mirror of
https://github.com/NVIDIA/OpenShell.git
synced 2026-10-15 05:09:44 +08:00
pull-request/2832
108
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Derek Carr
|
59479f492a |
feat(k8s): add namespace-per-workspace support (RFC 0011 Phase 3) (#2656)
* feat(k8s): add namespace-per-workspace support (RFC 0011 Phase 3) Implement three workspace namespace modes for the Kubernetes compute driver: shared (default, preserves current single-namespace behavior), managed (auto-creates/deletes namespaces per workspace), and operator (pre-provisioned namespaces with dynamic discovery via label selector or drop-in allowlist file). Key changes: - WorkspaceMode enum and namespace resolution in driver config - Managed namespace lifecycle with ServiceAccount and OpenShift SCC annotation propagation - Cluster-wide sandbox CR watchers for managed/operator modes - NamespaceValidator (Exact/Prefix/Allowlist) for SA token auth - Workspace-aware credential secret storage - Helm ClusterRole for multi-namespace RBAC - Gateway config, architecture, and reference docs Signed-off-by: Derek Carr <decarr@redhat.com> * test(k8s): add e2e tests for workspace namespace modes Add end-to-end tests for managed and operator workspace modes introduced in RFC 0011 Phase 3. The managed mode tests verify namespace creation with correct labels, ServiceAccount provisioning, sandbox CR placement, and namespace survival with remaining sandboxes. The operator mode tests verify rejection of unlabeled and nonexistent namespaces. The positive operator path (sandbox in labeled namespace) is known to fail due to an RBAC gap and will be addressed separately. Also fixes Helm 4 compatibility: move SPDX license headers inside conditional guards in 8 chart templates to prevent empty comment-only documents, and fix a trailing whitespace trimmer in clusterrole.yaml that concatenated the license header with apiVersion. Adds cleanup sweep in with-kube-gateway.sh to remove managed and operator namespaces before Helm uninstall, and mise tasks for running each mode independently. Signed-off-by: Derek Carr <decarr@redhat.com> * feat(k8s): add operator namespace label watcher Spawn a background kube::runtime::watcher in the K8s driver that watches namespaces matching the configured label selector and populates the OperatorNamespaceAllowlist at runtime. The driver owns the allowlist and exposes its Arc so the server can share the same set with the SA token authenticator. create_sandbox now gates pod creation on the allowlist in operator mode — workspaces whose namespace is not yet labeled are rejected at resource render time rather than silently proceeding. Workspace lifecycle itself is unaffected; only sandbox (resource) creation is gated. Signed-off-by: Derek Carr <decarr@redhat.com> * fix(k8s): harden operator mode and address review findings Close the fail-open gap in operator mode when only operator_namespace_file is configured: the allowlist is now created unconditionally in operator mode (fail-closed from startup). Implement the namespace file watcher using the notify crate, following the TLS hot-reload pattern (parent-directory watch, 1s debounce, ConfigMap symlink-swap safe). The file format is a JSON array of namespace name strings. Additional fixes from the 10-reviewer audit: - Change allowlist rejection from InvalidArgument to FailedPrecondition so callers know the request may succeed later once the namespace is provisioned. - NamespaceValidator::Allowlist now holds the OperatorNamespaceAllowlist newtype instead of a raw Arc<RwLock<BTreeSet>>, eliminating silent denial on RwLock poison. - Verify LABEL_MANAGED_BY and LABEL_GATEWAY_ID ownership before deleting a managed namespace. - Replace fixed 5s sleep in operator e2e test with a 30s poll loop. - Add Helm validation for workspaceMode values. - Fix Helm README type column and description for operator fields. - Add insert/remove methods to OperatorNamespaceAllowlist; label watcher now uses them instead of reaching through shared(). - Reject configs with both operator_namespace_label and operator_namespace_file set. Signed-off-by: Derek Carr <decarr@redhat.com> * feat(k8s): add workspace-level compute driver RPCs and harden RBAC Decouple namespace lifecycle from sandbox lifecycle by adding EnsureWorkspace/DeleteWorkspace RPCs to the ComputeDriver service. Namespace creation now happens before credential storage and namespace deletion happens on workspace delete, fixing credential storage in managed workspace mode. - Add EnsureWorkspace and DeleteWorkspace proto RPCs with implementations across all compute drivers (K8s managed delegates to ensure_namespace/delete_namespace_if_empty; others no-op) - Wire ensure_workspace into provider create/update/refresh paths so the namespace exists before the credential driver writes secrets - Wire delete_workspace into workspace deletion for cleanup - Remove delete_namespace_if_empty from sandbox deletion path - Scope ClusterRole secrets access to non-shared workspace modes - Add TODO for TLS cert hot-reload in sandbox gRPC client - Harden e2e tests with control-plane sandbox resolution assertions - Fix docker image save --platform flag for OCI index manifests Signed-off-by: Derek Carr <decarr@redhat.com> * fix(k8s): address re-review findings and add test coverage - Use server-side apply for TLS secret sync (fixes second sandbox creation failure when TLS is enabled) - Scope gateway-ID label selector unconditionally across all workspace modes (fixes operator reads/watches/deletes seeing foreign sandboxes) - Validate operator allowlist in EnsureWorkspace and DeleteWorkspace RPCs (prevents credential writes to namespaces outside the allowlist) - Extend ClusterRole secrets patch+delete to all non-shared modes with credential driver enabled (fixes operator credential storage RBAC) - Validate namespace ownership on 409 conflict in ensure_namespace (prevents adopting unowned namespaces in managed mode) - Replace delete_namespace_if_empty with unconditional delete_namespace letting Kubernetes cascade cleanup (fixes stuck terminating CRs) - Strengthen NetworkPolicy TODO to cover both managed and operator modes - Extract selector and ownership logic into testable free functions - Add unit tests for gateway-ID selectors and namespace ownership - Add Helm ClusterRole RBAC tests for operator credential driver Signed-off-by: Derek Carr <decarr@redhat.com> * ci(k8s): add workspace managed and operator mode e2e to CI Wire the existing e2e:kubernetes:workspace-managed and e2e:kubernetes:workspace-operator mise tasks into the branch-e2e workflow so they run alongside the other core Kubernetes e2e suites. Both are gated by run_core_e2e and included in the Core E2E result gate. Signed-off-by: Derek Carr <decarr@redhat.com> * test(k8s): add e2e tests for workspace namespace modes Add 7 new e2e tests covering workspace namespace lifecycle, TLS secret copying, ownership conflict detection, DNS-1123 validation, operator namespace preservation, and dynamic label watcher behavior. Fix async sandbox deletion race condition in existing tests by polling sandbox list instead of asserting immediately after delete. Signed-off-by: Derek Carr <decarr@redhat.com> * fix(k8s): grant secrets/patch unconditionally and backfill gateway-id labels Address two review findings: 1. RBAC: server-side apply (PATCH) is used for TLS secret sync in multi-namespace modes, but the ClusterRole only granted patch when the kubernetes-secrets credential driver was enabled. Grant patch unconditionally for non-shared modes since TLS sync always needs it; keep delete gated on the credential driver. 2. Upgrade safety: the new gateway-id label selector would orphan legacy Sandbox CRs that predate its introduction. Add a startup backfill in shared mode that patches any managed Sandbox CR missing the gateway-id label before the driver begins serving requests. Signed-off-by: Derek Carr <decarr@redhat.com> * fix(k8s): address workspace namespace review findings Signed-off-by: Derek Carr <decarr@redhat.com> * fix(k8s): address follow-up review findings Signed-off-by: Derek Carr <decarr@redhat.com> * fix(k8s): preserve workspace lookup after rebase Signed-off-by: Derek Carr <decarr@redhat.com> * fix(helm): allow managed secret creation Signed-off-by: Derek Carr <decarr@redhat.com> * fix(k8s): stop pods in workspace namespace Signed-off-by: Derek Carr <decarr@redhat.com> * test(k8s): scope pod deletion check to v1alpha1 Signed-off-by: Derek Carr <decarr@redhat.com> * fix(k8s): address workspace namespace review findings Signed-off-by: Derek Carr <decarr@redhat.com> --------- Signed-off-by: Derek Carr <decarr@redhat.com> |
||
|
krishicks
|
ae40cf6744 |
fix(gateway): respect OPENSHELL_BIND_ADDRESS in dev task (#2756)
This is convenient when you want to run a local gateway pointed at a remote compute driver so that the supervisor can reach across the network to the gateway which is listening on 0.0.0.0. Signed-off-by: Kris Hicks <khicks@nvidia.com> |
||
|
Drew Newberry
|
f12f3ef8d5 |
fix(macos): restore Homebrew sandbox callbacks (#2739)
* fix(macos): restore Docker gateway callbacks Signed-off-by: Drew Newberry <anewberry@nvidia.com> * fix(gateway): reuse reachable primary callback listener Signed-off-by: Drew Newberry <anewberry@nvidia.com> --------- Signed-off-by: Drew Newberry <anewberry@nvidia.com> |
||
|
Max DubrinskyandDrew Newberry
|
35fb27ef14 |
feat(sdk): add TypeScript SDK (@nvidia/openshell-sdk) (#2122)
* feat(sdk): add TypeScript SDK (@nvidia/openshell-sdk) First native, per-language SDK for the OpenShell gateway: a thin, idiomatic TypeScript client over proto-generated gRPC stubs (connect-es), no FFI. Covers the v0.1 surface — sandbox lifecycle (create/get/list/delete + waitReady/ waitDeleted), health, and streamed exec. - sdk/typescript/: package, client/transport/errors, protoc + protoc-gen-es codegen (gen/ gitignored, absorbed into dist/ at build), committed lockfile. - tasks/typescript.toml: sdk:ts install/proto/typecheck/build/ci/publish; sdk:ts:typecheck wired into `check`; sdk-typescript job in branch-checks (typecheck, build, and a --dry-run publish that validates the release path). - Enforce SPDX headers on .ts/.tsx/.mts/.cts (skip node_modules and gen/); back-fill docs/_components/jsx.d.ts and fern/components/CustomFooter.tsx. - release.py gains an npm version format; release-tag.yml publishes to GitHub Packages on tag, stamping the version (0.0.0 placeholder in git); prerelease builds publish under the `next` dist-tag, not `latest`. Ships as @nvidia/openshell-sdk on GitHub Packages pre-GA; public npm (@openshell/sdk) follows at GA with an unchanged public API. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * chore(sdk): adopt TypeScript 6, tidy @types/node range - typescript ^5.7.2 -> ^6.0.3 (6.0 is now `latest`; the old caret capped at 5.x) - @types/node ^24.0.0 -> ^24 (same range, tidier) No source changes; codegen, typecheck, and build pass on 6.0.3. Verified the emitted d.ts still type-check for downstream consumers on TypeScript 5.0.4 through 5.9.3, so this does not raise the SDK's consumer TS floor. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * refactor(sdk): group operations under a composable SandboxClient Reshape the client from flat methods (createSandbox, listSandboxes, exec) to a scoped SandboxClient reached as `client.sandbox.create/get/list/delete/exec` (+ waitReady/waitDeleted), mirroring the CLI's noun-verb model and the Python SDK's SandboxClient. SandboxClient is also usable standalone via SandboxClient.connect(); OpenShellClient composes it over a single shared transport, so future service/provider clients reuse one connection. health() stays top-level as a gateway call. No behavior change; types are unchanged. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * chore(sdk): generate TypeScript SDK stubs with buf Replace the protoc gen.sh with `buf generate` + buf.gen.yaml. `buf` (@bufbuild/buf) is a package devDependency and self-compiles the protos, so the TS SDK no longer depends on the mise-pinned protoc; it drives the same connect-es plugin. Generation stays limited to the client-surface closure (openshell/sandbox/datamodel) via the input paths. Output is byte-identical to the previous protoc + protoc-gen-es pipeline. Lays the groundwork for a shared buf.yaml (lint/breaking/LSP) as a follow-up. