vinzenz/hyperhive
vinzenz/hyperhive
Watch
0
0
Fork
You've already forked hyperhive
 0
Code Activity
hyperhive/hive-c0re
History
Repository files (latest commit first)
Filename Latest commit message Latest commit date
atlas 3cb7be29cf hive-c0re: bound the read-only container listing and systemctl status calls 
priv_client::list_containers now waits at most 30s (LIST_WAIT) for
hive-priv's `nixos-container list`, via the existing `within` helper. Its
periodic callers (crash_watch, agent_sockets, swarm_agent_status, wanted,
mcp_sockets, swarm_status via container_view) already treat an Err as a
skipped cycle, so a hung listing no longer stops their loops for good.

lifecycle::is_running / unit_state run `systemctl is-active` with
kill_on_drop and a 10s timeout (IS_ACTIVE_TIMEOUT); a timeout warns and
reads as not-running / UnitState::Other, the same as a systemctl that
could not be spawned. The new lifecycle::running_state returns None for
that case, and crash_watch keeps the agent's previous observation on None
so a hung systemctl is neither reported as a crash nor as a recovery.

`call` itself and every mutating op (start/stop/destroy/send-snapshot/
daemon-reload) stay unbounded.

Closes #5061
2026-10-09 15:46:55 +02:00
..
src hive-c0re: bound the read-only container listing and systemctl status calls 2026-10-09 15:46:55 +02:00
Cargo.toml hive-c0re stops owning scheduled prompts; agents lose the schedule tools 2026-10-04 21:23:30 +02:00
README.md hive-c0re: remove the approvals subsystem 2026-10-04 23:55:32 +02:00

README.md

hive-c0re

The unprivileged host daemon (runs as hive-core). Owns the sqlite broker, the reminder queue, the generic job-DAG queue, container lifecycle, gateway vhost wiring, forge/matrix config reconciliation, per-container stats, and the axum operator dashboard. Largest crate in the workspace — bin-only, no separate lib.

When to use it

Host-level, cross-container orchestration: spawning/rebuilding/ destroying agent containers, dashboard-visible state, wiring each agent's gateway vhost and reconciling its forge config-repo access and matrix room membership. The accounts themselves are swarm-controller's (docs/integrations/forge.md). Agent-side behavior (turn loop, MCP tools) lives in hive-agent/hive-agent-mcp instead — this daemon only talks to agents over the socket wire types in hive-sh4re.

Shape

Cohesive clusters live in directory submodules, each re-exported at the crate root (crate::broker::… keeps resolving regardless of which subdirectory a module actually lives in). One line each — read the module's own //! doc-comment for real detail, don't expect this file to track it:

  • dashboard/ — the operator dashboard (containers, schedules, logs, topology).
  • job_queue/ — the job-DAG queue + desired-state reconciliation (docs/scheduler/coordinator.md).
  • lifecycle/ — nixos-container lifecycle + per-agent config flake generation.
  • stores/ — sqlite-backed stores (broker, queues, audit, power).
  • workers/ — background sweeps (crash watch, autoupdate, knowledge sync).
  • agent_config/ — per-agent registries (tool groups, capabilities, resource limits, topology).
  • stats/ — dashboard metrics aggregation + OTEL export.
  • socket_server/ — the unix-socket request server shared by per-agent + manager sockets.
  • forge/ — optional Forgejo wiring (docs/integrations/forge.md).
  • coordinator.rs — top-level wiring for serve.
  • meta.rs, migrate.rs — the meta flake + schema/state migrations.
  • matrix.rs, gateway_nginx.rs, priv_client.rs — matrix provisioning, gateway vhosts, and the hive-priv client respectively.

See the top-level CLAUDE.md/docs/ index for the full reading-path map.