Skip to content

id phase-7: P7.5 deployments and real-network tests are owner-blocked with no tracking issue #794

Open

Description

@dezren39

Finding

P7.5 (deployments and real-network tests) is owner-blocked on external networks/accounts and is tracked only in plan/brief prose, not as a standalone issue.

Evidence

  • doc/2026-10-11T11-08-42Z_design_remaining_work/plan-phases.md §12 Phase 7, step P7.5: "Deployments and real-network tests (owner-blocked)" with "owner accounts and networks" in the Blocked-by column, tests "the C2 matrix; N29 on real networks".
  • doc/2026-10-12T21-22-53Z_phase_7_connectivity/2026-10-12T21-22-53Z_phase_7_connectivity.md P7.5 row: "owner-blocked ... accounts and networks"; the 2026-10-12T21:40:00Z note spells out what was not done: "Owner-blocked (P7.5): running either mode on a real host, with a Cloudflare Tunnel or a VPS, and checking /ping, /generate_204, and the systemd KillSignal behaviour. Not deployed here."
  • doc/2026-10-14T_platform_completion/acceptance-brief.md line 75 (unchecked): "P7.5/P3.7: real two-network and account-backed deployment matrix cannot be truthfully verified without the required networks/accounts. Keep the code-side test harness ready and leave the live matrix explicitly blocked, not falsely passed."

Why it matters

The relay kit (connectivity/relay/relay.kit.toml with tunnel and vps modes) has unit-level tests but has never been exercised on a real host. The live matrix stays explicitly blocked, which is correct, but there is no issue to re-open when the owner supplies networks/accounts.

What's needed

Owner-provided: two separate networks and the accounts (Cloudflare Tunnel and/or a VPS). Then run the C2 matrix and N29 real-network checks and record results, or close as superseded by later connectivity work.

Acceptance criteria

  • Owner-provided inputs exist and are identified: two separate networks, plus Cloudflare Tunnel and/or VPS account access (this is the blocker — see the intake review comment).
  • The relay kit is run on a real host in tunnel and/or vps mode, and /ping, /generate_204, and the systemd KillSignal behaviour are checked, with results recorded on this issue.
  • The C2 matrix and N29 real-network checks are run and recorded — OR this issue is closed as superseded by later connectivity work, with the superseding work linked.

Activity

  1. added
    kind/testTest coverage, test infrastructure, or test-only fixes
    status/blockedQueue or user escalation state
    priority/lowNice to have; do when convenient
    effort/lLarge: multi-day or open-ended; expect staged check-ins
    app/idPackage/app: id (pkgs/ or apps/ in this repo)
    user/reviewNeeds a user-look: Muse reviews first and handles what it can
    meta/enrichedIntake review done: acceptance criteria, links, and normalized labels added
    on Oct 26, 2026
  2. dezren39 commented on Oct 26, 2026

    @dezren39
    MemberAuthor

    Intake review

    Related: #795 (P3.7 — same owner-provided networks/accounts blocker; the acceptance brief groups P7.5/P3.7 at line 75). Plan §12 Phase 7 step P7.5 and the Phase 7 record's P7.5 row are cited in the body. Relay kit: connectivity/relay/relay.kit.toml (tunnel and vps modes).

    Still live (verified 2026-10-26): no merged PR references or resolves this issue; the relay kit has unit-level tests but has never been exercised on a real host. Not solved; do not close.

    Notes / risks: Genuinely owner-blocked (networks + accounts), so this is labelled status/blocked + user/review instead of status/needs-triage. The brief's instruction stands: keep the code-side harness ready and leave the live matrix explicitly blocked, not falsely passed. Nothing here is agent-actionable until the owner inputs exist.

    Suggested approach: When the inputs exist: deploy the relay kit on a real host (tunnel and/or vps), check /ping, /generate_204, and KillSignal behaviour, run the C2/N29 matrix, record results here and in the Phase 7 record, then close.

    🤖 Agent note · intake enrichment

    • Agent: fleet monitor (repo-fleet-monitor)
    • Session / ID: repo-fleet-monitor run, 2026-10-26 00:19 CDT (05:19 UTC)
    • Found via: monitor reconciliation
    • Task: Intake enrichment (§5 pass): acceptance criteria appended, labels normalized (blocked on owner-provided networks/accounts), still-live verified.
    • Expectation: effort/l once unblocked — real-host deployment + real-network matrix; until then, blocked on the owner inputs named above.
  3. added
    work/assignedQueue worker lifecycle state
    queue/claimedQueue worker lifecycle state
    and removed
    status/blockedQueue or user escalation state
    on Oct 28, 2026
  4. dezren39 commented on Oct 28, 2026

