fix(onboard): use deadline wait for gateway recovery - #6320
Merged
cv merged 10 commits intoJul 8, 2026
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
This PR updates the gateway recovery path in the CLI onboarding flow to use the shared waitUntilAsync helper instead of a hand-rolled polling loop, and improves timeout reporting to reflect the configured recovery wait budget.
Changes:
- Replaces the fixed
for-loop polling instartTargetGatewayForRecoverywithwaitUntilAsync, including a computed deadline budget. - Enhances timeout errors to include the configured wait budget and poll attempt/interval configuration.
- Adds a regression test asserting the configured attempt/interval budget is applied without sleeping after the final probe.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| src/lib/onboard/gateway-recovery.ts | Switches gateway recovery polling to waitUntilAsync, adds budget computation/formatting, and improves timeout error text. |
| src/lib/onboard/gateway-recovery.test.ts | Injects a sleepSeconds dep and adds a test validating the attempt/interval budget and sleep behavior. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Contributor
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
📝 WalkthroughWalkthroughGateway recovery now polls until a configured deadline using injectable clock, sleep, and readiness checks. The failure path reports the deadline budget and interval. Tests cover timeout, first-probe success, retry success, and zero-budget failure. ChangesGateway recovery timing refactor
Estimated code review effort: 3 (Moderate) | ~25 minutes Sequence Diagram(s)sequenceDiagram
participant startGatewayForRecovery
participant waitUntilAsync
participant Gateway probe
startGatewayForRecovery->>waitUntilAsync: poll until deadline
loop each attempt
waitUntilAsync->>Gateway probe: check connectivity, health, HTTP readiness
Gateway probe-->>waitUntilAsync: ready or not ready
waitUntilAsync->>waitUntilAsync: sleepSeconds(interval) unless final attempt
end
alt gateway ready
waitUntilAsync-->>startGatewayForRecovery: success
startGatewayForRecovery->>startGatewayForRecovery: set OPENSHELL_GATEWAY
else deadline exceeded
waitUntilAsync-->>startGatewayForRecovery: timeout
startGatewayForRecovery-->>startGatewayForRecovery: throw error with budget and interval
end
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
Signed-off-by: Ho Lim <subhoya@gmail.com>
HOYALIM
force-pushed
the
codex/issue-3768-gateway-recovery-deadline
branch
from
July 6, 2026 10:50
cad6567 to
c9f3979
Compare
wscurran
added
area: cli
Contributor
|
✨ Thanks for the PR. This fixes the gateway recovery path during onboarding by replacing the fixed polling loop with a deadline-based wait helper and improves timeout error messages. Ready for maintainer review. Related open issues: Related open issues: |
Collaborator
|
Hey, nice small refactor. A few things to work through before we merge: 1. The error message over-promises the wait budget. The message says "configured 6s recovery wait budget (3 attempts, 2s interval)" but the loop can actually exit at the maxAttempts cap before the deadline fires. On fast-failing probes it stops earlier than 6s, and on slow probes the reported budget is optimistic. Two ways to reconcile:
2. Scope vs the linked issue. #3768's acceptance criterion is broader than gateway recovery. There are still fixed poll loops in 3. Missing success-path test coverage. The new test pins the timeout shape and the no-final-sleep behavior, which is great. But there's no test for the actual happy path: immediate success without sleeping, retry-then-success, Small nit: the |
Collaborator
|
Please run the Dispatch: Optionally worth running |
Contributor
E2E Target Results — ✅ All selected jobs passedRun: 28950873219
|
…ss paths Address the advisor review on PR NVIDIA#6320. 1. Message/behavior reconciliation. The prior message said "configured Xs recovery wait budget" while waitUntilAsync can also terminate via maxAttempts, so the reported budget was an upper bound rather than the actual guarantee. Rewrite as "N recovery attempt(s) at Ns interval (max wait Xs)" so an operator scanning the error sees both dimensions: the attempt cap that likely fired, and the hard upper time bound the loop cannot exceed. 2. Add success-path coverage. Inject isGatewayHealthy and isGatewayHttpReady as optional deps (default to the production implementations) so caller-level tests can pin the happy path without standing up a real gateway. New tests: - "succeeds on the first healthy probe without sleeping and sets OPENSHELL_GATEWAY" - "succeeds after retrying past unhealthy probes and still sets OPENSHELL_GATEWAY" - "with NEMOCLAW_HEALTH_POLL_COUNT=0 fails fast without invoking the probe" (edge case: zero attempts must not silently pass) 3. Document the sleep unit adapter with a one-line comment noting waitUntilAsync passes durations in ms while the injected sleeper (sleepSeconds) takes seconds; adapt at the boundary only. 11/11 tests pass locally. Biome and the changed-file typecheck are clean. Pre-existing typecheck error in test/helpers/mcp-lifecycle- lock-properties.ts is unrelated to this branch's diff. Co-Authored-By: Ho Lim <holim@nvidia.com> Signed-off-by: Charan Jagwani <cjagwani@nvidia.com>
