starred/orca
1
0
Fork 0
mirror of https://github.com/stablyai/orca.git synced 2026-10-23 08:03:39 +00:00
Code Issues Packages Projects Releases Wiki Activity

test(serve): require exact retained terminal inventory

Browse Source
This commit is contained in:
Jinwoo-H
2026-08-23 18:50:29 -04:00
parent 3c069a4d73
commit db9d48e01b
2 changed files with 11 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
config/reliability-gates.jsonc
+6 -6
View File
@@ -2379,8 +2379,8 @@
"https://github.com/stablyai/orca/issues/16084",
"https://github.com/stablyai/orca/pull/16120"
],
"invariant": "Killing one exact offscreen Chromium renderer cannot replace or terminate the headless main process, runtime, daemon, or unrelated daemon-backed PTYs during the immediate fault or the reported nine-second incident interval. The host session inventory retains each original tab-to-PTY binding, and both original PTY processes continue accepting input.",
"oracle": "Start an isolated headless serve runtime with zero initial BrowserWindows, create two daemon-backed terminal fixtures, and record main, runtime, daemon, tab, PTY, fixture-process, WebContents, and renderer-process identities. Fence one marked offscreen page by WebContents, URL, and OS process, force that renderer to crash, and require its original process to die. Observe beyond the reported nine-second renderer-victim-to-unit-failure interval while repeatedly requiring no main-process close, unchanged main/runtime/daemon identities, both host tab-to-PTY bindings, and I/O through both original fixture processes.",
"invariant": "Killing one exact offscreen Chromium renderer cannot replace or terminate the headless main process, runtime, daemon, or unrelated daemon-backed PTYs during the immediate fault or the reported nine-second incident interval. The complete host terminal tab-to-PTY inventory remains identical to baseline, and both original PTY processes continue accepting input.",
"oracle": "Start an isolated headless serve runtime with zero initial BrowserWindows, create two daemon-backed terminal fixtures, and record main, runtime, daemon, complete terminal tab-to-PTY inventory, fixture-process, WebContents, and renderer-process identities. Fence one marked offscreen page by WebContents, URL, and OS process, force that renderer to crash, and require its original process to die. Observe beyond the reported nine-second renderer-victim-to-unit-failure interval while repeatedly requiring no main-process close, unchanged main/runtime/daemon identities, an unchanged complete host terminal inventory, and I/O through both original fixture processes.",
"commands": [
"pnpm exec electron-vite build --mode e2e",
"SKIP_BUILD=1 pnpm exec playwright test tests/e2e/headless-serve-renderer-kill-retention.spec.ts --config tests/playwright.config.ts --project electron-headless --workers=1"
@@ -2392,7 +2392,7 @@
"assertions": [
"the exact original renderer process dies after the fenced WebContents fault",
"main, runtime, and daemon identities remain unchanged and reachable",
"both original host tab-to-PTY bindings remain published",
"the complete host terminal tab-to-PTY inventory remains identical to baseline",
"both original PTY processes accept post-fault input"
]
}
@@ -2405,7 +2405,7 @@
"command": "SKIP_BUILD=1 pnpm exec playwright test tests/e2e/headless-serve-renderer-kill-retention.spec.ts --config tests/playwright.config.ts --project electron-headless --workers=1",
"result": "passed",
"durationSeconds": 33,
"summary": "The exact offscreen renderer process died while the headless main process, runtime, daemon, both original host tab-to-PTY bindings, and both original PTY processes remained live; repeated stability checkpoints beyond the reported nine-second interval retained post-fault I/O."
"summary": "The exact offscreen renderer process died while the headless main process, runtime, daemon, complete baseline terminal tab-to-PTY inventory, and both original PTY processes remained live; repeated stability checkpoints beyond the reported nine-second interval retained post-fault I/O."
}
],
"runtimeBudget": {
@@ -2418,7 +2418,7 @@
},
"redGreenEvidence": {
"status": "partial",
"evidence": "The earlier immediate-containment oracle passed on reported v1.4.184 and current main: renderer-only death did not reproduce runtime or session loss in the isolated host topology. The final candidate adds repeated checks through a 12-second stability window; that extension has candidate evidence only. The process or event that ended the reported unit remains unidentified. Reverting the candidate removes the characterization coverage without changing product behavior."
"evidence": "The earlier immediate-containment oracle passed on reported v1.4.184 and then-current main e3327c2f31: renderer-only death did not reproduce runtime or session loss in the isolated host topology. The final candidate, rebased onto main ab3b1d07cd, adds complete-inventory checks through a 12-second stability window; that extension has candidate evidence only. The process or event that ended the reported unit remains unidentified. Reverting the candidate removes the characterization coverage without changing product behavior."
},
"performanceBudget": {
"required": false,
@@ -2435,7 +2435,7 @@
"The test does not require the crashed browser surface to recover; browser-page recovery is a separate lifecycle contract.",
"SSH, WSL, folder-workspace, multi-client, mixed-version, and whole-service restart journeys are uncollected; no production or wire behavior changes in this gate."
],
"demotionRule": "Demote if the exact renderer fault changes main/runtime/daemon identity, removes either tab-to-PTY binding, replaces either fixture process, loses post-fault PTY I/O, or cannot identify the killed renderer unambiguously."
"demotionRule": "Demote if the exact renderer fault changes main/runtime/daemon identity, changes the complete terminal tab-to-PTY inventory, replaces either fixture process, loses post-fault PTY I/O, or cannot identify the killed renderer unambiguously."
},
{
"id": "runtime.websocket-heartbeat-cadence",
tests/e2e/headless-serve-renderer-kill-retention.spec.ts
+5 -2
View File
@@ -117,6 +117,9 @@ test('STA-5228 renderer death keeps headless runtime and PTYs alive', async ({ t
fixturePath,
path.join(scratch, 'pty-b.log')
)
const baselineInventory = await readHostTerminalInventory(call, worktreeId)
expect(baselineInventory.ptyIdByTabId[terminalA.tabId]).toBe(terminalA.ptyId)
expect(baselineInventory.ptyIdByTabId[terminalB.tabId]).toBe(terminalB.ptyId)
const proveHostState = async (marker: string): Promise<void> => {
const status = await launchedHost.client.call<RuntimeStatus>('status.get')
const processLiveness = await processIdentityLiveness([mainIdentity, daemonIdentity])
@@ -126,8 +129,8 @@ test('STA-5228 renderer death keeps headless runtime and PTYs alive', async ({ t
expect(readDaemonPid(userDataDir)).toBe(daemonPid)
expect(processLiveness.get(daemonPid)).toBe(true)
const inventory = await readHostTerminalInventory(call, worktreeId)
expect(inventory.ptyIdByTabId[terminalA.tabId]).toBe(terminalA.ptyId)
expect(inventory.ptyIdByTabId[terminalB.tabId]).toBe(terminalB.ptyId)
expect(inventory.tabIds.toSorted()).toEqual(baselineInventory.tabIds.toSorted())
expect(inventory.ptyIdByTabId).toEqual(baselineInventory.ptyIdByTabId)
await proveSameLivePty(call, terminalA, `${marker}_A`)
await proveSameLivePty(call, terminalB, `${marker}_B`)
}