The director's IP → country table (ORCA_RELAY_IP_COUNTRY_DB) is fetched and verified by the
deployment's image build and staged in apps/relay/ip-country. The directory holds only a README
here, so an image built from this repository alone is unchanged apart from that file.
* relay: edge rollout per region, country exclusion per region, region recorded on edge moves
director/edge.json gains regionRolloutPermille (a listed region is in scope at its own share),
regionExcludeCountries and regionExcludeUnknownCountry (checked after host allow, before account
allow and the share, for the requested region only). The leave ceiling must cover every regional
share; mayBeOnEdge's ceiling includes them. All absent: every decision is as before.
The director reads the ask's country from an offline IP table (ORCA_RELAY_IP_COUNTRY_DB, loaded at
boot) by the last X-Forwarded-For hop, keeps only the country, and logs per-minute region x country
counts (orca_relay_edge_region_country). Edge moves record the judged region in
relay_assignment_region_preferences inside their transaction; the evacuation sweep judges that
region when it has its own share, else 'any' (the highest in-scope share). The edge Worker's metric
place adds the country.
* relay: record edge regions only while a regional share is set; sweep judges a recorded region by its effective share
With no regionRolloutPermille key the edge path writes nothing extra and the sweep judges every
host 'any', as before. With one, edge moves record their region and the sweep judges a host with a
recorded region by that region's effective share (its own value, else the global share when
env-scoped, else out of scope); unrecorded hosts stay 'any'.
* relay: edge region record in its own table; dark sweep SQL unchanged; lenient country codes; hex IPv4-mapped lookups
Edge moves record the judged region in relay_edge_host_regions (created at director boot like
relay_edge_fence_pending), always on a move and on a kept host only when the region changed; the
row is deleted when the host is placed off the edge. The sweep reads only that table, and only
while a regional share is set, so a stale cell-path preference never drives drains and the dark
page query is unchanged. The cell-path region preferences are no longer written by the edge path.
Country codes in edge.json are upper-cased before validation, and hex-spelled IPv4-mapped IPv6
addresses are looked up as IPv4.
* relay: braces on every if/for body and T[] array types on the regional edge change (changed-lines lint)
- Directors with an edge read director/edge.json from the relay-control bucket through the shared
flag channel (new onApplied listener): mode on/off/down and rolloutPermille (0..1000). Boot
defaults are mode on and the env percent; every cohort decision reads the applied flags. Off
empties the cohort and hosts leave normally; down also skips the export at once (reason
mode_down). The permille bucket keeps every host of P percent in at P x 10 permille. A control
change re-arms the evacuation sweep within 5 s.
- A skipped export (120 s outage clock or mode down) removes the host's relay_devices rows before
placing it, so revokes made on the edge are never undone; logs a count only.
- Evacuation: 100 drains per tick with 10 in flight, each pass starting at a random point of the
keys and wrapping around; pages stay 500 rows and an idle pass reads each edge row once.
- Fence retries: 56 attempts (last about 25.5 h after the failure), 50 per sweep with 10 in
flight; fence_abandoned now carries outcome, cell and epoch like fence_failed.
Port of b369be6 (main-bound branch): failures more than 30 s apart restart the unavailable
clock, so one old failure cannot skip an export. The sweep's first pass is 45-75 s after start,
every 45-75 s while hosts are queued, else every 7.5-12.5 min.
Port of the main-bound hardening (stablyai/orca#39707) to the release branch:
- leaving hosts skip the edge export after 120 s of edge unavailability
(ORCA_RELAY_EDGE_EXPORT_SKIP_AFTER_MS), keeping Postgres devices;
- directors drain edge hosts that left the cohort (POST /__admin/drain);
- account-allowlisted hosts must also be in the edge regions;
- tests for the invite-socket edge branch.
Every host of a listed account (relay token sub) joins the edge cohort, in any region;
a denied host still stays off. For the team's own accounts before any share rollout.
The deployed edge keeps host data in its objects; without the key the director could
not push it and hosts would land on an edge holding none of their devices.
A host still holding old-cell units (a pairing invite holds one for its 10 min TTL) or whose push
the edge did not take now stays where it is and moves on a later ask. 503 stays for a held row
lock and for a desktop still connected to the edge on the way out.
