mirror of
https://github.com/stablyai/orca.git
synced 2026-10-06 00:02:43 +00:00
fix(ssh): stop reporting a confirmed kill when the SSH provider is gone (#14977)
* fix(ssh): stop reporting a confirmed kill when the SSH provider is gone A detached relay PTY is designed to outlive the provider that addressed it (it ignores SIGHUP and ships with an unlimited grace), so "the SSH provider is no longer registered" is lost contact, never evidence the remote process stopped. Both stop primitives in the PTY controller returned `true` from that branch, and every caller downstream reported the fabricated success: the CLI printed "PTY killed.", worker-stop settled the dispatch as stopped, and — because the stop "succeeded" — the unstopped-PTY gate never ran, so worktree removal walked straight past a live remote agent. `kill`/`stopAndWait` now still tombstone the local lease but report an unconfirmed stop and record why, using the three-verdict vocabulary the worktree teardown gate already spoke (`live` / `unverifiable` / `exited`), promoted out of that module into `src/shared/pty-liveness-verdict.ts`. The close receipt, the CLI wording, worker-stop and the removal gate all read that verdict instead of inferring an exit from silence. The same rule fixes the mirror-image defect: the aggregate inventory only enumerates registered providers, so a dropped relay clears `connected` for every remote PTY at once. The sweep now separates the provider answering "absent" (an exit) from no provider being able to answer (lost contact), so worker-stop stops claiming `exited` from a disconnect. The `connected` wire field is unchanged in meaning and shape. * fix(orchestration): apply the same honesty to the federation stop path The federation host runs its own copy of the worker observation and stop logic, with the same two defects: `inspectRemoteAttachment` read a dropped relay's `connected: false` as `exited`, and `federationStop` settled the dispatch as stopped from a close it never confirmed — relaying a fabricated success all the way home to the coordinator. Two guards also had to move so the honest verdict does not become a new refusal. `federationRead` gated on `status !== 'running'`, which would have rejected a connected terminal the moment a stop lost contact with it; it now gates on `status === 'exited'`, which is equivalent for every pre-existing status given the two guards beside it. Local `workerStop` likewise still attempts the close when the verdict is `unverifiable` — losing contact is a reason to report the outcome honestly, never a reason to stop trying. The show observations now carry the reason alongside the status, so a bare `unverifiable` is actionable. Both are new optional fields. * fix(ssh): preserve unconfirmed stop verdicts across consumers * fix(ssh): use canonical live verdict wording * fix(ssh): refuse wrong-host teardown verification * test(orchestration): confirm worker release teardown * fix(orchestration): negotiate honest worker stop receipts * fix(agent-teams): fence uncertain teammate respawns * fix(ssh): avoid duplicate missing-provider teardown * fix(orchestration): preserve archives across release retries * fix(ssh): preserve verdicts across synthetic kill exits * fix(ssh): preserve liveness evidence across teardown * fix(agent-teams): replace panes only after confirmed stop * fix(ssh): distinguish host exits from relay loss * fix(ssh): narrow concurrent inventory verdicts * fix(orchestration): serve archives after uncertain release * fix(orchestration): expose unverifiable read liveness * test(ssh): align liveness assertions with verdicts * fix(ssh): preserve host scope across inventory failures
This commit is contained in:
@@ -189,7 +189,7 @@ describe('ClaudeAgentTeamsService', () => {
|
||||
expect(api.closeTerminal).toHaveBeenLastCalledWith('teammate-2')
|
||||
})
|
||||
|
||||
it('keeps the placeholder handle when the respawn split fails', async () => {
|
||||
it('removes the pane when replacement split fails after a confirmed placeholder stop', async () => {
|
||||
const { service, teamId, token, leaderPane, api } = createServiceWithLeader()
|
||||
const request = (argv: string[], envPane = leaderPane) =>
|
||||
service.handleTmuxCompat({ teamId, token, envPane, argv }, api)
|
||||
@@ -212,10 +212,59 @@ describe('ClaudeAgentTeamsService', () => {
|
||||
request(['respawn-pane', '-k', '-t', '%2', '--', 'claude --agent-id a'])
|
||||
).resolves.toMatchObject({ ok: false, exitCode: 1 })
|
||||
|
||||
// the placeholder terminal is left intact and the fake pane id still resolves.
|
||||
expect(api.closeTerminal).not.toHaveBeenCalled()
|
||||
await request(['kill-pane', '-t', '%2'])
|
||||
expect(api.closeTerminal).toHaveBeenCalledWith('teammate-1')
|
||||
await expect(
|
||||
request(['list-panes', '-t', 'orca:0', '-F', '#{pane_id}'])
|
||||
).resolves.toMatchObject({ stdout: '%1\n' })
|
||||
})
|
||||
|
||||
it('keeps a pane registered when its process stop is unconfirmed', async () => {
|
||||
const { service, teamId, token, leaderPane, api } = createServiceWithLeader()
|
||||
const request = (argv: string[], envPane = leaderPane) =>
|
||||
service.handleTmuxCompat({ teamId, token, envPane, argv }, api)
|
||||
|
||||
await request(['split-window', '-t', leaderPane, '-h', '-P', '-F', '#{pane_id}'])
|
||||
vi.mocked(api.closeTerminal).mockResolvedValueOnce({
|
||||
handle: 'teammate-1',
|
||||
tabId: 'tab-1',
|
||||
ptyKilled: false
|
||||
})
|
||||
|
||||
await expect(request(['kill-pane', '-t', '%2'])).resolves.toMatchObject({
|
||||
ok: false,
|
||||
exitCode: 1
|
||||
})
|
||||
await expect(
|
||||
request(['list-panes', '-t', 'orca:0', '-F', '#{pane_id}'])
|
||||
).resolves.toMatchObject({ stdout: '%1\n%2\n' })
|
||||
})
|
||||
|
||||
it('does not launch a replacement when the placeholder stop is unconfirmed', async () => {
|
||||
const { service, teamId, token, leaderPane, api, splitCalls } = createServiceWithLeader()
|
||||
const request = (argv: string[], envPane = leaderPane) =>
|
||||
service.handleTmuxCompat({ teamId, token, envPane, argv }, api)
|
||||
|
||||
await request(['split-window', '-t', leaderPane, '-h', '-P', '-F', '#{pane_id}', 'cat'])
|
||||
vi.mocked(api.closeTerminal).mockResolvedValueOnce({
|
||||
handle: 'teammate-1',
|
||||
tabId: 'tab-1',
|
||||
ptyKilled: false,
|
||||
ptyStopVerdict: 'live'
|
||||
})
|
||||
|
||||
await expect(
|
||||
request(['respawn-pane', '-k', '-t', '%2', '--', 'claude --agent-id a'])
|
||||
).resolves.toMatchObject({ ok: false, exitCode: 1 })
|
||||
expect(api.closeTerminal).toHaveBeenNthCalledWith(1, 'teammate-1')
|
||||
expect(splitCalls).toHaveLength(1)
|
||||
|
||||
await expect(
|
||||
request(['respawn-pane', '-k', '-t', '%2', '--', 'claude --agent-id a'])
|
||||
).resolves.toMatchObject({ ok: false, exitCode: 1 })
|
||||
expect(splitCalls).toHaveLength(1)
|
||||
|
||||
await request(['kill-pane', '-t', '%2'])
|
||||
expect(api.closeTerminal).toHaveBeenLastCalledWith('teammate-1')
|
||||
})
|
||||
|
||||
it('refuses to respawn the leader pane', async () => {
|
||||
|
||||
@@ -4,6 +4,7 @@ import {
|
||||
tmuxSendKeysText,
|
||||
tmuxValue
|
||||
} from '../../shared/claude-agent-teams-tmux-compat'
|
||||
import { describeUnconfirmedAgentStop } from '../../shared/pty-liveness-verdict'
|
||||
import {
|
||||
formatContext,
|
||||
paneEnv,
|
||||
@@ -157,27 +158,31 @@ export class ClaudeAgentTeamsTmuxDispatcher {
|
||||
if (!command) {
|
||||
return ''
|
||||
}
|
||||
if (pane.respawnBlockedReason) {
|
||||
throw new Error(pane.respawnBlockedReason)
|
||||
}
|
||||
const origin =
|
||||
(pane.splitFromPane ? team.panes.get(pane.splitFromPane) : undefined) ??
|
||||
team.panes.get(team.leaderPane)!
|
||||
// Why: create the replacement before destroying the placeholder so a failed
|
||||
// split leaves the fake pane id pointing at a still-live terminal; on cleanup
|
||||
// failure, discard the new split and keep the placeholder registered.
|
||||
const previousHandle = pane.handle
|
||||
const split = await api.splitTerminal(origin.handle, {
|
||||
direction: pane.splitDirection ?? 'horizontal',
|
||||
command,
|
||||
env: paneEnv(team, pane.fakePaneId),
|
||||
envToDelete: ['TERM_PROGRAM'],
|
||||
activate: false
|
||||
})
|
||||
const close = await api.closeTerminal(previousHandle)
|
||||
if (!close.ptyKilled) {
|
||||
pane.respawnBlockedReason = describeUnconfirmedAgentStop(close)
|
||||
throw new Error(pane.respawnBlockedReason)
|
||||
}
|
||||
try {
|
||||
await api.closeTerminal(previousHandle)
|
||||
const split = await api.splitTerminal(origin.handle, {
|
||||
direction: pane.splitDirection ?? 'horizontal',
|
||||
command,
|
||||
env: paneEnv(team, pane.fakePaneId),
|
||||
envToDelete: ['TERM_PROGRAM'],
|
||||
activate: false
|
||||
})
|
||||
pane.handle = split.handle
|
||||
} catch (error) {
|
||||
await api.closeTerminal(split.handle).catch(() => {})
|
||||
this.removePane(team, pane)
|
||||
throw error
|
||||
}
|
||||
pane.handle = split.handle
|
||||
return ''
|
||||
}
|
||||
|
||||
@@ -266,14 +271,21 @@ export class ClaudeAgentTeamsTmuxDispatcher {
|
||||
if (pane.fakePaneId === team.leaderPane) {
|
||||
throw new Error('refusing to kill leader pane')
|
||||
}
|
||||
await api.closeTerminal(pane.handle)
|
||||
const close = await api.closeTerminal(pane.handle)
|
||||
if (!close.ptyKilled) {
|
||||
throw new Error(describeUnconfirmedAgentStop(close))
|
||||
}
|
||||
this.removePane(team, pane)
|
||||
return ''
|
||||
}
|
||||
|
||||
private removePane(team: AgentTeam, pane: TeamPane): void {
|
||||
team.panes.delete(pane.fakePaneId)
|
||||
team.paneOrder = team.paneOrder.filter((id) => id !== pane.fakePaneId)
|
||||
if (team.mainVertical?.lastColumnPane === pane.fakePaneId) {
|
||||
team.mainVertical.lastColumnPane =
|
||||
[...team.paneOrder].toReversed().find((id) => id !== team.leaderPane) ?? null
|
||||
}
|
||||
return ''
|
||||
}
|
||||
|
||||
private async lastPane(
|
||||
|
||||
@@ -59,6 +59,7 @@ export type TeamPane = {
|
||||
// respawn can recreate it in the same slot while preserving its fake pane id.
|
||||
splitFromPane?: string
|
||||
splitDirection?: 'horizontal' | 'vertical'
|
||||
respawnBlockedReason?: string
|
||||
}
|
||||
|
||||
export type AgentTeam = {
|
||||
|
||||
@@ -35,12 +35,12 @@ describe('terminal process incarnation liveness', () => {
|
||||
).resolves.toBe('live')
|
||||
await expect(
|
||||
runtime.inspectTerminalProcessIncarnationLiveness('remote:ssh-1:pty-1:inc-old', SSH_SCOPE)
|
||||
).resolves.toBe('dead')
|
||||
).resolves.toBe('exited')
|
||||
expect(listProcesses).toHaveBeenNthCalledWith(1, 'ssh-1')
|
||||
expect(listProcesses).toHaveBeenNthCalledWith(2, 'ssh-1')
|
||||
})
|
||||
|
||||
it('keeps missing or malformed identity and unavailable inventory unknown', async () => {
|
||||
it('keeps missing or malformed identity and unavailable inventory unverifiable', async () => {
|
||||
const listProcesses = vi
|
||||
.fn()
|
||||
.mockResolvedValueOnce([{ id: 'remote:ssh-1:pty-1', cwd: '', title: 'worker' }])
|
||||
@@ -52,13 +52,13 @@ describe('terminal process incarnation liveness', () => {
|
||||
|
||||
await expect(
|
||||
runtime.inspectTerminalProcessIncarnationLiveness(PROCESS_INCARNATION, SSH_SCOPE)
|
||||
).resolves.toBe('unknown')
|
||||
).resolves.toBe('unverifiable')
|
||||
await expect(
|
||||
runtime.inspectTerminalProcessIncarnationLiveness(PROCESS_INCARNATION, SSH_SCOPE)
|
||||
).resolves.toBe('unknown')
|
||||
).resolves.toBe('unverifiable')
|
||||
await expect(
|
||||
runtime.inspectTerminalProcessIncarnationLiveness(PROCESS_INCARNATION, SSH_SCOPE)
|
||||
).resolves.toBe('unknown')
|
||||
).resolves.toBe('unverifiable')
|
||||
})
|
||||
|
||||
it('does not inspect an unproven host scope', async () => {
|
||||
@@ -67,10 +67,10 @@ describe('terminal process incarnation liveness', () => {
|
||||
|
||||
await expect(
|
||||
runtime.inspectTerminalProcessIncarnationLiveness(PROCESS_INCARNATION, null)
|
||||
).resolves.toBe('unknown')
|
||||
).resolves.toBe('unverifiable')
|
||||
await expect(
|
||||
runtime.inspectTerminalProcessIncarnationLiveness(PROCESS_INCARNATION, '{"kind":"ssh"}')
|
||||
).resolves.toBe('unknown')
|
||||
).resolves.toBe('unverifiable')
|
||||
expect(listProcesses).not.toHaveBeenCalled()
|
||||
})
|
||||
|
||||
@@ -83,7 +83,7 @@ describe('terminal process incarnation liveness', () => {
|
||||
|
||||
await expect(
|
||||
runtime.inspectTerminalProcessIncarnationLiveness('local-pty:inc-1', JSON.stringify(scope))
|
||||
).resolves.toBe('dead')
|
||||
).resolves.toBe('exited')
|
||||
expect(listProcesses).toHaveBeenCalledWith(connectionId)
|
||||
})
|
||||
})
|
||||
|
||||
@@ -311,7 +311,7 @@ describe('terminal close and handle incarnation continuity', () => {
|
||||
|
||||
harness.retirePersistedTab()
|
||||
harness.acknowledged.resolve()
|
||||
await expect(closing).resolves.toMatchObject({ handle, tabId: TAB_ID, ptyKilled: true })
|
||||
await expect(closing).resolves.toMatchObject({ handle, tabId: TAB_ID, ptyKilled: false })
|
||||
expect(harness.kill).toHaveBeenCalledWith(PTY_ID)
|
||||
expect(harness.closeTerminal).not.toHaveBeenCalled()
|
||||
expect(harness.getSession().tabsByWorktree[WORKTREE_ID]).toEqual([])
|
||||
@@ -328,7 +328,7 @@ describe('terminal close and handle incarnation continuity', () => {
|
||||
expect(harness.getSession().tabsByWorktree[WORKTREE_ID]).toHaveLength(1)
|
||||
})
|
||||
|
||||
it('kills every live tab PTY after retirement when the renderer graph is stale', async () => {
|
||||
it('requests a stop for every live tab PTY after retirement when the renderer graph is stale', async () => {
|
||||
const harness = createHarness()
|
||||
harness.syncSplitFixtureGraph()
|
||||
const terminal = (await harness.runtime.listTerminals(`id:${WORKTREE_ID}`)).terminals.find(
|
||||
@@ -342,7 +342,7 @@ describe('terminal close and handle incarnation continuity', () => {
|
||||
|
||||
harness.retirePersistedTab()
|
||||
harness.acknowledged.resolve()
|
||||
await expect(closing).resolves.toMatchObject({ ptyKilled: true })
|
||||
await expect(closing).resolves.toMatchObject({ ptyKilled: false })
|
||||
expect(harness.kill).toHaveBeenCalledWith(PTY_ID)
|
||||
expect(harness.kill).toHaveBeenCalledWith(SIBLING_PTY_ID)
|
||||
})
|
||||
@@ -365,7 +365,64 @@ describe('terminal close and handle incarnation continuity', () => {
|
||||
expect(harness.kill).not.toHaveBeenCalled()
|
||||
})
|
||||
|
||||
it('falls back to kill when verified teardown rejects after retirement', async () => {
|
||||
it('reports an unconfirmed stop on the close receipt rather than a bare uncertain false', async () => {
|
||||
const harness = createHarness()
|
||||
const [{ handle }] = (await harness.runtime.listTerminals(`id:${WORKTREE_ID}`)).terminals
|
||||
// Mirrors pty.ts when the SSH provider is gone: the lease is tombstoned, the
|
||||
// stop reports false, and the PTY is marked as contact we lost — not a kill.
