fix(sandbox): export proxy env vars with full NO_PROXY and persist across reconnects#1025
Merged
kjw3 merged 5 commits intoNVIDIA:mainfrom Mar 27, 2026
Merged
Conversation
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:
📝 WalkthroughWalkthroughEntrypoint startup now derives proxy host/port from Changes
Sequence Diagram(s)sequenceDiagram
participant Entrypoint as Entrypoint Script
participant Env as Process Env
participant FS as User Home FS
participant Shell as Interactive/Login Shell
Entrypoint->>Env: read NEMOCLAW_PROXY_HOST/PORT (or defaults)
Entrypoint->>Env: export HTTP_PROXY/HTTPS_PROXY/NO_PROXY and lowercase vars
Entrypoint->>FS: if writable, write idempotent snippet to ~/.bashrc and ~/.profile
Shell->>FS: on interactive/login start, source ~/.bashrc/~/.profile
FS->>Shell: provide exported proxy envs to shell sessions
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes Poem
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (2)
scripts/nemoclaw-start.sh (1)
231-243: Potential duplicate appends to~/.profileon container restart.Line 242 appends the proxy snippet to
~/.profilewithout checking if it already exists. If the container is stopped and restarted (rather than recreated), the snippet will be appended again, resulting in duplicates.Consider making the append idempotent by checking for an existing marker:
♻️ Suggested idempotent append
elif [ -w "${HOME:-/sandbox}" ]; then - printf '\n%s\n' "$_PROXY_SNIPPET" >>"${HOME:-/sandbox}/.profile" + if ! grep -q 'nemoclaw-start.sh.*proxy' "${HOME:-/sandbox}/.profile" 2>/dev/null; then + printf '\n%s\n' "$_PROXY_SNIPPET" >>"${HOME:-/sandbox}/.profile" + fi fi🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@scripts/nemoclaw-start.sh` around lines 231 - 243, The script currently unconditionally appends the _PROXY_SNIPPET to "${HOME:-/sandbox}/.profile", causing duplicate entries on container restart; change the logic to be idempotent by detecting an existing marker or the snippet before appending: add a unique marker string (e.g. "# nemoclaw-proxy start") inside _PROXY_SNIPPET and, before the printf >>"${HOME:-/sandbox}/.profile" in the non-root branch, check the file for that marker (using grep or test) and only append if the marker is absent; keep the root branch writing to /etc/profile.d/nemoclaw-proxy.sh as-is.test/service-env.test.js (1)
175-208: Profile persistence test validates file writing but diverges from actual script logic.The wrapper (lines 180-194) is a simplified reimplementation that doesn't exercise the actual script's
id -ucheck or the root vs non-root branching. This tests the concept of writing a proxy snippet when a directory exists, but not the exact production logic.Consider extracting and sourcing the actual persistence block from the script (similar to
extractProxyVars) for higher fidelity, or document that this is an intentional simplification.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@test/service-env.test.js` around lines 175 - 208, The test "writes proxy snippet to /etc/profile.d when the directory exists" reimplements the persistence wrapper and therefore doesn't exercise the script's real branching (root check via id -u and root vs non-root behavior); update the test to source or invoke the actual persistence block instead of the ad-hoc wrapper—either reuse the existing extractProxyVars helper to build the exact snippet or load the real persistence logic (the same code that writes nemoclaw-proxy.sh and performs the id -u check) so the test covers root/non-root branches and matches production behavior (or add a clear comment explaining this is an intentional simplification if you prefer not to change the test).
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Nitpick comments:
In `@scripts/nemoclaw-start.sh`:
- Around line 231-243: The script currently unconditionally appends the
_PROXY_SNIPPET to "${HOME:-/sandbox}/.profile", causing duplicate entries on
container restart; change the logic to be idempotent by detecting an existing
marker or the snippet before appending: add a unique marker string (e.g. "#
nemoclaw-proxy start") inside _PROXY_SNIPPET and, before the printf
>>"${HOME:-/sandbox}/.profile" in the non-root branch, check the file for that
marker (using grep or test) and only append if the marker is absent; keep the
root branch writing to /etc/profile.d/nemoclaw-proxy.sh as-is.
In `@test/service-env.test.js`:
- Around line 175-208: The test "writes proxy snippet to /etc/profile.d when the
directory exists" reimplements the persistence wrapper and therefore doesn't
exercise the script's real branching (root check via id -u and root vs non-root
behavior); update the test to source or invoke the actual persistence block
instead of the ad-hoc wrapper—either reuse the existing extractProxyVars helper
to build the exact snippet or load the real persistence logic (the same code
that writes nemoclaw-proxy.sh and performs the id -u check) so the test covers
root/non-root branches and matches production behavior (or add a clear comment
explaining this is an intentional simplification if you prefer not to change the
test).
