From 62cda074c10dd1044940e816d48feaa93d7a3e09 Mon Sep 17 00:00:00 2001 From: sami Date: Thu, 10 Sep 2026 00:04:53 +0400 Subject: [PATCH] daemon: fold in CORE's sign-off on ADR 0013 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CORE accepted ADR 0013 as written, no amendments (core/docs/adr-0013-core-response.md, lane/core@c65e664). Folds in: - pause()/resume() idempotency contract, precisely: no-op success on an already-paused task, ALSO on a terminal task (pause racing completion isn't an error), resume() no-op on a non-paused task, the only error is task_not_found, and no state-change event fires for a no-op call. - tasks_starved pinned as {connecting, downloading} AND segments_active == 0 — a structural exclusion of retry_wait and auto-paused tasks rather than a special case, with the full state table CORE gave. - restart handling confirmed fully; two non-blocking notes from CORE about work-interruption during verifying/assembling. - "auto-pause" adopted as the term, no new wire/API surface. Status updated: accepted by CORE; PROTO's item 3 (permit `error` on event.task.state when state=="paused") is the one remaining blocker before daemon/src/sched/'s pause/resume logic can be written correctness-preservingly. Co-Authored-By: Claude Sonnet 5 Claude-Session: https://claude.ai/code/session_01Upd9WhG9oppieig5nRDLig --- docs/adr/0013-task-state-machine-ownership.md | 111 ++++++++++++------ 1 file changed, 76 insertions(+), 35 deletions(-) diff --git a/docs/adr/0013-task-state-machine-ownership.md b/docs/adr/0013-task-state-machine-ownership.md index 09b1736..612e08b 100644 --- a/docs/adr/0013-task-state-machine-ownership.md +++ b/docs/adr/0013-task-state-machine-ownership.md @@ -1,9 +1,13 @@ # ADR 0013 — Task state-machine ownership: CORE, DAEMON, and the shared `paused` -**Status:** proposed · **Date:** 2026-09-09 · **Lane:** DAEMON, drafted at PROTO's request -(D1 in `core/docs/proto-requests-m1.md`), for CORE and PROTO to review and sign off. -**Needs:** explicit accept from CORE and PROTO before `daemon/src/sched/` calls into CORE's -task-start API, since the calling convention below is derived from this split. +**Status:** accepted by CORE, PROTO's open item 3 outstanding · **Date:** 2026-09-09 +· **Lane:** DAEMON, drafted at PROTO's request (D1 in `core/docs/proto-requests-m1.md`). +**CORE's response:** `core/docs/adr-0013-core-response.md` (`lane/core`, commit `c65e664`) +— accepted as written, no amendments, plus the pinned `tasks_starved` definition and +`pause()`/`resume()` idempotency contract folded in below. +**Remaining blocker:** PROTO landing the `error`-on-`paused` widening (open item 3) — +`daemon/src/sched/`'s pause/resume logic is unblocked from CORE's side already, but not +buildable correctness-preservingly until that lands. ## Context @@ -77,19 +81,24 @@ Both sides can put a task in `paused`, for disjoint reasons: state-change callback — the same path every other CORE-driven transition uses. This is not a new mechanism. -The contract that makes this safe: +The contract that makes this safe, per CORE's sign-off (`core/docs/adr-0013-core-response.md`): -- **`pause()` is idempotent.** DAEMON calling `pause()` on a task CORE already - auto-paused is a no-op, not an error — DAEMON does not need to know CORE got there - first. Symmetrically, if CORE were ever mid-transition to `paused` when DAEMON's pause - request arrives, the result is still `paused`, once. +- **`pause(TaskId)` is idempotent: a no-op success** on a task already `paused` (however + it got there), and **also a no-op success on a terminal task** (`complete`/`failed`/ + `cancelled`) — a pause racing a completion is not an error. **`resume(TaskId)` is a + no-op success on a task that is not paused.