|
||||
harness.kill.mockReturnValue(false)
|
||||
harness.setVerifiedStopResult(false)
|
||||
harness.runtime.markPtyLivenessUnverifiable(PTY_ID, 'its SSH provider is no longer registered')
|
||||
|
||||
const closing = harness.runtime.closeTerminal(handle)
|
||||
await vi.waitFor(() => expect(harness.closeTerminalTab).toHaveBeenCalled())
|
||||
harness.retirePersistedTab()
|
||||
harness.acknowledged.resolve()
|
||||
|
||||
await expect(closing).resolves.toMatchObject({
|
||||
ptyKilled: false,
|
||||
ptyStopVerdict: 'unverifiable',
|
||||
ptyStopReason: 'its SSH provider is no longer registered'
|
||||
})
|
||||
expect(harness.kill).not.toHaveBeenCalled()
|
||||
})
|
||||
|
||||
it('downgrades a live verdict after issuing an unverified follow-up stop', async () => {
|
||||
const harness = createHarness()
|
||||
const [{ handle }] = (await harness.runtime.listTerminals(`id:${WORKTREE_ID}`)).terminals
|
||||
harness.setVerifiedStopResult(false)
|
||||
harness.runtime.markPtyLivenessLive(PTY_ID)
|
||||
|
||||
const closing = harness.runtime.closeTerminal(handle)
|
||||
await vi.waitFor(() => expect(harness.closeTerminalTab).toHaveBeenCalled())
|
||||
harness.retirePersistedTab()
|
||||
harness.acknowledged.resolve()
|
||||
|
||||
await expect(closing).resolves.toMatchObject({
|
||||
ptyKilled: false,
|
||||
ptyStopVerdict: 'unverifiable',
|
||||
ptyStopReason: 'a follow-up stop was issued but its outcome could not be verified'
|
||||
})
|
||||
expect(harness.kill).toHaveBeenCalledWith(PTY_ID)
|
||||
})
|
||||
|
||||
it('leaves a confirmed kill receipt free of any stop verdict', async () => {
|
||||
const harness = createHarness()
|
||||
const [{ handle }] = (await harness.runtime.listTerminals(`id:${WORKTREE_ID}`)).terminals
|
||||
harness.setVerifiedStopResult(true)
|
||||
|
||||
const closing = harness.runtime.closeTerminal(handle)
|
||||
await vi.waitFor(() => expect(harness.closeTerminalTab).toHaveBeenCalled())
|
||||
harness.retirePersistedTab()
|
||||
harness.acknowledged.resolve()
|
||||
|
||||
const close = await closing
|
||||
expect(close.ptyKilled).toBe(true)
|
||||
expect(close.ptyStopVerdict).toBeUndefined()
|
||||
expect(close.ptyStopReason).toBeUndefined()
|
||||
})
|
||||
|
||||
it('reports an unconfirmed stop when verified teardown rejects after retirement', async () => {
|
||||
const harness = createHarness()
|
||||
const [{ handle }] = (await harness.runtime.listTerminals(`id:${WORKTREE_ID}`)).terminals
|
||||
harness.setVerifiedStopResult(new Error('provider_unavailable'))
|
||||
@@ -375,7 +432,11 @@ describe('terminal close and handle incarnation continuity', () => {
|
||||
harness.retirePersistedTab()
|
||||
harness.acknowledged.resolve()
|
||||
|
||||
await expect(closing).resolves.toMatchObject({ ptyKilled: true })
|
||||
await expect(closing).resolves.toMatchObject({
|
||||
ptyKilled: false,
|
||||
ptyStopVerdict: 'unverifiable',
|
||||
ptyStopReason: 'provider_unavailable'
|
||||
})
|
||||
expect(harness.stopAndWait).toHaveBeenCalledWith(PTY_ID, {
|
||||
deadlineMs: expect.any(Number)
|
||||
})
|
||||
@@ -398,7 +459,7 @@ describe('terminal close and handle incarnation continuity', () => {
|
||||
harness.makeSessionUnavailable()
|
||||
harness.acknowledged.resolve()
|
||||
|
||||
await expect(closing).resolves.toMatchObject({ handle, tabId: TAB_ID, ptyKilled: true })
|
||||
await expect(closing).resolves.toMatchObject({ handle, tabId: TAB_ID, ptyKilled: false })
|
||||
expect(harness.closeTerminal).toHaveBeenCalledWith(TAB_ID)
|
||||
expect(harness.stopAndWait).toHaveBeenCalledWith(RUNTIME_OWNED_PTY_ID, {
|
||||
deadlineMs: expect.any(Number)
|
||||
|
||||
@@ -348,7 +348,7 @@ describe('remote runtime terminal split authority', () => {
|
||||
expect.objectContaining({ deadlineMs: expect.any(Number) })
|
||||
)
|
||||
expect(harness.kill).not.toHaveBeenCalled()
|
||||
expect(harness.retireRejectedPty).toHaveBeenCalledWith(SPLIT_PTY_ID)
|
||||
expect(harness.retireRejectedPty).toHaveBeenCalledWith(SPLIT_PTY_ID, true)
|
||||
})
|
||||
|
||||
it('revalidates a projected paired-runtime source after renderer adoption', async () => {
|
||||
@@ -374,7 +374,7 @@ describe('remote runtime terminal split authority', () => {
|
||||
expect.objectContaining({ deadlineMs: expect.any(Number) })
|
||||
)
|
||||
expect(harness.kill).toHaveBeenCalledWith(SPLIT_PTY_ID)
|
||||
expect(harness.retireRejectedPty).toHaveBeenCalledWith(SPLIT_PTY_ID)
|
||||
expect(harness.retireRejectedPty).toHaveBeenCalledWith(SPLIT_PTY_ID, false)
|
||||
})
|
||||
|
||||
it('preserves the split error when kill and retirement throw', async () => {
|
||||
@@ -398,6 +398,6 @@ describe('remote runtime terminal split authority', () => {
|
||||
|
||||
await expect(split).rejects.toThrow('terminal_split_source_not_found')
|
||||
expect(harness.kill).toHaveBeenCalledWith(SPLIT_PTY_ID)
|
||||
expect(harness.retireRejectedPty).toHaveBeenCalledWith(SPLIT_PTY_ID)
|
||||
expect(harness.retireRejectedPty).toHaveBeenCalledWith(SPLIT_PTY_ID, false)
|
||||
})
|
||||
})
|
||||
|
||||
@@ -664,6 +664,12 @@ import {
|
||||
type RetiredTerminalSurface
|
||||
} from './mobile-session-terminal-retirement'
|
||||
import { retireTerminalSurfaceFromPersistence } from './mobile-session-terminal-persistence-retirement'
|
||||
import {
|
||||
NO_OBSERVING_PROVIDER_REASON,
|
||||
SSH_EXIT_UNCONFIRMED_REASON,
|
||||
SSH_PROVIDER_UNREGISTERED_REASON,
|
||||
type PtyLivenessVerdict
|
||||
} from '../../shared/pty-liveness-verdict'
|
||||
import {
|
||||
advanceTerminalTopologyRevision,
|
||||
hasHostAuthoritativeTerminalMembership
|
||||
@@ -1568,6 +1574,14 @@ function isAgentSessionOperationOutcomeUnknown(error: unknown): boolean {
|
||||
)
|
||||
}
|
||||
|
||||
// Orphaned verdicts are bounded; active PTYs retain theirs until new evidence resolves them.
|
||||
const MAX_TRACKED_PTY_LIVENESS_VERDICTS = 256
|
||||
|
||||
type TrackedPtyLivenessVerdict = {
|
||||
verdict: PtyLivenessVerdict
|
||||
observedAt: number
|
||||
}
|
||||
|
||||
const AGENT_SESSION_OPERATION_PER_CLIENT_LIMIT = 512
|
||||
const AGENT_SESSION_OPERATION_GLOBAL_LIMIT = 4_096
|
||||
|
||||
@@ -1865,7 +1879,7 @@ type RuntimePtyController = {
|
||||
* False on doubt (absent session, SSH-scoped id, non-daemon provider). */
|
||||
attach?(ptyId: string): Promise<boolean>
|
||||
kill(ptyId: string): boolean
|
||||
retireRejectedPty?(ptyId: string): void
|
||||
retireRejectedPty?(ptyId: string, stopConfirmed: boolean): void
|
||||
stopAndWait?(
|
||||
ptyId: string,
|
||||
opts?: { keepHistory?: boolean; deadlineMs?: number }
|
||||
@@ -3206,6 +3220,12 @@ export class OrcaRuntimeService {
|
||||
// iterates them all. Listeners are cleaned up via subscriptionCleanups.
|
||||
private notificationListeners = new Set<(event: MobileNotificationEvent) => void>()
|
||||
private ptysById = new Map<string, RuntimePtyWorktreeRecord>()
|
||||
// Why a separate map: `connected` is a wire field that any inventory gap
|
||||
// clears, so it cannot distinguish an observed exit from lost contact. This
|
||||
// records the last liveness verdict we actually earned, and outlives the pty
|
||||
// record so a close/stop receipt can still say the stop was unconfirmed.
|
||||
private ptyLivenessVerdictByPtyId = new Map<string, TrackedPtyLivenessVerdict>()
|
||||
private ptyLivenessObservationSequence = 0
|
||||
private readonly pairedRendererSessionOwnedPtyIds = new Set<string>()
|
||||
private wslDistroByPtyId = new Map<string, string>()
|
||||
private titleObservationSequence = 0
|
||||
@@ -10579,6 +10599,7 @@ export class OrcaRuntimeService {
|
||||
incarnationId?: PtyIncarnationId,
|
||||
options: { awaitsRegistration?: boolean } = {}
|
||||
): void {
|
||||
this.forgetPtyLivenessVerdict(ptyId)
|
||||
if (options.awaitsRegistration !== false) {
|
||||
// Why: surface absence cannot distinguish an in-flight admission from a completed headless lifecycle.
|
||||
this.pendingPtyRegistrationIncarnations.set(ptyId, incarnationId ?? null)
|
||||
@@ -10612,6 +10633,7 @@ export class OrcaRuntimeService {
|
||||
isWsl?: boolean
|
||||
): void {
|
||||
this.assertPtyDidNotExitBeforeRegistration(ptyId, binding?.incarnationId)
|
||||
this.forgetPtyLivenessVerdict(ptyId)
|
||||
this.spawnPublishedPtys.add(ptyId)
|
||||
// Why: record the renderer pane identity at spawn time so a stalled graph
|
||||
// sync can't hide that a live PTY already backs a pending mobile create.
|
||||
@@ -14906,14 +14928,25 @@ export class OrcaRuntimeService {
|
||||
}
|
||||
}
|
||||
|
||||
onPtyExit(ptyId: string, exitCode: number, exitIncarnationId?: PtyIncarnationId): void {
|
||||
onPtyExit(
|
||||
ptyId: string,
|
||||
exitCode: number,
|
||||
exitIncarnationId?: PtyIncarnationId,
|
||||
options?: { hostExitConfirmed?: boolean }
|
||||
): void {
|
||||
const pty = this.ptysById.get(ptyId)
|
||||
if (exitIncarnationId && pty?.incarnationId && exitIncarnationId !== pty.incarnationId) {
|
||||
return
|
||||
}
|
||||
const preservesAbnormalSshSurface =
|
||||
this.isSshOwnedPtyId(ptyId) && pty?.connectionId != null && exitCode < 0
|
||||
this.isSshOwnedPtyId(ptyId) &&
|
||||
pty?.connectionId != null &&
|
||||
exitCode < 0 &&
|
||||
options?.hostExitConfirmed !== true
|
||||
if (preservesAbnormalSshSurface) {
|
||||
if (this.getPtyLivenessVerdict(ptyId)?.status !== 'unverifiable') {
|
||||
this.markPtyLivenessUnverifiable(ptyId, SSH_EXIT_UNCONFIRMED_REASON)
|
||||
}
|
||||
this.restoredOrchestrationAuthorityByPtyId.delete(ptyId)
|
||||
} else {
|
||||
this.retirePtyAgentLaunchAuthority(ptyId)
|
||||
@@ -15047,6 +15080,11 @@ export class OrcaRuntimeService {
|
||||
this.setPairedRendererSessionOwnership(pty.ptyId, false)
|
||||
pty.disconnectedAt = Date.now()
|
||||
pty.lastExitCode = exitCode
|
||||
if (exitCode >= 0 || options?.hostExitConfirmed === true) {
|
||||
// A real wait status from the owning host is the death certificate; the
|
||||
// synthetic -1 we emit on a failed/unroutable stop is not.
|
||||
this.forgetPtyLivenessVerdict(ptyId)
|
||||
}
|
||||
// Why: the exited process's live frames say nothing about a replacement.
