* refactor(db): port operations, settings, audit, activity, and misc service metadata to sqlc
Covers the cross-cutting surface: operations board reads, app settings,
notification producers, activity/audit inserts, builds upsert/cancel,
git clone credential resolution, invites, environments, registry auth,
project/service variables, HA resolver, boot sweeps, and admin user ops.
Dynamic filtered lists and Better Auth table probes stay raw by design.
Authored By: TDvorak <[email protected]>
* refactor(db): port apwhy gateway handlers and proxy data plane to sqlc
Final tranche — every static call site now uses generated queries. Raw SQL
remains only where the statement is assembled at runtime (filtered lists,
sparse updates, admin count closure) or probes Better Auth's auth_users
table, which lives outside the local schema.
Authored By: TDvorak <[email protected]>
Git-sourced deploys never materialized a checkout: sourcePath was
build_path resolved against the server cwd, so builds consumed the
backend's own directory. Remote placement could only run
registry-pulled images.
- internal/source: per-deployment checkout (init/fetch/checkout) at
branch or commit; GIT_ASKPASS keeps tokens out of argv, .git/config,
and the build context; dumb-transport and SHA-refusal fallbacks;
build_path validated inside the checkout.
- BuildManager.PackageContext: produces the same tar context Build
ships to the daemon — generated Dockerfile at tar root, .git
excluded. Fixes nixpacks passing the Dockerfile path as content.
- Agent artifact endpoint: GET /api/agents/artifacts/:id streams
build contexts and docker-save tarballs behind agent-token auth;
hex-validated ids, per-deployment cleanup, boot-time purge.
- Agent commands: build_image (docker build -, --build-arg values via
process env, not argv) and load_image (docker load for containr-*
rollback tags no registry serves).
- Deployment flow: remote git builds package context → artifact →
node docker-builds the deployment tag; remote rollbacks of local
tags export via SaveImage → artifact → node docker-loads.
Verified end-to-end: agent-pinned git service built on the node
(build_image → create_container → deployed) and rollback shipped a
docker-save tarball (load_image → deployed). Artifact auth returns
401/404 correctly; workspaces and blobs are removed after deploy.
Authored By: TDvorak <[email protected]>
Remote replicas publish an ephemeral host port; the agent reports the
assigned binding back in the create_container result and the backend
renders a per-service file-provider config into TRAEFIK_DYNAMIC_DIR
pointing Traefik at the node's mesh IP + port. Routes are rewritten on
every remote reconcile and removed on teardown or local placement.
Also: RetireService retires remote leftovers when a service returns to
local placement; remove_container is idempotent agent-side; tombstoned
replicas skip remove dispatch; the domains-stay-local ceiling is gone.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Operator-set node labels (node_agents.tags) gate which nodes a service
may schedule onto: services.placement_tags requires every listed tag.
auto and spread filter eligible agents by tags; tags are re-evaluated
each reconcile so retagging a node re-spreads replicas. Pin + tags and
local + tags rejected; empty array clears the requirement.
Surfaces: --placement-tags on services create/update, nodes update
--tags, MCP args, OpenAPI, node-detail tag editor, service-detail
placement-tags input.
Generated with [Devin](https://devin.ai)
Co-authored-by: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
services.spread distributes replicas across every online, schedulable
agent (replica i → agents[i mod n], deterministic). auto placement now
orders by real headroom — memory utilisation, cpu usage, then container
count — instead of count alone.
Lifecycle (start/stop/restart/delete) and GET /runtime are now driven
by container_instances inventory, so remote replicas are visible and
tear down wherever they run — covers spread, pins, and stale pin
leftovers. Spread and explicit pins are mutually exclusive.
Generated with [Devin](https://devin.ai)
Co-authored-by: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* feat(scheduling): multi-node service placement + remote container lifecycle
services.node_id pins a service to a node agent: explicit id, "auto"
(least-loaded online node), or "local"/empty for the local Docker host.
- DeploymentEngine gains a NodeRunner seam; ReconcileService routes
pinned services to the agent's command queue (create_container with
image/env/command/ports/volumes/labels/resource limits), tracked in
container_instances with deterministic per-replica ids.
- Stop/start/restart/delete fan out to the pinned node — no more leaked
remote containers on service delete.
- Agent create_container is idempotent (rm -f first) and honors restart
policy, labels, memory/cpus, published+ephemeral ports, volumes, env.
- GET /nodes/options exposes id/name/status to authed users; full agent
management stays admin-only.
- CLI services create/update --node, MCP node_id docs, OpenAPI
node_id/node_name + /nodes/options, service-detail Placement section.
Ceilings: remote nodes run registry-pulled images only — git builds,
Traefik domains, and remote sleep detection stay local until
build-on-node and per-node ingress land.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* feat(scheduling): node cordon/drain + schedulable gate
- node_agents.schedulable column; "auto" placement and explicit pins skip
cordoned nodes; already-pinned services keep running (reconcile is
unaffected — cordon is not eviction).
- POST /agents/:id/cordon|uncordon|drain (admin). Drain cordons, removes
every service container via the agent queue, and clears the services'
pins so they redeploy on the local host. Requires the node online —
an offline node can't acknowledge removals.
- CLI nodes cordon|uncordon|drain (drain confirm-gated), MCP
containr_nodes_cordon|uncordon|drain, OpenAPI paths + schedulable
field, node-detail page cordon badge + drain/cordon controls.
- Fix: re-registering an agent no longer resets auto_prune.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* feat(scheduling): remote registry pulls + post-run health check
- Dispatch payload carries the project owner's registry credentials for
the image host; the agent docker-logins for the pull and logs out
after, so private images deploy to remote nodes.
- Completed command payloads get their registry password scrubbed —
secrets don't linger in agent_commands rows.
- Agent verifies the container is still running ~1.5s after docker run;
crash-loops now fail the deployment with the container's last log
lines instead of reporting "running" optimistically.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Tomáš Dvořák