exec.rs's Reconcile arm was the one construction site declaring a
resource inline in an executor rather than in templates.rs. It now calls
templates::fanned_out_mechanical, which is where every other declaration
lives -- construction sites state their own holdings.
That also fixes a test which could not fail. The old one claimed a
Reconcile and then re-declared the fan-out itself, commented "same two
calls the scheduler makes, in the same order" -- a copy of production
inside the test. Had exec.rs stopped declaring the lease, it would have
kept passing. The replacement calls the real function and asserts the
declaration, with no DAG run at all.
The other half of the old test -- that a descendant re-enters its
ancestor's grant instead of taking a second unit of a cap-1 lease -- is
hive-jobq's, tested there by
child_borrows_ancestor_grant_released_when_subtree_done and
nested_borrowers_never_deadlock.
perm_change and the graceful rebuild chain walked their whole DAG to
collect node kinds in order; both now assert declared_shape. The
graceful one gets a sharper claim out of it -- signal and drain go
between the build and the stop, and nothing else changes -- which is
what distinguishes it from the non-graceful chain.
reparent_bulk needed the node's payload rather than its wiring, so
payload_of() reads it off the graph. The assertion is unchanged: one
node carries every move, because bulk atomicity is why a single node was
chosen.
resubmit_while_running_is_new_dag no longer claims a node to stage the
"while running" part. submit appends a container and inserts the
declared group; it never consults the state of any existing node, so a
running earlier DAG cannot change the outcome. The property is no dedup,
covered by identical_resubmit_is_a_distinct_dag -- this one keeps the
named scenario because a config bump mid-build is what people actually
worry about.
Two more tests drove DAGs to completion to watch an agent lease free up
-- one when a single agent's subgraph settled inside a still-running
multi-agent DAG, the other when a whole power op finished. Releasing a
grant once its owner's subtree is terminal is hive-jobq's, covered by
owner_holds_grant_for_its_whole_subtree,
child_borrows_ancestor_grant_released_when_subtree_done and
leaf_owner_goes_done_directly_and_releases.
Their host-side halves are declarations asserted elsewhere: that each
agent's subgraph is an independent root holding only its own lease is in
multi_agent_restart_is_one_dag_with_concurrent_per_agent_subgraphs, and
that a power op emits no tail node is in
cancelled_power_op_runs_no_compensating_node, which checks the DAG has
no pending nodes left at all.
Six tests drove two or three DAGs against each other to watch a build
slot or an agent lease serialize them. In every case the part that is
hive-c0re's -- which nodes declare which resource -- is now a
declared_resources table, and the part that is hive-jobq's -- what a
scheduler does with a contended unit -- is tested in hive-jobq
(build_slot_cap_limits_concurrency_and_release_unblocks,
unrelated_nodes_needing_the_same_resource_are_serialized, the two
sibling_borrowers cases, owner_holds_grant_for_its_whole_subtree, and
the fairness test added in the previous commit).
graceful_signal_and_drain_hold_no_build_slot is absorbed rather than
deleted: its claim is a declaration, so it now sits in the graceful-stop
shape test. Signal and Drain declare the agent lease and no build slot,
which is why a whole-hive graceful stop overlaps every agent's drain at
buildSlots = 1 -- the ceiling is one GRACEFUL_STOP_TIMEOUT in total, not
one per agent.
hive-c0re/src/job_queue/tests.rs: 42 tests to 36, 181 lines lighter.
fifo_fairness_for_the_slot lived in hive-c0re and submitted three
rebuilds, driving one to completion to watch the freed build slot go to
the earlier waiter. The guarantee it was checking is this crate's:
claim_one scans nodes in insertion order and takes the first satisfiable
one. Nothing here tested it -- the property hive-jobq provides was
asserted only downstream, through a host's templates.
a_contended_resource_goes_to_the_oldest_waiter tests it directly.
Mutation-checked: reversing the scan order fails it.
What is hive-c0re's is which nodes contend for the slot at all, and that
is a declaration, so its half is now a declared_resources table with
nothing running. The measured shape corrected an assumption on the way:
MetaSync takes the meta window only, and the agent lease starts at
StopForUpdate -- the first node that touches the container -- not at the
head of the chain. Prebuild deliberately holds no lease, which is what
lets it overlap another DAG on the same agent while the container is
still up.
"Uniform hold across the chain" needs no test of its own: a resource is
held for the acquirer's whole subtree, and the parent nesting is already
asserted in rebuild_chain_is_declared_serial.
swap_ok_runs_post_swap_before_reconcile drove a rebuild DAG to observe
that Reconcile waits for PostSwap. That ordering is not a dependency
between them: Reconcile deps AfterAny(Prebuild), and PostSwap sits
inside Prebuild's subtree, so Prebuild cannot satisfy the edge while
PostSwap is outstanding.
Renamed to say what it checks, and it asserts the parent chain plus that
edge instead of running anything. Kept as its own test rather than
folded into the chain table because the indirection is the easy thing to
break -- flattening the chain preserves every edge and still loses the
guarantee.
swap_failure_still_runs_reconcile is deleted. Its cascade claims are
hive-jobq's, and the "says so on the wire" half was nothing: snapshot
fills NodeView { state: node.state, .. }, a straight copy of the same
State type, so there is no host-side mapping that could disagree.
failed_reconcile_marks_dag_failed is deleted too: a one-node DAG whose
node fails, asserting the DAG reads Failed, is
failed_child_rolls_parent_up_to_failed restated through a c0re template.
Reproducing the runtime path is not needed to test what the runtime path
declares. What DeployApply grows is deploy_rebuild_nodes' output, and
that is a pure declaration -- so declare it directly and read the shape,
instead of running a deploy far enough to graft it.