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 61decdb9-e740-41db-9331-48f07277bead
📒 Files selected for processing (2)
scripts/nemoclaw-start.shtest/service-env.test.js
Contributor
Author
|
This PR builds on the approach from PR #704 by @nanookclaw, which correctly identified the need to export proxy env vars in |
…ross reconnects OpenShell injects NO_PROXY=127.0.0.1,localhost,::1 into the sandbox, missing inference.local and the gateway IP (10.200.0.1). This causes LLM inference requests to route through the egress proxy instead of going direct, and the proxy gateway IP itself gets proxied. Add proxy configuration block to nemoclaw-start.sh that: - Exports HTTP_PROXY, HTTPS_PROXY, and NO_PROXY with inference.local and the gateway IP included - Persists via /etc/profile.d/nemoclaw-proxy.sh (root) or ~/.profile (non-root fallback) so values survive OpenShell reconnect injection - Supports NEMOCLAW_PROXY_HOST / NEMOCLAW_PROXY_PORT overrides The non-root fallback ensures the fix works in environments like Brev where containers run without root privileges. Tested on DGX Spark (ARM64) and Brev VM (x86_64). Verified NO_PROXY contains inference.local and 10.200.0.1 inside the live sandbox after connect. Ref: NVIDIA#626, NVIDIA#704 Ref: NVIDIA#704 (comment)
senthilr-nv
force-pushed
the
fix/sandbox-proxy-no-proxy-persistence
branch
from
6baa176 to
e7fa924
Compare
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (1)
test/service-env.test.js (1)
175-208: Add coverage for non-root~/.profilefallback and repeat-run behavior.This test validates the
/etc/profile.dpath, but the PR also relies on the non-root fallback. Add one test for writing to~/.profileand one for repeated invocation behavior to guard regressions there.🧪 Suggested additional test shape
+ it("writes proxy snippet to ~/.profile in non-root fallback", () => { + // Arrange temp HOME and run wrapper with /etc/profile.d branch disabled + // Assert ~/.profile contains expected exports and inference.local entry + }); + + it("does not duplicate proxy snippet on repeated startup", () => { + // Run wrapper twice against same profile target + // Assert marker/export block count is 1 + });🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@test/service-env.test.js` around lines 175 - 208, Add two tests to test/service-env.test.js: one that simulates a non-root user by setting HOME to a temp directory and verifies the code writes the proxy snippet into ~/.profile (use readFileSync on join(process.env.HOME, ".profile") and assert presence of export HTTP_PROXY/HTTPS_PROXY/NO_PROXY and the host IP), and a second that calls the entrypoint twice (or runs the same wrapper twice like the existing test does with execFileSync on a temp script) to assert idempotency/no duplicate snippets in either /etc/profile.d/nemoclaw-proxy.sh or ~/.profile (check counts or absence of repeated blocks); locate the test scaffolding around the existing "writes proxy snippet to /etc/profile.d when the directory exists" spec and reuse its tmpdir/tmpFile setup and cleanup logic (profileDir, tmpFile, wrapper, execFileSync) to implement these two new specs.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@scripts/nemoclaw-start.sh`:
- Around line 239-243: The snippet writes _PROXY_SNIPPET to the non-root user's
profile using ${HOME:-/sandbox} before HOME is reset and uses >> which can
duplicate the snippet; change the logic to determine the actual non-root home
after the script resets ownership (or compute a target_home variable after
privilege drop) and then only persist the snippet idempotently by checking for
its presence (e.g. grep -Fq "$_PROXY_SNIPPET" "$target_home/.profile") and
writing it only if missing (or replace existing block), and continue to write
the root case to the profile.d file (use the same presence check for
/etc/profile.d target) so _PROXY_SNIPPET is not appended multiple times and the
path is resolved deterministically; reference the _PROXY_SNIPPET variable, the
.profile target, and the profile.d target in your changes.
In `@test/service-env.test.js`:
- Around line 161-173: The test titles are misleading: the it() descriptions say
"excludes" while the assertions check for inclusion; update the two it()
descriptions in the test file so they reflect inclusion/containment (e.g.,
change "NO_PROXY excludes loopback and inference.local" to "NO_PROXY includes
loopback and inference.local" and change "NO_PROXY excludes OpenShell gateway
IP" to "NO_PROXY includes OpenShell gateway IP") — keep the assertions
referencing extractProxyVars() and vars.NO_PROXY as-is.
---
Nitpick comments:
In `@test/service-env.test.js`:
- Around line 175-208: Add two tests to test/service-env.test.js: one that
simulates a non-root user by setting HOME to a temp directory and verifies the
code writes the proxy snippet into ~/.profile (use readFileSync on
join(process.env.HOME, ".profile") and assert presence of export
HTTP_PROXY/HTTPS_PROXY/NO_PROXY and the host IP), and a second that calls the
entrypoint twice (or runs the same wrapper twice like the existing test does
with execFileSync on a temp script) to assert idempotency/no duplicate snippets
in either /etc/profile.d/nemoclaw-proxy.sh or ~/.profile (check counts or
absence of repeated blocks); locate the test scaffolding around the existing
"writes proxy snippet to /etc/profile.d when the directory exists" spec and
reuse its tmpdir/tmpFile setup and cleanup logic (profileDir, tmpFile, wrapper,
execFileSync) to implement these two new specs.
🪄 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: a24817d1-2fc3-4417-820c-e14985633fcb
📒 Files selected for processing (2)
scripts/nemoclaw-start.shtest/service-env.test.js
Contributor
|
Thanks for working on this. I reran the validation entirely on a single Linux GPU host, and I don’t think this is ready to merge yet. What I tested on that one host:
What I found:
So the main blocker is still the same:
Because of that, I don’t think this closes the runtime mismatch yet. The next step is to fix the env injection / session startup path that OpenShell uses on connect, rather than relying on |
senthilr-nv
force-pushed
the
fix/sandbox-proxy-no-proxy-persistence
branch
from
3bb83cf to
6ebeea6
Compare
Contributor
There was a problem hiding this comment.
♻️ Duplicate comments (1)
test/service-env.test.js (1)
164-176:⚠️ Potential issue | 🟡 MinorTest names contradict the assertions.