Contributor
E2E Target Results — ✅ All selected jobs passedRun: 28952907265
|
cjagwani
added a commit
that referenced
this pull request
Jul 8, 2026
…6320 advisor) Advisor re-review kept blocking on the same concern: with maxAttempts still passed to waitUntilAsync alongside the deadline, a fast-failing probe sequence exits after N attempts even though the deadline still permits more polling. The reported "max wait Xs" language then under-promises the wait the system was willing to spend, and the loop regresses to the exact fixed-attempt behavior #3768 is meant to replace. Fix: 1. Drop `maxAttempts: recoveryPollCount` from the waitUntilAsync options. The loop is now purely deadline-driven; the interval and probe cost naturally bound the total attempt count. Under a 3 * 2s = 6s budget with fast-failing mocked probes, the loop now runs 3 probes and 3 sleeps until the top-of-loop deadline check terminates it, instead of the earlier 3 probes / 2 sleeps under the attempt cap. 2. Rewrite the error message to describe the actual deadline, not a hedged "budget + attempt cap" mix: "Gateway 'X' did not become ready within the configured Ns recovery deadline (Is poll interval)". 3. Inject Date.now via a new `now` deps hook so the deadline test can drive a captured virtual clock without vi.useFakeTimers globally patching timers (which hangs the async loop). The virtual clock advances only when the injected sleepSeconds mock is called, so time progresses deterministically at unit-test speed with zero real wall-clock waits. 4. Rewrite the timeout test to match the new deadline-only semantics: asserts 3 probes and 3 sleeps at 2s each under a 6s budget, and that the error message mentions "6s recovery deadline (2s poll interval)". 5. Adjust the zero-count edge test's message assertion to match the new deadline-only error text. 11/11 tests pass. Biome and changed-file typecheck are clean. The pre-existing typecheck error in test/helpers/mcp-lifecycle-lock- properties.ts is unrelated to this branch's diff. Signed-off-by: Charan Jagwani <cjagwani@nvidia.com>
…VIDIA#6320 advisor) Advisor re-review kept blocking on the same concern: with maxAttempts still passed to waitUntilAsync alongside the deadline, a fast-failing probe sequence exits after N attempts even though the deadline still permits more polling. The reported "max wait Xs" language then under-promises the wait the system was willing to spend, and the loop regresses to the exact fixed-attempt behavior NVIDIA#3768 is meant to replace. Fix: 1. Drop `maxAttempts: recoveryPollCount` from the waitUntilAsync options. The loop is now purely deadline-driven; the interval and probe cost naturally bound the total attempt count. Under a 3 * 2s = 6s budget with fast-failing mocked probes, the loop now runs 3 probes and 3 sleeps until the top-of-loop deadline check terminates it, instead of the earlier 3 probes / 2 sleeps under the attempt cap. 2. Rewrite the error message to describe the actual deadline, not a hedged "budget + attempt cap" mix: "Gateway 'X' did not become ready within the configured Ns recovery deadline (Is poll interval)". 3. Inject Date.now via a new `now` deps hook so the deadline test can drive a captured virtual clock without vi.useFakeTimers globally patching timers (which hangs the async loop). The virtual clock advances only when the injected sleepSeconds mock is called, so time progresses deterministically at unit-test speed with zero real wall-clock waits. 4. Rewrite the timeout test to match the new deadline-only semantics: asserts 3 probes and 3 sleeps at 2s each under a 6s budget, and that the error message mentions "6s recovery deadline (2s poll interval)". 5. Adjust the zero-count edge test's message assertion to match the new deadline-only error text. 11/11 tests pass. Biome and changed-file typecheck are clean. The pre-existing typecheck error in test/helpers/mcp-lifecycle-lock- properties.ts is unrelated to this branch's diff. Signed-off-by: Charan Jagwani <cjagwani@nvidia.com>
Contributor
E2E Target Results — ✅ All selected jobs passedRun: 28953876797
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (1)
src/lib/onboard/gateway-recovery.test.ts (1)
118-125: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winProbe count derived from a magic subprocess-call divisor.