That makes the finalize gate visible without any of the walking:
finalize_deploy declares AfterOk on both prebuild and reconcile, so
either root failing skips it, and reconcile hangs off prebuild with
AfterAny so a failed swap still reaches it.
deploy_dag_skips_finalize_but_still_tails_a_failed_graft is gone with
it. Its three declared claims are rows in that table, and its runtime
claims belong to hive-jobq: cascade on failure, roll-up, and first_error
digging past a group root that rolled up Failed while carrying no error
of its own -- which is why the DAG reports the swap's error rather than
nothing.
The grafting mechanism itself is also hive-jobq's and tested there: work
lands under the emitter before it settles, and the emitter parks in
Finishing so a downstream AfterAny gate stays shut while its new
children run.
deploy_dag_runs_phases_in_order_and_tails_a_failed_apply and
deploy_dag_skips_apply_but_still_runs_tail_when_verify_fails differed
only in where they injected the failure -- apply in one, verify in the
other -- and each drove the whole DAG to watch the compensation tail run
anyway.
Both follow from a single declared edge. The tail accepts
done|failed|skipped on apply, and skipped is exactly the state apply
lands in when verify failed and it never ran. Asserting that edge covers
both cases without running anything.
The runtime halves are hive-jobq's and tested there: a failed dep
cancels its AfterOk dependents while the AfterAny one still runs, and a
parent rolls up Failed from a failed child -- which is what stops an Ok
tail laundering a failed deploy into a success.
Mutation-checked: turning the tail's after_any(apply) into
after_ok(apply) fails the surviving test on that edge alone.
non_graceful_rebuild_has_no_signal_or_drain claimed and completed six
nodes to collect their kinds. Mutation-checked rather than trusted: with
the template's graceful flag flipped, it fails with signal and drain in
the list. An absence assertion is the easiest kind to make vacuous, so
it is the one that most needs the check.
reparent_shape claimed a node only to get an id for its resource
assertion. node_of() finds it by kind and asserts there is exactly one,
so the test cannot quietly start being about a different node.
multi_agent_restart asserted "both subgraphs start concurrently" by
claiming and seeing two. That is not one fact but two declared ones plus
a jobq guarantee: both heads are group roots with no deps, so nothing
orders them; and each declares only its own agent lease, so nothing
makes them contend. Whether a scheduler then runs independent,
resource-disjoint roots at once belongs to hive-jobq and is tested
there. This asserts the two declarations.
ErasedRecipe and erase() existed for one test, which put three power-op
cases in a single array. Three recipe closures have distinct types and
cannot share an array element type, so all three were boxed. The array
was the reason, not the specs.
The per-case assertion is now a helper taking the already-submitted DAG
id, so each case submits its own spec at its own concrete type. Nothing
is erased and nothing is boxed; production never needed either.
The final assertion is also stronger than the one it replaces.
claim_ready().is_empty() asks what is runnable at this instant, which a
node that is alive but blocked on a dependency passes -- exactly what a
leftover compensating node would look like. It now asserts the DAG has
no Pending nodes at all. It was also a mutating call inside an
assertion: claim_ready settles the graph.
Three template tests ran a whole DAG -- claim, complete, repeat -- to
observe an order that is fully determined the moment submit returns.
They now read the graph directly: kinds, parent nesting, and dep edges
with the outcome set each accepts.
rebuild_chain_claims_in_dep_order is renamed, because the old name was
wrong about the mechanism and reading it rather than the graph is how
you stay wrong: only half that chain is dep edges. stop_for_update and
swap declare no deps at all and are ordered by parent nesting -- a
node's sub-nodes run after its own logic. Both axes are asserted now,
since a template can break either independently.
declared_shape spells out each edge's accepted outcomes rather than
bucketing them into ok/any. Bucketing made spawn's three
ResolveApproval tails -- which differ only in accepted outcome -- render
as identical rows, which would have made the assertion a tautology.
Checked by mutation against the code under test rather than the
assertion: dropping .after_ok(signal) from the graceful-stop Drain fails
graceful_stop_shape with a diff naming the one missing edge.
error_is_truncated submitted a DAG, claimed its head, failed it with a
long string and read the error back out of a snapshot -- four moving
parts to observe one `&str -> String`. The DAG round-trip it depended on
is covered by its own tests either way.
Testing truncate_error directly also reaches the case the round-trip
never could: the cap is a byte length, so a multibyte char straddling it
would panic the slice. That boundary scan is the only non-obvious line
in the function and it had no coverage at all -- the old test used a
repeated ASCII 'x', where byte and char offsets coincide.
Also drops a stale claim from insert_job's doc: it has not recorded a
per-node node_rt since that map was deleted.
Two review findings from the previous round, re-checked against the
actual tree rather than against my notes.
`Scheduler::complete` was still a public wrapper whose entire body was
`self.finish(id, outcome)`. Its docstring argued the split was not a
redirect because both completion forms shared `finish` -- but sharing a
private helper is not a reason for two public names. `finish`'s body now
lives in `complete`, and `complete_growing` calls it. Same sharing, one
name, no redirect.
Growth on a failed node is now dropped by `complete_growing` instead of
by the host loop. Failure cancel-cascades to every pending child of the
completing node, and grown work is inserted as its children, so anything
appended here is Skipped by the next statement -- the insert is not
wrong, it is provably pointless. That is a consequence of this crate's
cascade rule, so this crate should be the one enforcing it; a host that
has to remember it can forget it. Behaviour is unchanged: hive-c0re
already dropped growth before calling, and now no longer has to.