|
||||
// A same-id respawn makes the leaf writable again before any new title,
|
||||
// so leaving this true would let push delivery type into the new process
|
||||
@@ -16929,17 +16967,17 @@ export class OrcaRuntimeService {
|
||||
async inspectTerminalProcessIncarnationLiveness(
|
||||
processIncarnation: string,
|
||||
serializedHostScope: string | null
|
||||
): Promise<'live' | 'dead' | 'unknown'> {
|
||||
): Promise<'live' | 'exited' | 'unverifiable'> {
|
||||
const hostScope = parseWorkerTerminalHostScope(serializedHostScope)
|
||||
if (!hostScope || !this.ptyController?.listProcesses) {
|
||||
return 'unknown'
|
||||
return 'unverifiable'
|
||||
}
|
||||
const listed = await withTimeoutResult(
|
||||
this.ptyController.listProcesses(hostScope.kind === 'ssh' ? hostScope.targetId : null),
|
||||
PTY_CONTROLLER_LIST_TIMEOUT_MS
|
||||
)
|
||||
if (!listed.ok) {
|
||||
return 'unknown'
|
||||
return 'unverifiable'
|
||||
}
|
||||
return classifyWorkerTerminalProcessIncarnation(processIncarnation, listed.value)
|
||||
}
|
||||
@@ -17787,6 +17825,68 @@ export class OrcaRuntimeService {
|
||||
return `${this.runtimeId}:${record.ptyId}:${record.ptyGeneration}`
|
||||
}
|
||||
|
||||
/**
|
||||
* Records that we lost contact with a PTY's owning host. Callers must never
|
||||
* read this as an exit: a detached relay PTY is designed to outlive the
|
||||
* provider that addressed it.
|
||||
*/
|
||||
markPtyLivenessUnverifiable(ptyId: string, reason: string): void {
|
||||
this.rememberPtyLivenessVerdict(ptyId, { status: 'unverifiable', reason })
|
||||
}
|
||||
|
||||
markPtyLivenessLive(ptyId: string): void {
|
||||
this.rememberPtyLivenessVerdict(ptyId, { status: 'live', ptyIds: [ptyId] })
|
||||
}
|
||||
|
||||
/** Null when nothing has been observed either way, so callers keep their own default. */
|
||||
getPtyLivenessVerdict(ptyId: string): PtyLivenessVerdict | null {
|
||||
return this.ptyLivenessVerdictByPtyId.get(ptyId)?.verdict ?? null
|
||||
}
|
||||
|
||||
getTerminalLivenessVerdict(handle: string): PtyLivenessVerdict | null {
|
||||
const record = this.getLivePtyForHandle(handle)?.record ?? this.handles.get(handle)
|
||||
return record?.ptyId ? this.getPtyLivenessVerdict(record.ptyId) : null
|
||||
}
|
||||
|
||||
private rememberPtyLivenessVerdict(ptyId: string, verdict: PtyLivenessVerdict): void {
|
||||
if (verdict.status === 'exited') {
|
||||
// An earned death certificate ends the question; nothing left to remember.
|
||||
this.ptyLivenessVerdictByPtyId.delete(ptyId)
|
||||
return
|
||||
}
|
||||
this.ptyLivenessVerdictByPtyId.delete(ptyId)
|
||||
this.ptyLivenessObservationSequence += 1
|
||||
this.ptyLivenessVerdictByPtyId.set(ptyId, {
|
||||
verdict,
|
||||
observedAt: this.ptyLivenessObservationSequence
|
||||
})
|
||||
while (this.ptyLivenessVerdictByPtyId.size > MAX_TRACKED_PTY_LIVENESS_VERDICTS) {
|
||||
let oldestOrphaned: string | null = null
|
||||
for (const candidate of this.ptyLivenessVerdictByPtyId.keys()) {
|
||||
if (
|
||||
!this.ptysById.has(candidate) &&
|
||||
!this.handleByPtyId.has(candidate) &&
|
||||
!this.leafExistsForPty(candidate)
|
||||
) {
|
||||
oldestOrphaned = candidate
|
||||
break
|
||||
}
|
||||
}
|
||||
if (!oldestOrphaned) {
|
||||
return
|
||||
}
|
||||
this.ptyLivenessVerdictByPtyId.delete(oldestOrphaned)
|
||||
}
|
||||
}
|
||||
|
||||
private forgetPtyLivenessVerdict(ptyId: string, observedNoLaterThan?: number): void {
|
||||
const tracked = this.ptyLivenessVerdictByPtyId.get(ptyId)
|
||||
if (observedNoLaterThan !== undefined && tracked && tracked.observedAt > observedNoLaterThan) {
|
||||
return
|
||||
}
|
||||
this.ptyLivenessVerdictByPtyId.delete(ptyId)
|
||||
}
|
||||
|
||||
getExactWorkerProviderSession(
|
||||
handle: string,
|
||||
observedAfter: number
|
||||
@@ -28845,13 +28945,34 @@ export class OrcaRuntimeService {
|
||||
let addressedPtyStopped = false
|
||||
const deadlineMs = Date.now() + EXPLICIT_TERMINAL_CLOSE_STOP_TIMEOUT_MS
|
||||
for (const ptyId of ptyIds) {
|
||||
let verifiedStopped = false
|
||||
try {
|
||||
verifiedStopped = (await this.ptyController?.stopAndWait?.(ptyId, { deadlineMs })) ?? false
|
||||
} catch {
|
||||
// Why: verified teardown is preferred, but its transport failure must not suppress the legacy provider kill.
|
||||
let stopped = false
|
||||
if (this.ptyController?.stopAndWait) {
|
||||
try {
|
||||
stopped = await this.ptyController.stopAndWait(ptyId, { deadlineMs })
|
||||
} catch (error) {
|
||||
this.markPtyLivenessUnverifiable(
|
||||
ptyId,
|
||||
error instanceof Error ? error.message : String(error)
|
||||
)
|
||||
}
|
||||
if (!stopped) {
|
||||
const verdict = this.getPtyLivenessVerdict(ptyId)
|
||||
const providerAlreadyRetiredPty =
|
||||
verdict?.status === 'unverifiable' &&
|
||||
verdict.reason === SSH_PROVIDER_UNREGISTERED_REASON
|
||||
if (!providerAlreadyRetiredPty) {
|
||||
this.ptyController.kill(ptyId)
|
||||
if (!verdict || verdict.status === 'live') {
|
||||
this.markPtyLivenessUnverifiable(
|
||||
ptyId,
|
||||
'a follow-up stop was issued but its outcome could not be verified'
|
||||
)
|
||||
}
|
||||
}
|
||||
}
|
||||
} else {
|
||||
stopped = this.ptyController?.kill(ptyId) ?? false
|
||||
}
|
||||
const stopped = verifiedStopped || (this.ptyController?.kill(ptyId) ?? false)
|
||||
if (ptyId === addressedPtyId) {
|
||||
addressedPtyStopped = stopped
|
||||
}
|
||||
@@ -29053,15 +29174,15 @@ export class OrcaRuntimeService {
|
||||
this.notifier.closeTerminal?.(tabId)
|
||||
}
|
||||
const ptyKilled = await this.stopExplicitlyClosedTabPtys(ptyIdsToKill, pty.pty.ptyId)
|
||||
return { handle, tabId, ptyKilled }
|
||||
return this.describeTerminalClose(handle, tabId, pty.pty.ptyId, ptyKilled)
|
||||
}
|
||||
if (siblingCount <= 1 && !surface && pty.pty.tabId && this.notifier?.closeTerminalTab) {
|
||||
const ptyIdsToKill = this.getPtyIdsForExplicitTabClose(pty.pty.worktreeId, tabId)
|
||||
await this.notifier.closeTerminalTab(tabId, { localPtyTeardownOwnedExternally: true })
|
||||
const ptyKilled = await this.stopExplicitlyClosedTabPtys(ptyIdsToKill, pty.pty.ptyId)
|
||||
return { handle, tabId, ptyKilled }
|
||||
return this.describeTerminalClose(handle, tabId, pty.pty.ptyId, ptyKilled)
|
||||
}
|
||||
const ptyKilled = this.ptyController?.kill(pty.pty.ptyId) ?? false
|
||||
const ptyKilled = await this.stopExplicitlyClosedTabPtys([pty.pty.ptyId], pty.pty.ptyId)
|
||||
if (!ptyKilled || siblingCount <= 1) {
|
||||
if (surface) {
|
||||
// Why: paired viewers keep ended streams mounted until the HUB publishes removal, so explicit close uses the durable host-tab transaction instead of viewer-local exit handling.
|
||||
@@ -29077,7 +29198,7 @@ export class OrcaRuntimeService {
|
||||
this.notifier?.closeTerminal(tabId)
|
||||
}
|
||||
}
|
||||
return { handle, tabId, ptyKilled }
|
||||
return this.describeTerminalClose(handle, tabId, pty.pty.ptyId, ptyKilled)
|
||||
}
|
||||
this.assertGraphReady()
|
||||
const { leaf } = this.getLiveLeafForHandle(handle)
|
||||
@@ -29100,7 +29221,36 @@ export class OrcaRuntimeService {
|
||||
if (siblingCount > 1 ? !ptyKilled : !this.notifier?.closeTerminalTab) {
|
||||
this.notifier?.closeTerminal(leaf.tabId, leaf.paneRuntimeId)
|
||||
}
|
||||
return { handle, tabId: leaf.tabId, ptyKilled }
|
||||
return this.describeTerminalClose(handle, leaf.tabId, leaf.ptyId ?? null, ptyKilled)
|
||||
}
|
||||
|
||||
/**
|
||||
* A close receipt must not read as a kill nobody performed: when the stop was
|
||||
* not confirmed, the receipt carries why so the CLI and callers can say so.
|
||||
*/
|
||||
private describeTerminalClose(
|
||||
handle: string,
|
||||
tabId: string,
|
||||
ptyId: string | null,
|
||||
ptyKilled: boolean
|
||||
): RuntimeTerminalClose {
|
||||
if (ptyKilled || !ptyId) {
|
||||
return { handle, tabId, ptyKilled }
|
||||
}
|
||||
const verdict = this.getPtyLivenessVerdict(ptyId)
|
||||
if (verdict?.status === 'unverifiable') {
|
||||
return {
|
||||
handle,
|
||||
tabId,
|
||||
ptyKilled,
|
||||
ptyStopVerdict: 'unverifiable',
|
||||
ptyStopReason: verdict.reason
|
||||
}
|
||||
}
|
||||
if (verdict?.status === 'live') {
|
||||
return { handle, tabId, ptyKilled, ptyStopVerdict: 'live' }
|
||||
}
|
||||
return { handle, tabId, ptyKilled }
|
||||
}
|
||||
|
||||
async closeTerminalTab(handle: string): Promise<RuntimeTerminalClose> {
|
||||
@@ -29331,7 +29481,7 @@ export class OrcaRuntimeService {
|
||||
}
|
||||
}
|
||||
try {
|
||||
this.ptyController.retireRejectedPty?.(result.id)
|
||||
this.ptyController.retireRejectedPty?.(result.id, stopped)
|
||||
} catch {
|
||||
// Best-effort cleanup; preserve the original split authority error.
|
||||
}
|
||||
@@ -31651,6 +31801,7 @@ export class OrcaRuntimeService {
|
||||
}
|
||||
const inventoryGeneration = this.ptyControllerInventorySequence + 1
|
||||
this.ptyControllerInventorySequence = inventoryGeneration
|
||||
const livenessObservationAtStart = this.ptyLivenessObservationSequence
|
||||
const providerKey = typeof connectionId === 'string' ? `ssh:${connectionId}` : 'local'
|
||||
if (connectionId === undefined) {
|
||||
this.ptyControllerAggregateInventoryGeneration = inventoryGeneration
|
||||
@@ -31760,6 +31911,8 @@ export class OrcaRuntimeService {
|
||||
const allLivePtyIds = new Set(sessions.map((session) => session.id))
|
||||
const selectedLivePtyIds = new Set<string>()
|
||||
for (const session of sessions) {
|
||||
// The owning inventory positively observed this PTY again; prior lost-contact doubt is stale.
|
||||
this.forgetPtyLivenessVerdict(session.id, livenessObservationAtStart)
|
||||
const sessionConnectionId =
|
||||
parseAppSshPtyId(session.id)?.connectionId ??
|
||||
(typeof connectionId === 'string' ? connectionId : null)
|
||||
@@ -31862,7 +32015,18 @@ export class OrcaRuntimeService {
|
||||
continue
|
||||
}
|
||||
if (!allLivePtyIds.has(pty.ptyId) && !this.leafExistsForPty(pty.ptyId)) {
|
||||
if (this.ptyController.hasPty?.(pty.ptyId) === true) {
|
||||
const currentVerdict = this.ptyLivenessVerdictByPtyId.get(pty.ptyId)
|
||||
if (
|
||||
currentVerdict &&
|
||||
currentVerdict.observedAt > livenessObservationAtStart &&
|
||||
currentVerdict.verdict.status === 'unverifiable'
|
||||
) {
|
||||
pty.connected = false
|
||||
pty.disconnectedAt ??= Date.now()
|
||||
continue
|
||||
}
|
||||
const observed = this.ptyController.hasPty?.(pty.ptyId)
|
||||
if (observed === true) {
|
||||
// Why: an SSH spawn can become addressable before an overlapping relay list includes it.
|
||||
allLivePtyIds.add(pty.ptyId)
|
||||
if (
|
||||
@@ -31873,10 +32037,19 @@ export class OrcaRuntimeService {
|
||||
}
|
||||
pty.connected = true
|
||||
pty.disconnectedAt = null
|
||||
this.forgetPtyLivenessVerdict(pty.ptyId)
|
||||
continue
|
||||
}
|
||||
pty.connected = false
|
||||
pty.disconnectedAt ??= Date.now()
|
||||
// Why: this list only enumerates registered providers, so a dropped relay
|
||||
// clears `connected` for every one of its PTYs at once. Only `false` here
|
||||
// is an observed absence; `null` means no provider could be asked.
|
||||
if (observed === false) {
|
||||
this.forgetPtyLivenessVerdict(pty.ptyId)
|
||||
} else if (observed === null) {
|
||||
this.markPtyLivenessUnverifiable(pty.ptyId, NO_OBSERVING_PROVIDER_REASON)
|
||||
}
|
||||
}
|
||||
}
|
||||
this.pruneDisconnectedPtyRecords()
|
||||
@@ -31958,6 +32131,7 @@ export class OrcaRuntimeService {
|
||||
if (pty) {
|
||||
pty.connected = true
|
||||
pty.disconnectedAt = null
|
||||
this.forgetPtyLivenessVerdict(ptyId)
|
||||
this.refreshPtyForegroundAgent(ptyId)
|
||||
}
|
||||
} else if (pty && !this.leafExistsForPty(ptyId)) {
|
||||
|
||||
@@ -64,10 +64,7 @@ export function requestWorkerTerminalRelease(
|
||||
this.db.exec('COMMIT')
|
||||
return { disposition: 'retained', resource, reason: 'ownership_transferred' }
|
||||
}
|
||||
if (
|
||||
resource.release_state === 'unknown' ||
|
||||
(resource.release_state === 'retained' && resource.retained_reason === 'user_requested')
|
||||
) {
|
||||
if (resource.release_state === 'retained' && resource.retained_reason === 'user_requested') {
|
||||
this.db.prepare('DELETE FROM worker_terminal_archives WHERE dispatch_id = ?').run(dispatchId)
|
||||
}
|
||||
this.db
|
||||
|
||||
@@ -39,7 +39,7 @@ export function parseWorkerTerminalHostScope(value: string | null): WorkerTermin
|
||||
export function classifyWorkerTerminalProcessIncarnation(
|
||||
processIncarnation: string,
|
||||
sessions: readonly PtyProcessInfo[]
|
||||
): 'live' | 'dead' | 'unknown' {
|
||||
): 'live' | 'exited' | 'unverifiable' {
|
||||
const possibleMatches = sessions.filter((session) =>
|
||||
processIncarnation.startsWith(`${session.id}:`)
|
||||
)
|
||||
@@ -57,6 +57,6 @@ export function classifyWorkerTerminalProcessIncarnation(
|
||||
return possibleMatches.some(
|
||||
(session) => !session.incarnationId || session.incarnationId !== session.incarnationId.trim()
|
||||
)
|
||||
? 'unknown'
|
||||
: 'dead'
|
||||
? 'unverifiable'
|
||||
: 'exited'
|
||||
}
|
||||
|
||||
@@ -0,0 +1,194 @@
|
||||
import { describe, expect, it, vi } from 'vitest'
|
||||
import { OrcaRuntimeService } from './orca-runtime'
|
||||
import { getDefaultWorkspaceSession } from '../../shared/constants'
|
||||
import { SSH_EXIT_UNCONFIRMED_REASON } from '../../shared/pty-liveness-verdict'
|
||||
|
||||
// The aggregate inventory only enumerates registered providers, so a dropped
|
||||
// relay clears `connected` for every one of its PTYs at once. Only the
|
||||
// provider's own answer separates an observed exit from lost contact.