probeCount = runCaptureCalls / 3hard-codes the current number ofrunCaptureOpenshellcalls per probe cycle (status + gateway info by name + gateway info current). If the production probe sequence adds/removes a subprocess call, this test silently miscounts without failing meaningfully, and it couples the test to implementation shape rather than the observable retry cadence.As per path instructions for
**/*.test.{ts,js,mts,mjs,cts,cjs}: "Prefer observable outcomes through the public boundary over source-text, private-shape, or mock-call assertions."Consider asserting on a more explicit signal — e.g., have the mocked
runCaptureOpenshellincrement a probe counter only on the first call of each cycle (args[0] === "status"), rather than deriving it via division.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/lib/onboard/gateway-recovery.test.ts` around lines 118 - 125, The test in gateway-recovery.test.ts is inferring probe count from a hard-coded subprocess-call divisor, which ties it to implementation details of runCaptureOpenshell. Update the assertion to count observable probe cycles directly by using the mocked runCaptureOpenshell and incrementing a probe counter only when the first call in each cycle occurs (for example, the status command), instead of dividing total mock calls by 3. Keep the expectation focused on the retry cadence rather than the internal call shape.Source: Path instructions
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/lib/onboard/gateway-recovery.test.ts`:
- Line 105: The new behavior-oriented test titles in gateway-recovery.test.ts
are missing the required local issue-reference suffix. Update the affected test
names in the relevant `it(...)` blocks to end with the same final `(`#3768`)`
suffix, keeping the titles behavior-focused while preserving the existing intent
in the test descriptions.
- Around line 105-129: The timeout test in gateway-recovery should prove it is
stopping on the recovery deadline, not just on a fixed attempt count. Update the
test around startGatewayForRecovery, makeVirtualClock, and the health-check
probe path so each probe advances the injected clock (for example via the mocked
subprocess/health check behavior), which makes deadline-driven polling and a
hidden maxAttempts limit diverge. Then assert on the resulting probe/sleep count
and timeout message in a way that only passes when the loop exits because the
configured deadline is reached.
---
Nitpick comments:
In `@src/lib/onboard/gateway-recovery.test.ts`:
- Around line 118-125: The test in gateway-recovery.test.ts is inferring probe
count from a hard-coded subprocess-call divisor, which ties it to implementation
details of runCaptureOpenshell. Update the assertion to count observable probe
cycles directly by using the mocked runCaptureOpenshell and incrementing a probe
counter only when the first call in each cycle occurs (for example, the status
command), instead of dividing total mock calls by 3. Keep the expectation
focused on the retry cadence rather than the internal call shape.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Enterprise
Run ID: 90bdbcb4-486d-4d5a-8864-f1bc33c26af8
📒 Files selected for processing (2)
src/lib/onboard/gateway-recovery.test.tssrc/lib/onboard/gateway-recovery.ts
🚧 Files skipped from review as they are similar to previous changes (1)
- src/lib/onboard/gateway-recovery.ts
…NVIDIA#6320) Advisor in-scope cleanup: `beforeEach` was left over from the earlier vi.useFakeTimers approach that the virtual-clock refactor replaced. Not used anywhere in the file. Signed-off-by: Charan Jagwani <cjagwani@nvidia.com>
