fix(inference): increase timeout for local providers to 180s by paritoshd-nv · Pull Request #1620 · NVIDIA/NemoClaw

paritoshd-nv · 2026-04-08T19:59:11Z

Local models on DGX Spark can exceed the default 60s inference timeout, especially with large prompts (~90k tokens of system context). Pass --timeout 180 to openshell inference set for ollama-local and vllm-local providers. The timeout is configurable via NEMOCLAW_LOCAL_INFERENCE_TIMEOUT env var. Also add timeout_secs to blueprint InferenceProfile for the nim-local and vllm profiles.

Closes #1588

Summary

Related Issue

Changes

Type of Change

Code change for a new feature, bug fix, or refactor.
Code change with doc updates.
Doc only. Prose changes without code sample modifications.
Doc only. Includes code sample changes.

Testing

npx prek run --all-files passes (or equivalently make check).
npm test passes.
make docs builds without warnings. (for doc-only changes)

Checklist

General

I have read and followed the contributing guide.
I have read and followed the style guide. (for doc-only changes)

Code Changes

Formatters applied — npx prek run --all-files auto-fixes formatting (or make format for targeted runs).
Tests added or updated for new or changed behavior.
No secrets, API keys, or credentials committed.
Doc pages updated for any user-facing behavior changes (new commands, changed defaults, new features, bug fixes that contradict existing docs).

Doc Changes

Follows the style guide. Try running the nemoclaw-contributor-update-docs agent skill to draft changes while complying with the style guide. For example, prompt your agent with "/nemoclaw-contributor-update-docs catch up the docs for the new changes I made in this PR."
New pages include SPDX license header and frontmatter, if creating a new page.
Cross-references and links verified.

Signed-off-by: Your Name your-email@example.com

Summary by CodeRabbit

New Features
- Added configurable inference timeouts for local providers with a default of 180 seconds; timeouts are applied to local inference requests so long-running operations can be controlled.
Tests
- Added tests to validate timeout behavior is applied when configured and omitted when not.

coderabbitai · 2026-04-08T19:59:24Z

📝 Walkthrough

Walkthrough

Adds optional per-profile inference timeouts and wires them through onboarding and CLI invocation: an optional timeout_secs on inference profiles, onboarding parses LOCAL_INFERENCE_TIMEOUT_SECS (default 180s) and openshell inference set calls include --timeout when configured.

Changes

Cohort / File(s)	Summary
Blueprint / Profiles `nemoclaw-blueprint/blueprint.yaml`	Added `timeout_secs: 180` to `components.inference.profiles.nim-local` and `components.inference.profiles.vllm`.
Runner (blueprint apply) `nemoclaw/src/blueprint/runner.ts`, `nemoclaw/src/blueprint/runner.test.ts`	Added optional `InferenceProfile.timeout_secs?: number`; `actionApply` builds `openshell inference set` args dynamically and appends `--timeout <value>` when present. Tests added to assert presence/absence of `--timeout`.
Onboarding script `bin/lib/onboard.js`	Parsed `LOCAL_INFERENCE_TIMEOUT_SECS` (env var `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT`, default 180, clamped to non-negative integer) and pass `--timeout` to `openshell inference set` for local providers (`vllm-local`, `ollama-local`).

Sequence Diagram(s)

sequenceDiagram
    participant Onboard as Onboard Script
    participant Openshell as openshell CLI
    participant Provider as Local Inference Provider

    Onboard->>Openshell: run "inference set" --provider <p> --model <m> [--timeout <t>]
    Openshell->>Provider: configure provider with model and optional timeout
    Provider-->>Openshell: ack
    Openshell-->>Onboard: operation result

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Poem

🐰 I nibbled flags and counted ticks,
Timeout seeds in tidy nooks,
Local models given longer breaths,
Commands now carry careful clocks,
Hoppity hop — setups finish quick!