Lines 164 and 173 describe "excludes" but the assertions verify inclusion with
toContain. Rename for clarity:- it("NO_PROXY excludes loopback and inference.local", () => { + it("NO_PROXY includes loopback and inference.local", () => {- it("NO_PROXY excludes OpenShell gateway IP", () => { + it("NO_PROXY includes OpenShell gateway IP", () => {🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@test/service-env.test.js` around lines 164 - 176, The two test descriptions are misleading: the specs in the test block for the NO_PROXY checks say "excludes" but the assertions on extractProxyVars().NO_PROXY use toContain, so either the assertions or the test names must be changed; update the test titles (the it(...) strings) to say "includes" (e.g., change "NO_PROXY excludes loopback and inference.local" and "NO_PROXY excludes OpenShell gateway IP" to "NO_PROXY includes ...") so they match the assertions that check inclusion, referencing the test cases that call extractProxyVars() and assert on vars.NO_PROXY.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Duplicate comments:
In `@test/service-env.test.js`:
- Around line 164-176: The two test descriptions are misleading: the specs in
the test block for the NO_PROXY checks say "excludes" but the assertions on
extractProxyVars().NO_PROXY use toContain, so either the assertions or the test
names must be changed; update the test titles (the it(...) strings) to say
"includes" (e.g., change "NO_PROXY excludes loopback and inference.local" and
"NO_PROXY excludes OpenShell gateway IP" to "NO_PROXY includes ...") so they
match the assertions that check inclusion, referencing the test cases that call
extractProxyVars() and assert on vars.NO_PROXY.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 05a59140-8d37-469c-aaac-34e0e85b01a1
📒 Files selected for processing (2)
scripts/nemoclaw-start.shtest/service-env.test.js
senthilr-nv
force-pushed
the
fix/sandbox-proxy-no-proxy-persistence
branch
from
6ebeea6 to
766294f
Compare
…ct sessions OpenShell's `sandbox connect` spawns `/bin/bash -i` (interactive, non-login), which sources ~/.bashrc — not ~/.profile or /etc/profile.d/*. The previous approach wrote to ~/.profile and /etc/profile.d/, neither of which is sourced by `bash -i`, so the narrow OpenShell-injected NO_PROXY persisted in live interactive sessions. Changes: - Write proxy snippet to ~/.bashrc (primary) and ~/.profile (login fallback) - Export both uppercase and lowercase proxy variants (NO_PROXY + no_proxy, HTTP_PROXY + http_proxy, etc.) — Node.js undici prefers lowercase no_proxy over uppercase NO_PROXY when both are set - Add idempotency guard to prevent duplicate blocks on container restart - Update tests: verify .bashrc writing, idempotency, bash -i override behavior, and lowercase variant correctness Tested on DGX Spark (ARM64) and Brev VM (x86_64) with full destroy + re-onboard + live `env | grep proxy` verification inside the sandbox shell via `openshell sandbox connect`. Ref: NVIDIA#626
senthilr-nv
force-pushed
the
fix/sandbox-proxy-no-proxy-persistence
branch
from
766294f to
ac2224f
Compare
Contributor
There was a problem hiding this comment.
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 `@scripts/nemoclaw-start.sh`:
- Around line 244-260: The current _write_proxy_snippet only skips writing when
_PROXY_MARKER already exists, leaving stale values in files; change the approach
to make the persisted block replaceable: update _PROXY_SNIPPET to include a
distinct end marker (e.g., _PROXY_MARKER_END), then modify _write_proxy_snippet
to detect the start marker and, if present, remove the existing block between
the start and end markers (or the start marker through the next blank line) and
then write the new _PROXY_SNIPPET; use the existing symbols _PROXY_MARKER,
_PROXY_SNIPPET and function _write_proxy_snippet to locate and perform the
replacement so subsequent runs refresh proxy values instead of deduping.
🪄 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: 9123b1a1-93de-4fa3-ab63-215e61de8366
📒 Files selected for processing (2)
scripts/nemoclaw-start.shtest/service-env.test.js
…rkers Use begin/end markers in .bashrc/.profile proxy snippet so _write_proxy_snippet replaces the block when PROXY_HOST/PORT change instead of silently keeping stale values. Adds test coverage for the replacement path. Addresses CodeRabbit review feedback on idempotency gap.
Contributor
There was a problem hiding this comment.
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 `@scripts/nemoclaw-start.sh`:
- Around line 253-266: The script sets _SANDBOX_HOME="${HOME:-/sandbox}" which
picks /root when running as root and causes _write_proxy_snippet to append to
root's dotfiles; change the logic that defines _SANDBOX_HOME so that when
running as UID 0 (or when HOME equals /root) it uses /sandbox instead of $HOME,
then continue calling _write_proxy_snippet for "${_SANDBOX_HOME}/.bashrc" and
"${_SANDBOX_HOME}/.profile"; update the assignment of _SANDBOX_HOME (and any
related conditional) so _write_proxy_snippet, _PROXY_MARKER and _PROXY_SNIPPET
operate on the actual sandbox user's home rather than /root.
🪄 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: bb628d55-ca34-4787-8f2f-92cbfbd93cda
📒 Files selected for processing (2)
scripts/nemoclaw-start.shtest/service-env.test.js
✅ Files skipped from review due to trivial changes (1)
- test/service-env.test.js
When the entrypoint runs as root, $HOME is /root — the proxy snippet was written to /root/.bashrc instead of the sandbox user's home. Use getent passwd to look up the sandbox user's home when running as UID 0; fall back to /sandbox if the user entry is missing. Addresses CodeRabbit review feedback on _SANDBOX_HOME resolution.
Contributor
Author
|
Thanks for the detailed repro — you were spot on. The original I traced through the OpenShell connect path to confirm the mechanism: The updated fix writes the full proxy config to I also addressed the CodeRabbit feedback in the latest commits:
Verified end-to-end on DGX Spark (ARM64) and Brev VM (x86_64) with fresh destroy -> re-onboard -> connect: env | grep -i proxy | sort
ALL_PROXY=http://10.200.0.1:3128
HTTPS_PROXY=http://10.200.0.1:3128
HTTP_PROXY=http://10.200.0.1:3128
NODE_USE_ENV_PROXY=1
NO_PROXY=localhost,127.0.0.1,::1,inference.local,10.200.0.1
grpc_proxy=http://10.200.0.1:3128
http_proxy=http://10.200.0.1:3128
https_proxy=http://10.200.0.1:3128
no_proxy=localhost,127.0.0.1,::1,inference.local,10.200.0.1Proxy bypass confirmed — the remaining |
kjw3
approved these changes
Mar 27, 2026
Contributor
|
Reran this on the latest branch tip, and the result changed in the right way. What I validated on Spark:
What I saw in the fresh sandbox after connect:
And inside the sandbox:
So with the Also, credit where it’s due: PR #704 by
This branch looks like the follow-through that closes that line of work. |
senthilr-nv
added a commit
to senthilr-nv/NemoClaw
that referenced
this pull request
Mar 28, 2026
inference.local is a virtual hostname that only the OpenShell proxy understands via CONNECT tunnel interception. Including it in NO_PROXY caused Node.js to bypass the proxy and connect directly, but since inference.local has no DNS/hosts entry, the connection hangs and the gateway surfaces "LLM request timed out" errors. Remove inference.local from NO_PROXY so requests route through the proxy at 10.200.0.1:3128, which intercepts the CONNECT and forwards to the configured inference provider. Fixes inference timeout regression introduced in NVIDIA#1025.