…VIDIA#3768) Address two CodeRabbit nits on PR NVIDIA#6320. 1. Test titles missing local issue reference (Major, quick win). Add the required `(NVIDIA#3768)` suffix to the four new tests so they match the repo's test-title-style contract. 2. Deadline-driven proof (Major, heavy lift). The prior timeout test used fast-failing probes that only advanced the clock via the sleep mock. Under that setup, a hidden `maxAttempts: recoveryPollCount` would have produced the same probe/sleep counts as a pure deadline, so the test could not tell them apart. Redesign the timeout test to make probes ALSO advance the virtual clock: NEMOCLAW_HEALTH_POLL_COUNT=10, NEMOCLAW_HEALTH_POLL_INTERVAL=1 with each probe consuming 1s of virtual time. Under the pure-deadline implementation the loop runs ~5 iterations before the 10s budget is exhausted (probes and sleeps consume 2s per iteration cumulatively). Under a hidden attempt cap of 10 the loop would run exactly 10 iterations. The new assertion `probeCount < 10` therefore only passes when the deadline (not an attempt cap) terminates the loop, which is the invariant NVIDIA#3768 asks for. Expose `advance(seconds)` on the makeVirtualClock helper so a mocked subprocess-probe implementation can move time forward while keeping the sleeper as the primary in-loop clock driver. 11/11 tests still pass. Biome and repo checks clean. Signed-off-by: Charan Jagwani <cjagwani@nvidia.com>
cjagwani
approved these changes
Jul 8, 2026
…IA#6320) The new deadline-driven timeout test added a mocked runCaptureOpenshell that used `if (probesRun * 3 === ...)` to advance the virtual clock once per probe. That single new `if` tripped codebase-growth-guardrails' "no new if statements in changed test files" gate. Move the branching into a named helper `advanceOnStatusCall` that uses a ternary, so the test body stays linear and the guardrail passes. 11/11 tests still pass. Signed-off-by: Charan Jagwani <cjagwani@nvidia.com>
Contributor
E2E Target Results — ✅ All selected jobs passedRun: 28955158984
|
CI's `Files were modified by following hooks` step tripped on the prior commit: Biome format collapsed the `advanceOnStatusCall` ternary from multi-line onto a single line. Apply the same formatter locally so the committed state matches what CI expects. Signed-off-by: Charan Jagwani <cjagwani@nvidia.com>
21 tasks
cv
pushed a commit
that referenced
this pull request
Jul 9, 2026
<!-- markdownlint-disable MD041 --> ## Summary Prepares the user documentation for NemoClaw v0.0.78 by replacing the unreleased section with release highlights and synchronizing the affected inference, lifecycle, messaging, and CLI reference pages with merged behavior. ## Changes - Publish the v0.0.78 release-notes section with links to the most specific user guides for each shipped behavior. - Document authoritative Deep Agents route health, Nemotron Ultra profile behavior, and Hermes compatible-endpoint context metadata. - Document forced rebuild recovery after total backup failure and the ownership-safe tunnel/full-stop behavior. - Keep command examples and shared agent variants aligned with the current OpenClaw, Hermes, and Deep Agents interfaces. Source mapping: - [#3787](#3787) -> `docs/about/release-notes.mdx`: Record reliable workspace template seeding during sandbox startup. - [#4960](#4960) -> `docs/about/release-notes.mdx`: Record safer detection of rewritten OpenClaw gateway processes. - [#5676](#5676) -> `docs/about/release-notes.mdx`: Record warning-tolerant agent-list JSON handling. - [#5857](#5857) -> `docs/about/release-notes.mdx`: Record synchronization of explicit OpenClaw main-agent model state. - [#5929](#5929) -> `docs/about/release-notes.mdx`: Record copyable SSH port-forward guidance for remote dashboards. - [#6068](#6068) -> `docs/about/release-notes.mdx`: Record custom-image plugin provenance reconciliation. - [#6116](#6116) -> `docs/about/release-notes.mdx`: Record live-loopback dashboard-forward recovery. - [#6122](#6122) -> `docs/about/release-notes.mdx`: Announce validated, round-trippable policy YAML output. - [#6211](#6211) -> `docs/manage-sandboxes/lifecycle.mdx`, `docs/reference/commands.mdx`, `docs/about/release-notes.mdx`: Explain the explicit no-backup `rebuild --force` recovery boundary. - [#6283](#6283) -> `docs/about/release-notes.mdx`: Record Hermes WebUI port alignment. - [#6293](#6293) -> `docs/inference/switch-inference-providers.mdx`, `docs/about/release-notes.mdx`: Document compatible-endpoint context-length probing for Hermes. - [#6320](#6320) -> `docs/about/release-notes.mdx`: Record bounded gateway-recovery waits. - [#6377](#6377) -> `docs/reference/commands.mdx`, `docs/about/release-notes.mdx`: Explain rebuild diagnostics and prepared MCP-destroy recovery. - [#6412](#6412) -> `docs/get-started/quickstart-langchain-deepagents-code.mdx`, `docs/about/release-notes.mdx`: Document authoritative agent-visible inference route health. - [#6421](#6421) -> `docs/about/release-notes.mdx`: Record the longer quiet-pull window for managed vLLM images. - [#6431](#6431) -> `docs/inference/model-capability-audit.mdx`, `docs/about/release-notes.mdx`: Document the version-pinned Nemotron Ultra profile plugin. - [#6439](#6439) -> `docs/about/release-notes.mdx`: Summarize the authenticated, pinned credential-capture helper boundary. - [#6450](#6450) -> `docs/manage-sandboxes/messaging-channels.mdx`, `docs/reference/commands.mdx`, `docs/about/release-notes.mdx`: Document host-forward cleanup and ownership-safe gateway-port release. - [#6474](#6474) -> `docs/manage-sandboxes/messaging-channels.mdx`, `docs/about/release-notes.mdx`: Record composable OpenClaw messaging runtime loaders. - [#6475](#6475) -> `docs/about/release-notes.mdx`: Record removal of the unavailable Kimi K2.6 production endpoint option. - [#6480](#6480) -> `docs/about/release-notes.mdx`: Record stderr routing for the plugin registration banner. - [#6481](#6481) -> `docs/about/release-notes.mdx`: Record post-pull Ollama model discovery checks. - [#6482](#6482) -> `docs/about/release-notes.mdx`: Record Ollama model warm-up after daemon restart. - [#6486](#6486) -> `docs/about/release-notes.mdx`: Publish the opt-in, thread-scoped Deep Agents auto-approval boundary. - [#6490](#6490) -> `docs/about/release-notes.mdx`: Record diagnostics for custom images missing the managed runtime. - [#6494](#6494) -> `docs/inference/model-capability-audit.mdx`, `docs/about/release-notes.mdx`: Document nonempty tool-call content preservation and placeholder rejection. - [#6497](#6497) -> `docs/get-started/quickstart-langchain-deepagents-code.mdx`, `docs/about/release-notes.mdx`: Document isolated Deep Agents route-probe output. - [#6506](#6506) -> `docs/get-started/quickstart-langchain-deepagents-code.mdx`, `docs/about/release-notes.mdx`: Document observability-preserving managed route probes. - [#6508](#6508) -> `docs/about/release-notes.mdx`: Link the new extension taxonomy and SDK-readiness reference from the release summary. Release-source verification: GitHub reports all 29 cited source PRs as merged with base `main`, and every merge commit is an ancestor of `origin/main` at `17bf9a6a9688b3b1d69cf4b37d3f23110acb055e`. No source-mapping mismatches were found. ## Type of Change - [ ] Code change (feature, bug fix, or refactor) - [ ] Code change with doc updates - [ ] Doc only (prose changes, no code sample modifications) - [x] Doc only (includes code sample changes) ## Quality Gates <!-- Check exactly one tests line and one docs line. Check other lines when applicable. Add every requested justification or approval reference. --> - [ ] Tests added or updated for changed behavior - [ ] Existing tests cover changed behavior — justification: - [x] Tests not applicable — justification: Documentation-only release-prep changes; `npm run docs` validates variants, routes, and Fern content. - [x] Docs updated for user-facing behavior changes - [ ] Docs not applicable — justification: - [ ] Sensitive paths changed (security, policy, credentials, preflight, onboarding, inference, runner, sandbox, or messaging) - [ ] Sensitive-path review completed or maintainer-approved waiver recorded — reviewer/approval link/justification: - [ ] Non-success, skipped, or missing CI check accepted by maintainer — check name, approval link, and follow-up issue: ## Verification <!