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 66.67% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately reflects the main change: increasing the timeout for local inference providers from the default 60s to 180s, which is the primary objective of the PR.
Linked Issues check	✅ Passed	The PR fully addresses issue `#1588` by making local inference timeout configurable via environment variable and setting it to 180s for ollama-local and vllm-local providers.
Out of Scope Changes check	✅ Passed	All changes directly support the timeout configuration objective: environment variable handling, CLI flag passing, blueprint profile updates, and corresponding test additions are all in scope.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

coderabbitai

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@nemoclaw/src/blueprint/runner.ts`:
- Around line 309-311: The code currently uses a truthy check for
inferenceCfg.timeout_secs which skips valid values like 0; update the
conditional around building inferenceArgs (the block that references
inferenceCfg and inferenceArgs in runner.ts) to check presence instead, e.g.
test that inferenceCfg.timeout_secs is not undefined/null
(inferenceCfg.timeout_secs !== undefined && inferenceCfg.timeout_secs !== null
or != null) before pushing "--timeout" and String(inferenceCfg.timeout_secs), so
zero is accepted but missing values are still skipped.

🪄 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: Pro

Run ID: dc6bd835-c9a9-4cc0-a321-3d303085823a

📥 Commits

Reviewing files that changed from the base of the PR and between e2bfdcf and b480fd8.

📒 Files selected for processing (4)

bin/lib/onboard.js
nemoclaw-blueprint/blueprint.yaml
nemoclaw/src/blueprint/runner.test.ts
nemoclaw/src/blueprint/runner.ts

Local models on DGX Spark can exceed the default 60s inference timeout, especially with large prompts (~90k tokens of system context). Pass --timeout 180 to openshell inference set for ollama-local and vllm-local providers. The timeout is configurable via NEMOCLAW_LOCAL_INFERENCE_TIMEOUT env var. Also add timeout_secs to blueprint InferenceProfile for the nim-local and vllm profiles. Closes NVIDIA#1588 Signed-off-by: Paritosh Dixit <paritoshd@nvidia.com>

coderabbitai

🧹 Nitpick comments (1)

nemoclaw/src/blueprint/runner.test.ts (1)
464-496: LGTM! Test correctly verifies --timeout argument passing.

The test properly sets up a blueprint with timeout_secs: 180, invokes actionApply, and verifies the mocked execa call includes the expected arguments. Good use of try/finally for environment variable cleanup.

One optional improvement: the current assertions verify that both "--timeout" and "180" exist in the argument array but don't confirm their adjacency. For stronger guarantees, you could check positioning:
♻️ Optional: More precise argument verification
       const inferenceCall = mockExeca.mock.calls.find(
         (c) => Array.isArray(c[1]) && c[1].includes("inference") && c[1].includes("set"),
       );
       if (!inferenceCall) throw new Error("inference set call not found");
-      expect(inferenceCall[1]).toContain("--timeout");
-      expect(inferenceCall[1]).toContain("180");
+      const args = inferenceCall[1] as string[];
+      const idx = args.indexOf("--timeout");
+      expect(idx).toBeGreaterThanOrEqual(0);
+      expect(args[idx + 1]).toBe("180");
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.

In `@nemoclaw/src/blueprint/runner.test.ts` around lines 464 - 496, The test
should assert that "--timeout" and "180" are adjacent in the mock execa args
rather than only both present; after locating the mocked call (variable
inferenceCall in the "passes --timeout when timeout_secs is set in profile" test
that invokes actionApply), get the argument array (inferenceCall[1]) and assert
that the index of "--timeout" is found and that the next element (index + 1)
equals "180" to guarantee adjacency.

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Nitpick comments:
In `@nemoclaw/src/blueprint/runner.test.ts`:
- Around line 464-496: The test should assert that "--timeout" and "180" are
adjacent in the mock execa args rather than only both present; after locating
the mocked call (variable inferenceCall in the "passes --timeout when
timeout_secs is set in profile" test that invokes actionApply), get the argument
array (inferenceCall[1]) and assert that the index of "--timeout" is found and
that the next element (index + 1) equals "180" to guarantee adjacency.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: abe34734-326b-4035-9520-b7e2299a0bbe

📥 Commits

Reviewing files that changed from the base of the PR and between b480fd8 and 22b3a89.