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * build(proto): add repo-level buf module with lint Declare proto/ as a single buf v2 module in a root buf.yaml so buf generate, lint, breaking, and the editor LSP resolve imports the same way. Lint uses STANDARD with six documented exceptions for deviations the current protos intentionally make: the flat proto/ layout with nested packages (DIRECTORY_SAME_PACKAGE, PACKAGE_DIRECTORY_MATCH) and the established API shape with unsuffixed services and reused request/response messages (RPC_REQUEST_RESPONSE_UNIQUE, RPC_REQUEST_STANDARD_NAME, RPC_RESPONSE_STANDARD_NAME, SERVICE_SUFFIX). Every other STANDARD rule now enforces on future protos. Breaking uses FILE. Code generation stays package-scoped in sdk/typescript/buf.gen.yaml since it binds to that package's connect-es plugin and output dir; its inputs are unchanged and regeneration is byte-identical. Wire the check in via a proto:lint mise task that runs buf from the SDK devDependencies. It is a dependency of both sdk:ts:ci (so the TypeScript SDK CI job enforces it) and the top-level lint aggregate (so local pre-commit covers it). Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * chore(sdk): publish as unscoped openshell-sdk on public npm Rename the package from @nvidia/openshell-sdk to the unscoped openshell-sdk and target public npm (registry.npmjs.org) instead of GitHub Packages. GitHub Packages requires a scope matching the owning org, and the @openshell scope is blocked by an unrelated existing package, so an unscoped name on public npm is the lowest-friction distribution path and needs no org approval. Rework the release-tag publish job to auth against registry.npmjs.org with NPM_TOKEN (the job now only needs packages: read to pull the CI image). Update the README install instructions and usage imports. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * chore(sdk): publish @nvidia/openshell-sdk to GitHub Packages Revert the unscoped-name switch. GitHub Packages only accepts scoped names matching the owning org, so shipping there first (which needs no external npm org or NPM_TOKEN, just the repo's GITHUB_TOKEN) requires the @nvidia scope. Keeping the @nvidia/openshell-sdk name also lets a later public-npm release use the same install specifier, so adding public npm becomes a second publish step rather than a rename. Restore the GitHub Packages publish auth in the release-tag job and the scoped install instructions in the README (keeping the buf codegen note). Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * feat(sdk-ts): add streaming exec, forward, ssh, provider, and config methods Grow SandboxClient to the surface the first two consumers need. execStream yields stdout/stderr chunks as they arrive and exec now drains it, keeping its buffered ExecResult and signature unchanged. execInteractive is the TTY + stdin transport primitive (start-first framing, output/write/resize/close/done, no terminal glue). forward binds a local TCP listener that tunnels each accepted connection into the sandbox for the process lifetime, minting and revoking a per-socket SSH session token around a forwardTcp bidi. Adds createSshSession / revokeSshSession, attach/detach/listProviders, and getConfig / setPolicy / setSetting (sandbox-scoped, network-policy-only, with an optional wait poll). Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * build(sdk-ts): add Biome and Vitest tooling The TypeScript SDK had no formatter or linter and no test runner. Add Biome (format + lint, generated src/gen excluded) enforcing 2-space indent, single quotes, semicolons, and a 120-column width, and reformat the existing hand-written sources accordingly. Add Vitest for unit tests. Wire sdk:ts:format, sdk:ts:lint, and sdk:ts:test mise tasks into the fmt/lint aggregates, the root test suite, and sdk:ts:ci so they run in CI. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * test(sdk-ts): cover the sandbox surface with in-memory transport tests Exercise SandboxClient against an in-memory OpenShell service built with createRouterTransport: request assembly and id resolution, u64/int64 rendered as strings, enum lowercasing, fromConnect code mapping, the exec/execStream drain plus a backward-compat check on exec, execInteractive start-first ordering and done resolution, and a forward() byte relay against a loopback echo with close() teardown. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * docs(sdk-ts): document the new surface and connect/upload/download boundaries Document execStream, execInteractive, forward, ssh sessions, providers, and config/policy in the SDK README, and record the intentional boundaries: interactive connect / PTY ownership, upload/download (no file-transfer RPC), and detached forwards stay out of scope. Note the Biome/Vitest dev commands. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * feat(sdk-ts): support mTLS client authentication Add clientCert and clientKey to ConnectOptions so the SDK can authenticate to the default local gateway, which uses mTLS user authentication. Without a client certificate and key the SDK could verify the server but never authenticate the caller, so it could not connect to the standard Docker, VM, Homebrew, or Linux-package gateway. Validate the pair as both-or-neither and pass cert and key through to the Node TLS options for https gateways. The h2c path is unchanged. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * chore(sdk-ts): drop the demo script and its tsx dependency Remove src/demo.ts, the demo npm script, the tsx devDependency, and the tsconfig build exclude for the demo. The demo was never part of the published package, and dropping it also removes the only place that logged part of an SSH session token. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * fix(sdk-ts)!: harden exec streaming, waits, SSH, and forwarding Address review feedback on the sandbox surface. - Make the streamed command exit code observable from idiomatic for-await: the terminal exit is now an in-band ExecStreamEvent ({ type: 'exit', exitCode }) rather than the async generator return value, which for-await discards. A stream that ends without an exit event now throws instead of reporting success. - Bound waitReady, waitDeleted, and the setPolicy wait by their timeout: each poll RPC carries a per-iteration deadline and the waits accept an AbortSignal, so a stalled call can no longer leave a wait pending forever. Add waitTimeoutSecs to SetPolicyOptions. - Validate the CreateSshSession response against the proto charset and range contract before returning it or using its token, since the values feed an OpenSSH ProxyCommand. - Respect socket backpressure when relaying forwarded responses: pause reading the gRPC stream when the local socket buffer is full and resume on drain so memory stays bounded. - Expose create-time sandbox policy: add policy and an advanced rawSpec passthrough to SandboxSpec so the safety boundary is expressible at creation and new spec fields do not require an SDK change. BREAKING CHANGE: execStream and the interactive exec output now yield a terminal { type: 'exit', exitCode } event; consumers iterating the stream must handle that arm. The exit code is no longer the async generator return value. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * feat(sdk-ts): export error contract, enum unions, and caller cancellation Address Tier-1 review feedback on the TypeScript SDK public surface (PR #2122). - Errors: export SdkError and SdkErrorCode so callers can use instanceof and exhaustively switch on .code. fromConnect preserves the originating ConnectError as .cause and its status as .connectCode, maps Aborted to a new 'aborted' code for optimistic-concurrency conflicts, and maps Canceled and DeadlineExceeded to 'canceled'. errorCode() behavior is unchanged. - Enums: replace the string-typed phase, status, scope, and policySource fields with lowercase literal unions (SandboxPhaseName, HealthStatus, SettingScopeName, PolicySourceName) backed by exhaustive Record maps. The unions are a hand-maintained mirror of the generated proto enums; a new drift test pins each literal to its generated member name. - Cancellation: accept an optional AbortSignal on exec, execInteractive, and forward, threaded into both sandbox resolution and the streaming RPC. forward tears down its local listener on abort. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * feat(sdk-ts): add raw escape hatch for uncurated gateway RPCs The curated sub-clients reduce proto messages to ergonomic subsets (for example get() drops created_at_ms, the full spec, conditions, runtime endpoints, and current_policy_version), and not every gateway RPC has a typed helper yet. Rather than ship methods that exist but throw, expose a generated client for the full surface. OpenShellClient.raw and SandboxClient.raw are generated clients covering every gateway RPC, returning the verbatim wire messages so proto distinctions the curated types smooth over are preserved. .transport exposes the shared connection for building extra clients over one socket. Generated request/response types are published at the new @nvidia/openshell-sdk/raw subpath. Curated methods stay the default; raw is the always-available floor. Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * fix(sdk-ts): address review feedback on exec, forward, and auth transport Settle exec's `done` promise before yielding the exit event so a consumer that breaks on exit no longer leaves it pending forever, and give it a lone rejection handler plus a finally-settle so a stream error or early abandon can never surface as an unhandled rejection or a hang. Attach an 'error' listener to each accepted forward socket synchronously, before forwardConnection awaits CreateSshSession; a peer reset in that window previously emitted an unhandled 'error' and crashed the process. Reject ambiguous or unsafe transport configs at buildTransport: oidcToken and edgeToken together (silently OIDC-only), and any auth token sent over plaintext http:// to a non-loopback host unless allowInsecureAuth is set. Wrap versionPin so a non-u64 expectedResourceVersion raises SdkError('invalid_config') instead of a raw BigInt SyntaxError, and raise the Node engine floor to >=20.3 for AbortSignal.any(). Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> * fix(sdk-ts): address review feedback Signed-off-by: Drew Newberry <anewberry@nvidia.com> * docs(sdk-ts): defer published sdk guide Signed-off-by: Drew Newberry <anewberry@nvidia.com> --------- Signed-off-by: Max Dubrinsky <mdubrinsky@nvidia.com> Signed-off-by: Drew Newberry <anewberry@nvidia.com> Co-authored-by: Drew Newberry <anewberry@nvidia.com> |
||
|
krishicks
|
245fe27589 |
fix(dev): separate Podman Machine loopback listeners (#2725)
On macOS, bind the standalone Podman gateway to IPv6 loopback while
registering localhost as the TLS endpoint. This keeps IPv4 loopback
available for the callback-only listener, matching the e2e fix in
commit
|
||
|
|
2f96c53b8c |
feat(gateway,cli): windows compilation support (#2496)
