From the CLI
Every deploy action is an aegis deploy-* command, run in the project
directory. Overseer and CI run the same commands.
First deployment
# 1. Create a project with ingress
aegis init my-app --components ingress
cd my-app
# 2. Run it locally
make serve
# 3. Set up deployment
aegis deploy-init --host your-server-ip
aegis deploy-setup
# 4. Deploy
aegis deploy
Or replace step 3 with aegis deploy-provision to make the server
(Cloud accounts and keys).
aegis deploy-setup runs once on a fresh server. It verifies SSH (adding
the host key if needed) and runs the project's scripts/server-setup.sh there,
which installs Docker, the Compose plugin, uv and build tools, opens ports
22, 80 and 443 in the firewall, sets up Docker log rotation, creates the
application directory, applies system tuning (file limits, TCP BBR), and under
4 GB of RAM adds a swapfile unless the box has swap. --public-key also
installs a key in the deploy user's authorized_keys, so you can skip
ssh-copy-id. The script comes with the ingress component: aegis add ingress
if it is missing.
Redeploying
aegis deploy # back up, sync, build, start, health check
aegis deploy --no-build # config-only change: reuse the built images
aegis deploy --rolling # no downtime
What each step does: How it works.
Monitoring
aegis deploy-status # docker compose ps on the server
aegis deploy-logs # follow every service's logs
aegis deploy-logs -s webserver # one service, with --no-follow to stop
aegis deploy-shell # a shell in the webserver (-s redis, -s traefik)
aegis deploy-restart # docker compose restart
aegis deploy-stop # docker compose down
Configuration
aegis deploy-init writes .aegis/deploy.yml:
| Field | Description |
|---|---|
server.host |
Server IP address or hostname |
server.user |
SSH user (default: root) |
server.path |
Application directory on the server (default: /opt/{project-name}) |
docker.context |
Docker context name (used by the generated Makefile's deploy-* targets) |
The backup and health_check settings are on
Backups and restore. Every field is optional beyond
server.host; a missing one takes its default.
Git-ignore deploy config
Add .aegis/ to your .gitignore to keep deployment config out of version control. deploy-init will remind you if it's missing.
Every command
| Command | Does |
|---|---|
deploy-init |
Write .aegis/deploy.yml |
deploy-setup |
Prepare a fresh server |
deploy-provision |
Make a server in your cloud account, and deploy |
deploy-destroy |
Remove the server it made |
deploy |
Deploy or redeploy |
deploy-backup |
Back up the server now |
deploy-backups |
List the server's backups |
deploy-rollback |
Restore a backup |
deploy-restore-drill |
Rehearse a restore of the off-site copy |
deploy-upload-db, deploy-upload-storage |
Send this machine's database, or the stored files the server lacks |
deploy-cd-setup |
Wire GitHub Actions deploys |
deploy-cd-env |
Push the production env for them |
deploy-agent |
Run what Overseer asks for |
deploy-status, deploy-logs, deploy-shell, deploy-exec |
Look at the server |
deploy-restart, deploy-stop |
Restart or stop its services |
ingress-enable |
Turn on HTTPS |