📒 Files selected for processing (4)

bin/lib/onboard.js
nemoclaw-blueprint/blueprint.yaml
nemoclaw/src/blueprint/runner.test.ts
nemoclaw/src/blueprint/runner.ts

✅ Files skipped from review due to trivial changes (1)

nemoclaw-blueprint/blueprint.yaml

🚧 Files skipped from review as they are similar to previous changes (2)

nemoclaw/src/blueprint/runner.ts
bin/lib/onboard.js

ericksoa

LGTM — clean, focused fix for local inference timeouts on DGX Spark. Correct use of !== undefined guard, configurable via env var, good test coverage.

## Summary - Document `nemoclaw credentials list` and `nemoclaw credentials reset` commands in commands reference (#1597) - Add `--dry-run` flag documentation for `policy-add` (#1276) - Update policy presets table: remove `docker` (#1647), add `brave` and `brew`, update HuggingFace endpoint (#1540) - Document `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT` env var for local providers (#1620) - Document `NEMOCLAW_PROXY_HOST`/`NEMOCLAW_PROXY_PORT` env vars (#1563) - Add troubleshooting entries for Docker group permissions (#1614), sandbox survival after gateway restart (#1587), and proxy configuration - Regenerate `nemoclaw-user-*` skills from updated docs ## Test plan - [x] `make docs` builds without warnings - [x] All pre-commit and pre-push hooks pass - [ ] Verify rendered pages in docs site preview 🤖 Generated with [Claude Code](https://claude.com/claude-code)  ## Summary by CodeRabbit * **New Features** * Added `nemoclaw credentials list` command to display stored credential names * Added `nemoclaw credentials reset <KEY>` command with `--yes` flag to remove credentials * Added `--dry-run` flag for policy-add to preview endpoint changes * New policy presets: `brave` and `brew` * New configuration options: `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT`, `NEMOCLAW_PROXY_HOST`, and `NEMOCLAW_PROXY_PORT` * **Documentation** * Expanded troubleshooting guides for Docker permissions, sandbox connectivity, local inference timeouts, and proxy configuration  Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

## Summary - Allow model and provider changes without rebuilding the sandbox image - The entrypoint patches `openclaw.json` at startup when `NEMOCLAW_MODEL_OVERRIDE` is set, then recomputes the config hash so integrity checks still pass - Same pattern as `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT` (PR #1620) ### New env vars | Env var | Purpose | When needed | |---------|---------|-------------| | `NEMOCLAW_MODEL_OVERRIDE` | Override `agents.defaults.model.primary` and provider model name | Any model switch | | `NEMOCLAW_INFERENCE_API_OVERRIDE` | Override inference API type (`openai-completions` or `anthropic-messages`) | Cross-provider switches only | ### Usage example (NVIDIA → Anthropic) ```bash # On host: configure gateway route openshell inference set --provider anthropic-prod --model claude-sonnet-4-6 --no-verify # Set env vars for the sandbox (via openshell or Docker) export NEMOCLAW_MODEL_OVERRIDE="anthropic/claude-sonnet-4-6" export NEMOCLAW_INFERENCE_API_OVERRIDE="anthropic-messages" # Restart sandbox — no image rebuild needed ``` ### Security - Env vars come from the host (Docker/OpenShell), not from inside the sandbox - Config integrity is verified first (detects build-time tampering), then override is applied - Config hash is recomputed after patching - Landlock locks the file after this function runs - Agent cannot set these env vars ## Related Issue Closes #759 ## Test plan - [ ] `npm test` passes (39 tests in nemoclaw-start.test.js) - [ ] Set `NEMOCLAW_MODEL_OVERRIDE` → sandbox starts with overridden model - [ ] Unset env var → sandbox starts with original baked model (no regression) - [ ] Set invalid model → sandbox starts but inference fails (expected) - [ ] Config hash passes integrity check on restart after override Signed-off-by: Prekshi Vyas <prekshiv@nvidia.com> 🤖 Generated with [Claude Code](https://claude.com/claude-code)  ## Summary by CodeRabbit * **New Features** * Env vars to override the active model and (optionally) the inference API at container startup; when used in privileged startup, the runtime config is updated and its integrity hash recomputed so startup verification aligns. * **Runtime safeguards** * Input validation, API allowlist, symlink protections, applies only in privileged mode, and no-op behavior when unset. * **Tests** * New unit and end-to-end tests covering override behavior, timing, hash recomputation, validation, and noop cases. * **Documentation** * Guidance added for cross-provider switching and the runtime override workflow.  --------- Signed-off-by: Prekshi Vyas <prekshiv@nvidia.com> Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: Carlos Villela <cvillela@nvidia.com>

