port
type: port
default: "443"
hidden: true
ports:
- port: 80
protocol: tcp
description: HTTP
- port: 443
protocol: tcp
description: HTTPS
volumes:
- name: data
path: /data
description: Caddy data
- name: config
path: /config
description: Caddy configuration
- name: site
path: /srv
description: Website files
requirements:
memory: 256MB
storage: 1GB
compose:
services:
caddy:
image: caddy:${VERSION:-2.7-alpine}
container_name: ${APP_NAME}
restart: unless-stopped
volumes:
- caddy_data:/data
- caddy_config:/config
- caddy_site:/srv
ports:
- "${HTTP_PORT}:80"
- "${HTTPS_PORT}:443"
volumes:
caddy_data:
caddy_config:
caddy_site:
post_install: |
# Create default Caddyfile
cat > /var/serverkit/apps/${APP_NAME}/config/Caddyfile << 'EOF'
:80 {
respond "Hello from Caddy!"
}
EOF