Repository navigation
Description
F12.10 — Fair scheduling and quotas
From the Phase 12 plan (.opencode/plans/phase-12-platform-flexibility.md, workstream D):
A hub scheduler that shares fuel and CPU fairly across worlds, with per-account quotas for worlds, storage, and fuel per second, and back-pressure for slow clients. Depends on F11.7 (#553) and F11.8 (#539).
Exit: a world that spins on fuel cannot raise another world's p99 latency by more than 10 percent, shown in a benchmark.
Coordination (2026-10-26 reprioritization, per Drew): every id roadmap item now has its own GitHub issue, and priority encodes the order: (1) the path to Phase 12 — now + priority/high; (2) Phase 12 itself — next; (3) Phase 13 and long-term ideas — priority/low. The dashboard mirrors this list (org Project "Developing Today" #4, Area id-train; Grist Program roadmap, Area id). If this item turns out to be blocked, wrong, or superseded, park or close it with a note rather than letting it stall the queue — dropping an issue is always allowed; silently stalling is not.
Activity
developing-today commentedon Oct 27, 2026
CODE-211
dezren39 commentedon Nov 1, 2026
🤖 Agent note · done
.opencode/plans/phase-12-platform-flexibility.md:49Triage verdict: CONFIRMED unimplemented; dependencies now closed; too large for a small_fix packet — stays as the tracked
nextroadmap itemDependencies — both closed (fresh check this run):
So F12.10's stated blockers are gone; nothing but the work itself remains.
Confirmed absent at
origin/mainb56ccda (laptop probes this run):grep -rn 'fair|per-account|per_account' pkgs/id/src --include='*.rs'→ zero hits;grep -rln quota pkgs/id --include='*.rs'→ zero source files.pkgs/id/src/world_limits.rsdefines server-adminRuntimeLimits(Wasm fuel/memory/payload caps; native rlimits + per-call deadline) and a per-worldWorldExecutionProfileclamped beneath those ceilings. There is no hub scheduler sharing fuel/CPU across worlds, no per-account quota accounting (worlds/storage/fuel-per-second), and no back-pressure mechanism for slow clients..opencode/plans/phase-12-platform-flexibility.md:49with the benchmark exit: a fuel-spinning world must not raise another world's p99 latency by >10%.Routing — why no follow-up packet from this triage: the queue's follow-up kind for implementation is
small_fix(45-minute timebox); this iseffort/ldesign+implementation (a scheduler + quota accounting + a p99 benchmark on the F11.8 harness). Enqueuing it as a small_fix would guarantee a park/evaporation cycle. The issue itself — open,next, priority/normal per the 2026-10-26 coordination — remains the tracking vehicle. Recommended next step when Phase 12 work is picked: a design slice first (scheduler placement: world actor vs hub; quota store; benchmark shape on #539's harness), then implementation slices sized to the queue.4 remaining items
dezren39 commentedon Nov 3, 2026
agent-note:v1
state: triaged (coverage pass)
Coverage pass 2026-11-03: reviewed in the fleet-wide per-issue pass (drift / overlap / already-handled / stale / blockers / ordering). Verdict: still live; queued in repo-health-code-pilot as triage at priority 40 (priority/normal). If this is already handled at current main, the triage closes it with evidence; if it needs splitting, the triage files the split.
Found via: coverage pass 2026-11-03
dezren39 commentedon Nov 3, 2026
Queue task 7637 is handed off to its bound OpenCode worker session
ses_e5e5c6e6828ffen7IP6Jr3vP3N.dezren39 commentedon Nov 3, 2026
Picked up queue task 7637 for a current-main re-triage of #1067. Fresh GitHub checks are underway; the prior triage verdict will be revalidated against current source and roadmap before routing.
dezren39 commentedon Nov 3, 2026
Triage verdict — still live; unimplemented; remains a planned Phase 12 item
Revalidated against current GitHub
mainat5c53272ff83bdf1ddffea2fb188c7d714182c9ab(2026-11-03T22:39:41Z). #1067 is OPEN. The stated prerequisites #553 (F11.7) and #539 (F11.8) are both CLOSED. No matching open PR was found.Current
pkgs/id/crates/id-core/src/world_limits.rsstill provides per-world Wasm/native ceilings, execution profiles, and a per-world memory budget; current code search found no quota implementation or fair-scheduler implementation. The Phase 12 plan still specifies hub-level fair sharing, per-account world/storage/fuel-per-second quotas, slow-client back-pressure, and the benchmark exit that a fuel-spinning world raises another world’s p99 latency by no more than 10%.planning/roadmaps/id.mdstill lists F12.10/#1067 as planned.Scope and routing: This is still an
effort/ldesign-and-implementation item, not asmall_fix. When scheduled, first define scheduler placement/ownership, fairness and per-account quota accounting, back-pressure behavior, and the contention benchmark; then split implementation and benchmark work into queue-sized slices. The F11.7/F11.8 dependencies are no longer blockers. Existingnextroadmap ordering remains appropriate; no follow-up small-fix packet was enqueued from this triage.Found via: queue task 7637; current-main GitHub/API review.
dezren39 commentedon Nov 3, 2026
Addendum after the initial source review: GitHub
mainadvanced to07168f959f65ffae0e6cd1756cdb01eb0d768d13at 2026-11-03T22:44:59Z. The three intervening commits changed onlypkgs/mcpx/...; nopkgs/idsource or roadmap files changed. The triage verdict above therefore remains current at this newer main.