`Scheduler::new_job` is left alone but documented for what it is: the
hole in `JobBuilder::new`'s pub(crate) wall, with no non-test caller
since claim_next mints a builder per running node. Closing it is a venue
question rather than a rename, so it stays for now.
c0re's run loop now goes through hive_jobq's claim_next seam, so the
host layer no longer needs its own claim/complete vocabulary.
exec::run_node takes (NodeId, &NodeKind) instead of a &Claim snapshot.
The agent already rides the payload, and the DAG id is a derived read
(JobQueue::dag_of) that only three arms want, so it is taken per-arm
rather than eagerly for every node. Two arms (WritePermFile, Reparent)
re-matched the kind behind a bail! that could never fire; the match arm
already destructures the payload, so they take it directly now.
Deleted from the c0re layer:
- struct Claim
- JobQueue::claim_ready
- JobQueue::complete_node / complete_node_growing
- scheduler::NodeDone / handle_completion
Completion happens inside the future claim_next hands back, so "ran the
node but forgot to complete it" is not expressible on the production
path any more. The node done / node failed logging moved with it -- it
lived in handle_completion but is not dead code.
claim_ready and the completion wrappers were left with no non-test
callers, so the tests carry them as ClaimReady / CompleteNode extension
traits over the crate primitives. JobQueue::new_job stays: run_worker
still mints an empty builder on a failed outcome.
`QueueInner` existed to hold the scheduler *and* a per-node side map. The
map is gone, so it was a struct around one field — and worse, a struct of
a type `hive_jobq` cannot drive: the crate's run-loop seam takes
`&Arc<Mutex<Scheduler<..>>>` specifically.
So `JobQueue` now holds `Arc<Mutex<Sched>>` directly, where `Sched` is
just `Scheduler<NodeKind, Resource>`. Its six methods become free
functions over `&Sched`; all six are `DagView` projections, i.e. the code
the endpoint rework is going to delete anyway, so this does not entrench
them.
This is the precondition for c0re calling `claim_next`, not that switch
itself — `run_worker` still claims through `claim_ready`. Landing it
separately keeps the type change reviewable on its own.
`QueueInner` was `{ sched, node_rt }`, where `node_rt` held exactly one
datum per node: the `build_logs` row id. It existed because a `hive_jobq`
node payload is immutable after insert while the log row is created when
the build starts — so the link could not ride the node.
Invert it: the log row names its node (`build_logs.node_id`, one migration
in the existing `schema_versions` framework). Same single-home property,
in the direction the type system allows.
`QueueInner` is now just the scheduler. That is the point: the queue holds
no per-node side map, so nothing has to be locked alongside the graph.
Deleted as a consequence, each surfaced by dead-code analysis after the
edit above rather than predicted:
- `NodeRuntime`, `node_rt`, `set_build_log_id`, and `build_log_id_of`
(which linear-scanned the map to match a wire `u64` against opaque
`NodeId`s). The lookup is an indexed query now.
- `struct Ctx`, entirely. It carried `coord` + `dag_id` + `node_id` into
the executors so the build-log callback could reach the queue; without
the callback, `coord`/`dag_id` were never read and `node_id` was already
on the `Claim` both executors receive.
- `QueueInner::node_running`, which existed only for `set_build_log_id`'s
"only while running" guard.
- The `Fn(i64)` callbacks on `prebuild_toplevel` / `swap_update` /
`priv_run_inner`, replaced by a `node_id: Option<u64>` passed down. The
id travels one way now instead of being registered back.
`meta.rs`'s `nix_logged` passes `None` deliberately: its callers reach it
from outside the queue as well as inside, and nothing reads the link for
them yet.
`id_for_node` takes `MAX(id)` rather than assuming uniqueness — a retried
node opens a second row and the panel wants the current attempt. The test
moved to where the behaviour lives and covers that, plus survival across
completion and non-collision with node-less rows.
A node no longer hands back a recipe for the scheduler to replay later. It
declares straight onto a builder it was given, and that builder is inserted
as part of completing the node.
Deleted: `pub type Declare`, `struct NodeOutput` (+ its hand-written `Debug`),
`JobQueue::append_subgraph`. Nothing added to `Dag` / `DagView`.
jobq gains `Scheduler::new_job()` (the only way to obtain a `JobBuilder`) and
`complete_growing(id, outcome, grown)`, which inserts under `id` and *then*
completes it, so a DAG cannot roll terminal while grown work is still pending.
`complete()` and `complete_growing()` share a private `finish()` rather than
one redirecting through the other. The DAG-gone guard lives beside the graph
now, where it cannot be skipped, instead of being a caller-side lookup.
The growth executors return data (`run_meta_lock -> (Vec<String>, RebuildOpts)`,
`run_reconcile -> Option<NodeKind>`) rather than taking the builder: a `&Job`
parameter is live for the whole function body, and `&RefCell<T>` is never
`Send`, so an async fn taking one cannot be spawned. `run_node` threads the
builder by value and hands it back.
A node can now declare work and then fail, which was previously inexpressible.
`grown` is dropped in that case — failure cancel-cascades downstream, so
inserting it would only add nodes to immediately cancel — and the log line
carries `grown_nodes` so the drop is visible.
Follow-up from #2872: the comment said 'the rest of the (much larger)
route table below is undocumented for now' -- that was true when the
annotation sweep started, not anymore. Trimmed to state the current
fact plainly instead of narrating the sweep's history (mara: is that
level of detail relevant here) -- this is a plain rustdoc comment on
ApiDoc, not part of the actual OpenAPI JSON (that comes from the
separate #[openapi(info(description = ...))] attribute below it), so
keeping it terse and mechanical is the right call.