|
||||
|
||||
const WORKTREE_ID = 'repo-1::/tmp/inventory-verdict'
|
||||
const REMOTE_PTY_ID = 'ssh:conn-1@@relay-9'
|
||||
|
||||
function deferred<T>() {
|
||||
let resolve!: (value: T) => void
|
||||
const promise = new Promise<T>((settle) => {
|
||||
resolve = settle
|
||||
})
|
||||
return { promise, resolve }
|
||||
}
|
||||
|
||||
function makeStore() {
|
||||
const session = getDefaultWorkspaceSession()
|
||||
return {
|
||||
getWorkspaceSession: vi.fn(() => session),
|
||||
setWorkspaceSession: vi.fn(),
|
||||
getRepos: vi.fn(() => [
|
||||
{
|
||||
id: 'repo-1',
|
||||
path: '/tmp/inventory-verdict',
|
||||
displayName: 'inventory-verdict',
|
||||
badgeColor: '#000000',
|
||||
addedAt: 0
|
||||
}
|
||||
]),
|
||||
getAllWorktreeMeta: vi.fn(() => ({})),
|
||||
getWorktreeMeta: vi.fn(() => undefined),
|
||||
setWorktreeMeta: vi.fn(),
|
||||
removeWorktreeMeta: vi.fn(),
|
||||
getSettings: vi.fn(() => ({ workspaceDir: '/tmp/workspaces' })),
|
||||
getProjects: vi.fn(() => [])
|
||||
}
|
||||
}
|
||||
|
||||
function makeRuntimeMissingFromInventory(
|
||||
hasPty: () => boolean | null,
|
||||
listProcesses: () => Promise<{ id: string; worktreeId: string }[]> = vi.fn(async () => [])
|
||||
): OrcaRuntimeService {
|
||||
const runtime = new OrcaRuntimeService(makeStore() as never)
|
||||
runtime.setPtyController({
|
||||
write: () => true,
|
||||
kill: () => true,
|
||||
hasPty,
|
||||
listProcesses,
|
||||
getForegroundProcess: async () => null
|
||||
} as never)
|
||||
runtime.attachWindow(1)
|
||||
runtime.syncWindowGraph(1, { tabs: [], leaves: [] })
|
||||
runtime.registerPty(REMOTE_PTY_ID, WORKTREE_ID, 'conn-1')
|
||||
return runtime
|
||||
}
|
||||
|
||||
describe('inventory sweep liveness verdicts', () => {
|
||||
it('records an abnormal SSH exit as unverifiable at the runtime boundary', () => {
|
||||
const runtime = makeRuntimeMissingFromInventory(() => null)
|
||||
|
||||
runtime.onPtyExit(REMOTE_PTY_ID, -1)
|
||||
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)).toEqual({
|
||||
status: 'unverifiable',
|
||||
reason: SSH_EXIT_UNCONFIRMED_REASON
|
||||
})
|
||||
})
|
||||
|
||||
it('preserves a more specific lost-contact reason across an abnormal SSH exit', () => {
|
||||
const runtime = makeRuntimeMissingFromInventory(() => null)
|
||||
runtime.markPtyLivenessUnverifiable(REMOTE_PTY_ID, 'inventory transport failed')
|
||||
|
||||
runtime.onPtyExit(REMOTE_PTY_ID, -1)
|
||||
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)).toEqual({
|
||||
status: 'unverifiable',
|
||||
reason: 'inventory transport failed'
|
||||
})
|
||||
})
|
||||
|
||||
it('accepts a current owning-host exit even when its numeric code is negative', () => {
|
||||
const runtime = makeRuntimeMissingFromInventory(() => null)
|
||||
runtime.markPtyLivenessUnverifiable(REMOTE_PTY_ID, 'inventory transport failed')
|
||||
|
||||
runtime.onPtyExit(REMOTE_PTY_ID, -1, undefined, { hostExitConfirmed: true })
|
||||
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)).toBeNull()
|
||||
})
|
||||
|
||||
it('records lost contact when no provider can answer for the PTY', async () => {
|
||||
const runtime = makeRuntimeMissingFromInventory(() => null)
|
||||
|
||||
await runtime.listTerminals(`id:${WORKTREE_ID}`)
|
||||
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)).toEqual({
|
||||
status: 'unverifiable',
|
||||
reason: 'no registered provider can observe its host'
|
||||
})
|
||||
})
|
||||
|
||||
it('records no doubt when the owning provider reports the PTY absent', async () => {
|
||||
const runtime = makeRuntimeMissingFromInventory(() => false)
|
||||
|
||||
await runtime.listTerminals(`id:${WORKTREE_ID}`)
|
||||
|
||||
// An observed absence is the death certificate callers already act on.
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)).toBeNull()
|
||||
})
|
||||
|
||||
it('clears lost-contact doubt when reconnect inventory observes the PTY live', async () => {
|
||||
let reconnected = false
|
||||
const runtime = makeRuntimeMissingFromInventory(
|
||||
() => null,
|
||||
vi.fn(async () => (reconnected ? [{ id: REMOTE_PTY_ID, worktreeId: WORKTREE_ID }] : []))
|
||||
)
|
||||
|
||||
await runtime.listTerminals(`id:${WORKTREE_ID}`)
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)?.status).toBe('unverifiable')
|
||||
|
||||
reconnected = true
|
||||
await runtime.listTerminals(`id:${WORKTREE_ID}`)
|
||||
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)).toBeNull()
|
||||
})
|
||||
|
||||
it('does not let a pre-drop inventory clear a newer lost-contact verdict', async () => {
|
||||
const inventory = deferred<{ id: string; worktreeId: string }[]>()
|
||||
const listProcesses = vi.fn(() => inventory.promise)
|
||||
const runtime = makeRuntimeMissingFromInventory(() => null, listProcesses)
|
||||
|
||||
const listing = runtime.listTerminals(`id:${WORKTREE_ID}`)
|
||||
await vi.waitFor(() => expect(listProcesses).toHaveBeenCalled())
|
||||
runtime.markPtyLivenessUnverifiable(REMOTE_PTY_ID, 'relay disconnected during stop')
|
||||
inventory.resolve([{ id: REMOTE_PTY_ID, worktreeId: WORKTREE_ID }])
|
||||
await listing
|
||||
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)).toEqual({
|
||||
status: 'unverifiable',
|
||||
reason: 'relay disconnected during stop'
|
||||
})
|
||||
})
|
||||
|
||||
it('does not let a partial inventory overwrite a concurrent provider failure', async () => {
|
||||
const inventory = deferred<{ id: string; worktreeId: string }[]>()
|
||||
const listProcesses = vi.fn(() => inventory.promise)
|
||||
const runtime = makeRuntimeMissingFromInventory(() => false, listProcesses)
|
||||
|
||||
const listing = runtime.listTerminals(`id:${WORKTREE_ID}`)
|
||||
await vi.waitFor(() => expect(listProcesses).toHaveBeenCalled())
|
||||
runtime.markPtyLivenessUnverifiable(REMOTE_PTY_ID, 'inventory transport failed')
|
||||
inventory.resolve([])
|
||||
await listing
|
||||
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)).toEqual({
|
||||
status: 'unverifiable',
|
||||
reason: 'inventory transport failed'
|
||||
})
|
||||
})
|
||||
|
||||
it('clears stale doubt when a new PTY lifecycle is positively registered', () => {
|
||||
const runtime = makeRuntimeMissingFromInventory(() => null)
|
||||
runtime.markPtyLivenessUnverifiable(REMOTE_PTY_ID, 'old incarnation lost contact')
|
||||
|
||||
runtime.onPtySpawned(REMOTE_PTY_ID, 'incarnation-2')
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)).toBeNull()
|
||||
|
||||
runtime.markPtyLivenessUnverifiable(REMOTE_PTY_ID, 'registration raced reconnect')
|
||||
runtime.registerPty(REMOTE_PTY_ID, WORKTREE_ID, 'conn-1', {
|
||||
tabId: 'tab-new',
|
||||
leafId: '00000000-0000-4000-8000-000000000001',
|
||||
incarnationId: 'incarnation-2'
|
||||
})
|
||||
expect(runtime.getPtyLivenessVerdict(REMOTE_PTY_ID)).toBeNull()
|
||||
})
|
||||
|
||||
it('retains unresolved verdicts for every still-addressable PTY', () => {
|
||||
const runtime = new OrcaRuntimeService(makeStore() as never)
|
||||
for (let index = 0; index < 257; index += 1) {
|
||||
const ptyId = `ssh:conn-1@@relay-${index}`
|
||||
runtime.registerPty(ptyId, WORKTREE_ID, 'conn-1')
|
||||
runtime.markPtyLivenessUnverifiable(ptyId, 'provider disconnected')
|
||||
}
|
||||
|
||||
expect(runtime.getPtyLivenessVerdict('ssh:conn-1@@relay-0')).toEqual({
|
||||
status: 'unverifiable',
|
||||
reason: 'provider disconnected'
|
||||
})
|
||||
})
|
||||
})
|
||||
@@ -6,6 +6,7 @@ import type { RemoteDispatchAttachmentRow } from '../../orchestration/types'
|
||||
import { defineMethod, type RpcMethod } from '../core'
|
||||
import { OptionalFiniteNumber, requiredString } from '../schemas'
|
||||
import { readExactWorkerOutput } from './orchestration-worker-output'
|
||||
import { describeUnconfirmedAgentStop } from '../../../../shared/pty-liveness-verdict'
|
||||
|
||||
const FederationDispatchParams = z.object({
|
||||
dispatchId: requiredString('Missing Dispatch ID')
|
||||
@@ -36,7 +37,11 @@ export const ORCHESTRATION_FEDERATION_CONTROL_METHODS: RpcMethod[] = [
|
||||
runtimeEpoch: runtime.getRuntimeId(),
|
||||
attachment: exposeRemoteAttachment(attachment),
|
||||
terminal: observation.exact ? observation.terminal : null,
|
||||
observation: { status: observation.status, exactWorker: observation.exact }
|
||||
observation: {
|
||||
status: observation.status,
|
||||
exactWorker: observation.exact,
|
||||
...(observation.reason ? { reason: observation.reason } : {})
|
||||
}
|
||||
}
|
||||
}
|
||||
}),
|
||||
@@ -46,7 +51,10 @@ export const ORCHESTRATION_FEDERATION_CONTROL_METHODS: RpcMethod[] = [
|
||||
handler: async (params, { runtime, authenticatedCallerFingerprint }) => {
|
||||
requireHomeAttachment(runtime, params.dispatchId, authenticatedCallerFingerprint)
|
||||
const observation = await inspectRemoteAttachment(runtime, params.dispatchId)
|
||||
if (!observation.exact || !observation.terminal || observation.status !== 'running') {
|
||||
// Why `=== 'exited'` rather than `!== 'live'`: the other non-live
|
||||
// statuses are already covered by the two guards, and an unverifiable
|
||||
// terminal is still readable — losing stop-contact is not an exit.
|
||||
if (!observation.exact || !observation.terminal || observation.status === 'exited') {
|
||||
throw new OrchestrationError(
|
||||
'worker_identity_changed',
|
||||
`Remote Dispatch ${params.dispatchId} no longer resolves to its exact process.`
|
||||
@@ -83,7 +91,18 @@ export const ORCHESTRATION_FEDERATION_CONTROL_METHODS: RpcMethod[] = [
|
||||
dispatchId: params.dispatchId,
|
||||
terminalHandle: observation.terminal.handle,
|
||||
workerState: attachment.state,
|
||||
terminalStatus: observation.status === 'exited' ? 'exited' : 'running',
|
||||
terminalStatus:
|
||||
observation.status === 'exited'
|
||||
? 'exited'
|
||||
: observation.status === 'unverifiable'
|
||||
? 'unknown'
|
||||
: 'running',
|
||||
terminalLiveness:
|
||||
observation.status === 'unverifiable'
|
||||
? 'unverifiable'
|
||||
: observation.status === 'exited'
|
||||
? 'exited'
|
||||
: 'live',
|
||||
attachedAt: attachment.created_at,
|
||||
source: params.source,
|
||||
cursor: params.cursor,
|
||||
@@ -134,6 +153,22 @@ export const ORCHESTRATION_FEDERATION_CONTROL_METHODS: RpcMethod[] = [
|
||||
}
|
||||
try {
|
||||
const close = await runtime.closeTerminal(observation.terminal.handle)
|
||||
if (!close.ptyKilled) {
|
||||
// The tab is retired but the process was never confirmed stopped, so
|
||||
// the coordinator must not be told this dispatch reached 'stopped'.
|
||||
const attachment = db.markRemoteAttachmentStopUnknown(
|
||||
params.dispatchId,
|
||||
describeUnconfirmedAgentStop(close)
|
||||
)
|
||||
return {
|
||||
dispatchId: params.dispatchId,
|
||||
state: attachment.state,
|
||||
alreadySettled: false,
|
||||
processAction: 'closed_agent_terminal',
|
||||
lastError: attachment.last_error,
|
||||
close
|
||||
}
|
||||
}
|
||||
const attachment = db.settleRemoteAttachmentStop(params.dispatchId)
|
||||
return {
|
||||
dispatchId: params.dispatchId,
|
||||
@@ -172,29 +207,44 @@ function requireHomeAttachment(
|
||||
return attachment
|
||||
}
|
||||
|
||||
async function inspectRemoteAttachment(runtime: OrcaRuntimeService, dispatchId: string) {
|
||||
async function inspectRemoteAttachment(
|
||||
runtime: OrcaRuntimeService,
|
||||
dispatchId: string
|
||||
): Promise<{
|
||||
terminal: Awaited<ReturnType<OrcaRuntimeService['showTerminal']>> | null
|
||||
exact: boolean
|
||||
status: 'unattached' | 'missing' | 'identity_changed' | 'live' | 'exited' | 'unverifiable'
|
||||
/** Set with `unverifiable`; names what we lost contact with. */
|
||||
reason?: string
|
||||
}> {
|
||||
const db = runtime.getOrchestrationDb()
|
||||
const attachment = db.getRemoteDispatchAttachment(dispatchId)
|
||||
if (!attachment?.terminal_handle) {
|
||||
return { terminal: null, exact: false, status: 'unattached' as const }
|
||||
return { terminal: null, exact: false, status: 'unattached' }
|
||||
}
|
||||
const terminal = await runtime.showTerminal(attachment.terminal_handle).catch(() => null)
|
||||
if (!terminal) {
|
||||
return { terminal: null, exact: false, status: 'missing' as const }
|
||||
return { terminal: null, exact: false, status: 'missing' }
|
||||
}
|
||||
const exact = db.isRemoteAttachmentProcessCurrent({
|
||||
dispatchId,
|
||||
paneKey: runtime.getTerminalPaneKey(attachment.terminal_handle),
|
||||
processIncarnation: runtime.getTerminalProcessIncarnation(attachment.terminal_handle)
|
||||
})
|
||||
if (!exact) {
|
||||
return { terminal, exact, status: 'identity_changed' }
|
||||
}
|
||||
// Why: the same rule as the local worker observation — the inventory only
|
||||
// iterates registered providers, so a dropped relay clears `connected` for
|
||||
// every remote PTY at once. Lost contact is not a death certificate.