TSavo
pushed a commit
to wopr-network/nemoclaw
that referenced
this pull request
Mar 28, 2026
* fix: improve gateway lifecycle recovery (NVIDIA#953) * fix: improve gateway lifecycle recovery * docs: fix readme markdown list spacing * fix: tighten gateway lifecycle review follow-ups * fix: simplify tokenized control ui output * fix: restore chat route in control ui urls * refactor: simplify ansi stripping in onboard * fix: shorten control ui url output * fix: move control ui below cli next steps * fix: swap hard/soft ulimit settings in start script (NVIDIA#951) Fixes NVIDIA#949 Co-authored-by: KJ <kejones@nvidia.com> * chore: add cyclomatic complexity lint rule (NVIDIA#875) * chore: add cyclomatic complexity rule (ratchet from 95) Add ESLint complexity rule to bin/ and scripts/ to prevent new functions from accumulating excessive branching. Starting threshold is 95 (current worst offender: setupNim in onboard.js). Ratchet plan: 95 → 40 → 25 → 15. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: ratchet complexity to 20, suppress existing violations Suppress 6 functions that exceed the threshold with eslint-disable comments so we can start enforcing at 20 instead of 95: - setupNim (95), setupPolicies (41), setupInference (22) in onboard.js - deploy (22), main IIFE (27) in nemoclaw.js - applyPreset (24) in policies.js Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: suppress complexity for 3 missed functions preflight (23), getReconciledSandboxGatewayState (25), sandboxStatus (27) --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: add host-side config and state file locations to README (NVIDIA#903) Signed-off-by: peteryuqin <peter.yuqin@gmail.com> * chore: add tsconfig.cli.json, root execa, TS coverage ratchet (NVIDIA#913) * chore: add tsconfig.cli.json, root execa, TS coverage ratchet Foundation for the CLI TypeScript migration (PR 0 of the shell consolidation plan). No runtime changes — config, tooling, and dependency only. - tsconfig.cli.json: strict TS type-checking for bin/ and scripts/ (noEmit, module: preserve — tsx handles the runtime) - scripts/check-coverage-ratchet.ts: pure TS replacement for the bash+python coverage ratchet script (same logic, same tolerance) - execa ^9.6.1 added to root devDependencies (used by PR 1+) - pr.yaml: coverage ratchet step now runs the TS version via tsx - .pre-commit-config.yaml: SPDX headers cover scripts/*.ts, new tsc-check-cli pre-push hook - CONTRIBUTING.md: document typecheck:cli task and CLI pre-push hook - Delete scripts/check-coverage-ratchet.sh Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * Apply suggestion from @brandonpelfrey * chore: address PR feedback — use types_or, add tsx devDep - Use `types_or: [ts, tsx]` instead of file glob for tsc-check-cli hook per @brandonpelfrey's suggestion. - Add `tsx` to devDependencies so CI doesn't re-fetch it on every run per CodeRabbit's suggestion. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix(ci): ignore GitHub "Apply suggestion" commits in commitlint * fix(ci): lint only PR title since repo is squash-merge only Reverts the commitlint ignores rule from the previous commit and instead removes the per-commit lint step entirely. Individual commit messages are discarded at merge time — only the squash-merged PR title lands in main and drives changelog generation. Drop the per-commit lint, keep the PR title check, and remove the now-unnecessary fetch-depth: 0. * Revert "fix(ci): lint only PR title since repo is squash-merge only" This reverts commit 1257a47. * Revert "fix(ci): ignore GitHub "Apply suggestion" commits in commitlint" This reverts commit c395657. * docs: fix markdownlint MD032 in README (blank line before list) * refactor: make coverage ratchet script idiomatic TypeScript - Wrap in main() with process.exitCode instead of scattered process.exit() - Replace mutable flags with .map()/.some() over typed MetricResult[] - Separate pure logic (checkMetrics) from formatting (formatReport) - Throw with { cause } chaining instead of exit-in-helpers - Derive CoverageThresholds from METRICS tuple (single source of truth) - Exhaustive switch on CheckStatus discriminated union * refactor: remove duplication in coverage ratchet script - Drop STATUS_LABELS map; inline labels in exhaustive switch - Extract common 'metric coverage is N%' preamble in formatResult - Simplify ratchetedThresholds: use results directly (already in METRICS order) instead of re-scanning with .find() per metric - Compute 'failed' once in main, pass into formatReport to avoid duplicate .some() scan * refactor: simplify coverage ratchet with FP patterns - Extract classify() as a named pure function (replaces nested ternary) - loadJSON takes repo-relative paths, eliminating THRESHOLD_PATH and SUMMARY_PATH constants (DRY the join-with-REPO_ROOT pattern) - Drop CoverageMetric/CoverageSummary interfaces (only pct is read); use structural type at the call site instead - Inline ratchetedThresholds (one-liner, used once) - formatReport derives fail/improved from results instead of taking a pre-computed boolean (let functions derive from data, don't thread derived state) - sections.join("\n\n") replaces manual empty-string pushing - Shorter type names (Thresholds, Status, Result) — no ambiguity in a single-purpose script * refactor: strip coverage ratchet to failure-only output prek hides output from commands that exit 0, so ok/improved reporting was dead code. Remove Status, Result, classify, formatResult, formatReport, and the ratcheted-thresholds suggestion block. The script now just filters for regressions and prints actionable