* chore(windows): gate Unix-only workspace code for MSVC Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * feat(windows): stub unsupported compute drivers Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * ci(windows): add MSVC mise build lane Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * docs(windows): document MSVC build-only design Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * docs(agent): add Windows MSVC build skill Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * feat(windows): add Windows build support Signed-off-by: Akber Raza <akberr@nvidia.com> * refactor(windows): consolidate Windows-specific dependencies and improve build logic Signed-off-by: Akber Raza <akberr@nvidia.com> * feat(windows): add libclang path resolution and update cargo commands with bundled Z3 features Signed-off-by: Akber Raza <akberr@nvidia.com> * chore(tooling): lock Windows tool artifacts Signed-off-by: Giedrius Burachas <gburachas@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * feat(windows): enhance libclang path resolution to support architecture-specific subdirectories Signed-off-by: Akber Raza <akberr@nvidia.com> * Fix Windows dependency gating after sync merge Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(z3): update Z3 header path requirements in Windows build documentation and scripts Signed-off-by: Akber Raza <akberr@nvidia.com> * docs(windows): relocate Windows MSVC build design to architecture/ Why: windows-msvc-build-design.mdx is a design document ("design decisions for the native Windows MSVC build lane"), but it lived in the published, user-facing docs/reference/ tree. Per AGENTS.md (Documentation) and architecture/README.md ("rfc/ vs architecture/"), design content belongs in architecture/ (or rfc/), not in published reference. It also shared Fern sidebar "position: 6" with the MXC compute-driver design page, colliding in the Reference nav ordering. What: - Move docs/reference/windows-msvc-build-design.mdx -> architecture/windows-msvc-build.md. - Strip the Fern publish frontmatter and add a plain H1, matching the other architecture docs. - Register it in the architecture doc index in architecture/README.md. - Repoint the inbound references (build-openshell-mxc-windows skill + reference, implement-openshell-mxc-driver skill) to the new path. With both design pages moved out of docs/reference/, the duplicate position-6 sidebar collision is resolved. Signed-off-by: Akber Raza <akberr@nvidia.com> * remove openshell-supervisor-network from unsupported driver package test exclusion list Signed-off-by: Akber Raza <akberr@nvidia.com> # Conflicts: # tasks/scripts/windows-msvc.ps1 * fix(interceptors): gate unix-only imports so the crate builds on Windows openshell-gateway-interceptors failed to compile on Windows (E0432: no UnixStream in tokio::net), breaking any Windows build of openshell-server (which depends on it unconditionally). The connect_unix_endpoint fn was already #[cfg(unix)]-gated, but the imports it uses (UnixStream, TokioIo, Uri, service_fn) were left ungated. Gate those four imports with #[cfg(unix)] too. No behavior change on unix; Windows now compiles (no errors, no unused-import warnings). Signed-off-by: Akber Raza <akberr@nvidia.com> * feat(windows): add native ARM64 test support Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(mise): skip Skaffold on Windows Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): harden ARM64 toolchain discovery Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): scope ARM64 toolchain preflight Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): restore compatibility after GitHub sync Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): avoid rate-limited Z3 source lookup Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(mise): skip Helm checks on Windows Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): support repository pre-commit checks Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): stabilize native MSVC validation Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): harden shared Z3 source cache Signed-off-by: Shailendra Singh <shailendras@nvidia.com> * fix(windows): avoid leaking MSVC flags into clang-cl Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): complete ARM64 migration audit Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): restore ARM64 Ninja discovery Signed-off-by: Akber Raza <akberr@nvidia.com> * refactor(windows): separate platform crate roots Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): restore proto include cfg gating Signed-off-by: Akber Raza <akberr@nvidia.com> * refactor: address lint errors * fix(windows): add preflight check for proxy auth file path * docs(windows): update GitHub checkout guidance Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): restore CI after dependency updates Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(mise): repair Windows sccache lock entry Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(windows): reconcile validation after rebase Signed-off-by: Akber Raza <akberr@nvidia.com> * refactor(server): exclude unsupported drivers on Windows Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com> * refactor(server): isolate platform driver config Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com> * fix(windows): repair unsupported driver contract test Signed-off-by: Akber Raza <akberr@nvidia.com> * fix(sandbox): remove stale dependencies Signed-off-by: Akber Raza <akberr@nvidia.com> * ci(windows): pin x64 workflow actions Signed-off-by: Akber Raza <akberr@nvidia.com> * ci(windows): align x64 Rust toolchain Signed-off-by: Akber Raza <akberr@nvidia.com> * ci(windows): align ARM64 workflow setup Signed-off-by: Akber Raza <akberr@nvidia.com> * refactor(windows): exclude unsupported runtime crates Signed-off-by: Akber Raza <akberr@nvidia.com> * refactor(windows): exclude unsupported crates at workspace boundary Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> * refactor(server): gate builtin driver config by platform Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> * fix(sandbox): restore crate documentation Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> * ci(windows): make build workflow manual Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> * ci(windows): temporarily enable pull request builds Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> * ci(windows): cache Rust dependencies Signed-off-by: Akber Raza <akberr@nvidia.com> * refactor(windows): remove unnecessary platform changes Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> * ci(windows): make build workflow manual Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> * fix(ci): synchronize mise lockfile Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> * fix(ci): normalize mise provenance metadata Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> * refactor(python): isolate Windows atomic replace retry Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> * fix(python): type Windows permission test errors Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> --------- Signed-off-by: Shailendra Singh <shailendras@nvidia.com> Signed-off-by: Akber Raza <akberr@nvidia.com> Signed-off-by: Giedrius Burachas <gburachas@nvidia.com> Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com> Signed-off-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> Co-authored-by: Shailendra Singh <shailendras@nvidia.com> Co-authored-by: Giedrius Burachas <gburachas@nvidia.com> Co-authored-by: Jamie King <jamiek@nvidia.com> Co-authored-by: Piotr Mlocek <pmlocek@nvidia.com> Co-authored-by: Piotr Mlocek <1116309+pimlock@users.noreply.github.com> |
||
|
Emilien MacchiandMrunal Patel
|
3e191558b3 |
feat(build): add glibc-static supervisor libc variant (#2682)
The supervisor binary runs inside sandbox images whose libc and glibc version are unknown at build time, so it must be statically linked. Add SUPERVISOR_LIBC to select between the default musl variant and a new glibc-static variant that builds the GNU target with +crt-static. glibc-static has no cross-compile path: zig cc accepts -static for *-linux-gnu targets and emits a dynamically linked binary anyway. The staging script therefore refuses a cross-arch request for that variant rather than silently degrading linkage, and requires a native per-architecture build. Add verify-static-binary.sh, run after every supervisor build in both the staging script and CI so linkage cannot regress unnoticed for either variant. It inspects via readelf (or greadelf/llvm-readelf) and fails closed rather than trusting the tool's exit status: every inspection must produce no diagnostics, the input must be an executable ELF (ET_EXEC, or ET_DYN with DF_1_PIE) whose PT_LOAD segments all lie within the file, whose dynamic table agrees with PT_DYNAMIC, and which carries no PT_INTERP and no DT_NEEDED. That rejects a dynamically linked, truncated, corrupt, non-ELF, or shared-object input that naive parsing would misread as static. Hosts without any inspector (e.g. macOS, which ships no binutils) skip with a warning; Linux, including CI, requires one and fails closed. No image or release workflow builds the glibc-static variant, so add a dedicated supervisor-static-validate workflow that builds it on both architectures and runs the verifier. rust-native-build.yml uses self-hosted runners, which reject pull_request-triggered jobs, so it validates in the merge queue and on pushes to main that touch the build inputs, plus a nightly schedule, so the GNU + crt-static build branch cannot regress unnoticed. The default is unchanged, so image, release, and CI behavior is identical. Selecting glibc-static statically links LGPL glibc into a redistributed binary, which is why it is opt-in. Signed-off-by: Mrunal Patel <mrunalp@gmail.com> Co-authored-by: Mrunal Patel <mrunalp@gmail.com> |
||
|
Evan LezarandDrew Newberry
|
d220d89468 |
feat(compute): negotiate gateway callback listeners (#2492)
* feat(compute): query gateway listener requirements Signed-off-by: Evan Lezar <elezar@nvidia.com> * feat(compute): add Podman listener requirements Signed-off-by: Evan Lezar <elezar@nvidia.com> * test(docker): use default gateway bind address Signed-off-by: Evan Lezar <elezar@nvidia.com> * fix(gateway): avoid wildcard primary listener Signed-off-by: Evan Lezar <elezar@nvidia.com> * fix(podman): validate callback listener discovery Signed-off-by: Evan Lezar <elezar@nvidia.com> * fix(server): support split dual-stack listeners Signed-off-by: Evan Lezar <elezar@nvidia.com> * fix(podman): support legacy rootless listener discovery Signed-off-by: Evan Lezar <elezar@nvidia.com> * test(e2e): accept loopback plaintext rejection Signed-off-by: Evan Lezar <elezar@nvidia.com> * docs(agent): add callback listener diagnostics Signed-off-by: Evan Lezar <elezar@nvidia.com> * fix(server): restrict compute callback listeners Signed-off-by: Evan Lezar <elezar@nvidia.com> * fix(podman): validate local callback port Signed-off-by: Evan Lezar <elezar@nvidia.com> * test(server): clarify callback listener contract Signed-off-by: Evan Lezar <elezar@nvidia.com> * fix(podman): require pasta for local callbacks Signed-off-by: Evan Lezar <elezar@nvidia.com> * docs(gateway): document RPM listener default Signed-off-by: Evan Lezar <elezar@nvidia.com> * refactor(server): keep listener provenance diagnostic-only Signed-off-by: Evan Lezar <elezar@nvidia.com> * fix(compute): preserve callback listener isolation Signed-off-by: Drew Newberry <anewberry@nvidia.com> * test(e2e): remove Podman callback relay Signed-off-by: Drew Newberry <anewberry@nvidia.com> * fix(packaging): preserve Podman callback loopback Signed-off-by: Drew Newberry <anewberry@nvidia.com> * ci(e2e): run VM smoke on nested-virt runner Signed-off-by: Drew Newberry <anewberry@nvidia.com> * ci(e2e): gate VM smoke on usable KVM Signed-off-by: Drew Newberry <anewberry@nvidia.com> * ci(e2e): probe KVM through VM driver Signed-off-by: Drew Newberry <anewberry@nvidia.com> * ci(e2e): tolerate hosted KVM denial Signed-off-by: Drew Newberry <anewberry@nvidia.com> * test(server): close traced futures before assertions Signed-off-by: Drew Newberry <anewberry@nvidia.com> * revert: remove tracing test stabilization Signed-off-by: Drew Newberry <anewberry@nvidia.com> --------- Signed-off-by: Evan Lezar <elezar@nvidia.com> Signed-off-by: Drew Newberry <anewberry@nvidia.com> Co-authored-by: Drew Newberry <anewberry@nvidia.com> |
||
|
Drew Newberry
|
1cbfc0d510 |
test(e2e): add reusable QEMU infrastructure for E2E tests (#2471)
* test(vm): add composable QEMU test guests Signed-off-by: Drew Newberry <anewberry@nvidia.com> * docs(vm): describe test VM directory structure Signed-off-by: Drew Newberry <anewberry@nvidia.com> * refactor(vm): replace shell catalog functions Signed-off-by: Drew Newberry <anewberry@nvidia.com> * test(vm): add Fedora release guest support Signed-off-by: Drew Newberry <anewberry@nvidia.com> * test(vm): enable rootless Podman socket Signed-off-by: Drew Newberry <anewberry@nvidia.com> * feat(test-guest): add OCI-backed image caching Signed-off-by: Drew Newberry <anewberry@nvidia.com> * perf(test-guest): accelerate cached guest startup Signed-off-by: Drew Newberry <anewberry@nvidia.com> * fix(test-guest): address review feedback Signed-off-by: Drew Newberry <anewberry@nvidia.com> * fix(test-guest): verify OCI cache provenance Signed-off-by: Drew Newberry <anewberry@nvidia.com> * fix(test-guest): harden cached guest reuse Signed-off-by: Drew Newberry <anewberry@nvidia.com> * fix(test-guest): refresh runtime setup state Signed-off-by: Drew Newberry <anewberry@nvidia.com> * feat(test-guest): support E2E runner inputs Signed-off-by: Drew Newberry <anewberry@nvidia.com> * fix(test-guest): harden runner and OCI reuse Signed-off-by: Drew Newberry <anewberry@nvidia.com> * feat(test-guest): prepare Podman E2E artifacts Signed-off-by: Drew Newberry <anewberry@nvidia.com> * fix(test-guest): address review findings Signed-off-by: Drew Newberry <anewberry@nvidia.com> * revert(test-guest): remove recent Podman artifact changes Signed-off-by: Drew Newberry <anewberry@nvidia.com> * fix(test-guest): canonicalize scp source paths Signed-off-by: Drew Newberry <anewberry@nvidia.com> * refactor(test-guest): provision artifacts with Ansible Signed-off-by: Drew Newberry <anewberry@nvidia.com> * feat(test-guest): populate missing caches on startup Signed-off-by: Drew Newberry <anewberry@nvidia.com> --------- Signed-off-by: Drew Newberry <anewberry@nvidia.com> |