|
||||
const verdict = runtime.getTerminalLivenessVerdict?.(attachment.terminal_handle) ?? null
|
||||
if (verdict?.status === 'unverifiable') {
|
||||
return { terminal, exact, status: 'unverifiable', reason: verdict.reason }
|
||||
}
|
||||
return {
|
||||
terminal,
|
||||
exact,
|
||||
status: exact
|
||||
? terminal.connected === false
|
||||
? ('exited' as const)
|
||||
: ('running' as const)
|
||||
: ('identity_changed' as const)
|
||||
status: verdict?.status !== 'live' && terminal.connected === false ? 'exited' : 'live'
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -0,0 +1,183 @@
|
||||
import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest'
|
||||
import { ORCHESTRATION_CONTRACT_VERSION } from '../../../../shared/protocol-version'
|
||||
import { OrcaRuntimeService } from '../../orca-runtime'
|
||||
import { OrchestrationDb } from '../../orchestration/db'
|
||||
import { ORCHESTRATION_METHODS } from './orchestration'
|
||||
|
||||
// The federation host runs its own copy of the observation and stop logic, so
|
||||
// it needs the same rule: lost contact with a worker's host is not an exit, and
|
||||
// a close it could not confirm must not be relayed home as a settled stop.
|
||||
|
||||
const HOME_FINGERPRINT = 'home-peer-fingerprint'
|
||||
const DISPATCH_ID = 'ctx_federation_verdict'
|
||||
const HANDLE = 'term_remote_worker'
|
||||
const PANE_KEY = 'tab_remote:bbbbbbbb-bbbb-4bbb-8bbb-bbbbbbbbbbbb'
|
||||
const INCARNATION = 'runtime:pty:7'
|
||||
const SSH_PROVIDER_GONE = 'its SSH provider is no longer registered'
|
||||
|
||||
describe('federation host liveness verdicts', () => {
|
||||
let db: OrchestrationDb
|
||||
let runtime: OrcaRuntimeService
|
||||
|
||||
beforeEach(() => {
|
||||
db = new OrchestrationDb(':memory:')
|
||||
runtime = new OrcaRuntimeService()
|
||||
runtime.setOrchestrationDb(db)
|
||||
vi.spyOn(runtime, 'getTerminalPaneKey').mockReturnValue(PANE_KEY)
|
||||
vi.spyOn(runtime, 'getTerminalProcessIncarnation').mockReturnValue(INCARNATION)
|
||||
vi.spyOn(runtime, 'showTerminal').mockResolvedValue({
|
||||
handle: HANDLE,
|
||||
worktreeId: 'repo::remote-worktree',
|
||||
connected: false,
|
||||
status: 'exited'
|
||||
} as never)
|
||||
db.createRemoteDispatchAttachment({
|
||||
dispatchId: DISPATCH_ID,
|
||||
taskId: 'task_remote',
|
||||
homePeerFingerprint: HOME_FINGERPRINT,
|
||||
protocolVersion: ORCHESTRATION_CONTRACT_VERSION,
|
||||
runtimeEpoch: runtime.getRuntimeId(),
|
||||
mutationReceipt: {
|
||||
callerFingerprint: HOME_FINGERPRINT,
|
||||
requestId: 'rpc_attach',
|
||||
method: 'orchestration.federationStart',
|
||||
payloadHash: 'hash'
|
||||
}
|
||||
})
|
||||
db.prepareRemoteAttachmentAuthority({
|
||||
dispatchId: DISPATCH_ID,
|
||||
paneKey: PANE_KEY,
|
||||
processIncarnation: INCARNATION,
|
||||
worktreeId: 'repo::remote-worktree',
|
||||
terminalHandle: HANDLE,
|
||||
setupState: 'not_applicable',
|
||||
effects: [{ kind: 'terminal', action: 'created', id: HANDLE }]
|
||||
})
|
||||
db.markRemoteAttachmentReady(DISPATCH_ID)
|
||||
})
|
||||
|
||||
afterEach(() => db.close())
|
||||
|
||||
async function call(name: string, params: Record<string, unknown>) {
|
||||
const method = ORCHESTRATION_METHODS.find((candidate) => candidate.name === name)
|
||||
if (!method) {
|
||||
throw new Error(`Method not found: ${name}`)
|
||||
}
|
||||
return method.handler(method.params!.parse(params), {
|
||||
runtime,
|
||||
authenticatedCallerFingerprint: HOME_FINGERPRINT
|
||||
} as never)
|
||||
}
|
||||
|
||||
it('reports lost contact as unverifiable rather than an observed exit', async () => {
|
||||
vi.spyOn(runtime, 'getTerminalLivenessVerdict').mockReturnValue({
|
||||
status: 'unverifiable',
|
||||
reason: SSH_PROVIDER_GONE
|
||||
})
|
||||
|
||||
await expect(
|
||||
call('orchestration.federationShow', { dispatchId: DISPATCH_ID })
|
||||
).resolves.toMatchObject({
|
||||
observation: { status: 'unverifiable', exactWorker: true, reason: SSH_PROVIDER_GONE }
|
||||
})
|
||||
})
|
||||
|
||||
it('uses the canonical live verdict for an observed process', async () => {
|
||||
vi.spyOn(runtime, 'showTerminal').mockResolvedValue({
|
||||
handle: HANDLE,
|
||||
worktreeId: 'repo::remote-worktree',
|
||||
connected: true,
|
||||
status: 'running'
|
||||
} as never)
|
||||
vi.spyOn(runtime, 'getTerminalLivenessVerdict').mockReturnValue({
|
||||
status: 'live',
|
||||
ptyIds: [HANDLE]
|
||||
})
|
||||
|
||||
await expect(
|
||||
call('orchestration.federationShow', { dispatchId: DISPATCH_ID })
|
||||
).resolves.toMatchObject({ observation: { status: 'live', exactWorker: true } })
|
||||
})
|
||||
|
||||
it('still reports a locally observed exit as exited', async () => {
|
||||
await expect(
|
||||
call('orchestration.federationShow', { dispatchId: DISPATCH_ID })
|
||||
).resolves.toMatchObject({ observation: { status: 'exited', exactWorker: true } })
|
||||
})
|
||||
|
||||
it('still serves output for a terminal we merely lost stop-contact with', async () => {
|
||||
// Why this matters: the read gate used to reject every status except live, which
|
||||
// would refuse a connected terminal the moment a stop lost contact with it.
|
||||
vi.spyOn(runtime, 'showTerminal').mockResolvedValue({
|
||||
handle: HANDLE,
|
||||
worktreeId: 'repo::remote-worktree',
|
||||
connected: true
|
||||
} as never)
|
||||
vi.spyOn(runtime, 'getTerminalLivenessVerdict').mockReturnValue({
|
||||
status: 'unverifiable',
|
||||
reason: SSH_PROVIDER_GONE
|
||||
})
|
||||
|
||||
const outcome = await call('orchestration.federationRead', {
|
||||
dispatchId: DISPATCH_ID
|
||||
}).catch((error: unknown) => error)
|
||||
|
||||
expect(outcome).not.toMatchObject({ code: 'worker_identity_changed' })
|
||||
})
|
||||
|
||||
it('does not relay an unconfirmed close home as a settled stop', async () => {
|
||||
vi.spyOn(runtime, 'getTerminalLivenessVerdict').mockReturnValue({
|
||||
status: 'unverifiable',
|
||||
reason: SSH_PROVIDER_GONE
|
||||
})
|
||||
const closeTerminal = vi.spyOn(runtime, 'closeTerminal').mockResolvedValue({
|
||||
handle: HANDLE,
|
||||
tabId: 'tab_remote',
|
||||
ptyKilled: false,
|
||||
ptyStopVerdict: 'unverifiable',
|
||||
ptyStopReason: SSH_PROVIDER_GONE
|
||||
})
|
||||
|
||||
const stopped = (await call('orchestration.federationStop', { dispatchId: DISPATCH_ID })) as {
|
||||
state: string
|
||||
lastError?: string
|
||||
}
|
||||
|
||||
// Losing contact is a reason to report honestly, never to stop trying.
|
||||
expect(closeTerminal).toHaveBeenCalledWith(HANDLE)
|
||||
expect(stopped.state).not.toBe('stopped')
|
||||
expect(stopped.lastError).toContain('could not be confirmed stopped')
|
||||
})
|
||||
|
||||
it('does not settle a bare false close as a stop', async () => {
|
||||
vi.spyOn(runtime, 'closeTerminal').mockResolvedValue({
|
||||
handle: HANDLE,
|
||||
tabId: 'tab_remote',
|
||||
ptyKilled: false
|
||||
})
|
||||
|
||||
const stopped = (await call('orchestration.federationStop', { dispatchId: DISPATCH_ID })) as {
|
||||
state: string
|
||||
lastError?: string
|
||||
}
|
||||
|
||||
expect(stopped.state).not.toBe('stopped')
|
||||
expect(stopped.lastError).toContain('could not be confirmed stopped')
|
||||
})
|
||||
|
||||
it('still settles a confirmed close as a stop', async () => {
|
||||
vi.spyOn(runtime, 'closeTerminal').mockResolvedValue({
|
||||
handle: HANDLE,
|
||||
tabId: 'tab_remote',
|
||||
ptyKilled: true
|
||||
})
|
||||
|
||||
const stopped = (await call('orchestration.federationStop', { dispatchId: DISPATCH_ID })) as {
|
||||
state: string
|
||||
processAction: string
|
||||
}
|
||||
|
||||
expect(stopped.state).toBe('stopped')
|
||||
expect(stopped.processAction).toBe('closed_agent_terminal')
|
||||
})
|
||||
})
|
||||
@@ -192,7 +192,7 @@ describe('orchestration federated worker output', () => {
|
||||
ok: true,
|
||||
result: {
|
||||
server: { environmentId: 'environment_windows', name: 'windows' },
|
||||
observation: { status: 'running', exactWorker: true },
|
||||
observation: { status: 'live', exactWorker: true },
|
||||
terminal: { handle: 'term_windows_worker' }
|
||||
}
|
||||
})
|
||||
|
||||
@@ -181,6 +181,7 @@ describe('orchestration federated setup evidence', () => {
|
||||
await expect(
|
||||
workerShow.handler(workerShow.params!.parse({ dispatch: started.dispatch.id }), { runtime })
|
||||
).resolves.toMatchObject({
|
||||
observation: { status: 'live', exactWorker: true },
|
||||
worker: {
|
||||
state: 'ready',
|
||||
stage: 'input_accepted',
|
||||
|
||||
@@ -152,7 +152,8 @@ describe('orchestration federation', () => {
|
||||
} as never)
|
||||
vi.spyOn(runtime, 'closeTerminal').mockResolvedValue({
|
||||
handle: 'term_windows_worker',
|
||||
closed: true
|
||||
tabId: 'tab-windows-worker',
|
||||
ptyKilled: true
|
||||
} as never)
|
||||
}
|
||||
|
||||
@@ -664,7 +665,7 @@ describe('orchestration federation', () => {
|
||||
|
||||
expect(shown).toMatchObject({
|
||||
ok: true,
|
||||
result: { observation: { status: 'running', exactWorker: true } }
|
||||
result: { observation: { status: 'live', exactWorker: true } }
|
||||
})
|
||||
expect(homeDb.getFederatedDispatch(dispatch.id)?.remote_runtime_epoch).not.toBe(oldEpoch)
|
||||
expect(homeDb.getFederatedDispatch(dispatch.id)?.peer_fingerprint).toBe(
|
||||
|
||||
@@ -99,7 +99,11 @@ export const ORCHESTRATION_WORKER_CONTROL_METHODS: RpcMethod[] = [
|
||||
server: { environmentId: server.environmentId, name: server.name },
|
||||
remoteRuntimeEpoch: remote.runtimeEpoch,
|
||||
terminal: remote.terminal,
|
||||
observation: remote.observation
|
||||
observation: {
|
||||
...remote.observation,
|
||||
// Legacy servers published `running`; normalize at the compatibility boundary.
|
||||
status: remote.observation.status === 'running' ? 'live' : remote.observation.status
|
||||
}
|
||||
}
|
||||
}
|
||||
if (worker.runtime_epoch && worker.runtime_epoch !== runtime.getRuntimeId()) {
|
||||
@@ -122,7 +126,12 @@ export const ORCHESTRATION_WORKER_CONTROL_METHODS: RpcMethod[] = [
|
||||
dispatch,
|
||||
worker: exposeWorker(worker),
|
||||
terminal: observation.exact ? observation.terminal : null,
|
||||
observation: { status: observation.status, exactWorker: observation.exact },
|
||||
observation: {
|
||||
status: observation.status,
|
||||
exactWorker: observation.exact,
|
||||
// Why: a bare `unverifiable` is not actionable without naming what we lost.