-- Check each applicable item only when supported by the requested evidence. Run targeted tests once per relevant change set and rerun after later edits or hook autofixes that can affect the tested behavior. Do not rerun hook-covered checks. --> - [x] PR description includes the DCO sign-off declaration and every commit appears as `Verified` in GitHub - [x] Normal `pre-commit`, `commit-msg`, and `pre-push` hooks passed, or `npm run check:diff` passed when hooks were skipped or unavailable - [x] Targeted behavior tests pass for the current change set, or tests are marked not applicable above — command/result or justification: Tests are not applicable to this documentation-only change set. - [ ] Applicable broad gate passed — `npm test` for broad runtime/test-harness changes; `npm run check` for repo-wide validation/coverage changes — command/result: - [x] Quality Gates section completed with required justifications or waivers - [x] No secrets, API keys, or credentials committed - [ ] `npm run docs` builds without warnings (doc changes only) — exited 0 with zero errors; Fern reported the existing unauthenticated redirect-check and light-mode contrast warnings. - [x] Doc pages follow the [style guide](https://github.com/NVIDIA/NemoClaw/blob/main/docs/CONTRIBUTING.md) (doc changes only) - [ ] New doc pages include SPDX header and frontmatter (new pages only) --- <!-- DCO sign-off is required in this PR description, and every commit must appear as Verified in GitHub. Run: git config user.name && git config user.email --> Signed-off-by: Charan Jagwani <cjagwani@nvidia.com> --------- Signed-off-by: cjagwani <cjagwani@nvidia.com>
Hadar301
pushed a commit
to Hadar301/NemoClaw-OpenShift
that referenced
this pull request
Jul 12, 2026
## Summary - replace the fixed gateway recovery polling loop with the shared deadline/attempt helper waitUntilAsync - report the actual recovery attempt count and interval alongside the upper wait bound in timeout errors - cover the configured attempt/interval budget without sleeping after the final probe - add caller-level tests for the immediate-success, retry-then-success, and zero-count-edge paths ## Scope This PR converts the gateway recovery polling loop only. NVIDIA#3768's broader acceptance criterion (all remaining fixed readiness polls in onboard become deadline-based) is not fully addressed here: fixed poll loops in `src/lib/onboard.ts` (around lines 2205 and 2328) using `NEMOCLAW_HEALTH_POLL_COUNT` / `NEMOCLAW_HEALTH_POLL_INTERVAL` remain. Follow-up PRs will address those. Refs NVIDIA#3768 ## Tests - `npx vitest run src/lib/onboard/gateway-recovery.test.ts` - `npm run build:cli` - `npm run typecheck:cli` - `npx biome check src/lib/onboard/gateway-recovery.ts src/lib/onboard/gateway-recovery.test.ts` <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Improved gateway recovery to use deadline-based readiness polling with consistent probe cadence. * Enhanced failure messaging to report the configured recovery deadline/budget when readiness isn’t reached. * Ensures readiness is detected promptly and the gateway readiness flag is set on success. * **Tests** * Expanded gateway recovery coverage with deterministic virtual time and injectable timing controls. * Added assertions for timeout budget calculations, exact probe/sleep cadence, success on immediate/after recovery, and fast-fail when polling is disabled. <!-- end of auto-generated comment: release notes by coderabbit.ai --> Signed-off-by: Ho Lim <holim@nvidia.com> Signed-off-by: Charan Jagwani <cjagwani@nvidia.com> --------- Signed-off-by: Ho Lim <subhoya@gmail.com> Signed-off-by: Charan Jagwani <cjagwani@nvidia.com> Co-authored-by: cjagwani <cjagwani@nvidia.com> Co-authored-by: Ho Lim <holim@nvidia.com>
Hadar301
pushed a commit
to Hadar301/NemoClaw-OpenShift
that referenced
this pull request
Jul 12, 2026
<!-- markdownlint-disable MD041 --> ## Summary Prepares the user documentation for NemoClaw v0.0.78 by replacing the unreleased section with release highlights and synchronizing the affected inference, lifecycle, messaging, and CLI reference pages with merged behavior. ## Changes - Publish the v0.0.78 release-notes section with links to the most specific user guides for each shipped behavior. - Document authoritative Deep Agents route health, Nemotron Ultra profile behavior, and Hermes compatible-endpoint context metadata. - Document forced rebuild recovery after total backup failure and the ownership-safe tunnel/full-stop behavior. - Keep command examples and shared agent variants aligned with the current OpenClaw, Hermes, and Deep Agents interfaces. Source mapping: - [NVIDIA#3787](NVIDIA#3787) -> `docs/about/release-notes.mdx`: Record reliable workspace template seeding during sandbox startup. - [NVIDIA#4960](NVIDIA#4960) -> `docs/about/release-notes.mdx`: Record safer detection of rewritten OpenClaw gateway processes. - [NVIDIA#5676](NVIDIA#5676) -> `docs/about/release-notes.mdx`: Record warning-tolerant agent-list JSON handling. - [NVIDIA#5857](NVIDIA#5857) -> `docs/about/release-notes.mdx`: Record synchronization of explicit OpenClaw main-agent model state. - [NVIDIA#5929](NVIDIA#5929) -> `docs/about/release-notes.mdx`: Record copyable SSH port-forward guidance for remote dashboards. - [NVIDIA#6068](NVIDIA#6068) -> `docs/about/release-notes.mdx`: Record custom-image plugin provenance reconciliation. - [NVIDIA#6116](NVIDIA#6116) -> `docs/about/release-notes.mdx`: Record live-loopback dashboard-forward recovery. - [NVIDIA#6122](NVIDIA#6122) -> `docs/about/release-notes.mdx`: Announce validated, round-trippable policy YAML output. - [NVIDIA#6211](NVIDIA#6211) -> `docs/manage-sandboxes/lifecycle.mdx`, `docs/reference/commands.mdx`, `docs/about/release-notes.mdx`: Explain the explicit no-backup `rebuild --force` recovery boundary. - [NVIDIA#6283](NVIDIA#6283) -> `docs/about/release-notes.mdx`: Record Hermes WebUI port alignment. - [NVIDIA#6293](NVIDIA#6293) -> `docs/inference/switch-inference-providers.mdx`, `docs/about/release-notes.mdx`: Document compatible-endpoint context-length probing for Hermes. - [NVIDIA#6320](NVIDIA#6320) -> `docs/about/release-notes.mdx`: Record bounded gateway-recovery waits. - [NVIDIA#6377](NVIDIA#6377) -> `docs/reference/commands.mdx`, `docs/about/release-notes.mdx`: Explain rebuild diagnostics and prepared MCP-destroy recovery. - [NVIDIA#6412](NVIDIA#6412) -> `docs/get-started/quickstart-langchain-deepagents-code.mdx`, `docs/about/release-notes.mdx`: Document authoritative agent-visible inference route health. - [NVIDIA#6421](NVIDIA#6421) -> `docs/about/release-notes.mdx`: Record the longer quiet-pull window for managed vLLM images. - [NVIDIA#6431](NVIDIA#6431) -> `docs/inference/model-capability-audit.mdx`, `docs/about/release-notes.mdx`: Document the version-pinned Nemotron Ultra profile plugin. - [NVIDIA#6439](NVIDIA#6439) -> `docs/about/release-notes.mdx`: Summarize the authenticated, pinned credential-capture helper boundary. - [NVIDIA#6450](NVIDIA#6450) -> `docs/manage-sandboxes/messaging-channels.mdx`, `docs/reference/commands.mdx`, `docs/about/release-notes.mdx`: Document host-forward cleanup and ownership-safe gateway-port release. - [NVIDIA#6474](NVIDIA#6474) -> `docs/manage-sandboxes/messaging-channels.mdx`, `docs/about/release-notes.mdx`: Record composable OpenClaw messaging runtime loaders. - [NVIDIA#6475](NVIDIA#6475) -> `docs/about/release-notes.mdx`: Record removal of the unavailable Kimi K2.6 production endpoint option. - [NVIDIA#6480](NVIDIA#6480) -> `docs/about/release-notes.mdx`: Record stderr routing for the plugin registration banner. - [NVIDIA#6481](NVIDIA#6481) -> `docs/about/release-notes.mdx`: Record post-pull Ollama model discovery checks. - [NVIDIA#6482](NVIDIA#6482) -> `docs/about/release-notes.mdx`: Record Ollama model warm-up after daemon restart. - [NVIDIA#6486](NVIDIA#6486) -> `docs/about/release-notes.mdx`: Publish the opt-in, thread-scoped Deep Agents auto-approval boundary. - [NVIDIA#6490](NVIDIA#6490) -> `docs/about/release-notes.mdx`: Record diagnostics for custom images missing the managed runtime. - [NVIDIA#6494](NVIDIA#6494) -> `docs/inference/model-capability-audit.mdx`, `docs/about/release-notes.mdx`: Document nonempty tool-call content preservation and placeholder rejection. - [NVIDIA#6497](NVIDIA#6497) -> `docs/get-started/quickstart-langchain-deepagents-code.mdx`, `docs/about/release-notes.mdx`: Document isolated Deep Agents