errors on failure. --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: Brandon Pelfrey <bpelfrey@nvidia.com> * fix: use CONNECT tunnel for WebSocket endpoints in Discord/Slack presets (NVIDIA#438) * fix: use CONNECT tunnel for WebSocket endpoints in Discord/Slack presets The egress proxy's HTTP idle timeout (~2 min) kills long-lived WebSocket connections when endpoints are configured with protocol:rest + tls:terminate. Switch WebSocket endpoints to access:full (CONNECT tunnel) which bypasses HTTP-level timeouts entirely. Discord: - gateway.discord.gg → access:full (WebSocket gateway) - Add PUT/PATCH/DELETE methods for discord.com (message editing, reactions) - Add media.discordapp.net for attachment access Slack: - Add wss-primary.slack.com and wss-backup.slack.com → access:full (Socket Mode WebSocket endpoints) Partially addresses NVIDIA#409 — the policy-level fix enables WebSocket connections to survive. The hardcoded 2-min timeout in openshell-sandbox still affects any protocol:rest endpoints with long-lived connections. Related: NVIDIA#361 (WhatsApp Web, same root cause) * fix: correct comment wording for media endpoint and YAML formatting * fix: standardize Node.js minimum version to 22.16 (NVIDIA#840) * fix: remove unused RECOMMENDED_NODE_MAJOR from scripts/install.sh Shellcheck flagged it as unused after the min/recommended merge. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: enforce full semver >=22.16.0 in installer scripts The runtime checks only compared the major Node.js version, allowing 22.0–22.15 to pass despite package.json requiring >=22.16.0. Use the version_gte() helper for full semver comparison in both installers. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: harden version_gte and align fallback message Guard version_gte() against prerelease suffixes (e.g. "22.16.0-rc.1") that would crash bash arithmetic. Also update the manual-install fallback message to reference MIN_NODE_VERSION instead of hardcoded "22". Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: update test stubs for Node.js 22.16 minimum and add Node 20 rejection test - Bump node stub in 'succeeds with acceptable Node.js' from v20.0.0 to v22.16.0 - Bump node stub in buildCurlPipeEnv from v22.14.0 to v22.16.0 - Add new test asserting Node.js 20 is rejected by ensure_supported_runtime --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: harden installer and onboard resiliency (NVIDIA#961) * fix: harden installer and onboard resiliency * fix: address installer and debug review follow-ups * fix: harden onboard resume across later setup steps * test: simplify payload extraction in onboard tests * test: keep onboard payload extraction target-compatible * chore: align onboard session lint with complexity rule * fix: harden onboard session safety and lock handling * fix: tighten onboard session redaction and metadata handling * fix(security): strip credentials from migration snapshots and enforce blueprint digest (NVIDIA#769) Reconciles NVIDIA#156 and NVIDIA#743 into a single comprehensive solution: - Filter auth-profiles.json at copy time via cpSync filter (from NVIDIA#743) - Recursive stripCredentials() with pattern-based field detection for deep config sanitization (from NVIDIA#156: CREDENTIAL_FIELDS set + CREDENTIAL_FIELD_PATTERN regex) - Remove gateway config section (contains auth tokens) from sandbox openclaw.json - Blueprint digest verification (SHA-256): recorded at snapshot time, validated on restore, empty/missing digest is a hard failure - computeFileDigest() throws when blueprint file is missing instead of silently returning null - Sanitize both snapshot-level and sandbox-bundle openclaw.json copies - Backward compatible: old snapshots without blueprintDigest skip validation - Bump SNAPSHOT_VERSION 2 → 3 Supersedes NVIDIA#156 and NVIDIA#743. * fix(sandbox): export proxy env vars with full NO_PROXY and persist across reconnects (NVIDIA#1025) * fix(sandbox): export proxy env vars with full NO_PROXY and persist across reconnects OpenShell injects NO_PROXY=127.0.0.1,localhost,::1 into the sandbox, missing inference.local and the gateway IP (10.200.0.1). This causes LLM inference requests to route through the egress proxy instead of going direct, and the proxy gateway IP itself gets proxied. Add proxy configuration block to nemoclaw-start.sh that: - Exports HTTP_PROXY, HTTPS_PROXY, and NO_PROXY with inference.local and the gateway IP included - Persists via /etc/profile.d/nemoclaw-proxy.sh (root) or ~/.profile (non-root fallback) so values survive OpenShell reconnect injection - Supports NEMOCLAW_PROXY_HOST / NEMOCLAW_PROXY_PORT overrides The non-root fallback ensures the fix works in environments like Brev where containers run without root privileges. Tested on DGX Spark (ARM64) and Brev VM (x86_64). Verified NO_PROXY contains inference.local and 10.200.0.1 inside the live sandbox after connect. Ref: NVIDIA#626, NVIDIA#704 Ref: NVIDIA#704 (comment) * fix(sandbox): write proxy config to ~/.bashrc for interactive reconnect sessions OpenShell's `sandbox connect` spawns `/bin/bash -i` (interactive, non-login), which sources ~/.bashrc — not ~/.profile or /etc/profile.d/*. The previous approach wrote to ~/.profile and /etc/profile.d/, neither of which is sourced by `bash -i`, so the narrow OpenShell-injected NO_PROXY persisted in live interactive sessions. Changes: - Write proxy snippet to ~/.bashrc (primary) and ~/.profile (login fallback) - Export both uppercase and lowercase proxy variants (NO_PROXY + no_proxy, HTTP_PROXY + http_proxy, etc.) — Node.js undici prefers lowercase no_proxy over uppercase