||
|
Evan Lezar
|
eb380d71a3 |
ci(e2e): probe VM gateway readiness (#2544)
Signed-off-by: Evan Lezar <elezar@nvidia.com> |
||
|
Philippe Martin
|
77e5c32217 |
feat(sandbox,gateway): route sandbox egress through corporate HTTP proxy (#2245)
* feat(sandbox,gateway): route sandbox egress through corporate HTTP proxy - Chain sandbox egress through a corporate HTTP proxy so outbound traffic from within the sandbox respects the host proxy settings - Forward sandbox proxy environment variables to the generated Podman config so the proxy is applied consistently to Podman-managed workloads Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): make corporate proxy routing operator-owned The operator-configured corporate egress proxy was injected under the conventional HTTPS_PROXY/HTTP_PROXY/NO_PROXY names as defaults beneath sandbox spec/template environment, so a sandbox creator could redirect egress at an arbitrary proxy or disable proxying with NO_PROXY=*. Route the boundary through reserved, supervisor-only variables (OPENSHELL_UPSTREAM_HTTPS_PROXY/HTTP_PROXY/NO_PROXY) written in the Podman driver's required-variable tier. Any sandbox-supplied value under a reserved name is stripped before the operator value is applied, so the supervisor never observes a reserved proxy variable the operator did not set. The supervisor now reads only the reserved names and ignores the conventional proxy variables the sandbox controls. Add the reserved proxy variables to the supervisor-only child-environment denylist so the corporate proxy URL and any embedded credentials are not inherited by the sandbox workload, which reaches egress through the local policy proxy and never needs them. Signed-off-by: Philippe Martin <phmartin@redhat.com> * feat(sandbox,podman): deliver corporate proxy credentials via secret file Proxy credentials were embedded inline in the proxy URL, so they were stored in gateway.toml and exposed in container metadata via 'podman inspect'. Reject inline 'user:pass@' credentials in https_proxy/http_proxy at startup (parsed with the url crate rather than hand-splitting), and add a proxy_auth_file option pointing at a 'user:pass' file. The driver stages that file as a per-sandbox root-only Podman secret, mounts it at a fixed path, and exports only the path in the reserved OPENSHELL_UPSTREAM_PROXY_AUTH_FILE variable, so the credential never appears in config, environment, or container metadata. Reading the file fails closed on a missing, empty, or control-character-bearing value. The supervisor reads the credential from the mounted file and builds the Proxy-Authorization: Basic header, rejecting control characters, and no longer derives credentials from URL userinfo. The auth-file path is added to the child-environment strip list, and the generated gateway.toml is written owner-only (mode 0600). Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): fail closed on invalid upstream proxy configuration The reserved OPENSHELL_UPSTREAM_* variables are an operator-owned egress boundary, but the supervisor treated present-but-invalid values as unset: an unsupported or malformed proxy URL was ignored with a warning, an unreadable auth file proceeded without credentials, and a malformed credential silently became unauthenticated. Any of these could quietly downgrade the corporate proxy boundary to direct dialing or unauthenticated proxy access. Make every configured-but-invalid proxy or auth setting fatal to supervisor proxy startup, emit an OCSF ConfigStateChange failure event before refusing, and share URL validation semantics between the Podman driver and the supervisor through a single validator in openshell-core (parse_upstream_proxy_url), so a value accepted at sandbox-create time can never be rejected in-container or vice versa. Inline user:pass@ URL credentials are now fatal in the supervisor too (previously warn-and-strip), matching the driver. Unset or empty variables still mean no proxy; only present-but-invalid values fail. Error paths never include credential content. Addresses the fail-closed review item on #2245. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): finish the fail-closed upstream proxy credential contract The upstream proxy URL already had a single shared validator, but the credential did not: the Podman driver rejected only CR/LF/NUL while the supervisor rejected every control character, so a credential accepted at sandbox-create time (e.g. one containing a tab) could still be rejected in-container. Present-but-whitespace reserved OPENSHELL_UPSTREAM_* values were also silently treated as unset, quietly downgrading the operator's egress boundary to direct dialing. Add parse_upstream_proxy_credential to openshell-core as the single source of truth for the documented user:pass credential form (non-empty user, no control characters, trimmed) and use it in both the Podman driver's secret staging and the supervisor's Proxy-Authorization header construction. Error variants carry no payload so credential content can never leak into messages. Make a present-but-empty reserved variable fatal to supervisor proxy startup instead of meaning "unset"; only fully unset variables disable the proxy. The driver correspondingly rejects an empty no_proxy at config time so it can never inject a value the supervisor refuses. Addresses the remaining fail-closed credential/config review item on Signed-off-by: Philippe Martin <phmartin@redhat.com> #2245. * fix(sandbox,podman): close remaining fail-open upstream proxy config paths Two configuration paths could still silently run without the proxy boundary the operator believed was in effect. A no_proxy bypass list configured without any https_proxy/http_proxy was accepted by both the driver and the supervisor and simply meant "dial everything directly". Reject it on both sides, exactly like the existing proxy_auth_file-without-proxy rule: an operator who wrote a bypass list assumed proxying was active, so accepting it hides a fail-open state. The gateway.sh dev script guarded proxy settings with [[ -n "${VAR:-}" ]], which conflates unset with explicitly-empty and dropped the latter before the gateway's validation could see it. Use ${VAR+x} instead so a set-but-empty variable is written into gateway.toml and rejected at startup by validate_proxy_config rather than silently discarded. Addresses the remaining fail-open configuration review item on #2245. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): reject upstream proxy URLs with path, query, or fragment parse_upstream_proxy_url accepted URLs like http://proxy.corp.com:8080/some/path and silently discarded everything after host:port, a lenience inherited from the original supervisor parser. A forward proxy is addressed by host:port only, so extra components indicate a misconfiguration (for example a pasted endpoint URL) and silently truncating them violates the present-but-invalid-is-fatal contract enforced everywhere else in this configuration surface. Reject a path, query, or fragment in the shared validator with a new UnexpectedComponent error. A bare trailing slash remains accepted because the url crate normalizes an absent http path to "/", making the two indistinguishable. Both the Podman driver (gateway startup) and the supervisor (sandbox startup) inherit the rule through the shared parser, keeping their semantics identical by construction. Addresses the proxy URL component review item on #2245. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): remove plain-HTTP upstream proxy support The http_proxy path tunneled plain-HTTP requests through the corporate proxy with CONNECT to port 80 and then sent origin-form requests down the tunnel. Conventional enterprise forward proxies expect plain HTTP as absolute-form requests sent directly over the proxy connection, and commonly refuse CONNECT to port 80, so the setting looked supported but failed against typical deployments. Tunneling also blinds the proxy to the one protocol it could inspect. Narrow the feature to TLS (CONNECT) egress only, which is the conventional and already-correct case: plain-HTTP requests now always dial the destination directly, and only client CONNECT tunnels chain through the corporate proxy. Remove the http_proxy config field, the --sandbox-http-proxy / OPENSHELL_SANDBOX_HTTP_PROXY driver surface, the reserved OPENSHELL_UPSTREAM_HTTP_PROXY variable, and the UpstreamScheme plumbing. The feature never shipped, so this is a clean removal; a stray http_proxy key in gateway.toml still fails loudly through the config's deny_unknown_fields. Removing the plain-HTTP proxy branch also removes its host-gateway special case; the architecture doc now documents the real host-gateway behavior (add driver-injected host aliases to the reserved NO_PROXY list) instead of an invariant the HTTPS path never implemented. Plain-HTTP forwarding through a corporate proxy can return later as absolute-form forwarding behind its own design review. Addresses the plain-HTTP forwarding review item on #2245. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): escape generated TOML and require explicit proxy URL form Escape backslashes, quotes, and control characters when gateway.sh writes proxy values into gateway.toml, so a hostile or unusual environment value cannot corrupt the config or inject extra keys. Restrict the upstream proxy URL grammar to the documented http://host:port form: a scheme-less value is no longer normalized to http:// and a missing port is no longer silently defaulted to 80. Docs, README, and CLI help now state the explicit-form requirement consistently. Also fix a test-only call of handle_tcp_connection that was missing the upstream_proxy argument added in an earlier commit. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): preserve tunneled bytes read with the CONNECT response The CONNECT handshake reads the corporate proxy's response in chunks, so the read that completes the header block can also contain the first tunneled payload bytes. Those bytes were discarded, silently corrupting the start of the tunnel for server-speaks-first destinations or proxies that coalesce writes. connect_via now returns a PrefixedStream that replays any bytes received past the response terminator before reading from the socket again; writes pass through unchanged. Direct dials wrap the stream with an empty prefix so downstream relay and TLS paths keep a single stream type, and tls_connect_upstream is generalized to any AsyncRead + AsyncWrite stream. Adds regression coverage for a combined response/payload read and for prefix replay ordering. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): gate cleartext proxy Basic auth behind an explicit opt-in Proxy-Authorization: Basic is base64 over the plain-TCP connection to the http:// corporate proxy, so anyone on the network path between the sandbox host and the proxy can recover the credential. Sending it is now an explicit operator decision instead of an implicit side effect of configuring proxy_auth_file. Add a proxy_auth_allow_insecure driver setting (CLI --sandbox-proxy-auth-allow-insecure, env OPENSHELL_SANDBOX_PROXY_AUTH_ALLOW_INSECURE), delivered to the supervisor as the reserved OPENSHELL_UPSTREAM_PROXY_AUTH_ALLOW_INSECURE variable. Fail-closed pairing on both sides: an auth file without the acknowledgement is rejected at gateway startup and at supervisor startup, as is the acknowledgement without an auth file or any value other than 'true'. gateway.sh writes the key only as a TOML boolean; a non-boolean value is emitted as a quoted string so the gateway rejects it at startup instead of risking injection. Documents the exposure prominently in the gateway config reference, the driver README, and the sandbox architecture doc. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): honor port qualifiers and resolved addresses in NO_PROXY Two divergences from the documented NO_PROXY contract: - A port-qualified entry (internal.corp:8443) was silently stripped to its hostname and bypassed the proxy for every port, excluding traffic the operator never listed. Entries now keep the optional :port qualifier (also on IP and CIDR entries) and only apply to that destination port; a trailing qualifier that is not a valid port stays part of the pattern instead of widening the entry. - IP and CIDR entries only matched IP-literal hosts, so a bypass like 10.0.0.0/8 never applied to hostnames resolving into that range. NO_PROXY evaluation now sees the validated resolved addresses: an IP/CIDR entry matching through resolution authorizes a direct dial of only the addresses it contains, so a bypass scoped to an internal range cannot widen into a direct dial of addresses outside it. Hostname-level matches (loopback, wildcard, domain entries, IP-literal hosts) keep authorizing all validated addresses. proxy_for is replaced by decision(host, port, resolved) returning either the proxy endpoint or the permitted direct-dial subset, and dial_upstream restricts the direct connect to that subset. Adds regression coverage for port-scoped bypasses on domain, IP, and CIDR entries, invalid port qualifiers, resolved-address matching, and split-resolution subset dialing. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): bind proxied CONNECT tunnels to validated addresses The CONNECT request sent to the corporate proxy carried the destination hostname, so the proxy resolved the name itself and the addresses that had passed SSRF and allowed_ips validation were discarded. Split-horizon DNS or rebinding at the proxy could then reach internal or otherwise unapproved destinations through a tunnel the supervisor logged as validated, and IP-range policy could not be enforced at all on proxied dials. CONNECT now targets a validated resolved address by default: the proxy performs no DNS resolution and the tunnel stays bound to the answer the supervisor checked. The hostname still travels inside the tunnel (TLS SNI, application Host), so destination servers behave normally. In split-horizon networks, operators point the gateway host at the corporate resolver so internal names validate to their internal addresses. For proxies whose ACLs filter on hostnames and reject IP CONNECT targets, a new proxy_connect_by_hostname opt-in (CLI --sandbox-proxy-connect-by-hostname, env OPENSHELL_SANDBOX_PROXY_CONNECT_BY_HOSTNAME, reserved OPENSHELL_UPSTREAM_PROXY_CONNECT_BY_HOSTNAME) restores hostname CONNECT, documented as re-opening proxy-side resolution and making the proxy's ACLs the effective egress control. Fail-closed pairing on both sides: the opt-in without a proxy, or any value other than 'true', is fatal. Adds regression coverage for the IP CONNECT request line (including IPv6 bracketing and hostname non-leakage), the hostname opt-in, and the config pairing rules in driver and supervisor. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): reject empty port after bracketed IPv6 proxy host http://[fd00::1]: passed the explicit-port check because the bracketed branch only tested for a colon after the bracket, then fell back to port 80 — violating the fail-closed http://host:port contract and potentially sending configured Basic credentials to an unintended service. Require a non-empty suffix after ]:, matching the unbracketed branch, and cover the case in the shared parser tests. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): fall back across validated addresses in proxied CONNECT The direct path hands TcpStream::connect the whole validated address list and it falls back across them, but the validated-IP CONNECT path attempted only the first address, so a dual-stack destination could fail through the corporate proxy even when a later validated address was reachable. connect_via_validated tries each validated address in order under one aggregate CONNECT_HANDSHAKE_TIMEOUT budget, returning the first success; when every attempt fails the error names the attempt count and carries the last failure. An empty address list is rejected up front. Adds regressions for first-fails/second-succeeds fallback (asserting both CONNECT request lines), the aggregate all-addresses failure message, and the empty-list rejection. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): strip new reserved proxy vars and complete docs/tests Follow-ups from review: - Add OPENSHELL_UPSTREAM_PROXY_AUTH_ALLOW_INSECURE and OPENSHELL_UPSTREAM_PROXY_CONNECT_BY_HOSTNAME to the supervisor-only strip list so workload child processes never inherit them, matching the documented contract for the other reserved proxy variables, and cover both in the supervisor-only variable test. - Document all five OPENSHELL_SANDBOX_* proxy variables in the mise run gateway help text, marked Podman-only and stating the auth-file/acknowledgement pairing, and complete the gateway-key list in the Podman README. - Add NO_PROXY composition coverage for bracketed IPv6 entries with port qualifiers, bare IPv6 entries, and IPv6 CIDR matching against an IPv6-literal host and against a hostname's resolved addresses, including the port-qualified CIDR form. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): cap each proxied CONNECT attempt within the shared budget A proxy that accepted the first CONNECT request but never responded consumed the entire aggregate handshake timeout, so later validated addresses were never tried and the hang defeated the multi-address fallback. Each attempt is now time-boxed to its fair share of the time remaining before the shared deadline (remaining / attempts_left): a hanging attempt is cut off with enough budget left for every remaining address, while time a fast failure does not use rolls over to later attempts and the total never exceeds CONNECT_HANDSHAKE_TIMEOUT. A timed-out attempt is recorded like any other failure, and the aggregate error distinguishes all-attempted from budget-exhausted runs. Adds a first-hangs/second-succeeds regression driven through a test-visible budget parameter so it runs in about a second instead of a real 30s window. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): deliver corporate proxy config on the supervisor argv The proxy settings were injected as reserved OPENSHELL_UPSTREAM_* container environment variables. The driver only wrote names the operator configured, but container runtimes layer the spec environment over ENV values baked into the sandbox image, so an image could supply NO_PROXY=*, enable hostname CONNECT, or point an unconfigured deployment at an attacker-controlled proxy whenever the operator left a field unset. The settings now travel as supervisor command-line arguments (--upstream-proxy, --upstream-no-proxy, --upstream-proxy-auth-file, --upstream-proxy-auth-allow-insecure, --upstream-proxy-connect-by-hostname) built by the driver from operator config. The driver sets the container entrypoint and command explicitly, so neither sandbox spec/template environment nor image ENV can influence argv, and an omitted flag genuinely means unconfigured — in every supervisor topology, since the supervisor no longer consults its environment for these settings at all. Credentials stay on the root-only secret mount; only the mount path appears on argv. The reserved environment names, their strip-list entries, and the env-based validation surface are removed. UpstreamProxyConfig::from_args replaces from_env, reusing the same shared fail-closed validation and pairing rules keyed by the CLI flag names. Driver tests now assert the argv contract, including that sandbox-supplied environment cannot add, remove, or redirect proxy flags; supervisor tests cover from_args mapping and its pairing rules. Signed-off-by: Philippe Martin <phmartin@redhat.com> * docs(sandbox): align proxy comments with the argv transport The argv migration left comments describing the configuration as reserved environment variables ("reserved value", "present-but-empty variable", "reserved upstream proxy variables"). Rephrase them as driver-supplied arguments and operator settings so the documented trust boundary matches the implementation. Comment-only change. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox): parse bracketed IPv6 authorities in client CONNECT targets parse_target split the CONNECT authority at the first colon, so an IPv6-literal target like [2001:db8::1]:443 always failed port parsing and IPv6-literal clients could never reach policy evaluation; a regression test even locked in that failure. Parse the RFC 3986 bracketed form and return the host bracket-free, matching what DNS resolution, SSRF validation, NO_PROXY matching, and the upstream CONNECT builder expect. Unclosed brackets, a missing or empty port after the bracket, and non-numeric ports are rejected; unbracketed behavior is unchanged. Replaces the failure-locking test with success coverage for bracketed targets and adds malformed-bracket rejection cases. Signed-off-by: Philippe Martin <phmartin@redhat.com> * test(podman): cover proxy-auth secret cleanup across lifecycle failures The per-sandbox proxy-auth credential secret is staged before the container is created and removed on cleanup, but no test proved the cleanup paths actually issue the secret removal. Add Podman-stub tests that drive create_sandbox to a container-create failure and to a start failure, and delete_sandbox for an out-of-band deletion, asserting each path issues the DELETE for the per-sandbox proxy-auth secret so a credential can never outlive the sandbox that owned it. Signed-off-by: Philippe Martin <phmartin@redhat.com> * docs: list corporate proxy keys in the Podman compute-driver overview The Fern Podman driver section enumerated its gateway.toml keys but omitted the corporate egress proxy settings. Add https_proxy, no_proxy, proxy_auth_file, proxy_auth_allow_insecure, and proxy_connect_by_hostname with a pointer to the gateway configuration reference for the full contract. No navigation change: the reference folder already includes the gateway configuration page. Signed-off-by: Philippe Martin <phmartin@redhat.com> * test(sandbox): cover the SSRF-to-TLS composition across the proxy tunnel Existing tests exercised validated-IP CONNECT and the upstream-TLS helper independently, but not the full boundary. Add an end-to-end regression that stands up a fake corporate proxy tunneling to a fake TLS server and drives the real path: connect_via_validated CONNECTs to the validated address, the proxy splices the tunnel, and tls_connect_upstream verifies the upstream certificate against the original hostname carried in SNI. It asserts the CONNECT authority is the validated IP and never the hostname, that verification succeeds for the matching hostname, and that a mismatched hostname is rejected — proving a rebinding or split-horizon substitution behind the proxy cannot pass certificate verification. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox,podman): bound proxy-auth reads, reject port 0, fix stale comment Three review findings: - CWE-400: the proxy-auth credential file was read with an unbounded read_to_string on both the driver (sandbox-create) and supervisor (startup) paths, so a huge file or a special file such as /dev/zero could exhaust memory. Add a shared bounded reader in openshell-core that rejects non-regular files, caps the size at 4 KiB, and reads at most that many bytes; the driver runs it via spawn_blocking. Covers oversized, special-file, and missing-path cases on both sides. - Reject an upstream proxy URL with port 0: it passed the explicit-port check and startup validation but is not a connectable TCP port, so every proxied dial would fail later. Add a typed ZeroPort error with shared-validator and Podman-config tests. - Reword a driver-config comment that still described a 'reserved variable' to match the argv transport. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(sandbox): open proxy-auth file non-blocking to reject FIFOs promptly read_upstream_proxy_credential_file opened the path with a blocking File::open before the regular-file check, so a configured FIFO with no writer would block open() indefinitely — hanging sandbox creation on the driver and supervisor startup. Open with O_NONBLOCK on Unix so the open returns immediately, then reject the non-regular file as before; O_NONBLOCK has no effect on the later read of a regular file. Adds a mkfifo regression asserting the reader returns promptly with a non-regular-file error instead of hanging. Signed-off-by: Philippe Martin <phmartin@redhat.com> * test(podman): cover corporate proxy egress across driver and supervisor The existing corporate-proxy tests construct config structs or call CONNECT helpers directly, so none of them detect a break in the wiring between layers: gateway TOML deserialization, the Podman argv and secret-mount semantics, supervisor CLI parsing, or policy denial before proxy contact. Add a Podman e2e that drives the whole chain against a fake authenticated forward proxy and asserts that an approved TLS request traverses it with a validated-IP CONNECT, a policy-denied destination is refused with 403 without ever reaching the proxy, credentials arrive through the mounted per-sandbox secret, and deleting the sandbox removes that secret. SupportContainer is a new harness fixture. Unlike ContainerHttpServer it probes readiness with a TCP connect rather than an HTTP GET, so it can host a forward proxy and TLS servers, and it exposes container logs and network IP for assertions. Signed-off-by: Philippe Martin <phmartin@redhat.com> * test(podman): restart the gateway on the proxy-config panic path The panic cleanup for the temporary corporate proxy configuration restored the gateway TOML but left the gateway process running with the temporary configuration still loaded, which could poison later test binaries in the same run. Nothing restarted it: the only ManagedGateway is the short-lived one inside restart_gateway, and its Drop only calls start, which does not reload config for an already-running gateway. Restore and synchronously stop/start the gateway in Drop, and set restored only after the normal restore and restart both succeed so a failed restart no longer suppresses the fallback. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(supervisor-network): derive the loopback proxy bypass from resolved IPs The automatic bypass treated the host string "localhost" as proof that the destination was loopback and returned every resolved address. A sandbox controls its own /etc/hosts and resolve_socket_addrs consults it before DNS, so a workload could map localhost to any policy-allowed address and dial it directly, escaping the operator proxy and the inspection and audit boundary it exists to provide. Check the resolved addresses instead: the name bypasses only when the resolution is non-empty and every address is loopback. A mixed answer is not partially honored, and an IP literal is still authoritative for itself. A spoofed localhost falls through to the entries below, so an explicit operator NO_PROXY entry is still honored. This matches the trust model detect_trusted_host_gateway already applies to the same hosts file, which validates the mapped address rather than the alias. Signed-off-by: Philippe Martin <phmartin@redhat.com> * fix(podman): clean up proxy-auth secret when container already deleted The delete_sandbox early-return path cleaned up the token secret but skipped the proxy-auth secret, leaking it on disk. Also update tests for recent API changes (Optional socket_path, workspace field, list-based container lookup). Signed-off-by: Philippe Martin <philippe@openshell.dev> Signed-off-by: Philippe Martin <phmartin@redhat.com> --------- Signed-off-by: Philippe Martin <phmartin@redhat.com> Signed-off-by: Philippe Martin <philippe@openshell.dev> |
||
|
Jim Meyer
|
745512e325 |
fix(build): raise open-file limit for host musl cross-compile on macOS (#2307)
Closes #2112 The host `cargo zigbuild` for `*-unknown-linux-musl` opens ~333 `.rlib` files at once during the static link, exceeding macOS's default soft limit of 256 and failing with `ProcessFdQuotaExceeded`. This blocked the docker/podman `mise run gateway` paths for macOS contributors; only the VM driver path guarded against it. Extract the VM path's `ensure_build_nofile_limit` guard into a shared `tasks/scripts/build-env.sh` and call it from the host-staging chokepoint (`stage-prebuilt-binaries.sh`), fixing docker, podman, and all docker:*/multiarch host cross-compiles at once. De-duplicate the VM script to source the shared helper. The guard is a no-op on Linux and when cargo-zigbuild is absent, so CI and Linux dev are unaffected. The limit is read from `OPENSHELL_BUILD_NOFILE_LIMIT` (default 8192), honoring the legacy `OPENSHELL_VM_BUILD_NOFILE_LIMIT` for back-compat. Also correct the stale comment in gateway-docker.sh (the cross-compile runs on the host, not inside Linux containers) and document the guard in architecture/build.md. Adds tasks/scripts/test-build-env.sh, wired into `mise run test` via `test:build-env`. Signed-off-by: Jim Meyer <jim@meyer4hire.com> |
||
|
Ignas Baranauskas
|
a72711697d |
chore: remove deprecated --keep flag from docs, scripts, and e2e tests (#2126)
* docs: remove deprecated --keep flag from tutorials and examples The --keep flag is deprecated, hidden, and a no-op since sandboxes are kept by default. Remove references from tutorial docs and example READMEs that explain it as a real feature. - Remove --keep from sandbox create commands - Remove --keep explanation text - Clarify that sandboxes are kept by default Signed-off-by: Ignas Baranauskas <ibaranau@redhat.com> * chore: remove deprecated --keep usage from scripts and e2e tests The --keep flag is a deprecated no-op since sandboxes are kept by default. Stop passing it in internal scripts, e2e test scripts, and example demo scripts. Signed-off-by: Ignas Baranauskas <ibaranau@redhat.com> --------- Signed-off-by: Ignas Baranauskas <ibaranau@redhat.com> |
||
|
Evan Lezar
|
a226806026 |
test(e2e): run gpu workloads from manifest (#1709)
* test(e2e): add workload manifest build flow Signed-off-by: Evan Lezar <elezar@nvidia.com> * test(e2e): add gpu workload validation tests Signed-off-by: Evan Lezar <elezar@nvidia.com> * ci(e2e): build gpu workloads before gpu e2e Signed-off-by: Evan Lezar <elezar@nvidia.com> --------- Signed-off-by: Evan Lezar <elezar@nvidia.com> |
||
|
Taylor Mutch
|
7e0cce405d |
fix(build): use zig archive tools for cross builds (#2014)
Signed-off-by: Taylor Mutch <taylormutch@gmail.com> |
||
|
Taylor Mutch
|
ba21bb32a2 |
feat(kubernetes): support agent-sandbox v1beta1 (#2009)
* feat(kubernetes): support agent-sandbox v1beta1 Signed-off-by: Taylor Mutch <taylormutch@gmail.com> * ci(kubernetes): test agent-sandbox api versions Signed-off-by: Taylor Mutch <taylormutch@gmail.com> * fix(kubernetes): retry agent-sandbox raw 404s Signed-off-by: Taylor Mutch <taylormutch@gmail.com> * fix(kubernetes): harden agent-sandbox api setup Signed-off-by: Taylor Mutch <taylormutch@gmail.com> * fix(kubernetes): cache agent-sandbox api version Signed-off-by: Taylor Mutch <taylormutch@gmail.com> * docs(kubernetes): document agent-sandbox upgrade behavior Signed-off-by: Taylor Mutch <taylormutch@gmail.com> * refactor(kubernetes): collapse agent-sandbox api selection Signed-off-by: Taylor Mutch <taylormutch@gmail.com> --------- Signed-off-by: Taylor Mutch <taylormutch@gmail.com> |
||
|
Simon Scatton
|
c7879a0a1d |
test(e2e): stop using custom e2e binary builds (#2000)
Remove the production-crate dev-settings feature so e2e tests build the same gateway, CLI, and supervisor binaries that we ship. Move settings coverage from dummy test-only keys to the production ocsf_json_enabled setting. Signed-off-by: Simon Scatton <sscatton@nvidia.com> |
||
|
Piotr Mlocek
|
82d03f1c98 | fix(linux): lower host glibc floor to 2.28 to support RHEL/Rocky 8 (#1934) | ||
|
Evan Lezar
|
b6428cb9cd |
fix(build): align container engine selection (#1944)
Signed-off-by: Evan Lezar <elezar@nvidia.com> |
||
|
Taylor Mutch
|
62aa5e3247 |
ci(branch-checks): align Python checks with pre-commit (#1908)
Signed-off-by: Taylor Mutch <taylormutch@gmail.com> |
||
|
Evan Lezar
|
8c0153438c |
test(e2e): add GPU workload image artifacts (#1484)
* test(e2e): add GPU workload image artifacts Closes #1476 Define local GPU workload image sources for smoke-pass, smoke-fail, and cuda-basic validation, plus a mise build task that tags images with the source revision and records the latest local image refs. Signed-off-by: Evan Lezar <elezar@nvidia.com> * test(e2e): add workload manifest build flow Signed-off-by: Evan Lezar <elezar@nvidia.com> * test(e2e): fix gpu workload README lint Signed-off-by: Evan Lezar <elezar@nvidia.com> * test(e2e): address gpu workload review feedback Signed-off-by: Evan Lezar <elezar@nvidia.com> --------- Signed-off-by: Evan Lezar <elezar@nvidia.com> |
||
|
Piotr Mlocek
|
6c8cf38b63 |
ci(docs): add docs website automation (#1788)
* ci(docs): add docs website automation * ci(docs): harden sync-docs inputs and add script tests Route free-form workflow_dispatch inputs through env vars to avoid shell injection, switch uv install to the approved setup-uv action, and add a pytest suite for sync_docs_website.py wired into the test task. |
||
|
Evan Lezar
|
9e805dc35d |
fix(build): use zigbuild for musl supervisor staging (#1850)
Signed-off-by: Evan Lezar <elezar@nvidia.com> |
||
|
Russell Bryant
|
3aba30c36f |
feat(telemetry): add build-time option to compile out telemetry (#1845)
* feat(telemetry): add build-time option to compile out telemetry Gate anonymous telemetry emission behind a default-on `telemetry` Cargo feature in openshell-core. The data model (enums, validation, emit_*/enabled* signatures) stays always-compiled, while the endpoint, HTTP client, queue, and emission code are feature-gated. With the feature off, enabled() returns false and emit_* are no-ops, so dependent crates compile unchanged and no telemetry endpoint, HTTP client, or emission code is included in the binary. chrono and reqwest become optional dependencies of openshell-core, dropped from its dependency graph when telemetry is disabled. Thread the switch through the workspace: every crate depends on openshell-core with default-features = false, and the default-on `telemetry` passthrough lives on the binary crates that emit or collect telemetry (openshell-server, openshell-sandbox, openshell-driver-vm). In-process drivers inherit it via resolver v2 feature unification. Build a telemetry-free binary with, e.g.: cargo build --release -p openshell-server --no-default-features The runtime OPENSHELL_TELEMETRY_ENABLED switch is unchanged for default builds. Signed-off-by: Russell Bryant <russell.bryant@gmail.com> * ci(telemetry): guard that telemetry can be compiled out Add tasks/scripts/verify-telemetry-compiled-out.sh, which inspects a built binary for telemetry markers (the telemetry endpoint host and client ID) that exist only when emission code is compiled in. The rust:verify:telemetry-off mise task builds the gateway with default features (positive control: markers must be present, so the absent checks can never be silently vacuous) and with --no-default-features (markers must be absent), and checks the --no-default-features sandbox binary as well. Wire the task into the Rust branch-checks job so a regression that reintroduces telemetry code into a --no-default-features build fails CI. Signed-off-by: Russell Bryant <russell.bryant@gmail.com> --------- Signed-off-by: Russell Bryant <russell.bryant@gmail.com> |
||
|
Zygmunt Krynicki
|
4025894a94 | chore(snap): remove early snap packaging (#1648) | ||
|
Roshni Malani
|
586c385bdf |
chore(k8s): use upstream agent-sandbox manifest in CI/e2e (#1657)
Drop the vendored deploy/kube/manifests/agent-sandbox.yaml. CI and e2e scripts now apply manifest.yaml directly from github.com/kubernetes-sigs/agent-sandbox releases, pinned via AGENT_SANDBOX_VERSION env (default v0.4.6, overridable) so internal runs stay reproducible. Public docs and the helm chart README continue to reference /releases/latest/download/ — no OpenShell/agent-sandbox support matrix exists yet (see #1649). Adds an air-gap note in docs/kubernetes/setup.mdx enumerating the manifest and image operators need to mirror to an internal registry. Signed-off-by: Roshni Malani <rmalani@nvidia.com> |
||
|
Piotr Mlocek
|
3d441e73ac |
fix(config): reject unknown fields in nested gateway config tables (#1666)
* fix(config): reject unknown fields in nested gateway config tables The gateway TOML loader silently ignored keys placed under the wrong table header. PR #1661 fixed one instance of this (supervisor_image under [openshell.gateway.gateway_jwt]) but the root cause remained: the nested gateway config tables did not deny unknown fields, so a misplaced key was accepted and dropped instead of erroring. Concretely, tasks/scripts/gateway.sh emitted `sandbox_namespace` right after the [openshell.gateway.gateway_jwt] heredoc, so it landed inside the gateway_jwt table rather than [openshell.gateway]. The k8s driver already receives the namespace via [openshell.drivers.kubernetes], so the stray line was dead config that parsed without complaint. Changes: - Add #[serde(deny_unknown_fields)] to the nested gateway config tables that are part of the config-file parse tree: TlsConfig, OidcConfig, MtlsAuthConfig, GatewayAuthConfig, GatewayJwtConfig. - Remove the misplaced sandbox_namespace line from gateway.sh. - Drop the unused Serialize/Deserialize derives from Config and ServiceRoutingConfig (see below). - Add a regression test asserting a key under the wrong nested table is rejected. |
||
|
Jesse Jaggars
|
019a986e6d |