Swagger UI's endpoint-list row already shows the HTTP method badge +
path for every row, so restating `METHOD /path` at the start of a
handler's own summary is pure duplication. Strips that self-referential
prefix from every summary that has it and re-capitalizes what follows
as a standalone sentence.
Left two false positives untouched: misc_api.rs's operator-inbox
summary cross-references a *different* sibling endpoint
(mark-all-read) for context, and topology.rs's SetParentForm struct
doc happens to mention its endpoint's path but isn't a handler summary
line. Both are legitimate, not redundant.
utoipa splits a handler's doc comment on the first blank `///` line:
everything before it becomes the OpenAPI `summary` (shown in Swagger
UI's collapsed endpoint-list row), everything after becomes the
`description` (only shown once that row is expanded). With no blank
line, the whole doc comment becomes the summary and the description is
empty — which is what every handler in hive-c0re/src/dashboard/ was
doing, so the all-endpoints list showed full multi-sentence prose next
to every route instead of a short one-liner.
For every `#[utoipa::path(...)]`-annotated handler across the 19 files
in that module:
- Inserted a blank `///` line after the first short sentence/clause so
utoipa's split produces a real summary + description, where the doc
comment had more to say. Left already-short single-clause docs alone
(nothing to split).
- Where a query struct derives `IntoParams`, moved param prose that
duplicated a field's own doc comment out of the handler doc (the
field already documents itself in the generated spec), or added a
field doc where the handler explained a param that had none.
No behavior changes — doc comments and `params()` description text
only. Verified `cargo build -p hive-c0re` (clean) and `nix fmt` (zero
changes) after.
Closes#2969
Swagger UI itself is nginx-hosted now (iris's 86a39c4c), so c0re
carrying its own vendored copy via utoipa-swagger-ui was a straight
duplicate — dropped the dependency (root Cargo.toml + hive-c0re's),
swapped the SwaggerUi::new(...).url(...) mount for a plain
/api/openapi.json GET route serving the same OpenApi doc as JSON.
Verified: cargo build/clippy/test -p hive-c0re clean, Cargo.lock
dropped utoipa-swagger-ui + utoipa-swagger-ui-vendored with no other
changes, nix fmt clean.
hive-c0re doesn't touch these files anymore (nginx hosts them
directly, see the previous commit) — hive-c0re/swagger-ui-theme/ was
a leftover from the original SWAGGER_UI_OVERWRITE_FOLDER build-hook
design, which this PR moved away from. New location matches the
existing top-level-directory-for-nix-packaged-assets convention
(branding/, claude-plugins/).
Pure rename, no content change: nix/packages/swagger-ui-theme.nix's
three file references updated, everything else picks it up from the
same content hash (nix build .#swagger-ui-theme resolves to the exact
same store path as before the move).
Extends the theme-only alias into the full shape mara asked for on the
PR thread:
1. nix/packages/swagger-ui-dist.nix — plain vendored Swagger UI 5.17.14
dist, sourced directly from the swagger-ui-dist npm package (same
release the Rust utoipa-swagger-ui-vendored crate ships, verified
via matching gitHead commit) rather than through Cargo.lock/cargo.
2. nix/packages/swagger-ui-theme.nix — overlays our 3 override files
(index.html, hyperhive-theme.css, and now swagger-initializer.js)
onto (1).
3. vhosts.nix's swaggerUiLocations now prefix-matches the whole
/api/docs/ tree (not just 2 exact-match files) straight from (2),
plus a `= /api/docs` redirect shim since hive-c0re's own redirect
is going away too. /api/openapi.json (outside this prefix) keeps
proxying to c0re unchanged — that's the one thing that stays
dynamic.
New file swagger-initializer.js needed hand-verification: the plain
vendored copy hardcodes the swagger.io petstore demo URL.
utoipa-swagger-ui normally rewrites it per-request from a {{config}}
placeholder its own build.rs injects — since hive-c0re won't be
serving this file at all once its SwaggerUi mount is removed, that
rewrite has to be baked in statically here instead. Derived by
actually running build.rs's own two transforms (strip the default
layout: line, splice the Config JSON in place of the url/deepLinking
block) against the real vendored file, not typed from scratch —
verified byte-for-byte against what format_config() would produce for
hive-c0re's actual single-URL config, and checked with node --check.
Coordinated with damocles: he's taking the corresponding hive-c0re
side (drop the utoipa-swagger-ui dependency + SwaggerUi::new(...)
mount, keep only the plain /api/openapi.json route) once this lands.
Verified: nix fmt clean; nix build .#swagger-ui-theme succeeds, output
byte-matches the checked-in override files and node --check passes on
swagger-initializer.js; a full nixosSystem eval of nixosModules.default
resolves both new locations (/api/docs/ aliased to the right store
path, = /api/docs redirecting) with auth threaded through.
Cut the process-narrative framing (operator quote, backstory) and
shortened per-section labels — the CSS was carrying more explanation
than declaration. Keep only what a future reader actually needs: the
override mechanism, the stylix-follows-live fact, and the two
non-obvious calls (method-colour exception, the one !important).
Loads the dashboard's own themed /static/colors.css directly (same
origin, see vhosts.nix dashboardProxyLocation) instead of redeclaring
the Catppuccin Mocha hex values locally. The theme now re-themes live
with the rest of the dashboard instead of drifting out of sync.
Addresses review feedback on PR.