## Summary - Allow model and provider changes without rebuilding the sandbox image - The entrypoint patches `openclaw.json` at startup when `NEMOCLAW_MODEL_OVERRIDE` is set, then recomputes the config hash so integrity checks still pass - Same pattern as `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT` (PR NVIDIA#1620) ### New env vars | Env var | Purpose | When needed | |---------|---------|-------------| | `NEMOCLAW_MODEL_OVERRIDE` | Override `agents.defaults.model.primary` and provider model name | Any model switch | | `NEMOCLAW_INFERENCE_API_OVERRIDE` | Override inference API type (`openai-completions` or `anthropic-messages`) | Cross-provider switches only | ### Usage example (NVIDIA → Anthropic) ```bash # On host: configure gateway route openshell inference set --provider anthropic-prod --model claude-sonnet-4-6 --no-verify # Set env vars for the sandbox (via openshell or Docker) export NEMOCLAW_MODEL_OVERRIDE="anthropic/claude-sonnet-4-6" export NEMOCLAW_INFERENCE_API_OVERRIDE="anthropic-messages" # Restart sandbox — no image rebuild needed ``` ### Security - Env vars come from the host (Docker/OpenShell), not from inside the sandbox - Config integrity is verified first (detects build-time tampering), then override is applied - Config hash is recomputed after patching - Landlock locks the file after this function runs - Agent cannot set these env vars ## Related Issue Closes NVIDIA#759 ## Test plan - [ ] `npm test` passes (39 tests in nemoclaw-start.test.js) - [ ] Set `NEMOCLAW_MODEL_OVERRIDE` → sandbox starts with overridden model - [ ] Unset env var → sandbox starts with original baked model (no regression) - [ ] Set invalid model → sandbox starts but inference fails (expected) - [ ] Config hash passes integrity check on restart after override Signed-off-by: Prekshi Vyas <prekshiv@nvidia.com> 🤖 Generated with [Claude Code](https://claude.com/claude-code)  ## Summary by CodeRabbit * **New Features** * Env vars to override the active model and (optionally) the inference API at container startup; when used in privileged startup, the runtime config is updated and its integrity hash recomputed so startup verification aligns. * **Runtime safeguards** * Input validation, API allowlist, symlink protections, applies only in privileged mode, and no-op behavior when unset. * **Tests** * New unit and end-to-end tests covering override behavior, timing, hash recomputation, validation, and noop cases. * **Documentation** * Guidance added for cross-provider switching and the runtime override workflow.  --------- Signed-off-by: Prekshi Vyas <prekshiv@nvidia.com> Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: Carlos Villela <cvillela@nvidia.com>

