fix(ssh): Windows hosts without Add-Type staging; runtime-store GC on Windows (#24149)

* fix(ssh): collect the pinned-Node runtime store on Windows hosts

Windows SSH hosts now run runtime-store GC instead of skipping it: one
PowerShell inventory reads .runtime-ref-node-<sha> and .runtime-node refs from
every version dir, and one Get-CimInstance Win32_Process query filtered on an
image path under runtimes\ adds process holds (never by image name; a failed
query keeps everything). Stale upload stages are swept with the same rule as
POSIX. Promotion and the post-upload hold check now take the store lock on
Windows too, and the lock's own commands run unwrapped there.

Windows relay version-dir liveness now honours .relay-pid (design D5): a live
PID answers ALIVE before any pipe is touched, a dead one (ESRCH) plus refusing
pipes is exited, anything else is unverifiable. The runtime probe adopts a
pinned node.exe an earlier vault reader left without a .verified marker after
running it.

* fix(ssh): Windows stage fencing and vault runtime go through the verified node.exe

Upload-stage file identity on Windows no longer compiles an Add-Type P/Invoke
helper when the relay runs on Orca's verified pinned node.exe: the stage
commands run a fixed fs.lstatSync(..., {bigint:true}) script through it. It
prints the legacy helper's vol:high:low lowercase hex, and identity files are
compared after normalising hex spelling, so old and new clients recover each
other's stages. Host-Node relays keep the legacy helper; the choice is
documented in windows-edr-posture.md.

The Windows OpenCode vault reader now installs the pinned runtime through
ensureRemoteOrcadNodeRuntime (official zip, host-side extraction, .verified,
store lock) instead of uploading a client-extracted node.exe, and the relay dir
gains a .runtime-ref-node-<sha> so store GC keeps the runtime the vault uses.

* test(ssh): run the Windows stage-identity and store-GC tests on the Windows lane

The legacy/node.exe identity compatibility test and the Win32_Process hold path
were gated to win32 but no CI lane ran them. Add both files to the Windows
package lane and a real running-node.exe hold test.

* test(ssh): tear down Windows-lane temp trees through removeTreeSync

* test(ssh): grant the store lock to the Windows OpenCode runtime setup test

The Windows promote now runs under runtimes/.store-lock, so the mocked host
must answer the lock's CreateNew step.