Not yet wired up -- these are the SWAGGER_UI_OVERWRITE_FOLDER payload
files (utoipa-swagger-ui's build-time overwrite hook), inert until
something points that env var at hive-c0re/swagger-ui-theme/ during
the crate's cargo build. See message to damocles for the nix-side ask.
hive-c0re/swagger-ui-theme/index.html: byte-identical to the vendored
swagger-ui-5.17.14 dist's own index.html (utoipa-swagger-ui-vendored
0.1.2), plus one added <link> to hyperhive-theme.css.
hive-c0re/swagger-ui-theme/hyperhive-theme.css: partial reskin per
mara's steer on hyperhive#2966 ("apply theme colors only", not a full
redesign) -- page background, topbar, borders, links, headings, form
controls recoloured to the Catppuccin Mocha palette already used
elsewhere (frontend/packages/shared/src/colors.css); Swagger's own
semantic HTTP-method badge colours and execute/cancel action-button
colours left untouched, they carry real meaning independent of
hyperhive branding.
Two review findings on the resources-at-construction change.
argus: `workers::auto_update`'s boot sweep constructs nodes through
`templates::node` too, and it was not converted. With the kind-derived
declaration gone, its sweep `MetaLock` and its per-agent `Reconcile`
silently declared no resources at all — so a boot reconcile no longer
held the agent lease and could race another DAG's container ops, and the
sweep's meta commit could land inside another node's staged deploy
window. Nothing failed to compile: removing an implicit behaviour from a
helper is invisible at every call site that relied on it.
The declarations now live in a pure `boot_nodes`, split out of
`submit_boot_tree` so they can be exercised without a `Coordinator`.
That path is the only place job nodes are built outside `job_queue/`,
which is exactly why it had no coverage; `boot_sweep_nodes_declare_
their_own_resources` closes that, asserting against declared graph edges
rather than against the kind.
mara: `templates::node` is a redundant redirect now that it no longer
derives resources — deleted, and its 43 call sites use `Job::node`
directly. The reasoning it documented moved to the module docs of
`templates.rs` and `resource.rs`, which is where it stays true.
Resources were derived from the node's kind: `templates::node` called
`NodeKind::resource_deps()`, which fanned out to `needs_build_slot` /
`needs_lease` / `needs_meta_window`. That made the requirement a property
of the *kind*, so a kind that happened to run under an ancestor already
holding the resource could get away with declaring nothing.
Three did. `Start`, `Stop` and `PostSwap` appear in none of the three
predicates, and that was only safe because one construction site fans
them out from inside a lease-holding `Reconcile` — a fact about today's
DAG shape, not about the nodes.
Each of the 41 construction sites now says what it holds. `Start` /
`Stop` / `PostSwap` declare the agent lease; per the contract that is a
re-entrant borrow, which a new test pins rather than argues.
`running_transients` reads the node's declared deps instead of
re-deriving from the kind. That closes the blank-pill gap: the pill went
blank during container start, stop and the post-swap tail because the
declaration was missing, not because the filter was wrong.
The deleted predicates carried the only written record of three design
decisions; each moved to the `Resource` variant it constrains rather than
dying with its function.
Three findings from the operator's review, all correct.
1. Two insert_job's. Graph::insert_job had no caller outside hive-jobq's
own tests -- production only ever went through Scheduler::insert_job.
It existed because the graph-level one got written first. Deleted; the
tests moved onto a Scheduler, which is where insertion belongs anyway.
2. insert_job was not atomic, and the previous commit made that worse: a
forward edge or forward parent surfaced mid-loop, leaving the nodes
before it in the graph, and resolve_wanted ran after every insert, so
an unknown handle failed once the whole job was already committed.
The module documented this under "Partial insertion" instead of fixing
it -- prose describing a hole is not a design.
All three are decidable from what the builder holds, so
check_declaration_order now runs before the first insert and the loop
indexes ids directly. A malformed job leaves the graph untouched.
What remains mid-insert is the graph's own rejection (out-of-group
dep, empty DepWhen); closing that needs a dry-run validate on Graph,
which is a separate change.
3. DagSpec no longer boxes its recipe: it is generic over the closure,
which travels from the template that built it straight into submit.
The box bought type inference, and paying for it costs annotations --
`|b: &Job|` at each declaration site (the field needs an HRTB, and an
unannotated closure binds one lifetime) and `+ use<>` on each
returning signature (or the opaque type captures the caller's borrows).
Erasure is still needed where several recipe shapes share one type:
the boxed Declare stays for the executor's append_subgraph, and a test
table uses an erase() helper.
The operator's instruction on the issue was "the closure returns an array
of guids, and enqueue_job returns the node ids in that order". What was
here instead returned a HashMap of everything inserted, and no caller used
the keys: submit dropped the return, insert_group did into_values(), and
the scheduler ignored what append_subgraph handed back. The guid-keyed
lookup was dead weight, and into_values() made that Vec arbitrarily
ordered -- harmless only because nothing read it.
insert_job now takes FnOnce(&JobBuilder) -> Vec<NodeGuid> and returns the
matching ids positionally. A handle from another job is UnknownNode rather
than a silent omission: the return is positional, so a short vector would
misalign every id after it.
c0re's Declare stays FnOnce(&Job) and the wrapper names no handles in one
place, rather than ending seven templates in an empty vector -- a DAG is
addressed by its container node, which submit inserts itself. That frees
insert_group from needing every id, so the node_rt pre-seeding goes too:
NodeRuntime is one Option field and every reader already tolerated a
missing entry (entry().or_default(), get().and_then(), iter().find()).
The tests are the argument for the shape: capturing a handle through a
mutable binding to look it up in the map afterwards collapses into
returning it and destructuring the result.
Follows the jobq change: a builder can no longer be constructed or
inserted outside `hive_jobq`, so `DagSpec` cannot hold one. It carries a
`Declare` — `Box<dyn FnOnce(&Job) + Send>` — and the queue runs it
against a builder jobq owns, at the moment it inserts.