…1620) Local models on DGX Spark can exceed the default 60s inference timeout, especially with large prompts (~90k tokens of system context). Pass --timeout 180 to openshell inference set for ollama-local and vllm-local providers. The timeout is configurable via NEMOCLAW_LOCAL_INFERENCE_TIMEOUT env var. Also add timeout_secs to blueprint InferenceProfile for the nim-local and vllm profiles. Closes NVIDIA#1588  ## Summary  ## Related Issue  ## Changes  ## Type of Change  - [x] Code change for a new feature, bug fix, or refactor. - [ ] Code change with doc updates. - [ ] Doc only. Prose changes without code sample modifications. - [ ] Doc only. Includes code sample changes. ## Testing  - [x] `npx prek run --all-files` passes (or equivalently `make check`). - [x] `npm test` passes. - [ ] `make docs` builds without warnings. (for doc-only changes) ## Checklist ### General - [x] I have read and followed the [contributing guide](https://github.com/NVIDIA/NemoClaw/blob/main/CONTRIBUTING.md). - [x] I have read and followed the [style guide](https://github.com/NVIDIA/NemoClaw/blob/main/docs/CONTRIBUTING.md). (for doc-only changes) ### Code Changes  - [x] Formatters applied — `npx prek run --all-files` auto-fixes formatting (or `make format` for targeted runs). - [x] Tests added or updated for new or changed behavior. - [x] No secrets, API keys, or credentials committed. - [ ] Doc pages updated for any user-facing behavior changes (new commands, changed defaults, new features, bug fixes that contradict existing docs). ### Doc Changes  - [ ] Follows the [style guide](https://github.com/NVIDIA/NemoClaw/blob/main/docs/CONTRIBUTING.md). Try running the `nemoclaw-contributor-update-docs` agent skill to draft changes while complying with the style guide. For example, prompt your agent with "`/nemoclaw-contributor-update-docs` catch up the docs for the new changes I made in this PR." - [ ] New pages include SPDX license header and frontmatter, if creating a new page. - [ ] Cross-references and links verified. ---  Signed-off-by: Your Name <your-email@example.com>  ## Summary by CodeRabbit * **New Features** * Added configurable inference timeouts for local providers with a default of 180 seconds; timeouts are applied to local inference requests so long-running operations can be controlled. * **Tests** * Added tests to validate timeout behavior is applied when configured and omitted when not.  Signed-off-by: Paritosh Dixit <paritoshd@nvidia.com>

## Summary - Document `nemoclaw credentials list` and `nemoclaw credentials reset` commands in commands reference (NVIDIA#1597) - Add `--dry-run` flag documentation for `policy-add` (NVIDIA#1276) - Update policy presets table: remove `docker` (NVIDIA#1647), add `brave` and `brew`, update HuggingFace endpoint (NVIDIA#1540) - Document `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT` env var for local providers (NVIDIA#1620) - Document `NEMOCLAW_PROXY_HOST`/`NEMOCLAW_PROXY_PORT` env vars (NVIDIA#1563) - Add troubleshooting entries for Docker group permissions (NVIDIA#1614), sandbox survival after gateway restart (NVIDIA#1587), and proxy configuration - Regenerate `nemoclaw-user-*` skills from updated docs ## Test plan - [x] `make docs` builds without warnings - [x] All pre-commit and pre-push hooks pass - [ ] Verify rendered pages in docs site preview 🤖 Generated with [Claude Code](https://claude.com/claude-code)  ## Summary by CodeRabbit * **New Features** * Added `nemoclaw credentials list` command to display stored credential names * Added `nemoclaw credentials reset <KEY>` command with `--yes` flag to remove credentials * Added `--dry-run` flag for policy-add to preview endpoint changes * New policy presets: `brave` and `brew` * New configuration options: `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT`, `NEMOCLAW_PROXY_HOST`, and `NEMOCLAW_PROXY_PORT` * **Documentation** * Expanded troubleshooting guides for Docker permissions, sandbox connectivity, local inference timeouts, and proxy configuration  Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