NO_PROXY when both are set - Add idempotency guard to prevent duplicate blocks on container restart - Update tests: verify .bashrc writing, idempotency, bash -i override behavior, and lowercase variant correctness Tested on DGX Spark (ARM64) and Brev VM (x86_64) with full destroy + re-onboard + live `env | grep proxy` verification inside the sandbox shell via `openshell sandbox connect`. Ref: NVIDIA#626 * fix(sandbox): replace stale proxy values on restart with begin/end markers Use begin/end markers in .bashrc/.profile proxy snippet so _write_proxy_snippet replaces the block when PROXY_HOST/PORT change instead of silently keeping stale values. Adds test coverage for the replacement path. Addresses CodeRabbit review feedback on idempotency gap. * fix(sandbox): resolve sandbox user home dynamically when running as root When the entrypoint runs as root, $HOME is /root — the proxy snippet was written to /root/.bashrc instead of the sandbox user's home. Use getent passwd to look up the sandbox user's home when running as UID 0; fall back to /sandbox if the user entry is missing. Addresses CodeRabbit review feedback on _SANDBOX_HOME resolution. --------- Co-authored-by: Carlos Villela <cvillela@nvidia.com> * fix(policies): preset application for versionless policies (Fixes NVIDIA#35) (NVIDIA#101) * fix(policies): allow preset application for versionless policies (Fixes NVIDIA#35) Fixes NVIDIA#35 Signed-off-by: Deepak Jain <deepujain@gmail.com> * fix: remove stale complexity suppression in policies --------- Signed-off-by: Deepak Jain <deepujain@gmail.com> Co-authored-by: Kevin Jones <kejones@nvidia.com> * fix: restore routed inference and connect UX (NVIDIA#1037) * fix: restore routed inference and connect UX * fix: simplify detected local inference hint * fix: remove stale local inference hint * test: relax connect forward assertion --------- Signed-off-by: peteryuqin <peter.yuqin@gmail.com> Signed-off-by: Deepak Jain <deepujain@gmail.com> Co-authored-by: KJ <kejones@nvidia.com> Co-authored-by: Emily Wilkins <80470879+epwilkins@users.noreply.github.com> Co-authored-by: Carlos Villela <cvillela@nvidia.com> Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: Peter <peter.yuqin@gmail.com> Co-authored-by: Brandon Pelfrey <bpelfrey@nvidia.com> Co-authored-by: Benedikt Schackenberg <6381261+BenediktSchackenberg@users.noreply.github.com> Co-authored-by: Lucas Wang <lucas_wang@lucas-futures.com> Co-authored-by: senthilr-nv <senthilr@nvidia.com> Co-authored-by: Deepak Jain <deepujain@users.noreply.github.com> Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
ksapru
pushed a commit
to ksapru/NemoClaw
that referenced
this pull request
Mar 30, 2026
…ross reconnects (NVIDIA#1025) * fix(sandbox): export proxy env vars with full NO_PROXY and persist across reconnects OpenShell injects NO_PROXY=127.0.0.1,localhost,::1 into the sandbox, missing inference.local and the gateway IP (10.200.0.1). This causes LLM inference requests to route through the egress proxy instead of going direct, and the proxy gateway IP itself gets proxied. Add proxy configuration block to nemoclaw-start.sh that: - Exports HTTP_PROXY, HTTPS_PROXY, and NO_PROXY with inference.local and the gateway IP included - Persists via /etc/profile.d/nemoclaw-proxy.sh (root) or ~/.profile (non-root fallback) so values survive OpenShell reconnect injection - Supports NEMOCLAW_PROXY_HOST / NEMOCLAW_PROXY_PORT overrides The non-root fallback ensures the fix works in environments like Brev where containers run without root privileges. Tested on DGX Spark (ARM64) and Brev VM (x86_64). Verified NO_PROXY contains inference.local and 10.200.0.1 inside the live sandbox after connect. Ref: NVIDIA#626, NVIDIA#704 Ref: NVIDIA#704 (comment) * fix(sandbox): write proxy config to ~/.bashrc for interactive reconnect sessions OpenShell's `sandbox connect` spawns `/bin/bash -i` (interactive, non-login), which sources ~/.bashrc — not ~/.profile or /etc/profile.d/*. The previous approach wrote to ~/.profile and /etc/profile.d/, neither of which is sourced by `bash -i`, so the narrow OpenShell-injected NO_PROXY persisted in live interactive sessions. Changes: - Write proxy snippet to ~/.bashrc (primary) and ~/.profile (login fallback) - Export both uppercase and lowercase proxy variants (NO_PROXY + no_proxy, HTTP_PROXY + http_proxy, etc.) — Node.js undici prefers lowercase no_proxy over uppercase NO_PROXY when both are set - Add idempotency guard to prevent duplicate blocks on container restart - Update tests: verify .bashrc writing, idempotency, bash -i override behavior, and lowercase variant correctness Tested on DGX Spark (ARM64) and Brev VM (x86_64) with full destroy + re-onboard + live `env | grep proxy` verification inside the sandbox shell via `openshell sandbox connect`. Ref: NVIDIA#626 * fix(sandbox): replace stale proxy values on restart with begin/end markers Use begin/end markers in .bashrc/.profile proxy snippet so _write_proxy_snippet replaces the block when PROXY_HOST/PORT change instead of silently keeping stale values. Adds test coverage for the replacement path. Addresses CodeRabbit review feedback on idempotency gap. * fix(sandbox): resolve sandbox user home dynamically when running as root When the entrypoint runs as root, $HOME is /root — the proxy snippet was written to /root/.bashrc instead of the sandbox user's home. Use getent passwd to look up the sandbox user's home when running as UID 0; fall back to /sandbox if the user entry is missing. Addresses CodeRabbit review feedback on _SANDBOX_HOME resolution. --------- Co-authored-by: Carlos Villela <cvillela@nvidia.com>