|
||||
...(observation.reason ? { reason: observation.reason } : {})
|
||||
},
|
||||
terminalResource: resource ? exposeWorkerTerminalResource(resource) : null
|
||||
}
|
||||
}
|
||||
@@ -176,7 +185,7 @@ export const ORCHESTRATION_WORKER_CONTROL_METHODS: RpcMethod[] = [
|
||||
)
|
||||
}
|
||||
const resource = db.getWorkerTerminalResourceByOwner(params.dispatch)
|
||||
if (resource && ['releasing', 'released'].includes(resource.release_state)) {
|
||||
if (resource && ['releasing', 'unknown', 'released'].includes(resource.release_state)) {
|
||||
return readArchivedWorkerOutput({
|
||||
db,
|
||||
dispatchId: params.dispatch,
|
||||
@@ -199,7 +208,18 @@ export const ORCHESTRATION_WORKER_CONTROL_METHODS: RpcMethod[] = [
|
||||
dispatchId: params.dispatch,
|
||||
terminalHandle: worker.agent_terminal_handle,
|
||||
workerState: worker.state,
|
||||
terminalStatus: observation.status === 'exited' ? 'exited' : 'running',
|
||||
terminalStatus:
|
||||
observation.status === 'exited'
|
||||
? 'exited'
|
||||
: observation.status === 'unverifiable'
|
||||
? 'unknown'
|
||||
: 'running',
|
||||
terminalLiveness:
|
||||
observation.status === 'unverifiable'
|
||||
? 'unverifiable'
|
||||
: observation.status === 'exited'
|
||||
? 'exited'
|
||||
: 'live',
|
||||
attachedAt: worker.created_at,
|
||||
source: params.source,
|
||||
cursor: params.cursor,
|
||||
|
||||
@@ -10,7 +10,9 @@ export async function inspectWorkerTerminal(
|
||||
): Promise<{
|
||||
terminal: Awaited<ReturnType<OrcaRuntimeService['showTerminal']>> | null
|
||||
exact: boolean
|
||||
status: 'unattached' | 'missing' | 'identity_changed' | 'running' | 'exited'
|
||||
status: 'unattached' | 'missing' | 'identity_changed' | 'live' | 'exited' | 'unverifiable'
|
||||
/** Set with `unverifiable`; names what we lost contact with. */
|
||||
reason?: string
|
||||
}> {
|
||||
const worker = db.getWorkerDispatch(dispatchId)
|
||||
if (!worker?.agent_terminal_handle) {
|
||||
@@ -25,10 +27,23 @@ export async function inspectWorkerTerminal(
|
||||
paneKey: runtime.getTerminalPaneKey(worker.agent_terminal_handle),
|
||||
processIncarnation: runtime.getTerminalProcessIncarnation(worker.agent_terminal_handle)
|
||||
})
|
||||
if (!exact) {
|
||||
return { terminal, exact, status: 'identity_changed' }
|
||||
}
|
||||
// Why: the aggregate inventory only iterates registered providers, so a dropped
|
||||
// relay clears `connected` for every remote PTY at once. Lost contact is not a
|
||||
// death certificate, and the verdict is the only field that can tell them apart.
|
||||
const verdict = runtime.getTerminalLivenessVerdict?.(worker.agent_terminal_handle) ?? null
|
||||
if (verdict?.status === 'unverifiable') {
|
||||
return { terminal, exact, status: 'unverifiable', reason: verdict.reason }
|
||||
}
|
||||
if (verdict?.status === 'live') {
|
||||
return { terminal, exact, status: 'live' }
|
||||
}
|
||||
return {
|
||||
terminal,
|
||||
exact,
|
||||
status: exact ? (terminal.connected === false ? 'exited' : 'running') : 'identity_changed'
|
||||
status: terminal.connected === false ? 'exited' : 'live'
|
||||
}
|
||||
}
|
||||
|
||||
@@ -71,7 +86,7 @@ export async function callFederatedWorkerShow(
|
||||
residualResources: unknown[]
|
||||
}
|
||||
terminal: unknown
|
||||
observation: { status: string; exactWorker: boolean }
|
||||
observation: { status: string; exactWorker: boolean; reason?: string }
|
||||
}> {
|
||||
return (await runtime.callOrchestrationWorkerServer(
|
||||
federated.environment_id,
|
||||
|
||||
@@ -83,6 +83,19 @@ describe('exact orchestration worker output', () => {
|
||||
expect(readTerminal).not.toHaveBeenCalled()
|
||||
})
|
||||
|
||||
it('reports unverifiable liveness without claiming the terminal is running', async () => {
|
||||
const result = await read({
|
||||
terminalStatus: 'unknown',
|
||||
terminalLiveness: 'unverifiable'
|
||||
})
|
||||
|
||||
expect(result.status).toEqual({
|
||||
worker: 'ready',
|
||||
terminal: 'unknown',
|
||||
liveness: 'unverifiable'
|
||||
})
|
||||
})
|
||||
|
||||
it('reads Grok through the shared Native Chat transcript decoder', async () => {
|
||||
await writeFile(
|
||||
transcriptA,
|
||||
|
||||
@@ -20,6 +20,7 @@ export async function readExactWorkerOutput(args: {
|
||||
terminalHandle: string
|
||||
workerState: string
|
||||
terminalStatus: RuntimeTerminalState
|
||||
terminalLiveness?: 'live' | 'unverifiable' | 'exited'
|
||||
attachedAt: string
|
||||
source?: OrchestrationWorkerReadSource
|
||||
cursor?: string | number
|
||||
@@ -100,7 +101,11 @@ export async function readExactWorkerOutput(args: {
|
||||
returnedMessageCount: transcript.messages.length
|
||||
},
|
||||
cursor: nextCursor,
|
||||
status: { worker: args.workerState, terminal: args.terminalStatus },
|
||||
status: {
|
||||
worker: args.workerState,
|
||||
terminal: args.terminalStatus,
|
||||
...(args.terminalLiveness ? { liveness: args.terminalLiveness } : {})
|
||||
},
|
||||
fallbackReason: null,
|
||||
warnings: transcript.warnings
|
||||
}
|
||||
@@ -145,7 +150,11 @@ async function readTerminalOutput(
|
||||
sourceIdentity,
|
||||
terminal: { ...terminal, tail: redactedTerminal.lines },
|
||||
cursor: nextCursor,
|
||||
status: { worker: args.workerState, terminal: terminal.status },
|
||||
status: {
|
||||
worker: args.workerState,
|
||||
terminal: args.terminalLiveness === 'unverifiable' ? args.terminalStatus : terminal.status,
|
||||
...(args.terminalLiveness ? { liveness: args.terminalLiveness } : {})
|
||||
},
|
||||
fallbackReason: null,
|
||||
warnings: redactedTerminal.warnings
|
||||
}
|
||||
|
||||
@@ -10,6 +10,7 @@ import {
|
||||
type WorkerTerminalTailArchive
|
||||
} from '../../orchestration/worker-output-archive'
|
||||
import type { OrcaRuntimeService } from '../../orca-runtime'
|
||||
import { describeUnconfirmedAgentStop } from '../../../../shared/pty-liveness-verdict'
|
||||
import { inspectWorkerTerminal } from './orchestration-worker-observation'
|
||||
import { orchestrationTimestampToMs } from './orchestration-worker-output'
|
||||
|
||||
@@ -212,7 +213,19 @@ async function completeWorkerTerminalReleaseOnce(
|
||||
}
|
||||
|
||||
try {
|
||||
await runtime.closeTerminal(resource.terminal_handle)
|
||||
const close = await runtime.closeTerminal(resource.terminal_handle)
|
||||
if (!close.ptyKilled) {
|
||||
const reason = describeUnconfirmedAgentStop(close)
|
||||
const unknown = db.markWorkerTerminalReleaseUnknown(resource.id, reason)
|
||||
return {
|
||||
dispatchId,
|
||||
state: 'release_unknown',
|
||||
processAction: 'closed_agent_terminal',
|
||||
archive: { source: archiveSource, status: archiveStatus },
|
||||
lastError: unknown.release_error ?? reason,
|
||||
recovery: `Inspect with: orca orchestration worker-show --dispatch ${dispatchId} --json — then repeat worker-release with the same --retry-request. Never substitute a broad terminal close.`
|
||||
}
|
||||
}
|
||||
} catch (error) {
|
||||
const reason = error instanceof Error ? error.message : String(error)
|
||||
if (/disposed|not connected|unavailable/i.test(reason)) {
|
||||
|
||||
@@ -0,0 +1,84 @@
|
||||
import { describe, expect, it, vi } from 'vitest'
|
||||
import type { OrcaRuntimeService } from '../../orca-runtime'
|
||||
import type { OrchestrationDb } from '../../orchestration/db'
|
||||
import type { WorkerTerminalResourceRow } from '../../orchestration/worker-terminal-ownership'
|
||||
import { completeWorkerTerminalRelease } from './orchestration-worker-release-completion'
|
||||
|
||||
describe('orchestration worker release liveness verdict', () => {
|
||||
it.each([
|
||||
{
|
||||
name: 'an explicit unverifiable verdict',
|
||||
close: {
|
||||
handle: 'term_worker',
|
||||
tabId: 'tab-worker',
|
||||
ptyKilled: false,
|
||||
ptyStopVerdict: 'unverifiable' as const,
|
||||
ptyStopReason: 'its SSH provider is no longer registered'
|
||||
},
|
||||
detail: 'its SSH provider is no longer registered'
|
||||
},
|
||||
{
|
||||
name: 'a bare unconfirmed close',
|
||||
close: { handle: 'term_worker', tabId: 'tab-worker', ptyKilled: false },
|
||||
detail: 'the stop outcome could not be verified'
|
||||
}
|
||||
])('does not release a worker after $name', async ({ close, detail }) => {
|
||||
const reason = 'its SSH provider is no longer registered'
|
||||
const resource = {
|
||||
id: 'resource-1',
|
||||
terminal_handle: 'term_worker',
|
||||
host_scope: JSON.stringify({ kind: 'ssh', targetId: 'target-1' }),
|
||||
archive_source: 'terminal',
|
||||
archive_status: 'captured',
|
||||
ownership_state: 'owned',
|
||||
release_state: 'requested'
|
||||
} as WorkerTerminalResourceRow
|
||||
const runtime = {
|
||||
showTerminal: vi.fn(async () => ({ handle: 'term_worker', connected: false })),
|
||||
getTerminalPaneKey: vi.fn(() => 'tab-worker:leaf-worker'),
|
||||
getTerminalProcessIncarnation: vi.fn(() => 'pty-worker:incarnation-1'),
|
||||
getTerminalLivenessVerdict: vi.fn(() => ({ status: 'unverifiable', reason })),
|
||||
getOrchestrationDispatchAuthority: vi.fn(() => ({
|
||||
hostScope: { kind: 'ssh', targetId: 'target-1' }
|
||||
})),
|
||||
closeTerminal: vi.fn(async () => close),
|
||||
notifyMessageArrived: vi.fn()
|
||||
} as unknown as OrcaRuntimeService
|
||||
const markWorkerTerminalReleaseUnknown = vi.fn((_resourceId: string, releaseError: string) => ({
|
||||
...resource,
|
||||
release_state: 'unknown',
|
||||
release_error: releaseError
|
||||
}))
|
||||
const db = {
|
||||
getWorkerDispatch: vi.fn(() => ({
|
||||
agent_terminal_handle: 'term_worker',
|
||||
created_at: '2026-08-16T00:00:00.000Z'
|
||||
})),
|
||||
isDispatchProcessCurrent: vi.fn(() => true),
|
||||
workerTerminalResourceHasIdentityConflict: vi.fn(() => false),
|
||||
getWorkerTerminalArchive: vi.fn(() => ({ kind: 'transcript_pin' })),
|
||||
commitWorkerTerminalArchiveForRelease: vi.fn(() => ({
|
||||
...resource,
|
||||
release_state: 'releasing'
|
||||
})),
|
||||
markWorkerTerminalReleaseUnknown
|
||||
} as unknown as OrchestrationDb
|
||||
|
||||
await expect(
|
||||
completeWorkerTerminalRelease({
|
||||
runtime,
|
||||
db,
|
||||
dispatchId: 'ctx-worker',
|
||||
resource
|
||||
})
|
||||
).resolves.toMatchObject({
|
||||
state: 'release_unknown',
|
||||
processAction: 'closed_agent_terminal',
|
||||
lastError: `The agent terminal was closed but its process could not be confirmed stopped: ${detail}.`
|
||||
})
|
||||
expect(markWorkerTerminalReleaseUnknown).toHaveBeenCalledWith(
|
||||
'resource-1',
|
||||
`The agent terminal was closed but its process could not be confirmed stopped: ${detail}.`
|
||||
)
|
||||
})
|
||||
})
|
||||
@@ -80,8 +80,9 @@ describe('orchestration worker release recovery', () => {
|
||||
})
|
||||
vi.spyOn(runtime, 'closeTerminal').mockResolvedValue({
|
||||
handle: 'term_worker',
|
||||
closed: true
|
||||
} as never)
|
||||
tabId: 'tab-worker',
|
||||
ptyKilled: true
|
||||
})
|
||||
vi.spyOn(runtime, 'notifyMessageArrived').mockImplementation(() => {})
|
||||
activeRunId = db.createRun({
|
||||
objective: 'Release recovery test Run',
|
||||
@@ -162,6 +163,36 @@ describe('orchestration worker release recovery', () => {
|
||||
expect(db.getWorkerTerminalResourceByOwner(dispatchId)?.release_state).toBe('releasing')
|
||||
})
|
||||
|
||||
it('preserves archived output when an unconfirmed release retry cannot find the terminal', async () => {
|
||||
setup()
|
||||
const { dispatchId } = await startSettledWorker()
|
||||
vi.mocked(runtime.closeTerminal).mockResolvedValueOnce({
|
||||
handle: 'term_worker',
|
||||
tabId: 'tab-worker',
|
||||
ptyKilled: false,
|
||||
ptyStopVerdict: 'unverifiable',
|
||||
ptyStopReason: 'its SSH provider is no longer registered'
|
||||
})
|
||||
|
||||
await expect(
|
||||
call('orchestration.workerRelease', { dispatch: dispatchId })
|
||||
).resolves.toMatchObject({ state: 'release_unknown' })
|
||||
expect(db.getWorkerTerminalArchive(dispatchId)).toBeDefined()
|
||||
|
||||
vi.mocked(runtime.showTerminal).mockRejectedValue(new Error('terminal_handle_stale'))
|
||||
await expect(
|
||||
call('orchestration.workerRelease', { dispatch: dispatchId })
|
||||
).resolves.toMatchObject({ state: 'release_unknown' })
|
||||
const read = (await call('orchestration.workerRead', { dispatch: dispatchId })) as {
|
||||
archived?: boolean
|
||||
terminal: { tail: string[] }
|
||||
}
|
||||
expect(read).toMatchObject({
|
||||
archived: true,
|
||||
terminal: { tail: ['worker output line 1', 'worker output line 2'] }
|
||||
})
|
||||
})
|
||||
|
||||
it('never touches resources without requested releases', async () => {
|
||||
setup()
|
||||
await startSettledWorker()
|
||||
|
||||
@@ -93,7 +93,8 @@ describe('orchestration worker release', () => {
|
||||
})
|
||||
vi.spyOn(runtime, 'closeTerminal').mockResolvedValue({
|
||||
handle: 'term_worker',
|
||||
closed: true
|
||||
tabId: 'tab-worker',
|
||||
ptyKilled: true
|
||||
} as never)
|
||||
vi.spyOn(runtime, 'notifyMessageArrived').mockImplementation(() => {})
|
||||
activeRunId = db.createRun({
|
||||
@@ -242,7 +243,7 @@ describe('orchestration worker release', () => {
|
||||
it('reconciles a dead external terminal without closing a process', async () => {
|
||||
setup()
|
||||
const { dispatchId } = await startSettledWorker('succeeded', { terminal: 'term_worker' })
|
||||
inspectProcessLiveness.mockResolvedValue('dead')
|
||||
inspectProcessLiveness.mockResolvedValue('exited')
|
||||
|
||||
await expect(
|
||||
call('orchestration.workerRelease', { dispatch: dispatchId })
|
||||
@@ -268,7 +269,7 @@ describe('orchestration worker release', () => {
|
||||
raw
|
||||
.prepare('UPDATE worker_terminal_resources SET prior_owner_dispatch_ids = ? WHERE id = ?')