With ORCA_RELAY_EDGE_ADMIN_SIGNING_KEY set (its own key, never the assignment lease key), a move
onto the edge pushes the host's relay_devices and a location hint under the row lock and commits
only after the edge accepts; an edge holding a newer epoch is re-minted above once. Before any
lane may re-place an edge row, the director exports (refused while the desktop is connected; the
edge freezes the host), replaces the host's Postgres devices, places it as today, then fences the
edge with the new cell and epoch. Idle regional rehome skips edge rows. Unset key: unchanged.
Directors with ORCA_RELAY_EDGE_CELL_URL seed an existing-only 'edge' cell (registered
with the admission selector so deploy workflows carry it) and answer a host cohort
(hash bucket, allow/deny lists, region scope) on the edge before the reserve and
database lanes. The epoch rises only on a cell change; a host still holding units on
its old cell is told to retry. Phone resolve and the invite move answer edge rows
without moving anything; old cells answer WRONG_CELL to phones of moved hosts.
Unset URL keeps today's behaviour.
* fix(relay): exit promptly on SIGTERM instead of waiting out docker's stop timeout
On SIGTERM the relay drained sessions and called server.close(), which waits
for every open socket. A TCP connection that never sends a byte (or sends
partial headers) is not idle, and server.close() also stops the header-timeout
checks, so nothing ever closed it. The process stayed alive until docker's
--stop-timeout 300 SIGKILLed it: about five minutes of every cell VM
replacement during a same-cap roll.
Shutdown now tracks accepted sockets (including upgraded WebSockets), keeps
the existing drain order, destroys sockets still open after a 5 s grace, waits
(bounded) for queued pool clients, closes the database and exits 0. An unref'd
10 s backstop exits even if the database close hangs. SIGTERM/SIGINT use
process.on behind a shutting-down guard so a repeated signal no longer kills
the process with 143.
The new blackbox test fails on the previous index.ts (silent TCP and
partial-headers peers keep it alive past the test bound) and passes now.
* style(relay): satisfy changed-lines lint gate in SIGTERM exit change
* fix(relay): give a freshly booted reserve cell a dead-man boot grace until its first director contact
A reserve cell's dead-man starts its 60-120 s window when the switch file applies, but no
placing director can reach a new process until its load balancer routes to it. On the
2026-11-01 production same-cap rolls that took up to about 110 s after the switch file
applied; c10 (104 s window) and c16 (71 s) tripped 1-4 s before the first director poll
landed, latched into db, and failed their roll after a clean drain and restart.
Until the process's first contact (a reserver=1 feed poll or a real booking), the window is
the hashed window plus 180 s, matching autohealing's initial delay. After the first contact
nothing changes. A process that never hears from a placing director still falls back to db,
up to three minutes later than before.
* test(relay): assert the server wires the dead-man boot grace
* docs(relay): state the boot gap behind the dead-man grace as inferred, not measured
* fix(relay): cap HTTP request bodies before they are buffered, including chunked ones
Every relay POST route checked only the Content-Length header before calling req.json(), so a chunked request with no Content-Length was buffered and parsed whole. /v1/resolve does that before any authentication. Enforce Hono's bodyLimit app-wide at RELAY_PROTOCOL_LIMITS.maxHttpBodyBytes (4 KiB, the limit every route already declares), with the cell reserve and demote routes keeping their own 16 KiB limit; push already does the same.
* test(relay): prove a 6 KiB reserve body reaches its route
The runtime stages shipped node:24.22.1-alpine (Alpine 3.24.2) as built, so CVE-2026-85091 (zlib) and CVE-2026-107891 (musl) were present in all three images. Move to node:24.22.2-alpine (Alpine 3.24.3) and apk upgrade the runtime stage, as the api and auth images already do, so later Alpine fixes land at build time.
The Asia drain and departing-redial runs asserted a 20ms-sampled mean of lock-waiting director
backends under 0.5; it flaked at 0.56 and 0.57 on loaded runners. A focused test now holds the
source cell row and proves a reconnect placement still lands on a target (and that holding the
target rows does block one); the load runs keep their outcome assertions and report the mean.
Cap each splice direction at 4,096 queued frames (closes 4429 with
trigger queue-frame-limit) and charge a 64-byte per-entry overhead
against the shared process queued-byte budget so empty frames are not free.