** The only error either returns is + `task_not_found`. CORE emits no state-change event for a no-op call — DAEMON reads the + resulting state from the normal state-change callback / `TaskDetail`, never from + `pause()`'s or `resume()`'s return value. If CORE is mid-transition into `paused` (its + own auto-pause) when DAEMON's `pause()` arrives, the task ends up `paused` once, with + exactly one state-change event. - **DAEMON persists *why* a task is paused**, in the `tasks` table, not in `TaskState` itself (the wire type stays a flat enum — this is DAEMON-local bookkeeping, not a contract change). A `pauseReason` distinguishing at least `user`, `schedule`, `queue_stopped`, `admission_reconcile`, and CORE's `error.code` when auto-paused. This - is what makes §3's resume rule possible — but it needs a contract fix first: see the - gap in open item 3 below. `error.code` is not currently carried on a transition into - `paused` at all. + is what makes §3's resume rule possible — but it needs a contract fix first: see "A + contract gap this ADR surfaces" below. `error.code` is not currently carried on a + transition into `paused` at all. - **CORE does not need to track why it's paused past the current occurrence.** Once paused, CORE's job is done; DAEMON is the only side that later decides whether to resume, and DAEMON is also the only side with persistent storage to remember the @@ -118,14 +127,28 @@ and asserts `tasks_starved == 0` in steady state past a bounded delay. A task in `retry_wait` also holds zero segments, deliberately, for up to 60 s (`docs/04` §7's backoff cap) — and it is **not** admission-starved, it is CORE's own policy holding it idle. -Resolution: `retry_wait` must be distinguishable from true starvation without DAEMON -guessing from timing. CORE's `starved_tasks()` (ADR 0011) excludes any task CORE itself -holds at zero by policy — `retry_wait` and CORE-initiated `paused` both fall outside -`tasks_starved` by construction, because that accessor is about the segment allocator -failing to grant a slot to a task that wants one, not about a task that isn't asking. A -task in `retry_wait` still counts against `connection.maxConcurrentDownloads` from -DAEMON's side (it is running, not requeued) but contributes nothing to the segment -allocator's guarantee pass until it re-enters `connecting`. +**Pinned by CORE's sign-off**, tightening ADR 0011's definition rather than special-casing +it: `tasks_starved` counts only tasks whose `TaskState` is `connecting` or `downloading` +**and** `segments_active == 0` — precisely "the allocator has not granted a slot to a task +that is asking for one." + +| Task state | In `tasks_starved`? | Why | +|---|---|---| +| `probing` | no | uses the probe pool (ADR 0011 §5), not the segment budget | +| `connecting`, 0 segments | **yes** | admitted + probed, waiting on the allocator's first grant — the real starvation case | +| `connecting`/`downloading`, ≥1 segment | no | a segment in its own `connecting` sub-state counts as held (ADR 0011 amendment A3) | +| `downloading`, 0 segments | **yes** | held slots and lost them all (e.g. every segment failed and is being re-requested) — transient, still real | +| `retry_wait` | no | CORE's backoff timer holds it at zero *deliberately*; not asking the allocator for anything until it re-enters `connecting` | +| `paused` (either-initiated) | no | not asking for a slot | +| `new`, `queued`, `assembling`, `verifying`, terminal | no | outside the counted state set | + +So `retry_wait` and auto-paused tasks are excluded **structurally, by not being in the +`{connecting, downloading}` state set** — not via a special case that could rot as the +engine evolves. `starved_tasks()` returns exactly the TaskIds in this count; +`starved_since(id)` is defined only for them. A task in `retry_wait` still counts against +`connection.maxConcurrentDownloads` from DAEMON's side (it is running, not requeued) but +contributes nothing to the segment allocator's guarantee pass until it re-enters +`connecting`. ### 5. Restart — CORE holds no persistent state, DAEMON reloads to `queued` @@ -143,6 +166,18 @@ where to actually resume from `.veloxpart.meta` and re-validates with `If-Range` `paused` tasks reload as `paused`, with their `pauseReason` intact, and are not auto-admitted — §3 applies identically after a restart as it does live. +**Confirmed by CORE, fully.