* feat(sandbox): default to official Alpine sandbox image
default_sandbox_image() now returns docker.io/library/alpine:3.22, a generic
version-qualified official image, so a fresh install no longer depends on the
community sandbox image catalog. All compute drivers (docker, podman,
kubernetes, vm) inherit this fallback.
Part of #3116.
Signed-off-by: Akram
Signed-off-by: Akram <akram.benaissi@gmail.com>
* feat(deploy): default deployment configs to the official Alpine sandbox image
Update the shared gateway default_image, Helm chart values, the standalone
Kubernetes manifest, and the dev gateway task scripts to use
docker.io/library/alpine:3.22 instead of the community base image, consistent
with default_sandbox_image(). GPU e2e image-build base is left unchanged (CUDA
needs a glibc base).
Part of #3116.
Signed-off-by: Akram
Signed-off-by: Akram <akram.benaissi@gmail.com>
* feat(driver): default to numeric non-root identity for USER-less images
With the default sandbox image now Alpine, images that declare no OCI USER
must start instead of being rejected. When the image declares no USER and
the policy requests none, the Podman and Docker drivers now supply a numeric
non-root identity (DEFAULT_SANDBOX_UID/GID = 1000) instead of rejecting,
matching the numeric-identity behavior of the Kubernetes and VM drivers. The
supervisor's resolved-identity path runs the sandbox as a synthesized
non-root account without the account existing in the image. Images that
declare a USER keep the OCI resolution path unchanged.
Part of #3116.
Signed-off-by: Akram <akram.benaissi@gmail.com>
Signed-off-by: Evan Lezar <elezar@nvidia.com>
* test(conformance): use Alpine workload image
Signed-off-by: Evan Lezar <elezar@nvidia.com>
* refactor(policy): drop community image /app path from default policy
The restrictive default policy granted read-only access to /app, a directory
that only existed in the community base image. A generic Alpine default has no
/app, so remove it. Landlock best-effort already ignores absent paths; this
just stops advertising a community-specific layout in the default.
Part of #3116.
Signed-off-by: Akram
Signed-off-by: Akram <akram.benaissi@gmail.com>
* docs(config): document Alpine default images
Signed-off-by: Evan Lezar <elezar@nvidia.com>
* fix(podman): report early sandbox termination
Signed-off-by: Evan Lezar <elezar@nvidia.com>
* fix(podman): initialize rootless workspace ownership
Signed-off-by: Evan Lezar <elezar@nvidia.com>
* fix(sandbox): qualify NVIDIA Ubuntu default
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
* fix(podman): initialize rootful default workspace
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
* feat(sftp): add native sandbox adapter
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
* fix(sftp): gate runtime helper support to Linux
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
* fix(sftp): support standard OpenSSH file operations
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
* fix(sftp): harden rename and special file handling
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
* refactor(runtime): remove community image dependencies
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
* test(e2e): build provider readiness tool fixture
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
* fix(e2e): use a dedicated Noble fixture for Docker tests
Signed-off-by: Evan Lezar <elezar@nvidia.com>
---------
Signed-off-by: Akram
Signed-off-by: Akram <akram.benaissi@gmail.com>
Signed-off-by: Evan Lezar <elezar@nvidia.com>
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
Co-authored-by: Evan Lezar <elezar@nvidia.com>
Co-authored-by: Drew Newberry <anewberry@nvidia.com>
The opt-in glibc-static supervisor variant from #2682 is gone: the Nix
release builds (#2977) removed it from CI and the supervisor Dockerfile,
and the RFC 0012 sandbox split (#2942) removed it from the staging script.
The split also moved the binary that runs inside workload images to
openshell-sandbox; the supervisor now runs from its own image and is
dynamically linked. A few places still describe the old model:
- skills/debug-openshell-cluster told operators to check that
supervisor_image contains a static /openshell-supervisor. Ask instead for
a supervisor from the matching release whose loader and shared libraries
are available inside that image, and give a `--version` check that shows
both. The static requirement for /openshell-sandbox is unchanged.
- verify-static-binary.sh justified its check with the supervisor and the
glibc-static variant. Describe the property it enforces instead, with
the musl sandbox runtime as the main example.