## Summary - Allow model and provider changes without rebuilding the sandbox image - The entrypoint patches `openclaw.json` at startup when `NEMOCLAW_MODEL_OVERRIDE` is set, then recomputes the config hash so integrity checks still pass - Same pattern as `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT` (PR NVIDIA#1620) ### New env vars | Env var | Purpose | When needed | |---------|---------|-------------| | `NEMOCLAW_MODEL_OVERRIDE` | Override `agents.defaults.model.primary` and provider model name | Any model switch | | `NEMOCLAW_INFERENCE_API_OVERRIDE` | Override inference API type (`openai-completions` or `anthropic-messages`) | Cross-provider switches only | ### Usage example (NVIDIA → Anthropic) ```bash # On host: configure gateway route openshell inference set --provider anthropic-prod --model claude-sonnet-4-6 --no-verify # Set env vars for the sandbox (via openshell or Docker) export NEMOCLAW_MODEL_OVERRIDE="anthropic/claude-sonnet-4-6" export NEMOCLAW_INFERENCE_API_OVERRIDE="anthropic-messages" # Restart sandbox — no image rebuild needed ``` ### Security - Env vars come from the host (Docker/OpenShell), not from inside the sandbox - Config integrity is verified first (detects build-time tampering), then override is applied - Config hash is recomputed after patching - Landlock locks the file after this function runs - Agent cannot set these env vars ## Related Issue Closes NVIDIA#759 ## Test plan - [ ] `npm test` passes (39 tests in nemoclaw-start.test.js) - [ ] Set `NEMOCLAW_MODEL_OVERRIDE` → sandbox starts with overridden model - [ ] Unset env var → sandbox starts with original baked model (no regression) - [ ] Set invalid model → sandbox starts but inference fails (expected) - [ ] Config hash passes integrity check on restart after override Signed-off-by: Prekshi Vyas <prekshiv@nvidia.com> 🤖 Generated with [Claude Code](https://claude.com/claude-code)  ## Summary by CodeRabbit * **New Features** * Env vars to override the active model and (optionally) the inference API at container startup; when used in privileged startup, the runtime config is updated and its integrity hash recomputed so startup verification aligns. * **Runtime safeguards** * Input validation, API allowlist, symlink protections, applies only in privileged mode, and no-op behavior when unset. * **Tests** * New unit and end-to-end tests covering override behavior, timing, hash recomputation, validation, and noop cases. * **Documentation** * Guidance added for cross-provider switching and the runtime override workflow.  --------- Signed-off-by: Prekshi Vyas <prekshiv@nvidia.com> Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: Carlos Villela <cvillela@nvidia.com>

