Release v0.1.0-rc.10

CRE v0.1.0-rc.10

The Cluster Readiness Engine (CRE) is a Kubernetes controller that
certifies GPU clusters before production use. It runs real training
and communication workloads across topology-aware node groups,
measures goodput and bandwidth, detects hardware failures, and
quarantines bad nodes. See the
README
for the full feature list.

Install nvcrectl

While this repository is internal, GitHub serves release assets only through
authenticated API downloads — plain curl against releases/download/...
returns a 404 "Not Found" page instead of the script, even with a token.
Fetch the installer with the gh CLI (authenticate with gh auth login):

gh release download v0.1.0-rc.10 \
  --repo dsx-ai-factory/cluster-readiness-engine \
  --pattern installer --output - | bash -s -- -v v0.1.0-rc.10

Once the repository is public, the plain curl form works. The version is
named explicitly because releases/latest resolves only to the newest
stable release, and every release so far is a pre-release. Once v0.1.0
is tagged, the shorter form installs the newest stable one:

curl -sSL https://github.com/dsx-ai-factory/cluster-readiness-engine/releases/download/v0.1.0-rc.10/installer | bash -s -- -v v0.1.0-rc.10
curl -sSL https://github.com/dsx-ai-factory/cluster-readiness-engine/releases/latest/download/installer | bash

Helm Chart

helm install cluster-readiness-engine \
  oci://ghcr.io/dsx-ai-factory/cluster-readiness-engine \
  --version v0.1.0-rc.10 \
  --namespace cluster-readiness-engine \
  --create-namespace

The controller image for this release is
ghcr.io/dsx-ai-factory/cluster-readiness-engine/manager:v0.1.0-rc.10.

nvcrectl CLI

Download the binary for your platform from the assets below and place it on your $PATH.

Platform	Binary
Linux amd64	nvcrectl-linux-amd64
Linux arm64	nvcrectl-linux-arm64
macOS amd64	nvcrectl-darwin-amd64
macOS arm64 (Apple Silicon)	nvcrectl-darwin-arm64

Verify your download

Every asset is listed in checksums.txt with its SHA-256 digest.
After you download an asset, verify it:

sha256sum --check --ignore-missing checksums.txt

Third-Party Notices

Third-party Go module licenses are listed in
THIRD_PARTY_NOTICES.md
and attached as a release asset.

Support

• Ask questions in GitHub Discussions.
• Report bugs with the issue templates.
• Report security issues per SECURITY.md, never through GitHub issues.

What's Changed

Dependency Updates

• build(deps): bump the docker-images group across 1 directory with 2 updates by @dependabot[bot] in #204
• build(deps): bump the kubernetes group with 4 updates by @dependabot[bot] in #209
• build(deps): bump github.com/stretchr/testify from 1.11.1 to 1.12.1 in the go-dependencies group across 1 directory by @dependabot[bot] in #210
• build(deps): bump the actions group across 1 directory with 3 updates by @dependabot[bot] in #211

Other Changes

• fix(podlogs): bound pod log stream duration by @kaynetu in #169
• chore(helm): remove unused nemo-4-training LogProfile by @ndipebot in #173
• test: convert Tier 1 + Tier 2 table-driven tests to testutil golden files by @ndipebot in #172
• Add initial documentation and Fern configuration by @pdmack in #2
• chore: migrate module path and repo references to dsx-ai-factory org by @ndipebot in #206
• docs: fix TMF-reported doc/code drift, add WorkloadRun quickstarts, operations section, and gangScheduler docs by @ndipebot in #186
• feat(cli): add progress heartbeat to workloadrun run --wait by @ndipebot in #187
• fix(setup): surface Helm release health in setup status by @ndipebot in #188
• fix(setup): report retained resources after reset by @ndipebot in #189
• feat(cli): disambiguate duplicate categories in live watch output by @ndipebot in #190
• fix(cli): derive --platform validation from a single shared platform list by @ndipebot in #191
• fix(catalog): Azure GB200/GB300 use InfiniBand — align all-reduce with sibling collectives by @ndipebot in #192
• fix(report): surface threshold and hardware failure reasons in failed groups by @ndipebot in #193
• fix(cli): derive certification --wait timeout from category job budgets by @ndipebot in #200
• fix(catalog): add oci+gb300 override to nccl-loopback-nvswitch, document intentional gaps by @ndipebot in #201
• fix(release): verify published assets and stamped versions; document authenticated installer download by @ndipebot in #203
• fix(controller): resolve MPI launcher MNNVL from architecture default by @ndipebot in #208
• fix(platform): pin MPI transport to TCP for AWS GB300 WorkloadRuns by @ndipebot in #197
• fix(setup): converge setup init retries after a partial Trainer install by @ndipebot in #198
• feat(ci): verify relative markdown links in README.md and docs/ by @ndipebot in #202
• fix(controller): freeze GoodputMeasurement status at Job terminal state by @ndipebot in #205
• refactor!: rename ncre to nvcre across the repo by @ndipebot in #217

New Contributors

• @pdmack made their first contribution in #2

Full Changelog: v0.1.0-rc.9...v0.1.0-rc.10