fix(gateway): place supervisor_image under podman driver TOML table (#1661)
The gateway.sh script appended supervisor_image after the [openshell.gateway.gateway_jwt] table header, so TOML parsed it as a gateway_jwt field. The Podman driver never saw the override and fell back to the default ghcr.io/nvidia/openshell/supervisor:latest. Move supervisor_image into [openshell.drivers.podman] where the driver config deserializer expects it. |
||
|
Taylor Mutch
|
7cea9d9b2b |
fix(gateway): align package TLS bootstrap path (#1601)
* fix(gateway): align package TLS bootstrap path Closes #1593 Default package-managed gateway services to a stable local TLS directory and use that same value for certificate generation and runtime startup. Signed-off-by: Taylor Mutch <taylormutch@gmail.com> * test(packaging): validate package asset paths exist Signed-off-by: Taylor Mutch <taylormutch@gmail.com> * ci(e2e): pin mise in kubernetes job Signed-off-by: Taylor Mutch <taylormutch@gmail.com> --------- Signed-off-by: Taylor Mutch <taylormutch@gmail.com> |
||
|
Alexander Watson
|
e98ea3ee93 | feat(policy): add agentic approval loop (#1528) | ||
|
Naveen Malik
|
f1ed347a40 |
fix(driver-podman): bind gateway to 0.0.0.0 in rootless mode (#1623)
Rootless Podman sandbox containers reach the host through pasta's local connection bypass, which translates L2 frames to L4 host sockets. The dev gateway script binds to 127.0.0.1 by default, which is not routable through pasta. Auto-detect rootless mode and bind to 0.0.0.0 so sandbox containers can connect to the gateway. - Auto-detect rootless Podman in gateway.sh and export OPENSHELL_BIND_ADDRESS=0.0.0.0 when not explicitly set - Add e2e:podman:rootless mise task and CI matrix entry to validate rootless Podman networking end-to-end - CI creates a non-root user inside the privileged container to trigger Podman's rootless code paths (pasta, user namespace isolation) Signed-off-by: Naveen Malik <nmalik@redhat.com> |
||
|
krishicks
|
fa84e437ab |
fix(gateway): configure local dev auth (#1575)
This makes it so you can run the dev gateway and sandbox with: ``` mise run gateway # in another shell mise run sandbox ``` Signed-off-by: Kris Hicks <khicks@nvidia.com> |
||
|
Mesut Oezdil
|
88508a01be |
fix(scripts): replace mapfile with bash 3.2-compatible read loop in helm-k3s-local (#1539)
macOS ships bash 3.2 which lacks mapfile/readarray. Replace all three occurrences in configure_ghcr_credentials, cluster_has_image, and cluster_image_platform with a portable while-read loop, consistent with the fix applied to docker-build-image.sh in #1334. |
||
|
Taylor Mutch
|
7d38aa8b8e |
fix(homebrew): repair local driver bootstrap state (#1527)
* fix(homebrew): repair local driver bootstrap state * fix(bootstrap): satisfy default SAN doc lint |
||
|
Piotr Mlocek
|
0dc08a185e | fix(release): build host Linux binaries with glibc floor (#1490) | ||
|
Drew Newberry
|
603b3e27fa | docs: update NemoClaw/OpenClaw references (#1529) | ||
|
Taylor Mutch
|
a3b16c18ab | feat(auth): per-sandbox authentication to gateway (#1404) | ||
|
Florent BENOIT
|
b93a3d8661 | fix(scripts): use portable lowercase in normalize_bool for Bash 3.2 (#1493) | ||
|
John T. Myers
|
dbba580e84 |
fix(security): refresh CI and gateway image dependencies (#1432)
Refresh the CI image tool pins so Go-built tools are rebuilt with patched Go releases and move the sandbox Python runtime to 3.14.5. Rebase the gateway runtime to a pinned distroless Debian 13 image with glibc 2.41-12+deb13u3 while preserving the existing UID/GID 1000 runtime identity for upgrade compatibility. Update rustls-webpki to 0.103.13 and clarify Linux k3d guidance now that k3d is not installed through mise on Linux. Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com> |
||
|
Drew Newberry
|
f257ed0193 |
refactor(packaging): rely on gateway runtime defaults (#1415)
* fix(packaging): use gateway TOML config in packages * refactor(packaging): rely on gateway runtime defaults |
||
|
Evan Lezar
|
555680cefc |
fix(docker): fall back to host arch for local builds (#1420)
(cherry picked from commit 392ab51f58fc18e03e721cdfa21f666091c67b80) Signed-off-by: Evan Lezar <elezar@nvidia.com> |
||
|
Drew Newberry
|
f819f7dcb5 | fix(vm): restore sandboxes after gateway restart (#1407) | ||
|
Drew Newberry
|
910d3f09f2 |
feat(vm): boot sandboxes from ext4 root disks (#1263)
* feat(vm): boot sandboxes from ext4 root disks * feat(vm): boot sandboxes with writable overlay disk * fix(sandbox): improve sandbox provisioning behavior * fix(vm): repair sandbox image provisioning * fix(vm): clean up rebase fallout * perf(vm): cache empty overlay images * perf(vm): skip repeated sandbox ownership repair * fix(vm): address lifecycle review issues * fix(vm): repair gateway vm startup * wip |
||
|
Taylor MutchandDrew Newberry
|
b61a98dbad |
feat(gateway): add TOML configuration file (RFC 0003) (#1317)
* feat(gateway): add TOML configuration file (RFC 0003)
Introduces an opt-in --config / OPENSHELL_GATEWAY_CONFIG flag that loads a
TOML file with gateway-wide settings and per-driver tables. Source
precedence is CLI > env > file > built-in default, implemented via clap's
ValueSource so existing flags and env vars keep their priority.
Driver crates (kubernetes, docker, podman, vm) now derive Deserialize on
their config structs. SupervisorSideloadMethod gains Deserialize with
kebab-case rename. A per-driver inheritance allowlist on the loader side
overlays [openshell.gateway] shared defaults (default_image,
supervisor_image, image_pull_policy, guest_tls_*, ssh_handshake_skew_secs,
client_tls_secret_name, host_gateway_ip, enable_user_namespaces) onto
each [openshell.drivers.<name>] table before deserialization.
The Helm chart renders a new gateway-config ConfigMap and mounts it at
/etc/openshell/gateway.toml. The migrated OPENSHELL_* env entries are
dropped from the StatefulSet — only the Secret-backed
OPENSHELL_SSH_HANDSHAKE_SECRET remains. database_url stays on --db-url.
Adds examples/gateway/gateway.example.toml and updates architecture/gateway.md
with the source precedence and inheritance rules.
* docs(gateway): drop ssh_handshake_skew_secs and ssh_handshake_secret from examples
Both fields are scheduled for removal. Remove the example values and the
env-only note so the gateway.toml example and the architecture doc stop
recommending settings that will not exist much longer.
* docs(rfc): correct OPENSHELL_CONFIG to OPENSHELL_GATEWAY_CONFIG in RFC 0003
* docs(gateway): add per-driver TOML example configurations
Adds focused single-driver examples next to the comprehensive
gateway.example.toml: kubernetes, docker, podman, and microvm. Each one
demonstrates the realistic settings for that driver plus how shared
[openshell.gateway] defaults inherit into the driver table.
A new unit test (`checked_in_examples_parse`) loads every example through
the config_file loader so schema drift fails CI rather than silently
shipping a broken example.
* refactor(gateway): drop image_pull_policy from shared inheritance
Kubernetes and Podman use mutually-incompatible vocabularies for the same
TOML key:
- Kubernetes: `Always | IfNotPresent | Never` (free-form string passed
verbatim to the K8s API).
- Podman: `always | missing | never | newer` (strict lowercase enum
deserialised into `ImagePullPolicy`).
No value means the same thing in both drivers. Sharing the key at
`[openshell.gateway]` scope and inheriting it into every active driver's
table meant any value safe for one driver was either wrong or silently
dropped for the other (`IfNotPresent` → `ImagePullPolicy::Missing` after
`.unwrap_or_default()`). Operators run one driver per gateway, so the
"shared default" never pays for itself.
Make `image_pull_policy` driver-local:
- Remove the field from `GatewayFileSection` and from
`inheritable_keys()` for both Kubernetes and Podman.
- Drop the file→`RunArgs` merge for the gateway-scope key.
- Stop unconditionally clobbering the driver value with
`config.sandbox_image_pull_policy` in the runtime wiring — only apply
the CLI/env override when it was set (and, for Podman, only when it
parses into the lowercase enum).
- Move the key under `[openshell.drivers.kubernetes]` and
`[openshell.drivers.podman]` in every example, the RFC, the
architecture doc, and the Helm-rendered gateway ConfigMap.
The supervisor pull policy follows the same shape: it is K8s-only and
moves into `[openshell.drivers.kubernetes]` alongside `image_pull_policy`
in the Helm template.
* fix(gateway): address review feedback on TOML configuration
Resolves the P1 and P2 issues raised in PR #1317:
- Helm gateway ConfigMap moves `grpc_endpoint` under
`[openshell.drivers.kubernetes]` so the default install no longer fails
the gateway's `deny_unknown_fields` schema check.
- `kubernetes_config_from_file` and `podman_config_from_file` only let
the gateway-wide CLI/env `grpc_endpoint` overwrite the driver-table
value when it was actually supplied, preserving file-only configs.
- Kubernetes driver default `image_pull_policy` is now empty (was Podman
vocabulary "missing"), so default deployments let the Kubernetes API
apply its own policy instead of being rejected.
- New `disable_tls` gateway field plumbs `.Values.server.disableTls`
through the TOML ConfigMap instead of relying on env vars dropped from
the StatefulSet.
- StatefulSet pod template now carries a `checksum/gateway-config`
annotation so `helm upgrade` rolls pods when the ConfigMap changes.
- Auxiliary listener resolution preserves the full `SocketAddr` from
`health_bind_address` / `metrics_bind_address`, so a loopback-pinned
health port is not silently relocated onto the public bind address.
- `ssh_session_ttl_secs` from the file is now applied to `Config` (it
was previously accepted by the loader but never read).
New regression coverage: cli-level merge tests for the new fields plus
helm-unittest assertions for the ConfigMap shape, checksum annotation,
and `disable_tls` rendering.
* docs(gateway): consolidate gateway TOML examples into docs reference
Replaces the per-driver example files under examples/gateway/ with a
single published reference page at docs/reference/gateway-config.mdx
covering source precedence, layout, the full example, and the four
per-driver examples (Kubernetes, Docker, Podman, microVM). Drew flagged
during PR #1317 review that the examples belong with the user-facing
docs rather than in a sibling examples/ directory.
The cross-references in architecture/gateway.md and RFC 0003 are updated
to point at the new docs page; the round-trip test in config_file.rs is
removed (schema coverage stays on the inline parses_full_example test
and per-field merge tests — doc-snippet drift belongs in a separate
docs-lint, not in a cross-tree Rust unit test).
* refactor(core): move DEFAULT_K8S_NAMESPACE into K8s driver
The constant is Kubernetes-specific (used only by KubernetesComputeConfig's
Default impl) and does not belong in openshell-core. Relocate it to the
driver crate that owns the K8s vocabulary; openshell-core retains only
truly cross-cutting defaults.
* refactor(core): move Podman bridge default into Podman driver
DEFAULT_NETWORK_NAME is Podman vocabulary, consumed only by the Podman
driver. Also drops the unused DEFAULT_IMAGE_PULL_POLICY constant.
* docs(auth): scrub remaining SSH handshake secret references
Sweeps the trailing mentions left after the rebase: the gateway
config-file module doc, the Helm gateway-config ConfigMap header,
the gateway-config.mdx env-only note, and the RPM systemd unit
comment for init-gateway-env.sh.
* docs(gateway): clarify OPENSHELL_GRPC_ENDPOINT applies to all drivers
The previous comment implied the callback endpoint was Kubernetes-only,
but the value is propagated to every compute driver (Kubernetes, Docker,
Podman, VM) and must be reachable from wherever the sandbox runs.
* refactor(gateway): move driver options into config (#1394)
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
* fix(e2e): regenerate gateway config via TOML for docker + podman harnesses
The gateway CLI flags moved into TOML config tables in 560550d2 (#1394),
which made every existing e2e/with-{docker,podman}-gateway.sh invocation
fail with "unexpected argument '--sandbox-namespace'" (and a long tail
of similar driver-specific options) before the gateway could even bind.
Replace the obsolete CLI flags with a synthesized
`[openshell.drivers.<driver>]` table written to `${STATE_DIR}/gateway.toml`
and passed via the new `--config` flag. Only the gateway-wide flags
that survived 560550d2 (bind-address, port, drivers, db-url, tls-*,
disable-tls, log-level, health-port) stay on the command line.