## Summary `NEMOCLAW_SANDBOX_READY_TIMEOUT` has been a recognised env var since #2849, but no documentation accompanied it — `docs/reference/commands.md`, `docs/reference/troubleshooting.md`, and the inference / deployment guides only mention the companion `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT` (added in #1620 and documented at that time). Operators hitting `Sandbox '<name>' was created but did not become ready within 180s` have no doc-grep path to the workaround, and the two timeouts are easy to conflate. This closes the documentation gap left by #2849. Originally tried under #3435; closed because that PR mis-framed the docs as resolving #3344 / #3416 (the root cause of both was the GPU policy bug fixed in #3436, not a timeout misconfiguration). The docs themselves still have value as a follow-up to the env-var introductions, so reopening as a new PR with the correct framing. ## Related Issue  ## Changes - `docs/reference/commands.md`: add `NEMOCLAW_SANDBOX_READY_TIMEOUT` and `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT` to the Onboard Timeouts table. - `docs/reference/troubleshooting.md`: new troubleshooting entry "Sandbox onboard times out with 'did not become ready within Ns'" that distinguishes the readiness wait from the inference-probe budget, with a worked example. - `docs/inference/use-local-inference.md`: cross-link the two timeouts from the existing `NEMOCLAW_LOCAL_INFERENCE_TIMEOUT` section so readers of either knob land on the other. - `docs/deployment/deploy-to-remote-gpu.md`: new "First-Run Readiness Budget" section calling out DGX Station / cloud-VM / large-quantised-model conditions that exceed the default and showing how to raise it. No code changes — the readiness behaviour is unchanged. ## Type of Change - [ ] Code change (feature, bug fix, or refactor) - [ ] Code change with doc updates - [x] Doc only (prose changes, no code sample modifications) - [ ] Doc only (includes code sample changes) ## Verification - [ ] `npx prek run --all-files` passes - [ ] `npm test` passes - [ ] Tests added or updated for new or changed behavior - [ ] No secrets, API keys, or credentials committed - [ ] Docs updated for user-facing behavior changes - [x] `make docs` builds without warnings (doc changes only) - [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) --- Signed-off-by: Tinson Lai <tinsonl@nvidia.com>  ## Summary by CodeRabbit * **Documentation** * Added a “First-Run Readiness Budget” note for remote GPU hosts explaining longer initial sandbox build/upload times and advice to increase NEMOCLAW_SANDBOX_READY_TIMEOUT. * Clarified that NEMOCLAW_LOCAL_INFERENCE_TIMEOUT applies to inference-server validation while sandbox readiness uses NEMOCLAW_SANDBOX_READY_TIMEOUT (default 180s). * Expanded examples for exporting both timeouts and onboarding timeout messaging. * Added troubleshooting guidance and inspection steps when sandbox readiness timeouts delete partial sandboxes.  [![Review Change Stack](https://storage.googleapis.com/coderabbit_public_assets/review-stack-in-coderabbit-ui.svg)](https://app.coderabbit.ai/change-stack/NVIDIA/NemoClaw/pull/3440)   --------- Signed-off-by: Tinson Lai <tinsonl@nvidia.com> Co-authored-by: Carlos Villela <cvillela@nvidia.com>

coderabbitai Bot reviewed Apr 8, 2026

View reviewed changes

Comment thread nemoclaw/src/blueprint/runner.ts Outdated

paritoshd-nv force-pushed the fix/update-local-inference-timeout branch from b480fd8 to 22b3a89 Compare April 8, 2026 20:11

coderabbitai Bot reviewed Apr 8, 2026

View reviewed changes

paritoshd-nv requested review from cv and ericksoa April 8, 2026 20:25

ericksoa approved these changes Apr 8, 2026

View reviewed changes

ericksoa merged commit b8a5245 into NVIDIA:main Apr 8, 2026
13 of 24 checks passed

prekshivyas mentioned this pull request Apr 8, 2026

feat(sandbox): runtime model override via env vars #1633

Merged

5 tasks

miyoungc mentioned this pull request Apr 9, 2026

docs: catch up documentation for v0.0.10 changes #1654

Merged

3 tasks

laitingsheng mentioned this pull request May 13, 2026

docs(onboard): document NEMOCLAW_SANDBOX_READY_TIMEOUT #3440

Merged

12 tasks

wscurran added the bug-fix PR fixes a bug or regression label Jun 8, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix(inference): increase timeout for local providers to 180s#1620

fix(inference): increase timeout for local providers to 180s#1620
ericksoa merged 1 commit into
NVIDIA:mainfrom
paritoshd-nv:fix/update-local-inference-timeout

paritoshd-nv commented Apr 8, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot commented Apr 8, 2026 •

edited

Loading

Walkthrough

Changes

Sequence Diagram(s)

Estimated code review effort

Poem

❌ Failed checks (1 warning)

Uh oh!

coderabbitai Bot left a comment

Uh oh!

Uh oh!

coderabbitai Bot left a comment

Uh oh!

ericksoa left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

paritoshd-nv commented Apr 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Related Issue

Changes

Type of Change

Testing

Checklist

General

Code Changes

Doc Changes

Summary by CodeRabbit

Uh oh!

coderabbitai Bot commented Apr 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Sequence Diagram(s)

Estimated code review effort

Poem

❌ Failed checks (1 warning)

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

ericksoa left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

paritoshd-nv commented Apr 8, 2026 •

edited

Loading

coderabbitai Bot commented Apr 8, 2026 •

edited

Loading