|
||||
.run('{invalid', resource?.id)
|
||||
inspectProcessLiveness.mockResolvedValue('dead')
|
||||
inspectProcessLiveness.mockResolvedValue('exited')
|
||||
|
||||
await expect(
|
||||
call('orchestration.workerRelease', { dispatch: dispatchId })
|
||||
@@ -297,7 +298,7 @@ describe('orchestration worker release', () => {
|
||||
setup()
|
||||
const { dispatchId } = await startSettledWorker()
|
||||
await call('orchestration.workerTerminalUserInput', { paneKey: workerPaneKey })
|
||||
inspectProcessLiveness.mockResolvedValue('dead')
|
||||
inspectProcessLiveness.mockResolvedValue('exited')
|
||||
|
||||
await expect(
|
||||
call('orchestration.workerRelease', { dispatch: dispatchId })
|
||||
@@ -324,7 +325,7 @@ describe('orchestration worker release', () => {
|
||||
} else {
|
||||
db.abandonWorkerDispatch(dispatchId)
|
||||
}
|
||||
inspectProcessLiveness.mockResolvedValue('dead')
|
||||
inspectProcessLiveness.mockResolvedValue('exited')
|
||||
|
||||
await expect(
|
||||
call('orchestration.workerRelease', { dispatch: dispatchId })
|
||||
@@ -706,7 +707,7 @@ describe('orchestration worker release', () => {
|
||||
expect(transferred?.terminal_handle).toBe('term_reminted')
|
||||
expect(db.getWorkerTerminalResourceByOwner(first.dispatchId)).toBeUndefined()
|
||||
|
||||
inspectProcessLiveness.mockResolvedValueOnce('dead')
|
||||
inspectProcessLiveness.mockResolvedValueOnce('exited')
|
||||
const oldRelease = (await call('orchestration.workerRelease', {
|
||||
dispatch: first.dispatchId
|
||||
})) as { state: string; reason?: string }
|
||||
@@ -727,7 +728,7 @@ describe('orchestration worker release', () => {
|
||||
const first = await startSettledWorker('succeeded')
|
||||
const second = await startWorker({ terminal: 'term_reminted' })
|
||||
settle(second.taskId, second.dispatchId, 'succeeded')
|
||||
inspectProcessLiveness.mockResolvedValue('dead')
|
||||
inspectProcessLiveness.mockResolvedValue('exited')
|
||||
|
||||
await expect(
|
||||
call('orchestration.workerRelease', { dispatch: first.dispatchId })
|
||||
|
||||
@@ -61,7 +61,7 @@ export const ORCHESTRATION_WORKER_RELEASE_METHODS: RpcMethod[] = [
|
||||
(await runtime.inspectTerminalProcessIncarnationLiveness(
|
||||
processIncarnation,
|
||||
resource.host_scope
|
||||
)) === 'dead'
|
||||
)) === 'exited'
|
||||
) {
|
||||
const reconciled = db.settleDeadWorkerTerminalRelease({
|
||||
requestingDispatchId: params.dispatch,
|
||||
|
||||
@@ -0,0 +1,57 @@
|
||||
import { describe, expect, it, vi } from 'vitest'
|
||||
import { ORCHESTRATION_CONTRACT_RUNTIME_CAPABILITY } from '../../../../shared/protocol-version'
|
||||
import type { OrcaRuntimeService } from '../../orca-runtime'
|
||||
import type { OrchestrationDb } from '../../orchestration/db'
|
||||
import { ORCHESTRATION_WORKER_STOP_METHODS } from './orchestration-worker-stop'
|
||||
|
||||
describe('federated worker stop capability', () => {
|
||||
it('does not trust a legacy server stop receipt', async () => {
|
||||
const markWorkerStopUnknown = vi.fn((_dispatchId: string, reason: string) => ({
|
||||
state: 'stop_unknown',
|
||||
last_error: reason
|
||||
}))
|
||||
const db = {
|
||||
getFederatedDispatch: vi.fn(() => ({
|
||||
dispatch_id: 'ctx_remote',
|
||||
environment_id: 'environment_linux',
|
||||
environment_name: 'linux',
|
||||
peer_fingerprint: 'peer-linux'
|
||||
})),
|
||||
beginWorkerStop: vi.fn(() => ({ disposition: 'started' })),
|
||||
markWorkerStopUnknown
|
||||
} as unknown as OrchestrationDb
|
||||
const callOrchestrationWorkerServer = vi.fn(async (_environmentId, method) => {
|
||||
if (method === 'status.get') {
|
||||
return { capabilities: [ORCHESTRATION_CONTRACT_RUNTIME_CAPABILITY] }
|
||||
}
|
||||
return { state: 'stopped', processAction: 'closed_agent_terminal' }
|
||||
})
|
||||
const runtime = {
|
||||
getOrchestrationDb: () => db,
|
||||
resolveOrchestrationWorkerServer: () => ({
|
||||
environmentId: 'environment_linux',
|
||||
name: 'linux',
|
||||
peerFingerprint: 'peer-linux'
|
||||
}),
|
||||
callOrchestrationWorkerServer
|
||||
} as unknown as OrcaRuntimeService
|
||||
const method = ORCHESTRATION_WORKER_STOP_METHODS[0]!
|
||||
|
||||
await expect(
|
||||
method.handler(method.params!.parse({ dispatch: 'ctx_remote' }), {
|
||||
runtime,
|
||||
orchestrationMutation: {
|
||||
callerFingerprint: 'coordinator',
|
||||
requestId: 'request_stop',
|
||||
method: 'orchestration.workerStop',
|
||||
payloadHash: 'hash'
|
||||
}
|
||||
})
|
||||
).resolves.toMatchObject({ state: 'stop_unknown', processAction: 'none' })
|
||||
expect(markWorkerStopUnknown).toHaveBeenCalledWith(
|
||||
'ctx_remote',
|
||||
'Connected server linux cannot prove the worker stop outcome.'
|
||||
)
|
||||
expect(callOrchestrationWorkerServer).toHaveBeenCalledTimes(1)
|
||||
})
|
||||
})
|
||||
@@ -0,0 +1,158 @@
|
||||
import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest'
|
||||
import { OrcaRuntimeService } from '../../orca-runtime'
|
||||
import { OrchestrationDb } from '../../orchestration/db'
|
||||
import { ORCHESTRATION_METHODS } from './orchestration'
|
||||
|
||||
// The aggregate terminal inventory only iterates registered providers, so a
|
||||
// dropped relay clears `connected` for every remote PTY at once. That is lost
|
||||
// contact, not a death certificate.
|
||||
describe('worker-stop against a terminal we lost contact with', () => {
|
||||
let db: OrchestrationDb
|
||||
let runtime: OrcaRuntimeService
|
||||
|
||||
beforeEach(() => {
|
||||
db = new OrchestrationDb(':memory:')
|
||||
runtime = new OrcaRuntimeService()
|
||||
runtime.setOrchestrationDb(db)
|
||||
vi.spyOn(runtime, 'getTerminalPaneKey').mockReturnValue(
|
||||
'tab_worker:bbbbbbbb-bbbb-4bbb-8bbb-bbbbbbbbbbbb'
|
||||
)
|
||||
vi.spyOn(runtime, 'getTerminalProcessIncarnation').mockReturnValue('runtime:pty:1')
|
||||
vi.spyOn(runtime, 'showTerminal').mockResolvedValue({
|
||||
handle: 'term_worker',
|
||||
worktreeId: 'repo::worktree',
|
||||
connected: false,
|
||||
status: 'exited'
|
||||
} as never)
|
||||
})
|
||||
|
||||
afterEach(() => db.close())
|
||||
|
||||
async function call(name: string, params: Record<string, unknown>) {
|
||||
const method = ORCHESTRATION_METHODS.find((candidate) => candidate.name === name)
|
||||
if (!method) {
|
||||
throw new Error(`Method not found: ${name}`)
|
||||
}
|
||||
return method.handler(method.params!.parse(params), { runtime })
|
||||
}
|
||||
|
||||
function createWorker() {
|
||||
const run = db.createRun({
|
||||
objective: 'Verdict',
|
||||
coordinatorHandle: 'term_coord',
|
||||
coordinatorPaneKey: 'tab_coord:aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa'
|
||||
})
|
||||
const task = db.createTask({ spec: 'stop worker', runId: run.id })
|
||||
const started = db.createStartingWorkerDispatch({
|
||||
taskId: task.id,
|
||||
startOptions: {},
|
||||
runtimeEpoch: runtime.getRuntimeId()
|
||||
})
|
||||
db.prepareStartingWorkerAuthority({
|
||||
dispatchId: started.dispatch.id,
|
||||
handle: 'term_worker',
|
||||
paneKey: 'tab_worker:bbbbbbbb-bbbb-4bbb-8bbb-bbbbbbbbbbbb',
|
||||
processIncarnation: 'runtime:pty:1',
|
||||
worktreeId: 'repo::worktree',
|
||||
setupState: 'not_applicable',
|
||||
effects: [{ kind: 'terminal', action: 'created', id: 'term_worker' }]
|
||||
})
|
||||
db.markWorkerDispatchReady(started.dispatch.id)
|
||||
return started.dispatch
|
||||
}
|
||||
|
||||
it('reports the process as unverifiable, not exited, when the link dropped', async () => {
|
||||
vi.spyOn(runtime, 'getTerminalLivenessVerdict').mockReturnValue({
|
||||
status: 'unverifiable',
|
||||
reason: 'its SSH provider is no longer registered'
|
||||
})
|
||||
const closeTerminal = vi.spyOn(runtime, 'closeTerminal').mockResolvedValue({
|
||||
handle: 'term_worker',
|
||||
tabId: 'tab_worker',
|
||||
ptyKilled: false,
|
||||
ptyStopVerdict: 'unverifiable',
|
||||
ptyStopReason: 'its SSH provider is no longer registered'
|
||||
})
|
||||
const dispatch = createWorker()
|
||||
|
||||
await expect(
|
||||
call('orchestration.workerShow', { dispatch: dispatch.id })
|
||||
).resolves.toMatchObject({
|
||||
observation: {
|
||||
status: 'unverifiable',
|
||||
exactWorker: true,
|
||||
reason: 'its SSH provider is no longer registered'
|
||||
}
|
||||
})
|
||||
|
||||
const stopped = (await call('orchestration.workerStop', { dispatch: dispatch.id })) as {
|
||||
state: string
|
||||
processAction: string
|
||||
lastError: string
|
||||
}
|
||||
// Losing contact is a reason to report honestly, never to stop trying.
|
||||
expect(closeTerminal).toHaveBeenCalledWith('term_worker')
|
||||
expect(stopped.processAction).toBe('closed_agent_terminal')
|
||||
expect(stopped.state).toBe('stop_unknown')
|
||||
expect(stopped.lastError).toContain('could not be confirmed stopped')
|
||||
expect(stopped.lastError).not.toContain('exited')
|
||||
})
|
||||
|
||||
it('does not settle a bare false close as stopped', async () => {
|
||||
vi.spyOn(runtime, 'showTerminal').mockResolvedValue({
|
||||
handle: 'term_worker',
|
||||
worktreeId: 'repo::worktree',
|
||||
connected: true,
|
||||
status: 'running'
|
||||
} as never)
|
||||
vi.spyOn(runtime, 'closeTerminal').mockResolvedValue({
|
||||
handle: 'term_worker',
|
||||
tabId: 'tab_worker',
|
||||
ptyKilled: false
|
||||
})
|
||||
const dispatch = createWorker()
|
||||
|
||||
const stopped = (await call('orchestration.workerStop', { dispatch: dispatch.id })) as {
|
||||
state: string
|
||||
lastError: string
|
||||
}
|
||||
|
||||
expect(stopped.state).toBe('stop_unknown')
|
||||
expect(stopped.lastError).toContain('could not be confirmed stopped')
|
||||
})
|
||||
|
||||
it('uses the canonical live verdict for an observed process', async () => {
|
||||
vi.spyOn(runtime, 'showTerminal').mockResolvedValue({
|
||||
handle: 'term_worker',
|
||||
worktreeId: 'repo::worktree',
|
||||
connected: true,
|
||||
status: 'running'
|
||||
} as never)
|
||||
vi.spyOn(runtime, 'getTerminalLivenessVerdict').mockReturnValue({
|
||||
status: 'live',
|
||||
ptyIds: ['runtime:pty:1']
|
||||
})
|
||||
const dispatch = createWorker()
|
||||
|
||||
await expect(
|
||||
call('orchestration.workerShow', { dispatch: dispatch.id })
|
||||
).resolves.toMatchObject({
|
||||
observation: { status: 'live', exactWorker: true }
|
||||
})
|
||||
})
|
||||
|
||||
it('still reports a locally observed exit as exited', async () => {
|
||||
const dispatch = createWorker()
|
||||
|
||||
await expect(
|
||||
call('orchestration.workerShow', { dispatch: dispatch.id })
|
||||
).resolves.toMatchObject({
|
||||
observation: { status: 'exited', exactWorker: true }
|
||||
})
|
||||
|
||||
const stopped = (await call('orchestration.workerStop', { dispatch: dispatch.id })) as {
|
||||
lastError: string
|
||||
}
|
||||
expect(stopped.lastError).toBe('The recorded worker process is exited; no terminal was closed.')
|
||||
})
|
||||
})
|
||||
@@ -2,6 +2,9 @@ import { z } from 'zod'
|
||||
import { OrchestrationError } from '../../orchestration/orchestration-error'
|
||||
import { defineMethod, type RpcMethod } from '../core'
|
||||
import { requiredString } from '../schemas'
|
||||
import { describeUnconfirmedAgentStop } from '../../../../shared/pty-liveness-verdict'
|
||||
import { ORCHESTRATION_WORKER_STOP_VERDICT_RUNTIME_CAPABILITY } from '../../../../shared/protocol-version'
|
||||
import type { RuntimeStatus } from '../../../../shared/runtime-types'
|
||||
import {
|
||||
inspectWorkerTerminal,
|
||||
resolvePinnedFederatedServer
|
||||
@@ -29,6 +32,24 @@ export const ORCHESTRATION_WORKER_STOP_METHODS: RpcMethod[] = [
|
||||
return settledReceipt(params.dispatch, begun.worker.state)
|
||||
}
|
||||
try {
|
||||
const status = (await runtime.callOrchestrationWorkerServer(
|
||||
server.environmentId,
|
||||
'status.get',
|
||||
undefined,
|
||||
30_000
|
||||
)) as RuntimeStatus
|
||||
if (
|
||||
!status.capabilities?.includes(ORCHESTRATION_WORKER_STOP_VERDICT_RUNTIME_CAPABILITY)
|
||||
) {
|
||||
return unknownReceipt(
|
||||
params.dispatch,
|
||||
db.markWorkerStopUnknown(
|
||||
params.dispatch,
|
||||
`Connected server ${server.name} cannot prove the worker stop outcome.`
|
||||
),
|
||||
'none'
|
||||
)
|
||||
}
|
||||
const remote = (await runtime.callOrchestrationWorkerServer(
|
||||
server.environmentId,
|
||||
'orchestration.federationStop',
|
||||
@@ -101,7 +122,12 @@ export const ORCHESTRATION_WORKER_STOP_METHODS: RpcMethod[] = [
|
||||
)
|
||||
}
|
||||
const observation = await inspectWorkerTerminal(runtime, db, params.dispatch)
|
||||
if (!observation.exact || observation.status !== 'running') {
|
||||
// Why `unverifiable` still proceeds: losing contact is a reason to report
|
||||
// the outcome honestly, never a reason to stop trying to stop the worker.