- stage-prebuilt-binaries.sh kept an unreachable gnu-static case in
target_triple.
No behavior change: resolve_component only ever selects gnu or musl.
Signed-off-by: Emilien Macchi <emacchi@redhat.com>
* feat(kubernetes): support HA gateway rebalancing
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
* perf(server): cache peer connections, tokens, and owner lookups
Every forwarded relay rebuilt its setup from scratch: an owner lookup, a
blocking read of the peer token, a TLS connect to the owning replica, and
a TokenReview plus Pod GET on the receiving side. Sandbox service routing
does this per HTTP request, so the apiserver calls scaled with traffic.
Cache all of it on ServerState:
- peer channels pooled per endpoint, so relays multiplex over one
connection instead of redialing
- peer tokens keyed by SHA-256, expiring at min(ttl, token exp) so a hit
cannot accept an expired token
- owner records for 3s against a 45s ownership TTL, still freshness
checked before use
Entries are evicted when a relay fails. Also raise HTTP/2
max_concurrent_streams to 1024, since pooling funnels every relay between
two replicas onto one connection and hyper's default of 200 sits below
the 256 pending-relay budget.
Signed-off-by: divesh <dgude@nvidia.com>
* perf(server): pool upstream connections for sandbox services
Each HTTP request to a sandbox service opened its own supervisor relay,
paying a new TCP connection and HTTP/1 handshake every time. Worse, it
counted against the 32 in-flight relay cap, so a service handling more
than 32 concurrent requests failed outright.
Pool idle upstreams per endpoint and port, up to 8 each for 15s. Reuse is
safe because the pool only returns a connection hyper reports as ready,
and HTTP/1 cannot start a request until the previous body has drained.
Upgrades are never pooled since they take the connection over, and a
failed send evicts that endpoint. Pruning is bounded per key, with the
full sweep limited to once per 30s.
Signed-off-by: divesh <dgude@nvidia.com>
* fix(server): address HA gateway review findings (#3449)
- Let a gateway own supervisor sessions without a peer endpoint. Requiring
one whenever the store is PostgreSQL broke every single-instance
PostgreSQL deployment, because no sandbox supervisor could connect.
A cross-replica request to an owner that advertises no endpoint now fails
immediately naming the cause, instead of retrying until the wait timeout.
- Close a supervisor session on heartbeat only when another replica owns it,
or after renewals fail for the ownership TTL. A database error no longer
drops every session heartbeating during an outage.
- Clamp owner record ages at zero so a skewed or corrupt stored timestamp
cannot produce a negative age.
- Bound the cross-object advisory lock with a lock timeout, so a stuck holder
fails instead of blocking every mutation in the fleet.
- Refuse to start when a peer endpoint is configured on a multi-replica
backend but peer authentication is unavailable, and warn when a
multi-replica backend has no peer endpoint at all.
- Reject a plaintext peer endpoint when the gateway serves TLS.
- Skip the sandbox watch poller on single-replica backends, where the local
update bus already sees every write.
- Rate-limit the peer owner cache sweep so an insert no longer scans the
whole map under the lock.
- Retry GET and HEAD on a pooled upstream the sandbox closed, instead of
returning 502, and drop an emptied endpoint from the pool right away.
- Document the gateway peer environment variables and the post-rollout
ownership skew operators should expect.
Signed-off-by: divesh <dgude@nvidia.com>
* fix(server): harden HA supervisor ownership
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
---------
Signed-off-by: Drew Newberry <anewberry@nvidia.com>
Signed-off-by: divesh <dgude@nvidia.com>
Co-authored-by: Drew Newberry <anewberry@nvidia.com>
Co-authored-by: divesh <dgude@nvidia.com>
Co-authored-by: Divesh Chowdary <47188680+FrostGod@users.noreply.github.com>
* fix(vm): unpack registry images correctly and validate prepared disks
The registry image-prep path expected `umoci raw unpack` to produce a
bundle-style rootfs/ subdirectory, but it extracts the image filesystem
directly into the target. Every registry prep therefore failed after a
successful unpack. Guest init exit codes do not survive the libkrun
boundary, so the failure looked like success and the broken disk was
cached, making every later sandbox for that image fail with "prepared
image disk missing /image-rootfs".