route-probe output. - [NVIDIA#6506](NVIDIA#6506) -> `docs/get-started/quickstart-langchain-deepagents-code.mdx`, `docs/about/release-notes.mdx`: Document observability-preserving managed route probes. - [NVIDIA#6508](NVIDIA#6508) -> `docs/about/release-notes.mdx`: Link the new extension taxonomy and SDK-readiness reference from the release summary. Release-source verification: GitHub reports all 29 cited source PRs as merged with base `main`, and every merge commit is an ancestor of `origin/main` at `17bf9a6a9688b3b1d69cf4b37d3f23110acb055e`. No source-mapping mismatches were found. ## Type of Change - [ ] Code change (feature, bug fix, or refactor) - [ ] Code change with doc updates - [ ] Doc only (prose changes, no code sample modifications) - [x] Doc only (includes code sample changes) ## Quality Gates <!-- Check exactly one tests line and one docs line. Check other lines when applicable. Add every requested justification or approval reference. --> - [ ] Tests added or updated for changed behavior - [ ] Existing tests cover changed behavior — justification: - [x] Tests not applicable — justification: Documentation-only release-prep changes; `npm run docs` validates variants, routes, and Fern content. - [x] Docs updated for user-facing behavior changes - [ ] Docs not applicable — justification: - [ ] Sensitive paths changed (security, policy, credentials, preflight, onboarding, inference, runner, sandbox, or messaging) - [ ] Sensitive-path review completed or maintainer-approved waiver recorded — reviewer/approval link/justification: - [ ] Non-success, skipped, or missing CI check accepted by maintainer — check name, approval link, and follow-up issue: ## Verification <!-- Check each applicable item only when supported by the requested evidence. Run targeted tests once per relevant change set and rerun after later edits or hook autofixes that can affect the tested behavior. Do not rerun hook-covered checks. --> - [x] PR description includes the DCO sign-off declaration and every commit appears as `Verified` in GitHub - [x] Normal `pre-commit`, `commit-msg`, and `pre-push` hooks passed, or `npm run check:diff` passed when hooks were skipped or unavailable - [x] Targeted behavior tests pass for the current change set, or tests are marked not applicable above — command/result or justification: Tests are not applicable to this documentation-only change set. - [ ] Applicable broad gate passed — `npm test` for broad runtime/test-harness changes; `npm run check` for repo-wide validation/coverage changes — command/result: - [x] Quality Gates section completed with required justifications or waivers - [x] No secrets, API keys, or credentials committed - [ ] `npm run docs` builds without warnings (doc changes only) — exited 0 with zero errors; Fern reported the existing unauthenticated redirect-check and light-mode contrast warnings. - [x] Doc pages follow the [style guide](https://github.com/NVIDIA/NemoClaw/blob/main/docs/CONTRIBUTING.md) (doc changes only) - [ ] New doc pages include SPDX header and frontmatter (new pages only) --- <!-- DCO sign-off is required in this PR description, and every commit must appear as Verified in GitHub. Run: git config user.name && git config user.email --> Signed-off-by: Charan Jagwani <cjagwani@nvidia.com> --------- Signed-off-by: cjagwani <cjagwani@nvidia.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Scope
This PR converts the gateway recovery polling loop only. #3768's broader acceptance criterion (all remaining fixed readiness polls in onboard become deadline-based) is not fully addressed here: fixed poll loops in
src/lib/onboard.ts(around lines 2205 and 2328) usingNEMOCLAW_HEALTH_POLL_COUNT/NEMOCLAW_HEALTH_POLL_INTERVALremain. Follow-up PRs will address those.Refs #3768
Tests
npx vitest run src/lib/onboard/gateway-recovery.test.tsnpm run build:clinpm run typecheck:clinpx biome check src/lib/onboard/gateway-recovery.ts src/lib/onboard/gateway-recovery.test.tsSummary by CodeRabbit
Signed-off-by: Ho Lim holim@nvidia.com
Signed-off-by: Charan Jagwani cjagwani@nvidia.com