|
||||
if (
|
||||
!observation.exact ||
|
||||
(observation.status !== 'live' && observation.status !== 'unverifiable')
|
||||
) {
|
||||
return unknownReceipt(
|
||||
params.dispatch,
|
||||
db.markWorkerStopUnknown(
|
||||
@@ -113,6 +139,15 @@ export const ORCHESTRATION_WORKER_STOP_METHODS: RpcMethod[] = [
|
||||
}
|
||||
try {
|
||||
const close = await runtime.closeTerminal(handle)
|
||||
if (!close.ptyKilled) {
|
||||
// The tab is retired, but the agent process was never confirmed stopped —
|
||||
// settling here is the false success this receipt exists to prevent.
|
||||
return unknownReceipt(
|
||||
params.dispatch,
|
||||
db.markWorkerStopUnknown(params.dispatch, describeUnconfirmedAgentStop(close)),
|
||||
'closed_agent_terminal'
|
||||
)
|
||||
}
|
||||
const worker = db.settleWorkerStop(params.dispatch)
|
||||
runtime.notifyMessageArrived(`dispatch:${params.dispatch}`, 'status')
|
||||
return {
|
||||
|
||||
@@ -30,7 +30,8 @@ describe('orchestration worker recovery', () => {
|
||||
})
|
||||
vi.spyOn(runtime, 'closeTerminal').mockResolvedValue({
|
||||
handle: 'term_worker',
|
||||
closed: true
|
||||
tabId: 'tab-worker',
|
||||
ptyKilled: true
|
||||
} as never)
|
||||
})
|
||||
|
||||
@@ -80,7 +81,7 @@ describe('orchestration worker recovery', () => {
|
||||
call('orchestration.workerShow', { dispatch: dispatch.id })
|
||||
).resolves.toMatchObject({
|
||||
worker: { state: 'ready' },
|
||||
observation: { status: 'running', exactWorker: true },
|
||||
observation: { status: 'live', exactWorker: true },
|
||||
terminal: { handle: 'term_worker' }
|
||||
})
|
||||
await expect(
|
||||
|
||||
@@ -1,18 +1,20 @@
|
||||
import type { IPtyProvider } from '../providers/types'
|
||||
import type { OrcaRuntimeService } from './orca-runtime'
|
||||
import {
|
||||
UNSTOPPED_PTY_DETAIL_SEPARATOR,
|
||||
UNSTOPPED_PTY_LIVE_DETAIL_PREFIX,
|
||||
UNSTOPPED_PTY_REMOVAL_PREFIX
|
||||
} from '../../shared/worktree/removal'
|
||||
import {
|
||||
NO_OBSERVING_PROVIDER_REASON,
|
||||
type PtyLivenessVerdict
|
||||
} from '../../shared/pty-liveness-verdict'
|
||||
import { settleBeforeDeadline } from './settle-before-deadline'
|
||||
|
||||
// Floor for the verification window when the sweep ran on a very short budget.
|
||||
export const WORKTREE_TEARDOWN_VERIFY_GRACE_MS = 2_000
|
||||
|
||||
export type UnstoppedPtyVerdict =
|
||||
| { status: 'exited' }
|
||||
| { status: 'live'; ptyIds: string[] }
|
||||
| { status: 'unverifiable'; reason: string }
|
||||
export type UnstoppedPtyVerdict = PtyLivenessVerdict
|
||||
|
||||
/**
|
||||
* Re-lists the provider's processes to decide what a failed stop RPC actually
|
||||
@@ -56,6 +58,45 @@ export async function verifyUnstoppedPtys(
|
||||
return stillLive.length > 0 ? { status: 'live', ptyIds: stillLive } : { status: 'exited' }
|
||||
}
|
||||
|
||||
/**
|
||||
* A stop that lost contact with the PTY's own host stays unverifiable: the
|
||||
* surviving provider's inventory is silent about a host it cannot reach, and
|
||||
* silence is not evidence of an exit. Force Delete is still the escape hatch.
|
||||
*/
|
||||
export function unverifiableStopVerdict(
|
||||
failedPtyIds: readonly string[],
|
||||
runtime: OrcaRuntimeService | undefined
|
||||
): UnstoppedPtyVerdict | null {
|
||||
for (const ptyId of failedPtyIds) {
|
||||
const verdict = runtime?.getPtyLivenessVerdict?.(ptyId)
|
||||
if (verdict?.status === 'unverifiable') {
|
||||
return verdict
|
||||
}
|
||||
}
|
||||
return null
|
||||
}
|
||||
|
||||
export async function resolveUnstoppedPtyVerdict(
|
||||
failedPtyIds: readonly string[],
|
||||
provider: IPtyProvider,
|
||||
sweepBudgetMs: number,
|
||||
providerObservesOwningHost: boolean,
|
||||
runtime?: OrcaRuntimeService
|
||||
): Promise<UnstoppedPtyVerdict> {
|
||||
if (failedPtyIds.length === 0) {
|
||||
return { status: 'exited' }
|
||||
}
|
||||
if (!providerObservesOwningHost) {
|
||||
return (
|
||||
unverifiableStopVerdict(failedPtyIds, runtime) ?? {
|
||||
status: 'unverifiable',
|
||||
reason: NO_OBSERVING_PROVIDER_REASON
|
||||
}
|
||||
)
|
||||
}
|
||||
return verifyUnstoppedPtys(failedPtyIds, provider, sweepBudgetMs)
|
||||
}
|
||||
|
||||
/** Names the blocking PTYs so a wedged removal is diagnosable, not just refused. */
|
||||
export function describeUnstoppedPtys(
|
||||
worktreeId: string,
|
||||
|
||||
@@ -175,6 +175,80 @@ describe('destructive teardown when a PTY stop cannot be proven', () => {
|
||||
expect(error.message).not.toContain('w1@@gone-2')
|
||||
})
|
||||
|
||||
// Why: the client's own provider list is silent about an SSH host, so a stop
|
||||
// it could not confirm must stay unverifiable instead of reading as exited —
|
||||
// otherwise removal walks straight past a live remote agent.
|
||||
it("blocks removal when the stop lost contact with the PTY's own host", async () => {
|
||||
// The surviving provider lists nothing for a host it cannot reach, so without
|
||||
// the recorded verdict this empty list reads as "exited" and removal walks
|
||||
// straight past a live remote agent.
|
||||
const localProvider = createProviderStub(async () => [])
|
||||
listRegisteredPtysMock.mockReturnValue([])
|
||||
const runtime = {
|
||||
stopTerminalsForWorktree: async (
|
||||
_worktreeId: string,
|
||||
opts: {
|
||||
stopPty?: (
|
||||
ptyId: string,
|
||||
stop: () => boolean | Promise<boolean>
|
||||
) => Promise<{ stopped: boolean; owner: boolean }>
|
||||
}
|
||||
) => {
|
||||
await opts.stopPty?.('ssh:conn-1@@relay-9', () => false)
|
||||
return { stopped: 0 }
|
||||
},
|
||||
getPtyLivenessVerdict: (ptyId: string) =>
|
||||
ptyId === 'ssh:conn-1@@relay-9'
|
||||
? { status: 'unverifiable', reason: 'its SSH provider is no longer registered' }
|
||||
: null
|
||||
}
|
||||
|
||||
await expect(
|
||||
killAllProcessesForWorktree('w1', {
|
||||
runtime: runtime as never,
|
||||
localProvider,
|
||||
resolvedConnectionId: 'conn-1',
|
||||
includeProviderInventory: false,
|
||||
includeLocalRegistry: false,
|
||||
requirePhysicalStop: true
|
||||
})
|
||||
).rejects.toThrow(
|
||||
/could not verify[\s\S]*ssh:conn-1@@relay-9[\s\S]*SSH provider is no longer registered[\s\S]*--force/
|
||||
)
|
||||
})
|
||||
|
||||
it('does not infer a remote exit from fallback local inventory without a cached verdict', async () => {
|
||||
const localProvider = createProviderStub(async () => [])
|
||||
listRegisteredPtysMock.mockReturnValue([])
|
||||
const runtime = {
|
||||
stopTerminalsForWorktree: async (
|
||||
_worktreeId: string,
|
||||
opts: {
|
||||
stopPty?: (
|
||||
ptyId: string,
|
||||
stop: () => boolean | Promise<boolean>
|
||||
) => Promise<{ stopped: boolean; owner: boolean }>
|
||||
}
|
||||
) => {
|
||||
await opts.stopPty?.('ssh:conn-1@@relay-10', () => false)
|
||||
return { stopped: 0 }
|
||||
},
|
||||
getPtyLivenessVerdict: () => null
|
||||
}
|
||||
|
||||
await expect(
|
||||
killAllProcessesForWorktree('w1', {
|
||||
runtime: runtime as never,
|
||||
localProvider,
|
||||
resolvedConnectionId: 'conn-1',
|
||||
includeProviderInventory: false,
|
||||
includeLocalRegistry: false,
|
||||
requirePhysicalStop: true
|
||||
})
|
||||
).rejects.toThrow(/could not verify[\s\S]*no registered provider can observe its host/)
|
||||
expect(localProvider.listProcesses).not.toHaveBeenCalled()
|
||||
})
|
||||
|
||||
it('reports unverifiable separately from live when the process list fails', async () => {
|
||||
const localProvider = createProviderStub(async () => {
|
||||
throw new Error('daemon socket closed')
|
||||
|
||||
@@ -521,7 +521,7 @@ describe('killAllProcessesForWorktree', () => {
|
||||
expect(localProvider.shutdown).toHaveBeenCalledTimes(1)
|
||||
})
|
||||
|
||||
it('accepts a failed Windows stop when a fresh inventory proves the PTY exited', async () => {
|
||||
it('accepts a failed stop when exact-owner inventory supersedes cached uncertainty', async () => {
|
||||
const worktreeId = 'repo-1::C:/Users/User/orca/workspaces/repo/feature'
|
||||
const ptyId = `${worktreeId}@@windows-pty`
|
||||
const stopTerminalsForWorktree = vi.fn(
|
||||
@@ -538,7 +538,11 @@ describe('killAllProcessesForWorktree', () => {
|
||||
})
|
||||
)
|
||||
const runtime = {
|
||||
stopTerminalsForWorktree
|
||||
stopTerminalsForWorktree,
|
||||
getPtyLivenessVerdict: vi.fn(() => ({
|
||||
status: 'unverifiable',
|
||||
reason: 'the provider disconnected during stop'
|
||||
}))
|
||||
} as unknown as Parameters<typeof killAllProcessesForWorktree>[1]['runtime']
|
||||
let inventoryCount = 0
|
||||
const localProvider = createProviderStub(async () => {
|
||||
@@ -568,6 +572,45 @@ describe('killAllProcessesForWorktree', () => {
|
||||
expect(localProvider.listProcesses).toHaveBeenCalledTimes(2)
|
||||
})
|
||||
|
||||
it('keeps cached uncertainty when no provider can observe the owning host', async () => {
|
||||
const worktreeId = 'repo-1::/remote/worktree'
|
||||
const ptyId = `${worktreeId}@@remote-pty`
|
||||
const stopTerminalsForWorktree = vi.fn(
|
||||
async (
|
||||
_worktreeId: string,
|
||||
options: {
|
||||
stopPty: (
|
||||
ptyId: string,
|
||||
stop: () => boolean
|
||||
) => Promise<{ stopped: boolean; owner: boolean }>
|
||||
}
|
||||
) => ({
|
||||
stopped: (await options.stopPty(ptyId, () => false)).owner ? 1 : 0
|
||||
})
|
||||
)
|
||||
const runtime = {
|
||||
stopTerminalsForWorktree,
|
||||
getPtyLivenessVerdict: vi.fn(() => ({
|
||||
status: 'unverifiable',
|
||||
reason: 'no registered provider can observe its host'
|
||||
}))
|
||||
} as unknown as Parameters<typeof killAllProcessesForWorktree>[1]['runtime']
|
||||
const fallbackProvider = createProviderStub(async () => [])
|
||||
listRegisteredPtysMock.mockReturnValue([])
|
||||
|
||||
await expect(
|
||||
killAllProcessesForWorktree(worktreeId, {
|
||||
runtime,
|
||||
resolvedConnectionId: 'missing-connection',
|
||||
localProvider: fallbackProvider,
|
||||
includeProviderInventory: false,
|
||||
includeLocalRegistry: false,
|
||||
requirePhysicalStop: true
|
||||
})
|
||||
).rejects.toThrow('no registered provider can observe its host')
|
||||
expect(fallbackProvider.listProcesses).not.toHaveBeenCalled()
|
||||
})
|
||||
|
||||
it('keeps duplicate sweeps behind the runtime physical-stop promise', async () => {
|
||||
let releasePhysicalStop: () => void = () => undefined
|
||||
const physicalStop = new Promise<boolean>((resolve) => {
|
||||
|
||||
@@ -15,8 +15,7 @@ import {
|
||||
describeError,
|
||||
describeFailedPtySweep,
|
||||
describeUnstoppedPtys,
|
||||
verifyUnstoppedPtys,
|
||||
type UnstoppedPtyVerdict
|
||||
resolveUnstoppedPtyVerdict
|
||||
} from './unstopped-pty-verification'
|
||||
|
||||
// Why: normal inventories still coalesce into one process scan, while a stale
|
||||
@@ -231,10 +230,15 @@ export async function killAllProcessesForWorktree(
|
||||
[...stopAttempts].map(async ([ptyId, stopped]) => [ptyId, await stopped] as const)
|
||||
)
|
||||
const failedPtyIds = stopResults.filter(([, stopped]) => !stopped).map(([ptyId]) => ptyId)
|
||||
const verdict: UnstoppedPtyVerdict =
|
||||
failedPtyIds.length === 0
|
||||
? { status: 'exited' }
|
||||
: await verifyUnstoppedPtys(failedPtyIds, deps.localProvider, sweepBudgetMs)
|
||||
const verdict = await resolveUnstoppedPtyVerdict(
|
||||
failedPtyIds,
|
||||
deps.localProvider,
|
||||
sweepBudgetMs,
|
||||
deps.includeProviderInventory !== false ||
|
||||
(deps.resolvedConnectionId === undefined &&
|
||||
deps.resolvedRuntimeEnvironmentId === undefined),
|
||||
deps.runtime
|
||||
)
|
||||
if (verdict.status === 'exited') {
|
||||
for (const ptyId of failedPtyIds) {
|
||||
clearStoppedPtyState(ptyId, deps.onPtyStopped)
|
||||
@@ -363,13 +367,10 @@ async function sweepRegistryForWorktree(
|
||||
}
|
||||
|
||||
function clearStoppedPtyState(ptyId: string, onPtyStopped?: (ptyId: string) => void): void {
|
||||
if (!onPtyStopped) {
|
||||
return
|
||||
}
|
||||
try {
|
||||
// Why: daemon shutdown does not always fan a local pty:exit event back
|
||||
// through pty.ts, but removed worktrees must immediately drop memory rows.
|
||||
onPtyStopped(ptyId)
|
||||
onPtyStopped?.(ptyId)
|
||||
} catch {
|
||||
/* cleanup is best-effort and must not block git-level removal */
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user