laitingsheng
pushed a commit
that referenced
this pull request
Apr 2, 2026
…ross reconnects (#1025) * fix(sandbox): export proxy env vars with full NO_PROXY and persist across reconnects OpenShell injects NO_PROXY=127.0.0.1,localhost,::1 into the sandbox, missing inference.local and the gateway IP (10.200.0.1). This causes LLM inference requests to route through the egress proxy instead of going direct, and the proxy gateway IP itself gets proxied. Add proxy configuration block to nemoclaw-start.sh that: - Exports HTTP_PROXY, HTTPS_PROXY, and NO_PROXY with inference.local and the gateway IP included - Persists via /etc/profile.d/nemoclaw-proxy.sh (root) or ~/.profile (non-root fallback) so values survive OpenShell reconnect injection - Supports NEMOCLAW_PROXY_HOST / NEMOCLAW_PROXY_PORT overrides The non-root fallback ensures the fix works in environments like Brev where containers run without root privileges. Tested on DGX Spark (ARM64) and Brev VM (x86_64). Verified NO_PROXY contains inference.local and 10.200.0.1 inside the live sandbox after connect. Ref: #626, #704 Ref: #704 (comment) * fix(sandbox): write proxy config to ~/.bashrc for interactive reconnect sessions OpenShell's `sandbox connect` spawns `/bin/bash -i` (interactive, non-login), which sources ~/.bashrc — not ~/.profile or /etc/profile.d/*. The previous approach wrote to ~/.profile and /etc/profile.d/, neither of which is sourced by `bash -i`, so the narrow OpenShell-injected NO_PROXY persisted in live interactive sessions. Changes: - Write proxy snippet to ~/.bashrc (primary) and ~/.profile (login fallback) - Export both uppercase and lowercase proxy variants (NO_PROXY + no_proxy, HTTP_PROXY + http_proxy, etc.) — Node.js undici prefers lowercase no_proxy over uppercase NO_PROXY when both are set - Add idempotency guard to prevent duplicate blocks on container restart - Update tests: verify .bashrc writing, idempotency, bash -i override behavior, and lowercase variant correctness Tested on DGX Spark (ARM64) and Brev VM (x86_64) with full destroy + re-onboard + live `env | grep proxy` verification inside the sandbox shell via `openshell sandbox connect`. Ref: #626 * fix(sandbox): replace stale proxy values on restart with begin/end markers Use begin/end markers in .bashrc/.profile proxy snippet so _write_proxy_snippet replaces the block when PROXY_HOST/PORT change instead of silently keeping stale values. Adds test coverage for the replacement path. Addresses CodeRabbit review feedback on idempotency gap. * fix(sandbox): resolve sandbox user home dynamically when running as root When the entrypoint runs as root, $HOME is /root — the proxy snippet was written to /root/.bashrc instead of the sandbox user's home. Use getent passwd to look up the sandbox user's home when running as UID 0; fall back to /sandbox if the user entry is missing. Addresses CodeRabbit review feedback on _SANDBOX_HOME resolution. --------- Co-authored-by: Carlos Villela <cvillela@nvidia.com>
lakamsani
pushed a commit
to lakamsani/NemoClaw
that referenced
this pull request
Apr 4, 2026
…ross reconnects (NVIDIA#1025) * fix(sandbox): export proxy env vars with full NO_PROXY and persist across reconnects OpenShell injects NO_PROXY=127.0.0.1,localhost,::1 into the sandbox, missing inference.local and the gateway IP (10.200.0.1). This causes LLM inference requests to route through the egress proxy instead of going direct, and the proxy gateway IP itself gets proxied. Add proxy configuration block to nemoclaw-start.sh that: - Exports HTTP_PROXY, HTTPS_PROXY, and NO_PROXY with inference.local and the gateway IP included - Persists via /etc/profile.d/nemoclaw-proxy.sh (root) or ~/.profile (non-root fallback) so values survive OpenShell reconnect injection - Supports NEMOCLAW_PROXY_HOST / NEMOCLAW_PROXY_PORT overrides The non-root fallback ensures the fix works in environments like Brev where containers run without root privileges. Tested on DGX Spark (ARM64) and Brev VM (x86_64). Verified NO_PROXY contains inference.local and 10.200.0.1 inside the live sandbox after connect. Ref: NVIDIA#626, NVIDIA#704 Ref: NVIDIA#704 (comment) * fix(sandbox): write proxy config to ~/.bashrc for interactive reconnect sessions OpenShell's `sandbox connect` spawns `/bin/bash -i` (interactive, non-login), which sources ~/.bashrc — not ~/.profile or /etc/profile.d/*. The previous approach wrote to ~/.profile and /etc/profile.d/, neither of which is sourced by `bash -i`, so the narrow OpenShell-injected NO_PROXY persisted in live interactive sessions. Changes: - Write proxy snippet to ~/.bashrc (primary) and ~/.profile (login fallback) - Export both uppercase and lowercase proxy variants (NO_PROXY + no_proxy, HTTP_PROXY + http_proxy, etc.) — Node.js undici prefers lowercase no_proxy over uppercase NO_PROXY when both are set - Add idempotency guard to prevent duplicate blocks on container restart - Update tests: verify .bashrc writing, idempotency, bash -i override behavior, and lowercase variant correctness Tested on DGX Spark (ARM64) and Brev VM (x86_64) with full destroy + re-onboard + live `env | grep proxy` verification inside the sandbox shell via `openshell sandbox connect`. Ref: NVIDIA#626 * fix(sandbox): replace stale proxy values on restart with begin/end markers Use begin/end markers in .bashrc/.profile proxy snippet so _write_proxy_snippet replaces the block when PROXY_HOST/PORT change instead of silently keeping stale values. Adds test coverage for the replacement path. Addresses CodeRabbit review feedback on idempotency gap. * fix(sandbox): resolve sandbox user home dynamically when running as root When the entrypoint runs as root, $HOME is /root — the proxy snippet was written to /root/.bashrc instead of the sandbox user's home. Use getent passwd to look up the sandbox user's home when running as UID 0; fall back to /sandbox if the user entry is missing. Addresses CodeRabbit review feedback on _SANDBOX_HOME resolution. --------- Co-authored-by: Carlos Villela <cvillela@nvidia.com>