`NodeOutput.append_subgraph` becomes `Vec<Declare>` for the same reason,
and this is where the shape was always heading: that field's doc already
said an executor "cannot reach the queue, so it hands the declaration
back", while its type was a `Vec<Job>` the executor had built itself.
The rejected `build_nodes -> Vec<NodeSpec>` was the first version of that
escape hatch; a recipe is the last one, because there is no job-shaped
value to hand over at all.
Templates and the power-op assemblers move their owned data into the
closure and are otherwise unchanged — `rebuild_nodes`, `node` and the
tail helpers already took `&Job` and returned handles, so only each
template's outermost frame moved.
Two `Debug` impls are hand-written: a closure has nothing to show, and
its nodes do not exist until the queue runs it. `NodeOutput` reports how
many subgraphs were emitted, `DagSpec` its source and reason.
`append_subgraph`'s `is_empty()` early-return is gone — you cannot ask a
recipe whether it will declare anything without running it. It now
inserts and returns an empty id list if nothing was declared, which
takes the queue lock in a case that previously skipped it.
The two in-DAG-growth tests build `Declare`s now, so they exercise the
shape an executor actually produces rather than one only a test could
construct. 45 job-queue tests unchanged and passing.
The submit-time petgraph `toposort` this described is gone — a cycle
needs an edge pointing at a node declared later, and a handle only
exists for a node already declared. Say why the validation pass is
absent rather than leaving a description of one that isn't there.
`templates.rs`'s module doc was 35 lines and over the comment-block
lint's max; it now points here for the reasoning instead of restating
it, and drops the power-op paragraph that `submit.rs` already owns.
Every template built a `Vec<NodeSpec>` whose edges and parents were
positional indices into that vector, so a shape was expressed as
arithmetic: `base + 1`, `stop_root + 2`, `sfu + 1`, and a
`reconcile_index()` helper that read the emitted vector's length to find
out where its own last node had landed. `concat_subgraphs` existed
solely to rebase one per-agent subgraph's indices onto another's.
Templates now declare into a `hive_jobq::JobBuilder` and hold the
handles they get back, so an edge names the node it waits on. The
arithmetic is gone, and with it:
- `NodeSpec` and the job-queue's own index-based `Dep`.
- `insert_group`'s index resolution — it wraps `Scheduler::insert_job`.
- `concat_subgraphs` — per-agent chains share one builder and each keeps
its own root, so independence is structural rather than computed.
- `reconcile_index` and `dep_index`.
- `templates::validate` and its petgraph toposort. It rejected dangling
deps and cycles; both are now unrepresentable, since a handle only
exists for an already-declared node and every edge therefore points
backwards. (petgraph stays in the tree for `agent_config::topology`.)
`NodeOutput.append_subgraph` becomes `Vec<Job>`: an executor cannot
reach the queue, so it hands back declarations and the scheduler inserts
them under its own lock. That is what the in-DAG growth path always
wanted — a transferable declaration, not a vector of specs.
Resource declaration is unchanged in behaviour: the `templates::node`
helper applies `NodeKind::resource_deps()` at the construction site, so
every node still declares what its kind needs. Moving that declaration
to the call sites is #2818's job; this leaves it one place to delete.
Three tests went with the guard they covered — they hand-built malformed
specs out of indices, which is the representation that made those shapes
possible. Two more now read a DAG's shape off the queue rather than out
of a spec vector, which is where it is observable. The remaining 45
job-queue tests are unchanged and still pass: lease serialization,
roll-up, cancel-cascade, in-DAG growth and per-agent concurrency all
behave as before.
It never produced a `Dep::Node`, so returning `Vec<Dep<Resource>>` made
every caller match a variant that cannot occur. `running_transients`
paid for it with a two-arm match to pull the agent out of a lease edge.
`Vec<(Resource, u32)>` says the same thing in the type, and is what the
job builder's `.needs_units(name, count)` takes — the insertion path
wraps it back into a `Dep::Resource` at the one place that still speaks
in edges.
`forge_git_url` spliced `core:<token>@` between scheme and authority, and
that URL is a process argument. `/proc/<pid>/cmdline` is mode 0444 —
world-readable — so the core admin token, which provisions every agent's
forge account, was published to any local user for the lifetime of each
git child. Seven call sites built such a URL.
The credential now travels in the environment instead:
`git_command_authed` sets `http.extraHeader` via `GIT_CONFIG_*`, which
git reads exactly like a config file, and `/proc/<pid>/environ` is 0400 —
owner-only. Same credential, materially smaller audience. The remote is a
plain `http://forge/<org>/<repo>.git`, and `forge_git_url` no longer takes
a token, so the old shape cannot be rebuilt by accident.
`knowledge`'s clone was the one place a credentialed URL was stored as a
named remote — git persists the clone URL into `.git/config`, so the
token sat on disk and every later `pull` authenticated from there. That
is the case `forge::repos::push_config` documents as forbidden ("the
tokenised URL ... deliberately never stored as a named remote"). `pull`
now rewrites `origin` to the plain URL first, which also scrubs the
persisted token from existing deployments, and authenticates from the
environment when a token is available. The repo is public, so the pull
still works without one.
Three call sites also stopped spawning `Command::new("git")` directly,
so they honour the `HYPERHIVE_GIT` path the NixOS module bakes in and
the `kill_on_drop` every other git spawn gets.
The two URL-shape tests now assert the *absence* of a credential, and a
new one decodes the header back to `core:<token>` — without that, a
malformed header would leave every forge operation silently anonymous
with the other assertions still green.
forge_admin interpolated its whole argument vector into the error
context, and two callers pass a live operator password in that vector
(user create --password, user change-password). Any failure of those
commands wrote the password to hive-c0re's log in cleartext -- and the
likeliest trigger is forgejo rejecting a weak password, so the secret
got logged precisely because forgejo refused it.