VM E2E started hitting this after the bootstrap image moved to
nvcr.io/nvidia/base/ubuntu:24.04: `--from base` no longer matches the
bootstrap image, so it now goes through registry prep.
- Accept umoci's direct extraction layout in the guest prep script.
- Build the image rootfs under a partial directory and rename it to
/image-rootfs only after every prep step succeeds.
- Check the prepared disk for /image-rootfs before caching it. On
failure, leave the cache untouched and report the prep console tail.
- Size the prep disk to hold the payload and the unpacked rootfs at the
same time. The community base image needs 1.40 GB + 3.32 GB, which
did not fit in the old payload*3 + 512 MiB.
Fixes#2358
Co-authored-by: s2cube <26961336+s2cube@users.noreply.github.com>
Signed-off-by: Emilien Macchi <emacchi@redhat.com>
* test(e2e): run tool-dependent VM tests from the community base image
The host_gateway_alias and vm_corporate_proxy workloads run curl and
python3. The VM driver now defaults to nvcr.io/nvidia/base/ubuntu:24.04,
which ships neither, so these tests fail in VM E2E with "command not
found". Request the community base image explicitly with `--from base`.
Docker, Podman, and Kubernetes E2E already default to that image, so
their behavior is unchanged.
Signed-off-by: Emilien Macchi <emacchi@redhat.com>
---------
Signed-off-by: Emilien Macchi <emacchi@redhat.com>
Co-authored-by: s2cube <26961336+s2cube@users.noreply.github.com>
* feat(cli): promote profile commands to top level
Add profile discovery and management commands with shared handlers for
the existing provider entry points. List a flat catalog across scopes
and follow continuation tokens through full and short pages.
Describe metadata, credentials, endpoints, TLS inspection, and MCP access
settings while preserving complete JSON/YAML definitions. Cover parser
equivalence, scope forwarding, pagination, and inspection settings with
focused unit and compiled-CLI integration tests.
Update docs, public skills, examples, and E2E command invocations.
Refs #2588
Signed-off-by: Shiju <shiju@nvidia.com>
* fix(cli): remove redundant workspace selector qualification
Use the imported WorkspaceSelector in the provider integration helper so
the target passes Clippy with warnings denied.
Signed-off-by: Shiju <shiju@nvidia.com>
---------
Signed-off-by: Shiju <shiju@nvidia.com>
Previously, Network Activity could be constructed without a source or
destination endpoint, allowing connection, accept, relay, and configuration
events to violate the OCSF 1.8 endpoint constraint.
Now, NetworkActivityBuilder requires a source or destination endpoint at
compile time. Connection failures identify the workload peer or genuine
transparent destination, listener failures identify the listening endpoint,
and mediation-lane failures use Application Lifecycle rather than fabricated
network endpoints. Malformed forward requests use HTTP Activity with a
method-only request, generated 400 response, and workload peer.
Additionally, Unix relay-channel events use Base Event, policy-validation
warnings use Config State Change, and the unused bypass monitor is removed
because the current isolation architecture no longer uses it.
Signed-off-by: Kris Hicks <khicks@nvidia.com>
BREAKING CHANGE: Policy YAML and public policy APIs now use the protobuf-shaped openshell.policy.v1.PolicyDocument contract.
Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com>
* fix(policy): reject unknown endpoint security modes
Closes#3046
Validate TLS, enforcement, and access values across policy and provider profile ingress, and prevent runtime parsing from falling back to audit for unknown enforcement values.
Signed-off-by: Krzysztof Malczuk <kmalczuk@redhat.com>
* fix(policy)!: use enums for endpoint security modes
Replace the public TLS, enforcement, and access strings with protobuf enums and carry the typed values through policy composition, provider profiles, drivers, and runtime conversion.
Preserve the documented YAML spellings, reject unknown and invalid numeric enum values consistently, and update generated Go bindings, SDK conversions, tests, and policy documentation.
Signed-off-by: Krzysztof Malczuk <kmalczuk@redhat.com>
---------
Signed-off-by: Krzysztof Malczuk <kmalczuk@redhat.com>
* feat(e2e): run kubernetes suite on cargo-nextest with JUnit/HTML reports
Switch e2e:kubernetes (and all its variants) from `cargo test` to
`cargo nextest run` for per-test process isolation and output consistent
with the other nextest-based CI runs.
