vinzenz/hyperhive
vinzenz/hyperhive
Watch
0
0
Fork
You've already forked hyperhive
 0
Code Activity

fix passive-voice vale errors in docs + hive-otel module-eval test for the new journald receiver

Browse source
This commit is contained in:
damocles 2026-10-13 11:30:27 +02:00 • committed by mara
commit 8e58ad8717
4 changed files with 22 additions and 16 deletions
Download patch file Download diff file Expand all files Collapse all files

6
docs/agent-lifecycle/config-changes.md
View file

@ -364,10 +364,10 @@ enough to drive an immediate claude turn (delivered as a regular
`system` inbox message, same `recv` path as any other message). It `system` inbox message, same `recv` path as any other message). It
currently has one variant, `NeedsUpdate { agent }` (sub-agent's currently has one variant, `NeedsUpdate { agent }` (sub-agent's
recorded flake rev is stale, operator rebuilds from the dashboard) — recorded flake rev is stale, operator rebuilds from the dashboard) —
and nothing in the tree constructs it; the push path and nothing in the tree constructs it; this removal dropped the push path
(`Coordinator::notify_agent`) was removed once its last caller (`Coordinator::notify_agent`) once its last caller
(`notify_manager`, used only for the crash event below) went away. (`notify_manager`, used only for the crash event below) went away.
`HelperEvent` is kept as the type for whatever comes along that `HelperEvent` stays as the type for whatever comes along that
genuinely needs to interrupt an agent's turn loop, not because genuinely needs to interrupt an agent's turn loop, not because
anything live uses it today. anything live uses it today.

11
docs/scheduler/observability.md
View file

@ -327,17 +327,15 @@ Cadence follows `metricIntervalMs` (default 60s). Transport is OTLP/HTTP
first hop is unauthenticated for every producer on this host, and the upstream first hop is unauthenticated for every producer on this host, and the upstream
credential stays on the swarm tier. credential stays on the swarm tier.
<!-- vale write-good.Passive = YES -->
## Container-crash alerting (hive-c0re) ## Container-crash alerting (hive-c0re)
The crash watcher (`crash_watch.rs`, a 10s poll of every managed container) The crash watcher (`crash_watch.rs`, a 10s poll of every managed container)
classifies a container that stopped without an operator-initiated action as classifies a container that stopped without an operator-initiated action as
a crash and `tracing::warn!`s `"container crash detected"` — nothing else: a crash and `tracing::warn!`s `"container crash detected"` — nothing else:
no broker event, no per-hive dashboard banner. Both of those used to exist; no broker event, no per-hive dashboard banner. Both of those used to exist;
they were removed in favour of this. this change removed them in favour of the log line.
When `otel.enable` is set, the hive tier's own collector runs a `journald` When `otel.enable` is on, the hive tier's own collector runs a `journald`
receiver scoped to `hive-c0re.service` (no bind mount needed — unlike an receiver scoped to `hive-c0re.service` (no bind mount needed — unlike an
agent's journal, this collector already shares a filesystem with the unit it agent's journal, this collector already shares a filesystem with the unit it
reads, being a host systemd service rather than a container) and forwards reads, being a host systemd service rather than a container) and forwards
@ -351,8 +349,9 @@ Requires `services.journald.storage = "persistent"` (the NixOS default) on
every hive host with `otel.enable` set — asserted at eval time. every hive host with `otel.enable` set — asserted at eval time.
⚠️ **A hive with `otel.enable` off has no operator-visible crash signal at ⚠️ **A hive with `otel.enable` off has no operator-visible crash signal at
all** — the per-hive dashboard banner this replaces is gone, and there is no all** — the per-hive dashboard banner this replaces no longer exists, and
fallback path. Turning on `otel.enable` is what makes a crash reach anyone. there is no fallback path. Turning on `otel.enable` is what makes a crash
reach anyone.
## Agent-emitted per-turn metrics (`hive-agent`) ## Agent-emitted per-turn metrics (`hive-agent`)

6
docs/turn-loop/mcp.md
View file

@ -66,9 +66,9 @@ is opt-in via the P3RM1SS10NS tab.
declared, but no call site constructs it) as regular inbox messages declared, but no call site constructs it) as regular inbox messages
(same `recv` path; body is a JSON object with an `event` discriminant (same `recv` path; body is a JSON object with an `event` discriminant
field). A container crash used to be the other variant field). A container crash used to be the other variant
(`ContainerCrash`, routed to `ruth`) but that path is gone — a crash (`ContainerCrash`, routed to `ruth`) but this tree no longer has that
now only produces a log line + swarm-level grafana alert, no agent path — a crash now only produces a log line + swarm-level grafana
turn at all (`docs/scheduler/observability.md`). Lifecycle alert, no agent turn at all (`docs/scheduler/observability.md`). Lifecycle
transitions the job-queue scheduler or crash watcher drive directly — transitions the job-queue scheduler or crash watcher drive directly —
stop/kill, destroy, a flake-rev login or logout state change — reach stop/kill, destroy, a flake-rev login or logout state change — reach
no individual agent: they publish onto a swarm-wide NATS no individual agent: they publish onto a swarm-wide NATS

15
nix/module-eval/hive-otel.nix
View file

@ -80,14 +80,21 @@ let
# The tier in the middle. Its OTLP receiver takes both signals on one # The tier in the middle. Its OTLP receiver takes both signals on one
# port, so without this pipeline an agent's push is answered 404 on # port, so without this pipeline an agent's push is answered 404 on
# `/v1/logs` — and a forwarder retrying into a 404 is indistinguish- # `/v1/logs` — and a forwarder retrying into a 404 is indistinguish-
# able from one with nothing to send. Compared against the metrics # able from one with nothing to send. `journald` joined the receiver
# pipeline's exporters rather than a name, so the two signals cannot # list for this collector's own host journal (hive-c0re's crash-watch
# drift to different destinations. # log line, see ./otel.nix) — both have to be present, in either
# order, so this stays a set check rather than a list-equality one.
# Compared against the metrics pipeline's exporters rather than a
# name, so the two signals cannot drift to different destinations.
name = "the hive collector forwards logs upstream, not only metrics"; name = "the hive collector forwards logs upstream, not only metrics";
ok = ok =
(hiveOtelPipelines ? logs) (hiveOtelPipelines ? logs)
&& hiveOtelPipelines.metrics.exporters != [ ] && hiveOtelPipelines.metrics.exporters != [ ]
&& hiveOtelPipelines.logs.receivers == [ "otlp" ] &&
lib.sort builtins.lessThan hiveOtelPipelines.logs.receivers == [
"journald"
"otlp"
]
&& hiveOtelPipelines.logs.exporters == hiveOtelPipelines.metrics.exporters; && hiveOtelPipelines.logs.exporters == hiveOtelPipelines.metrics.exporters;
} }
{ {