Redacting the value after --password would repeat the bug the issue is
about: redact_password_line matched one keyword and a differently named
secret walked past it. A denylist fails open, silently, and the next
secret-bearing flag would leak until someone extended the list.
describe_forge_admin keeps the leading verb path and stops at the first
flag, so "user create --username iris --password ..." is reported as
"forgejo admin user create". The verbs are a closed set this crate
chooses itself; argument values never are, so a new flag is excluded by
construction. Nothing useful is lost -- the context says which operation
failed, and the underlying error already carries forgejo's own message
about why.
The same pattern in hive-priv is deliberately untouched: that crate runs
as root and the redactor's shape is still an open question on the issue.
This change holds under either answer.
mint_token interpolated forgejo's raw stdout into its anyhow context on
the parse-failure path, and on that call stdout carries the access token
that was just created. The happy path below it is careful to log only
the user and token names; the error path handed the secret over whole.
It fires exactly when forgejo's output format drifts, which is the same
drift that breaks extract_token in the first place -- so the "help me
debug this" context printed the secret it had failed to find.
Report the shape of the output (bytes, lines) instead of its contents.
That is what diagnoses a version drift anyway: you want to know forgejo
printed something with no token-shaped word in it, not the bytes.
Redaction at a logging call site does not cover the error path.
with_context and bail! are output channels too.
Phase 4 (repoint every container onto `meta#<n>`) and phase 5 (rename
the `root` container to `h-root`) were marker-guarded one-shots for
layouts no live hive still has: containers are rendered onto `meta#<n>`
at creation, and the `h-` prefix has been the naming for far longer than
any deployment predates. A one-shot nobody can still trigger is dead
weight, so both are gone along with `repoint_container`,
`rename_manager_container`, `CONTAINER_TIMEOUT` and the two marker paths.
Phase 6 was not obsolete, only misplaced. Ruth's tool groups are now
seeded by `ensure_root_agent` on the one path that creates her, rather
than re-asserted on every hive-c0re boot. The skip-if-already-set guard
survives the move: a destroy+recreate under the same name must not reset
an operator's chosen group set back to MANAGER_DEFAULT.
That also settles a latent bug. Phase 4's marker check was a `return`,
not a skip, so on any hive carrying the marker phases 5 and 6 never ran
at all — the tool-group backfill, whose whole job was preventing a silent
privilege downgrade, has not executed here in a long time. Moving it to
create-time removes the question rather than answering it.
What stays is convergence: three unguarded, idempotent phases that re-run
each boot and no-op once their state is right. The module doc now names
the three categories so the next person can tell which kind they're
adding.
The doc still described a `label`/`deliberate_stop` parameter pair
inherited from `transient_guard`, which this function replaced and whose
signature it does not share — it only takes `name`.
Rewritten to say what it does and, more usefully, what must not come
through it: the queue answers the same question from the node itself via
`NodeKind::takes_container_down`, so this path is only for the two
operations that have no node behind them yet.
mara on !2910: "also remove imperative path for the things that are not
nodes yet, file follow up issue to fix that".
`TransientGuard`, the stored map and both manual set/clear are gone.
`transient_snapshot()` is derived and nothing else — destroy and
migration show no pill, because there is no node to derive one from. The
pill returns for free when they become nodes.
What those three guards were actually doing, though, was suppressing the
crash watcher, not drawing a pill. `migrate.rs` said so in its own
comment: without it, `crash_watch` fires `ContainerCrash` for every
migrated agent and the manager tries to recover containers that were
stopped on purpose. Destroy is the same — the container disappears
deliberately and nothing in the graph says so.
Deleting them outright would therefore have traded a dashboard pill for
false crash alerts on every destroy and every migration. So the
suppression survives as its own thing, `suppress_crash_watch`, with a
name that says what it is. It is still RAII, and still held for the
operation rather than stamped once, because the crash watcher's grace
window is finite and a destroy is not — a single tombstone would expire
mid-operation. The drop stamps the tombstone, covering the poll that
lands just after.
That leaves RAII in the codebase for exactly one purpose instead of two.
Untangling the pill from the suppression is what made the transient layer
deletable at all.
Follow-up issue for making destroy + migration real queue nodes to
follow; at that point this guard goes too.
Checked with clippy (`--all-targets -D warnings`), `cargo test -p
hive-c0re -p hive-jobq` (322 + 41 passed) and `nix fmt`.
mara on !2910: "why is set_transient still a thing if it completely
derives from nodes?"
It was still a thing because the scheduler mirrored the derived set into
a stored map that every consumer read — derived state computed once and
then cached, with the reconciliation loop existing only to keep the cache
honest. `transient_snapshot()` now derives: `running_transients()` off the
live graph, with the handful of entries that have no node behind them
(destroy, migration) overlaid on top. There is no cached copy left to go
stale or disagree with what is running.
`set_transient` / `clear_transient` split by what they actually do:
`set_manual_transient` / `clear_manual_transient` own the stored map for
the no-node callers, and `emit_transient_set` / `emit_transient_cleared`
publish the edges both paths need.
Two things had to survive, and both are edges rather than state:
- The dashboard's `TransientSet` / `TransientCleared` events. The
scheduler carries the previous derived value and emits the diff.
- The crash watcher's grace window. `recent_transient_within` answers
"was a transient cleared just now?", which is what stops a deliberate
stop from reading as a crash on the next 10s poll — a derived read of
current state cannot answer it, so the clear still stamps. The
scheduler keeps `deliberate_stop` alongside the label precisely so it
is available at clear time: the node it came from is, by definition, no
longer running to be asked.