- Add a dedicated `e2e-kubernetes` nextest profile with a JUnit report and a
generous slow-timeout (60s flag, 5-min terminate) suited to live-cluster
tests; kept separate from `ci` so its JUnit path and timeouts don't affect
the workspace run.
- Pin `--target-dir` for the run so the profile's relative JUnit path resolves
to the repo-root results/ regardless of any inherited CARGO_TARGET_DIR
(nextest ignores absolute JUnit paths).
- Render the JUnit XML to a standalone HTML report via xsltproc and a committed
XSLT stylesheet (best-effort; never masks the test exit code).
- Name each report via `OPENSHELL_E2E_REPORT_NAME` (default `e2e-kubernetes`),
used verbatim for both the `results/<name>.{xml,html}` filenames and the HTML
heading. Tasks that invoke the script multiple times in one run set a distinct
name per invocation so the reports no longer clobber the single fixed path:
the credential-driver runs write results/e2e-kubernetes-secrets.xml and
-vault.xml, and e2e:kubernetes:agent-sandbox-versions writes
results/e2e-kubernetes-agent-sandbox-v1beta1.xml and -v1alpha1.xml.
- Declare cargo-nextest in mise [tools] so the task runs without the Nix shell.
- Ignore the results/ output directory.
The results/ reports do not leak information. They are gitignored and no
workflow uploads them as artifacts, so they stay on the ephemeral CI runner
and are discarded when it is torn down. The HTML template renders only test
names, status, timings, and failure messages (no captured stdout/stderr).
Moving from `cargo test -- --nocapture` to nextest's captured, failure-only
output also reduces what lands in the retained, viewable console logs.
Signed-off-by: Jorge Garcia Oncins <jgarciao@redhat.com>
* chore(e2e): revert per-lane report names for agent-sandbox-versions
The agent-sandbox-versions task runs two lanes sequentially against the
same cluster: v0.5.0 (v1beta1 storage version) then v0.4.6 (v1alpha1).
On a reused cluster the second lane fails when kubectl applies the older
CRD, because Kubernetes refuses to drop v1beta1 from spec.versions while
it remains in status.storedVersions (the storage-version downgrade
guardrail). This is a pre-existing issue with the v0.4.6 lane, unrelated
to the nextest reporting work.
The per-lane OPENSHELL_E2E_REPORT_NAME additions do not address that
downgrade failure, so revert them to keep this PR scoped to the nextest
change. Agent Sandbox 0.4.x is also superseded (1.0.0 is published);
dropping or bumping the v1alpha1 lane is left as a follow-up.
Signed-off-by: Jorge Garcia Oncins <jgarciao@redhat.com>
---------
Signed-off-by: Jorge Garcia Oncins <jgarciao@redhat.com>
The libkrun and libkrunfw projects moved from the containers/ GitHub
org to libkrun/. Update all clone URLs and version pins accordingly.
libkrun v1.19.4 introduces krun_add_virtiofs4 with a permissions
semantics parameter, enabling correct file ownership mapping on
host-to-guest bind mounts — required for issue #2585.
- libkrun: v1.17.4 → v1.19.4 (libkrun/libkrun)
- libkrunfw: 463f717b → v5.6.1 (libkrun/libkrunfw)
- Centralise LIBKRUN_REF in pins.env instead of hardcoding in each
build script
- Source pins.env in the macOS build script for consistency with the
Linux build script
- Remove the init/init make target step — since libkrun/libkrun@05c4eb7
the init blob moved into the init_blob crate and is built by cargo
Ref: https://github.com/NVIDIA/OpenShell/issues/2585
Signed-off-by: Florent Benoit <fbenoit@redhat.com>
BREAKING CHANGE: Policy YAML must use the openshell.policy.v1 field shape; legacy scalar matcher and tool shorthands are no longer accepted.
Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com>
* docs(fern): sync announcement configuration
Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com>
* fix(docs): scope announcements to synced channel
Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com>
* docs(fern): use dev as source version
Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com>
* docs(fern): use channel-neutral logo link
Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com>
---------
Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com>
Piotr Mlocek