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com>
Groundwork for per-company Slack apps. Each Slack app gets its own row with encrypted secrets, and Slack posts to /slack/apps/<app id>/..., where the address only selects which secret to check; a request acts only in a workspace where that app is installed. Rate limits are per address and per app, so one workspace can't hold up another's posts. Schema changes are additive. The existing staging app keeps working on the old route until a documented cutover. Inert for users.
The always-on Discord gateway worker now runs the same background sweeps as the API (queued posts, held starts, cleanup), so Discord replies go out at once instead of waiting for the scaled-to-zero API to wake. Posts are claimed with leases, so running the sweeps in both services never sends one twice.
Freezes the connector's shipped table definitions: later schema changes must be additive statements. Adds a baseline fixture and an upgrade test (baseline, then current schema, compared with a fresh database on SQLite and Postgres), plus the statement-shape check relay already uses. No runtime change.
Adds Discord as a second connector adapter on the shared core: signed interactions for answer buttons, one gateway worker (same image, single leader by database lock, commits on the lock's connection, checkpoints for resume), thread posts through the shared post queue, owner-only control, and teammate context only from linked accounts. Inert until the Discord settings are configured. A narrow gitleaks allowlist entry covers connector Discord env var names in config errors.
The served-relay blackbox test closed the phone and data sockets from the client and sent the resume confirm before the relay had processed the close, so on a busy runner the relay correctly answered device-resume-confirmed and the test failed (about 1.8% under load). The test now closes only the phone and waits for the relay to close the data leg with 4408, which happens in the same step that marks the connection inactive. Test-only.
Moves routing, machine choice, held starts, the post queue, prompt cards and owner controls out of the Slack files into a provider-neutral connector core, with Slack as the first adapter behind one adapter type. Conversation records and delivery ids are keyed by source, so Slack and Discord ids cannot collide. Behaviour-neutral for Slack apart from a slack: delivery-id prefix and a 403 (not 404) from the reply route when Slack is off. Staging's connector database is reset after deploy.
A connector conversation now names its source (slack, discord or linear), and a posted reply reports a neutral messageId. Each machine's hello lists the sources it accepts, and the cloud only routes a Slack conversation to machines that list Slack. Machine events from other sources are filtered out where they arrive, and a reply for a non-Slack conversation is refused, so a Discord or Linear conversation can never post into a Slack thread with the same id. The unused turn-bound cancel fields are removed (they return with Stop). Inert: no shipped desktop uses the contract, and Slack is off unless configured.
When a Slack-started agent asks for approval or asks one pick-one question, the thread gets a card with one button per option. Only the person who asked for the work can answer; anyone else is told so privately. The answer travels through the existing inbox and delivery path, and the card closes once answered or no longer pending. Other question formats (grouped, pick-several, typed answers) still point to Orca; a follow-up adds them. Connector service only; Slack is off unless configured.
Log-only, Cloud Run director only. Each runtime line gains per-window deltas of the instance's /proc/net/snmp TCP counters (segments in and out, retransmits, errors, resets), so a stall can be checked against packet re-sends. Unreadable or reset windows are omitted.
The connector cloud now assigns each machine connection's revision instead of trusting the desktop's saved number. Acks, events and polls from an older connection get 409 hello_required, and the desktop announces itself again. Also records the cloud's request size limit in the shared contract and adds the current turn to the machine snapshot so a stale turn end cannot clear a newer prompt. Connector service only; not deployed to production.
Adds the Slack side of the Orca Cloud connector: verified Slack event ingress, owner-only routing, thread starts held in the shared inbox, a post queue for replies and status, and agent replies into the thread. Off unless the connector is configured with Slack secrets; no production deployment.
Director only. When one pooled connection is far slower than its peers (median over 25 ms and 3x theirs, or third-slowest statement in 60 s over 200 ms and 5x every peer's slowest), it is destroyed on release and the pool dials a fresh one. At most one per pool per 10 s; never mid-transaction; no change in connection count. Fixes the 10-26 director stalls where one connection stalled the sticky lane.
Moves the push gateway's database layer into a shared cloud/packages/cloud-database package (service name and schema passed in). Push's behaviour, log names, SQLite file and tests are unchanged.
<!-- orca-pr-loc -->
<!-- Programmatic LoC summary. Do not edit by hand; rewritten on every commit. -->
| | Files | Added | Deleted | Net |
| :--- | ---: | ---: | ---: | ---: |
| Test | 4 | $\color{#1a7f37}{\Huge{\mathbf{+}}}$406 | $\color{#cf222e}{\Huge{\mathbf{−}}}$4 | $\color{#1a7f37}{\Huge{\mathbf{+}}}$402 |
| Prod | 6 | $\color{#1a7f37}{\Huge{\mathbf{+}}}$165 | $\color{#cf222e}{\Huge{\mathbf{−}}}$10 | $\color{#1a7f37}{\Huge{\mathbf{+}}}$155 |
<!-- /orca-pr-loc -->
## ELI5
Relay servers ("cells") normally get their seat bookings from the front door (the director). If the director goes quiet for a minute or two, each cell switches itself to an older mode that talks to the shared Postgres database for every connection (the "dead-man" trip). Today every cell switches at once, so the database budget must keep room for every cell's full connection pool. This PR adds a per-cell switch that caps how many cells may trip at the same time. A cell that cannot trip yet stays in its normal mode and keeps serving the desktops it already has.
## What Changed
- New cell flag `deadManTripSlots` (integer 0-64). Absent, which is every cell today, keeps the current behaviour: trip as soon as the dead-man window (60-120 s) passes.
- With the flag set, a cell whose window has passed first claims a slot in Postgres. The claim is one short transaction: it takes an advisory lock without waiting, counts the other reserve cells whose own `relay_cell_admit_effective` row says `db`, and writes this cell's row to `db` only if that count is below the flag. Only then does the cell trip (void bookings, re-register its hosts, run on its full pool).
- A cell that gets no slot is "held". It stays in reserve mode, logs `orca_relay_cell_reserve_dead_man_held` once, keeps writing `reserve` to its row, and asks again every 30 s.
- A slot is held while the holder's own row says `db`, its switch still says `reserve`, and the row was refreshed in the last 10 minutes. A holder that restarts or is re-armed writes `reserve` and frees its slot. One that died tripped frees it after 10 minutes. Heartbeats are not used: they reach Postgres through the director, so in a director outage they all go stale and every slot would read as free.
- A tripped cell skips its 15 s row refresh while its pool has waiters. So that a long brownout cannot age its row out of the count, a tripped cell's row older than 5 minutes is refreshed even when the pool is busy.
- A trip stays latched after the director returns, as today. The tripped cells keep their slots until an operator re-arms them with a new reserve write. Until then, a second outage trips nobody new: the other cells stay held. That is the intended bound, but re-arming after an outage now also frees the slots.
- The claim goes through `CellAdmitEffectiveWriter`, which already serialises every write of that row. A `reserve` write still in flight therefore cannot land after the claim and hide the slot from the next claimant.
- Cells keep their existing per-cell window spread (60-120 s by cell-id hash), so the earliest windows claim first.
What a held cell does during a director outage (unchanged code paths, now for longer than the window):
- Desktops already connected keep their sockets. Nothing is closed or re-registered.
- A desktop that drops and redials at the same epoch is admitted from memory (`seatedHereAt`). Any other hello falls through to the normal database check, as it does in reserve mode today.
- The database placement path places no new desktops on it, because its row says `reserve`. It also never re-places the held cell's desktops, because `liveReserveCell` keys on the switch (`relay_cell_admit_modes`), not on this row.
- New placements during the outage go to cells already in database mode (c32, c33 and c34 today) plus the tripped slot holders.
## Why
The database budget (Cloud SQL `orca-cloud-auth-db`, shared with auth, max_connections 500, 490 usable) currently reserves every relay cell's full pool. The only reason for that is the dead-man trip of the whole fleet at once. Spare is 35 on main and about 3 once C36/C37 land, which caps the number of Asia cells.
Budget arithmetic (`relay-cloud-sql-connection-budget.mjs` on orca-cloud main: directors 15, auth 20, API 50, rollout overlap 65, maintenance 5, explicit reserve 10):
| | Cells line | Operating | Spare |
|---|---:|---:|---:|
| Main today: 6 db/legacy cells at 10, 16 US at 8, 7 Asia at 16 | 300 | 455 | 35 |
| Plus C36/C37 at 16 | 332 | 487 | 3 |
| Same fleet, `deadManTripSlots=2`, Asia reserve cells budgeted at 12 (measured peak) | 76 + 128 + 8x12 + 2x4 = 308 | 463 | 27 |
| Plus two more Asia cells at 12 | 332 | 487 | 3 |
So with two slots, two more Asia cells fit in the room C36/C37 leave today. Two caveats keep this honest:
- The bound holds only once every reserve cell runs this image with the same flag value. An older image ignores the flag and trips freely, though its `db` row still takes a slot from the capped cells. With mixed values the cap is the largest value set.
- Counting reserve cells at their measured peak is a softer guarantee than counting their configured pool: the pool is still 16, and a reserve cell could burst past 12. Making it a hard bound needs a pool ceiling that depends on the mode (reserve-mode ceiling 12, the full 16 only for a slot holder or a deliberate db cell). That is a follow-up; this PR does not change pool sizes.
Alternatives considered:
- Longer or ranked trip delays only spread the trips out. A trip is latched until the switch is rewritten, so after the longest delay every cell is still in database mode. That bounds the burst, not the budget.
- A static per-cell "may trip" list in the switch file needs no database coordination, but a listed cell that is down or draining wastes its slot, and an operator has to keep the list current.
- Removing the fallback (cells never trip) is the target design, but it waits on the director stall fix and two weeks of clean running. `deadManTripSlots=0` gives exactly that per cell, so it is the same switch.
Deliberate db flips (for example, flipping reserve cells to db for an image roll) are outside the cap. Budget each Asia cell flipped that way at its full 16. With a two-cell Asia roll batch running during an outage, the slot row above is 316, which leaves room for one more Asia cell rather than two.
Rollout: the flag is dark. Setting it is an `Operate Relay Production Cell Flags` write, per cell, after every reserve cell runs this image.
## Linked Issue
N/A (internal relay capacity work).
## Visual Proof
N/A: relay server behaviour only, no UI.
## Testing
- `cell-reserve-dead-man.test.ts` (new): trips as before without the flag; held until a permit, then latched; slot 0 never trips; a permit for a silence that contact ended, or for an earlier switch generation, is dropped.
- `cell-admit-effective-writer.test.ts`: the claim waits for an in-flight reserve write; a refused claim keeps the reserve row fresh and retries after 30 s; no claim while the pool has waiters or the switch is off reserve; a tripped row is refreshed past a saturated pool after 5 minutes, a reserve row is not.
- `cell-reserve-dead-man-slots.blackbox.test.ts` (new, real cell on SQLite): a director outage with the only slot taken by another cell that writes nothing but its own row: the cell stays reserve across several claims, re-registers nothing, keeps its desktops connected, and lets a redialing desktop back in from memory; once the peer writes `reserve`, the cell trips and re-registers. Also: slot 0 rides out 30 minutes of outage without tripping; no flag trips as before.
- `cell-dead-man-trip-slot-postgres.test.ts` (new, PostgreSQL): eight cells claiming together get exactly two slots; a live holder keeps its slot; a stale one frees exactly one. Removing the advisory lock fails this test.
- Mutation checks: making the slot count always 0, or removing the lock, each fails a test.
- Full relay suite, including the PostgreSQL suites against a local PostgreSQL 17: 173 files, 1564 passed, 3 skipped. The pr.yml static-analysis steps pass locally.
- Two rounds of adversarial review. Round 1 found that the first version counted holders by director-written heartbeats (fixed). Round 2 found the brownout aging (fixed) and the latched slots after an outage (documented above).
- [x] I manually tested these changes locally
- [x] Automated tests added/updated, or explained why not below
## Review
## Agent skill upstream boundary
- [x] Not applicable, or this change copies or mechanically translates no upstream skill-installer source, tests, fixtures, registry entries, path tables, comments, or documentation.
## Notes
No new table or column. The claim reads `relay_cell_admit_effective` and `relay_cell_admit_modes` (one row per cell). Under a fleet-wide trip, held cells add one short transaction each per 30 s; the advisory lock is held for about one round trip.
## Checklist
- [x] This PR is small and focused
- [x] I explained what changed and why (ELI5, the user-facing before/after, the mechanism, and why over the alternatives)
- [x] Before/after screenshots or videos attached for UI changes, or `N/A` with reason
- [x] Self-reviewed for correctness, security, and performance
- [x] Cross-platform, SSH/remote, and path/shortcut impact considered (or N/A)
- [x] `pnpm lint`, `pnpm typecheck`, `pnpm test`, and `pnpm build` pass (or CI will cover; local preferred)