b68b74d687
A new `swarm-builder` service, in its own container on the host that sets
`deploy.builder.enable` (defaulted on by `allSwarmServices` where the host
has the builder's store leaf, like the review classifier).
When an `agent-configs/<agent>` PR is opened, reopened or synchronized,
swarm-controller queues a `build_agent` jobq node. The node sets a
`swarm-builder` commit status to pending on the PR head, sends
`{agent, rev, pr}` to the builder over NATS request-reply
(`$SWARM.builder.build`), and writes success (store path), failure (last
line nix printed) or error (no builder / no reply) on the head. The node
fails when the build does, so the Jobs page matches the PR.
The builder builds nix/builder/check-flake's `check` with every input
overridden: hyperhive and nixpkgs to the host's own
`c0re.hyperhiveFlake`/`c0re.nixpkgsFlake` (what a hive there renders into
its meta flake), and the agent to the PR head on the forge. Evaluation is
pure and runs in the container; builds go to the host nix daemon as an
untrusted client. The private config repo is fetched with the token of a
new read-only `builder` forge account (in the `hives` team), which the
controller mints into `swarm/services/builder/forge/token`; git gets it
from a credential helper reading the child's environment, so it is in no
argument or URL.
Credentials: a `builder` store leaf minted by glue-bao-tls on the store
host, a `swarm-builder` reader policy (forge token + its own OIDC client
secret), an authelia machine client, and a `--builder-client` grant in the
queue's auth-callout responder. The controller gains a grant on the token
path and the secret publisher a deny on it.
This is the agent-config build check atlas scoped on #5081 (slice 3):
nothing is pushed to a cache, nothing is written to KV, and a red status
gates no deploy. The check composes agent-base + the config without the
values a hive adds (CA, docs, claude-code path, ports, input follows), so
it can disagree with a hive's own build.
31 lines
1.5 KiB
Markdown
31 lines
1.5 KiB
Markdown
# swarm-builder
|
|
|
|
The swarm builder's build service, in the `swarm-builder` container on the
|
|
host that sets `deploy.builder.enable`. The design is
|
|
[`docs/swarm/builder.md`](../docs/swarm/builder.md).
|
|
|
|
swarm-controller sends one request per config PR head: an agent, a commit and
|
|
the PR. This service builds that agent config on the agent base this host was
|
|
deployed with, through the host's nix daemon, and answers with the system
|
|
closure's store path or the tail of what `nix` printed. The controller turns
|
|
the answer into a `swarm-builder` commit status on the PR's head.
|
|
|
|
## What a build is
|
|
|
|
- `nix build` of `nix/builder/check-flake`'s `check` output, with its three
|
|
inputs overridden: `hyperhive` and `nixpkgs` to this host's own sources, and
|
|
`agent` to `git+https://<forge>/agent-configs/<agent>?ref=refs/pull/<pr>/head&rev=<rev>`.
|
|
- The evaluation runs here, pure; the builds run in the host daemon's sandbox
|
|
as an untrusted client, so a config can neither set daemon options nor read
|
|
this service's environment.
|
|
- Nothing is pushed and nothing is recorded: the closure stays in the host's
|
|
store until it is collected.
|
|
|
|
## Credentials
|
|
|
|
The container holds one credential: its own store certificate, which nspawn
|
|
hands in at start without a copy on disk. With it the service reads its
|
|
queue client secret at start and its read-only forge token at every build.
|
|
`git` gets the token through a credential helper reading the child's
|
|
environment, never in a URL or an argument. Its store role reads exactly
|
|
those two paths.
|