** `start(TaskId)` transparently checks for a valid +`.veloxpart.meta` sidecar (CORE's stage 5), re-validates with `If-Range` (`docs/04` §5), +and either resumes from the recorded offsets or restarts if the sidecar is missing or +fails validation. DAEMON does nothing special beyond rewriting CORE-owned states to +`queued` and re-admitting, as this section already said. + +**Two notes from CORE, not objections to the ADR:** the *transition* into `paused` is +honoured from any CORE state per §1's table, but the *work interruption* is best-effort — +a pause during `verifying` discards the in-progress hash and re-hashes from the start on +resume (cheap, bounded); a pause during `assembling` (HLS/DASH mux) is an M4 concern and +may not be cleanly interruptible mid-mux. Neither changes this ADR's API shape. + ## Consequences - `daemon/src/sched/` calls `start(TaskId)` exactly once per admission (`queued → @@ -151,10 +186,10 @@ auto-admitted — §3 applies identically after a restart as it does live. - The `tasks` table needs a `pauseReason` column (DAEMON-local; not a wire type) before `sched/`'s pause/resume logic can be written correctly — flagging as a concrete follow-up, not blocking this ADR's acceptance. -- CORE's `starved_tasks()` (ADR 0011) must exclude `retry_wait` and CORE-auto-paused tasks - by construction; if that isn't already true in CORE's accessor, it needs to be before - DAEMON relies on the starvation invariant, since otherwise every backoff cycle would - trip DAEMON's governor-invariant warning as a false positive. +- CORE's `starved_tasks()` (ADR 0011) excludes `retry_wait` and CORE-auto-paused tasks by + construction — pinned in CORE's sign-off as the `{connecting, downloading} ∧ + segments_active == 0` definition in §4's table — so DAEMON's governor-invariant warning + will not false-positive on a normal backoff cycle. - No `contracts/` change. `TaskState`, `event.task.state`, and `previousState` are already sufficient; this ADR is entirely about which process calls which function when. @@ -190,15 +225,21 @@ follow-up (minor: widening an existing field's presence condition, per tell CORE-paused from user-paused on the wire in general — `error: null` on a DAEMON-initiated pause is sufficient, since DAEMON already knows it just did that. -## Open, for CORE and PROTO to confirm or amend +## Resolution of the four open items -1. Does CORE's `starved_tasks()` already exclude `retry_wait` and auto-paused tasks, or - does this ADR ask for a behavior change there (§4)? -2. Is `pause()` idempotent today, or does calling it on an already-paused task currently - return an error CORE needs to relax? -3. PROTO: land the `error`-on-`paused` widening above before DAEMON writes the resume - logic in §3 — otherwise DAEMON has no correctness-preserving way to implement it and - would have to guess from timing, which is the exact failure mode ADR 0011 was written - to rule out for a different pair of governors. -4. Naming: is "auto-pause" the term CORE already uses internally, or does CORE have an - existing name for this that this ADR should adopt instead of introducing a new one? +1. **`starved_tasks()` excludes `retry_wait`/auto-paused by construction — confirmed.** + Pinned as a design commitment in CORE's sign-off (§4's table above): the counted set is + `{connecting, downloading} ∧ segments_active == 0`, so exclusion is structural, not a + special case. +2. **`pause()`/`resume()` idempotency — confirmed as specified in §2 above**, plus two + details this ADR hadn't anticipated: a no-op call also succeeds against a *terminal* + task (pause racing completion isn't an error), and no state-change event fires for a + no-op — DAEMON reads resulting state from the callback/`TaskDetail`, never from the + call's return value. +3. **Still open — PROTO.** CORE's half is ready (`ErrorInfo` is populated on every + auto-pause transition today's callback path would carry, using the existing B1 + taxonomy); the wire only needs `error` permitted when `state == "paused"`. This is the + one remaining blocker on `daemon/src/sched/`'s pause/resume logic. +4. **CORE adopts "auto-pause".** No new wire or API term — the discriminator stays + `state == paused` plus the presence of an `Error` (present ⇒ CORE-initiated, absent ⇒ + DAEMON-initiated), exactly as §2 already specified.