---------

Co-authored-by: m4air <m4air@m4airs-Air.localdomain>
Co-authored-by: m4air <m4air@Mac.localdomain>
This commit is contained in:
OrcaWin
2026-10-01 03:25:42 -07:00
committed by GitHub
co-authored by m4air m4air
parent dfdcfcf61f
commit 14d4bb2e2a
26 changed files with 1196 additions and 519 deletions
+27 -54
View File
@@ -1,16 +1,11 @@
import { randomBytes } from 'node:crypto'
import { copyFile, link, mkdtemp, rm } from 'node:fs/promises'
import { dirname, join } from 'node:path'
import { waitForPromiseWithSignal } from '../../shared/abort-signal-reason'
import { ORCAD_NODE_RUNTIME_WINDOWS_EXECUTABLE } from '../../shared/orcad-artifacts'
import type { SshConnection } from './ssh-connection'
import type { RemoteRuntimeStep } from './orcad-remote-node-runtime'
import {
preparePinnedNodeForVault,
type PinnedNodeVaultUpload
} from './ssh-relay-opencode-pinned-node'
import { preparePinnedNodeForVault } from './ssh-relay-opencode-pinned-node'
import { RUNTIME_REF_NODE_PREFIX } from './remote-node-runtime-store-inventory'
import { execCommand, isUnconfirmedSshCommandTermination } from './ssh-relay-deploy-helpers'
import { uploadRelayDirectory, writeRelayFile } from './ssh-relay-install-transfers'
import { writeRelayFile } from './ssh-relay-install-transfers'
import {
createRelayUploadStageNamespace,
relayUploadStageSftpNamespaceMapping
@@ -28,7 +23,6 @@ import {
import {
parseOpenCodeRuntimeResult,
probeOpenCodeNodeSqliteCommand,
promoteOpenCodeRuntimeCommand,
publishOpenCodeRuntimeReferenceCommand
} from './ssh-relay-opencode-runtime-commands'
@@ -46,6 +40,8 @@ const installations = new WeakMap<
type SetupOptions = {
nodePath: string
/** The relay's verified pinned node.exe, if any; stage fencing then needs no Add-Type (D5). */
verifiedNodePath?: string
relayDir: string
signal?: AbortSignal
cacheRoot?: string
@@ -156,21 +152,28 @@ async function install(
throw new Error('The host did not complete its SQLite read probe.')
}
let executable = node.executable
let upload: PinnedNodeVaultUpload | undefined
let runtimeRef: { path: string; sha256: string } | undefined
let identityNode = options.verifiedNodePath
if (node.status === 'unsupported') {
const pinned = await preparePinnedNodeForVault({
conn,
host,
nodePath: options.nodePath,
relayDir: options.relayDir,
cacheRoot: options.cacheRoot,
referencePath: joinRemotePath(host, options.relayDir, RUNTIME_REFERENCE_NAME),
signal,
exec,
remote
})
executable = pinned.executable
upload = pinned.upload
identityNode ??= pinned.executable
runtimeRef = {
path: joinRemotePath(
host,
options.relayDir,
`${RUNTIME_REF_NODE_PREFIX}${pinned.runtimeSha256}`
),
sha256: pinned.runtimeSha256
}
}
if (!executable) {
throw new Error('The host did not identify its SQLite executable.')
@@ -179,12 +182,13 @@ async function install(
const relativePool = `${RELAY_REMOTE_DIR}/${RELAY_UPLOAD_STAGE_POOL_NAME}`
const poolDir = joinRemotePath(host, remoteHome, relativePool)
const owner = createRelayInstallMarkerFileName()
await exec(recoverOneStaleRelayUploadStageCommand(host, poolDir))
const identity = identityNode ? { node: identityNode } : undefined
await exec(recoverOneStaleRelayUploadStageCommand(host, poolDir, undefined, identity))
const stage = parseReservedRelayUploadStage(
host,
poolDir,
owner,
await exec(reserveRelayUploadStageCommand(host, poolDir, owner))
await exec(reserveRelayUploadStageCommand(host, poolDir, owner, identity))
)
const stageDir = stage.slotDir
const namespace = createRelayUploadStageNamespace(`${relativePool}/${stage.slotName}`, owner)
@@ -194,40 +198,6 @@ async function install(
: undefined
let cleanupAllowed = true
try {
if (upload) {
const { localRuntime } = upload
const localStage = await mkdtemp(join(dirname(localRuntime), '.vault-upload-'))
try {
const binaryName = ORCAD_NODE_RUNTIME_WINDOWS_EXECUTABLE
const localBinary = join(localStage, binaryName)
await link(localRuntime, localBinary).catch(() => copyFile(localRuntime, localBinary))
signal.throwIfAborted()
await remote(() =>
uploadRelayDirectory(conn, localStage, joinRemotePath(host, stageDir, 'payload'), host, {
signal,
sftpNamespace: mapping()
})
)
const promoted = parseOpenCodeRuntimeResult(
await exec(
promoteOpenCodeRuntimeCommand({
host,
nodePath: options.nodePath,
stagedBinary: joinRemotePath(host, stageDir, 'payload', binaryName),
executable,
expectedHash: upload.expectedHash,
repairToken: token
})
)
)
if (promoted.status !== 'ready' || !promoted.executable) {
throw new Error('The host did not verify the uploaded SQLite runtime.')
}
executable = promoted.executable
} finally {
await rm(localStage, { recursive: true, force: true }).catch(() => {})
}
}
const referenceName = RUNTIME_REFERENCE_NAME
const stagedReference = joinRemotePath(host, stageDir, 'payload', referenceName)
signal.throwIfAborted()
@@ -244,7 +214,8 @@ async function install(
nodePath: options.nodePath,
stagedReference,
reference: joinRemotePath(host, options.relayDir, referenceName),
token
token,
runtimeRef
})
)
)
@@ -254,11 +225,13 @@ async function install(
throw error
} finally {
if (cleanupAllowed && !signal.aborted) {
await exec(cleanupOwnedRelayUploadStageCommand(host, stage, owner)).catch((error) => {
if (isUnconfirmedSshCommandTermination(error)) {
throw error
await exec(cleanupOwnedRelayUploadStageCommand(host, stage, owner, identity)).catch(
(error) => {
if (isUnconfirmedSshCommandTermination(error)) {
throw error
}
}
})
)
}
}
}