    @dezren39
    MemberAuthor

    🤖 Agent note · picked up

    • Agent: queue worker (repo-worker-code), Lane A shift
    • Session / ID: queue task 2224 · repo-queue-worker-shift 2026-10-28 14:00 CDT
    • Found via: queue claim (issue carries user/review — reviewed by the worker first, per protocol)
    • Task: Triage issue id phase-7: P7.5 deployments and real-network tests are owner-blocked with no tracking issue #794 — verify the P7.5 owner-blocked state at current HEAD and route it.
    • Expectation: s — done = facts verified, the exact owner inputs named, escalation recorded with the read attached
  5. dezren39 commented on Oct 28, 2026

    @dezren39
    MemberAuthor

    🤖 Agent note · escalated

    • Agent: queue worker (repo-worker-code), Lane A shift
    • Session / ID: queue task 2224 · repo-queue-worker-shift 2026-10-28 14:00 CDT
    • Found via: queue claim; issue carried user/review, reviewed here first
    • Task: Triage of id phase-7: P7.5 deployments and real-network tests are owner-blocked with no tracking issue #794 complete; the blocker is owner-provided infrastructure, escalated with specifics below
    • Expectation: Drew provides (or declines) the inputs listed; on provision, the live matrix run is a straightforward worker task

    Triage read — facts verified at HEAD (2026-10-28)

    • The relay kit is present and current: relay.kit.toml exists at HEAD both in the design tree (doc/2026-10-11T03-44-33Z_design_groups_events_delegation/connectivity/relay/, alongside tunnel.toml, vps.toml, and iroh-relay.service) and as a test fixture (pkgs/id/tests/fixtures/relay/relay.kit.toml). Nothing about the kit is stale — it has simply never run on a real host, exactly as the issue says.
    • The blocker is genuinely external: every cited record (plan §12 P7.5, the Phase 7 connectivity note of 2026-10-12, acceptance brief line 75) says the same thing — the live matrix "cannot be truthfully verified without the required networks/accounts". No worker-side action substitutes for that; running it on one network or a sandbox would be the false pass the brief warns against.
    • Not superseded: no later connectivity work has run this matrix. (Phase 13's F13.3, id: Phase 13 F13.3 SSH-over-iroh reachability (and iroh-over-SSH fallback) #785 SSH-over-iroh, is a different item and still open.) The close-as-superseded exit in the criteria is not available on current evidence.
    • Shared blocker: sibling id phase-3: P3.7 real-network holepunch and n0 QAD tests on two networks are owner-blocked with no tracking issue #795 (P3.7, real-network holepunch/QAD tests) is blocked on the same owner inputs and is still open with the same status — one provision unblocks both issues.

    What Drew needs to provide (the whole ask)

    1. Two separate networks the test hosts can sit on (e.g. home LAN + a VPS's network, or two VPS in different providers/regions).
    2. Account access for at least one relay mode: a Cloudflare account/zone for Tunnel mode, and/or a VPS for vps mode.

    With those, the run is mechanical: deploy the kit in tunnel and/or vps mode on a real host, check /ping, /generate_204, and the systemd KillSignal behaviour, run the C2 matrix + N29 checks, record results here and on #795. If Drew would rather not spend the accounts/time, the honest alternative is an explicit descope decision recorded here — the kit and fixtures stay in-tree either way.

  6. added
    user/escalatedQueue or user escalation state
    queue/blockedQueue or user escalation state
    status/blockedQueue or user escalation state
    and removed
    user/reviewNeeds a user-look: Muse reviews first and handles what it can
    queue/claimedQueue worker lifecycle state
    on Oct 28, 2026
  7. dezren39 commented on Oct 28, 2026

    @dezren39
    MemberAuthor

    Owner constraints recorded (Drew, 2026-10-28, from the #795 ruling)

  8. added
    mode/hybridOnline work plus a physical step (power, cabling, KVM access to a real device)
    queue/queuedHas a queued task in the Supabase agent queue
    on Nov 1, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

app/idPackage/app: id (pkgs/ or apps/ in this repo)effort/lLarge: multi-day or open-ended; expect staged check-inskind/testTest coverage, test infrastructure, or test-only fixesmeta/enrichedIntake review done: acceptance criteria, links, and normalized labels addedmode/hybridOnline work plus a physical step (power, cabling, KVM access to a real device)priority/lowNice to have; do when convenientqueue/blockedQueue or user escalation statequeue/queuedHas a queued task in the Supabase agent queuestatus/blockedQueue or user escalation stateuser/escalatedQueue or user escalation statework/assignedQueue worker lifecycle state

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions