hephaestus/docs/how-to/self-update.md
Erich Blume 626c796e6c
All checks were successful
Build / validate (pull_request) Successful in 5m45s
feat(heph): bake daemon mode/hub/oidc/self-update-interval into the service
`heph daemon start`/`restart` previously hardcoded `hephd --mode local` and
only wired the bare `--self-update` bool — the poll interval and all spoke/hub
sync config (`--hub-url`, `--http-addr`, `--oidc-*`) could not be set on the
managed service without hand-editing the plist/unit (which a later
start/restart would clobber).

Generate the hephd arg vector from a DaemonConfig and add the corresponding
`heph daemon start/restart` flags: --mode, --hub-url, --http-addr,
--oidc-issuer, --oidc-audience, --oidc-client-id, and
--self-update-interval-secs. Regenerating now reads the existing service file
and preserves any flags not passed (start as well as restart), so a bare
invocation never silently drops baked config.

Closes the "pass through --self-update-interval-secs" and "bake hub/spoke
config into the generated service" backlog tasks.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-08 13:25:15 -07:00

2.1 KiB

title modified tags
hephd self-update 2026-06-04
how-to

hephd self-update

hephd can keep itself current: it polls the forge for a newer release and, when one appears, rebuilds and restarts onto it — unattended. It is opt-in and off by default.

Enable it

On the managed service:

heph daemon start --self-update

That generates a launchd/systemd service that runs hephd --self-update and gives it a PATH that can find cargo. Override the 6h poll cadence with --self-update-interval-secs <secs>:

heph daemon start --self-update                              # default: poll every 6h
heph daemon start --self-update --self-update-interval-secs 3600

Both start and restart preserve an already-baked setting (the flag and its interval), so a bare invocation won't silently disable it — pass --self-update again only to turn it on later. To run the daemon directly instead:

hephd --self-update                              # default: poll every 6h
hephd --self-update --self-update-interval-secs 3600

How it works

  1. Each interval, hephd GETs the forge's releases/latest and compares the tag against its own version (the one heph --version reports).
  2. On a newer release it runs cargo install --locked --git <public-https-url> --tag vX.Y.Z for heph/hephd/heph-tui/heph-quickadd. hephaestus is a public repo, so this is an anonymous clone — no credentials.
  3. On a successful install it exits cleanly; the service manager (launchd KeepAlive / systemd Restart=always) brings the new binary up.

A failed poll or build is logged and the daemon keeps running on its current version — self-update never takes the daemon down.

Requirements & notes

  • The Rust toolchain (cargo) must be installed for the service user; the update builds from source.
  • Off by default — nothing happens unless --self-update is passed.
  • The first real cross-version upgrade is observable on the first release cut after enabling it.