Both scripts get a small `toml_string` helper to properly TOML-quote
the values (the previous `%q` printf format produced bash-escape, not
TOML-escape). The Docker harness also corrects two field names that
diverged from the driver schema: `docker_network_name` →
`network_name`, and the supervisor binary/image plumbing now reads
through to `supervisor_bin` / `supervisor_image` in the same table.
The Podman harness drops `--ssh-gateway-port` (deleted in 560550d2 —
gRPC + SSH are multiplexed on the same port now) and substitutes
`network_name` + `gateway_port` for the obsolete `--sandbox-namespace`
(which the Podman driver never had as a typed field).
* fix(core): swap bind-only 0.0.0.0 SSH gateway host for cluster URL host
CLI's resolve_ssh_gateway treated 0.0.0.0 as a loopback "keep as-is"
when the cluster URL was also loopback, so the SSH proxy connected to
0.0.0.0:port. The unspecified address is never a valid connect target
and is not present in any TLS cert SAN, which produced BadCertificate
TLS handshake failures during `openshell sandbox create -- ...` in
docker/podman e2e (e.g. bypass_detection).
Resolution: when the server returns 0.0.0.0 or :: as the gateway host
and both endpoints are loopback, fall back to the cluster URL's host
(which the CLI is already using to reach the gateway, so it must
resolve and match the cert).
* fix(e2e): repair podman harness on macOS
Podman 5.x with the applehv/libkrun provider no longer creates the legacy
~/.local/share/containers/podman/machine/podman.sock symlink, and
`podman system service` is a Linux-only subcommand — the macOS client
delegates the API service to the VM. Both assumptions in the harness
were stale, so the script tried to start a temporary service that podman
rejected with "unknown flag: --time".
- Discover the macOS socket via `podman machine inspect` instead of the
hardcoded path.
- On Darwin, fail fast with a "start podman machine" message rather than
attempting the Linux-only `podman system service` fallback.
- Write socket_path into [openshell.drivers.podman] so the in-process
driver picks up the discovered socket; the driver reads TOML only
after the config refactor (560550d2), so OPENSHELL_PODMAN_SOCKET alone
was no longer enough.
* fix(server): use clone_from for TLS client CA assignment
clippy 1.95.0 rejects assigning the result of `Clone::clone()` to an
existing variable under `-D warnings` (`assigning_clones`). Switch to
`clone_from(&...)` to satisfy the lint and avoid the redundant
allocation.
---------
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
Co-authored-by: Drew Newberry <anewberry@nvidia.com>
|
||
|
Florent BENOIT
|
a1fb9bd954 |
fix(scripts): replace mapfile with bash 3.2-compatible read loop (#1334)
macOS ships bash 3.2 which lacks mapfile (introduced in bash 4.0), causing docker-build-image.sh and stage-prebuilt-binaries.sh to fail on stock macOS developer machines. |
||
|
Taylor Mutch
|
7a0c444445 |
refactor!(auth): drop SSH handshake secret (#1274)
* refactor!(auth): drop SSH handshake secret in favor of mTLS The OPENSHELL_SSH_HANDSHAKE_SECRET / x-sandbox-secret mechanism was misnamed: it does not authenticate SSH (which flows over the RelayStream gRPC RPC and is gated by mTLS plus supervisor Unix-socket permissions). It only gated a small set of sandbox-to-gateway control-plane RPCs, and production deployments already enforce mTLS on that channel — so the shared secret was redundant. Replace the secret check with an mTLS-presence marker. Sandbox-class methods (ReportPolicyStatus, PushSandboxLogs, GetSandboxProviderEnvironment, SubmitPolicyAnalysis, GetSandboxConfig, GetInferenceBundle) accept callers without a Bearer token; the gRPC mTLS handshake is the trust boundary. Dual-auth methods treat Bearer-present as full-scope CLI access and Bearer-absent as sandbox-restricted scope via validate_sandbox_caller_update. Drops the secret from all drivers (K8s, Podman, VM), the sandbox gRPC interceptor, the Helm chart (values + pre-install hook + StatefulSet env), the RPM bootstrap script, the man pages, and the debug-openshell-cluster skill. Also removes the never-read ssh_handshake_skew_secs flag and config field. BREAKING CHANGE: --ssh-handshake-secret / OPENSHELL_SSH_HANDSHAKE_SECRET and --ssh-handshake-skew-secs / OPENSHELL_SSH_HANDSHAKE_SKEW_SECS are removed from the gateway, sandbox, and all driver binaries. The openshell-ssh-handshake K8s Secret is no longer managed by the chart; operators may delete the orphan. Deployments using --disable-gateway-auth must enforce caller authentication at the fronting proxy, since the gateway no longer validates a per-request secret on sandbox-class methods. Refs OS-174. * docs(auth): scrub residual SSH handshake secret references Sweep across docs, e2e scripts, the Podman driver README/NETWORKING notes, the RPM/Helm/setup guides, the gateway man page, and RFC 0003 to remove instructions and examples that still referenced OPENSHELL_SSH_HANDSHAKE_SECRET / --ssh-handshake-secret / ssh_handshake_skew_secs. The mechanism is gone; nothing should still suggest setting it. Negative-assertion regression tests are kept so the env var cannot silently be re-introduced. * test(drivers): drop SSH handshake secret negative-assertion tests The supporting code, env vars, CLI flags, and config plumbing are gone — these tests asserted absence of strings that no longer have any path to being set. Remove the guards from the Docker, Podman, Kubernetes, and VM driver test modules. |
||
|
Drew Newberry
|
0471c6d2ab | fix(gateway): keep vm driver opt-in (#1375) | ||
|
Taylor MutchandDrew Newberry
|
ba77967fc4 |
refactor(docker): split gateway/supervisor Dockerfiles and use native rust builds (#1316)
* wip
* refactor(docker): use native rust builds for split gateway/supervisor images
Drop the in-Docker BUILD_FROM_SOURCE path so both images consume only
prebuilt binaries staged natively via tasks/scripts/stage-prebuilt-binaries.sh.
This mirrors what CI does and reuses the host's cargo target cache and
sccache across rebuilds.
- Dockerfile.gateway: nvcr.io/nvidia/distroless/cc:v4.0.4 base (the 4.0.0
tag does not exist on nvcr.io; the registry uses a v prefix). GNU-linked
binary copied to /usr/local/bin.
- Dockerfile.supervisor: scratch base, static musl binary. Static linkage
lets the image stay scratch while still being executable as a Kubernetes
init container.
- skaffold.yaml: each artifact invokes tasks/scripts/docker-build-image.sh,
which stages the binary natively (cargo / cargo-zigbuild) and then builds
the image. Drops the cross-build.sh dependency from the supervisor build.
- seccomp.rs: add a local SYS_kexec_file_load constant for musl/aarch64.
libc 0.2.185 omits the symbol from its musl/aarch64 bindings, so the
supervisor's seccomp filter previously failed to compile for that target.
- architecture/build.md: describe the native-first pipeline and per-image
runtime choices.
Local validation: gateway image 101MB (was 194MB), supervisor image 21.7MB.
helm:skaffold:run deploys cleanly; the static musl supervisor binary runs
correctly in a non-glibc agent container.
* refactor(docker): tighten binary perms via --chown + 0550
Replace `COPY --chmod=755` with `COPY --chown=<user> --chmod=0550` in
the gateway and supervisor Dockerfiles. The binary is no longer
world-readable or world-executable; ownership is pinned to the runtime
user.
- Gateway uses `--chown=nvs:nvs` + `USER nvs:nvs`, matching the only
non-root user defined in `nvcr.io/nvidia/distroless/cc` (UID 1000) and
the Helm chart's `securityContext.runAsUser: 1000`, which overrides
the Dockerfile USER at runtime.
- Supervisor uses numeric `--chown=65534:65534` because the scratch base
has no `/etc/passwd` for name resolution. The supervisor image is
only consumed by the init-container copy-self path; the destination
pod's runAsUser governs execute access.
Validated by deploying to a local k3d cluster via `helm:skaffold:run`
and confirming the gateway StatefulSet reaches 1/1 Running.
* ci(gpu): repoint GPU probe image lookup at Dockerfile.gateway
The previous awk parsed `FROM <image> AS gateway` from the now-deleted
`Dockerfile.images`. The new `Dockerfile.gateway` uses an ARG with a
default (`ARG GATEWAY_BASE_IMAGE=nvcr.io/nvidia/distroless/cc:v4.0.4`)
and `FROM ${GATEWAY_BASE_IMAGE} AS gateway`, so the old script returns
nothing.
Parse the ARG default value directly so the GPU prerequisites check
keeps using the gateway base image as a `nvidia-smi` probe target.
* ci(gpu): pin GPU probe to nvcr.io/nvidia/base/ubuntu:noble
The previous probe parsed the gateway base image out of the Dockerfile,
relying on the fact that the gateway ran on `nvcr.io/nvidia/base/ubuntu`
and that NVIDIA Container Toolkit CDI injection would populate
`nvidia-smi` and the supporting libs at runtime. The new gateway base
(`nvcr.io/nvidia/distroless/cc`) lacks `ldconfig`, a populated
`/usr/bin`, and the broader filesystem layout CDI injection assumes,
so it cannot serve as a GPU probe.
Pin the probe image explicitly to the NVIDIA-managed Ubuntu base. The
probe is independent of the gateway runtime and survives future base
swaps.
* fix(e2e-gpu): pass GPU probe image via env, drop Dockerfile.images parse
The Rust e2e test `gpu_request_for_each_discovered_device_matches_plain_container`
was still parsing the deleted `Dockerfile.images` to derive its GPU probe
image, panicking with `No such file or directory` after this PR's
Dockerfile split.
Move the probe image to a single source of truth in the workflow
(`OPENSHELL_E2E_GPU_PROBE_IMAGE` env at the job level) and require the
e2e test to read it from there. No silent codebase default — the test
panics with a pointer at the workflow if the env is missing, so it
fails loudly rather than drifting from CI.
The prereq probe step in the workflow now consumes the same env, so
the probe image is declared exactly once.
* fix(docker): keep supervisor binary root-owned for rootless Podman
The Podman driver mounts the supervisor image read-only into the sandbox
container at /opt/openshell/bin and runs that container as UID 0, but
deliberately drops DAC_OVERRIDE for hardening (container.rs:419). With
--chown=65534:65534 --chmod=0550 the binary was r-xr-x--- owned by UID
65534, so the container's UID 0 fell into "other" with no read or exec
access and the supervisor crashed on start (ContainerExited code 1).
Docker and Kubernetes both retain DAC_OVERRIDE, so root could still
exec the file — which is why this regression only surfaced in the
Podman e2e job.
Drop --chown from the supervisor COPY so the binary stays root-owned.
Keep --chmod=0550: the security win was dropping world-execute, not
changing the owner. The chown bought nothing here because the container
is always UID 0 regardless of driver, but it actively broke the only
driver that drops DAC_OVERRIDE.
---------
Co-authored-by: Drew Newberry <anewberry@nvidia.com>
|
||
|
Florent BENOIT
|
764d930680 |
fix(vm): use bash 3.2-safe empty array expansion in supervisor build script (#1311)
macOS ships bash 3.2, which treats empty array expansion ("${arr[@]}")
as an unbound variable error under set -u. Unlike set -e, this error
is not caught by if-constructs — it kills the shell immediately with
no output, causing mise run vm:supervisor to fail silently.
Use the ${arr[@]+"${arr[@]}"} idiom so the script works on both
bash 3.2 (macOS default) and bash 4+.
Signed-off-by: Florent Benoit <fbenoit@redhat.com>
|
||
|
Zygmunt Krynicki
|
dfd47683e7 | (feat) early snap support (#1238) |