`TransientState::since` becomes wall-clock and, for derived entries, is
the node's own `started_at` — the true start of the operation rather than
the moment a watcher first noticed it, which is what the old
guard-creation timestamp actually measured.
`running_transients` returns a named `RunningTransient` rather than a
4-tuple; two of its fields are strings and one is a bool whose meaning is
not guessable at a call site.
Note for anyone reaching for a timestamp here: chrono is vendored with
`default-features = false`, so there is no `Utc::now()`. The workspace
convention is `wire_time::now_unix()` / `from_secs()`.
Checked with clippy (`--all-targets -D warnings`), `cargo test -p
hive-c0re -p hive-jobq` (322 + 41 passed) and `nix fmt`.
mara on !2910: "transient guard as well - should be removable now?" — for
the queue path, yes.
`set_transient`'s own doc explained why the RAII guard existed: a
cancelled future must not leak an imperatively-set transient and pin the
dashboard on "rebuilding…" forever. That cannot happen to a derived set.
`running_transients()` is recomputed from the graph every loop, so a node
that stops running stops appearing — there is nothing to own and nothing
to leak.
So the scheduler no longer holds a guard per pill. It keeps the previous
derived value and publishes the transitions, which is the one thing a
derived read cannot express: the dashboard wants `TransientSet` /
`TransientCleared` edges, and the crash watcher wants the *moment* a pill
cleared, since its grace window is what stops an operator stop from
reading as a crash.
That also retires a hazard rather than restating it. The old code carried
a warning that stale guards had to be dropped before new ones were
created, because `TransientGuard::drop` clears by agent with no notion of
which label it was clearing — so a same-agent label change could clear
the pill it had just set. With no guards there is no ordering to get
wrong; clears are emitted before sets so a relabel reads as
clear-then-set rather than two overlapping pills.
`set_transient` / `clear_transient` become `pub(crate)`. The guard stays
for destroy and migration, which have no node behind them and where the
cancellation concern is real.
Checked with clippy (`--all-targets -D warnings`), `cargo test -p
hive-c0re -p hive-jobq` (322 + 41 passed) and `nix fmt`.
mara on !2910: "rename now, we will see if we can remove it later when
some of the users have been removed or work differently."
Nothing is held. The old name described a transient the DAG declared and
kept for its whole lifetime — precisely the thing this PR replaces — so
it outlived its own meaning the moment the derivation landed. The value
is recomputed from the running set on every call.
Kept as a function rather than inlined at its single call site, per the
above: removing it is a later step that depends on its users changing,
not something this PR should force.
Rename plus its two references (the call in `reconcile_transients` and
the module doc link). No behaviour change; the doc comment records what
the old name meant so the rename doesn't erase the reason for it.
Checked with clippy (`--all-targets -D warnings`), `cargo test -p
hive-c0re -p hive-jobq` (322 + 41 passed) and `nix fmt`.
The dashboard pill was declared once per DAG at submit time, so a rebuild
reported `rebuilding` for its entire life — through the prebuild, the
stop, the swap, the tail and the reconcile. It named the intent of the
request, not what was happening.
It is now read off the nodes actually running. A node lights a pill when
it is `Running` and declares the agent's own resource. Declaring is the
test, not targeting: `Prebuild` and `MetaSync` name an agent but are
lease-exempt on purpose (the container keeps serving), so they must not
light one. It is also not the lease *owner* — `resource_state()` answers
"who holds the slot", which is a different question from "what is
running", and a descendant that borrows an ancestor's grant never
appears in that map.
`TransientKind` is gone entirely rather than being re-derived. The label
is the node's own wire tag (`NodeKind::as_str`) — the same vocabulary
`NodeView.kind` already ships, so a pill and a DAG node name an operation
identically and there is no second taxonomy to keep in step. Work with no
node behind it (destroy, migration) supplies its own literal.
`DagSpec::transient`, `Claim::transient`, `DagMeta::transient` and
`NodeKind::Dag`'s `transient` field all go with it.
## the safety half, which is deliberately not the display half
`crash_watch::is_deliberate_stop` used to match a `TransientKind` to
decide whether a vanished container was intentional or a crash. That made
a pill's display vocabulary decide an alerting question, so renaming or
adding a label would silently move the alerting boundary.
`TransientState` now carries two independent fields: `label` (rendered,
nothing branches on it) and `deliberate_stop` (read only by the crash
watcher). The producer sets the second, because the producer is the only
thing that knows — it is not recoverable from the first.
For queue work that value is `NodeKind::takes_container_down()`, and it
is emphatically not "holds a lease": `Create` and `Start` hold the
agent's lease exactly like `Stop` does, and a container dying *while
starting* is a real crash that must keep reporting as one. The default is
`false` on purpose — a wrong `false` costs a spurious crash event, a
wrong `true` swallows a real crash silently.
## known cost, accepted on the issue
A restart no longer reads `restarting`. No `NodeKind` is unique to a
restart — `restart_chain` reuses `Signal` / `StopForUpdate` / `Drain` /
`Reconcile` — because "restart" is a property of the DAG's shape, not of
any node. A restart now reads `signal` / `stop_for_update`, then the
agent returns.
`Start` / `Stop` / `PostSwap` run inside a lease-holding ancestor and
re-declare nothing, so they light no pill and the agent reads idle for
those windows. Closing that is the resources-where-constructed work
(#2818), not this change.
Checked with clippy (`--all-targets -D warnings`), `cargo test -p
hive-c0re -p hive-jobq` (321 + 40 passed) and `nix fmt`.