gemini2026
pushed a commit
to gemini2026/NemoClaw
that referenced
this pull request
Apr 14, 2026
…ross reconnects (NVIDIA#1025) * fix(sandbox): export proxy env vars with full NO_PROXY and persist across reconnects OpenShell injects NO_PROXY=127.0.0.1,localhost,::1 into the sandbox, missing inference.local and the gateway IP (10.200.0.1). This causes LLM inference requests to route through the egress proxy instead of going direct, and the proxy gateway IP itself gets proxied. Add proxy configuration block to nemoclaw-start.sh that: - Exports HTTP_PROXY, HTTPS_PROXY, and NO_PROXY with inference.local and the gateway IP included - Persists via /etc/profile.d/nemoclaw-proxy.sh (root) or ~/.profile (non-root fallback) so values survive OpenShell reconnect injection - Supports NEMOCLAW_PROXY_HOST / NEMOCLAW_PROXY_PORT overrides The non-root fallback ensures the fix works in environments like Brev where containers run without root privileges. Tested on DGX Spark (ARM64) and Brev VM (x86_64). Verified NO_PROXY contains inference.local and 10.200.0.1 inside the live sandbox after connect. Ref: NVIDIA#626, NVIDIA#704 Ref: NVIDIA#704 (comment) * fix(sandbox): write proxy config to ~/.bashrc for interactive reconnect sessions OpenShell's `sandbox connect` spawns `/bin/bash -i` (interactive, non-login), which sources ~/.bashrc — not ~/.profile or /etc/profile.d/*. The previous approach wrote to ~/.profile and /etc/profile.d/, neither of which is sourced by `bash -i`, so the narrow OpenShell-injected NO_PROXY persisted in live interactive sessions. Changes: - Write proxy snippet to ~/.bashrc (primary) and ~/.profile (login fallback) - Export both uppercase and lowercase proxy variants (NO_PROXY + no_proxy, HTTP_PROXY + http_proxy, etc.) — Node.js undici prefers lowercase no_proxy over uppercase NO_PROXY when both are set - Add idempotency guard to prevent duplicate blocks on container restart - Update tests: verify .bashrc writing, idempotency, bash -i override behavior, and lowercase variant correctness Tested on DGX Spark (ARM64) and Brev VM (x86_64) with full destroy + re-onboard + live `env | grep proxy` verification inside the sandbox shell via `openshell sandbox connect`. Ref: NVIDIA#626 * fix(sandbox): replace stale proxy values on restart with begin/end markers Use begin/end markers in .bashrc/.profile proxy snippet so _write_proxy_snippet replaces the block when PROXY_HOST/PORT change instead of silently keeping stale values. Adds test coverage for the replacement path. Addresses CodeRabbit review feedback on idempotency gap. * fix(sandbox): resolve sandbox user home dynamically when running as root When the entrypoint runs as root, $HOME is /root — the proxy snippet was written to /root/.bashrc instead of the sandbox user's home. Use getent passwd to look up the sandbox user's home when running as UID 0; fall back to /sandbox if the user entry is missing. Addresses CodeRabbit review feedback on _SANDBOX_HOME resolution. --------- Co-authored-by: Carlos Villela <cvillela@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
3 participants
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
OpenShell injects
NO_PROXY=127.0.0.1,localhost,::1into the sandbox, missinginference.localand the gateway IP (10.200.0.1). This causes LLM inference requests to be routed through the egress proxy instead of going direct, and the proxy gateway IP itself gets proxied.This PR adds a proxy configuration block to
nemoclaw-start.shthat:HTTP_PROXY,HTTPS_PROXY,NO_PROXY(and lowercase equivalents) with the full bypass list~/.bashrc(primary) and~/.profile(login-shell fallback) so values surviveopenshell sandbox connectre-injection_SANDBOX_HOMEdynamically viagetent passwdwhen running as root, falling back to/sandboxNEMOCLAW_PROXY_HOST/NEMOCLAW_PROXY_PORToverrides for environments where the gateway address differsWhy ~/.bashrc?
openshell sandbox connectspawnsbash -i(interactive, non-login) viassh.rs. The connect path callsenv_clear()and then injects the narrowNO_PROXY. Sincebash -isources~/.bashrcon startup, writing the full proxy config there restores the correct values after OpenShell's injection.Problem
When a user asks the agent to fetch a URL (e.g.
web_fetchtool), the sandbox routes the request through the egress proxy incorrectly because:NO_PROXYis too narrow —inference.localis missing, so LLM inference traffic is wrongly routed through the proxy10.200.0.1) is missing fromNO_PROXY, creating a potential proxy loopopenshell sandbox connectre-injects the narrowNO_PROXY, overwriting any manual fixundiciprefers lowercaseno_proxyover uppercase — without lowercase exports, the bypass list is ignoredChanges
scripts/nemoclaw-start.sh: Added proxy environment block with fullNO_PROXYlist,~/.bashrc+~/.profilepersistence with begin/end markers, dynamic sandbox home resolutiontest/service-env.test.js: Added unit tests for proxy variable extraction,NO_PROXYcompleteness, lowercase variants,.bashrc/.profilepersistence, idempotency, stale value replacement, and bash-i override simulationTesting
NO_PROXYandno_proxyin live sandbox shellRelated
getaddrinfo EAI_AGAIN) is a separate issue tracked in fix: sandbox DNS resolution via CoreDNS proxy in sandbox pod #732