NVIDIA Fleet Intelligence Agent v1.0.1-rc.3

This release includes a host agent for GPU telemetry collection and attestation.

Available Packages

• DEB packages: amd64 and arm64 for Debian/Ubuntu systems
• RPM packages: x86_64 and aarch64 for RHEL/CentOS/Fedora systems
• Binary archives: Standalone binaries for both architectures

Installation

Download the appropriate package for your system from the assets below. For installation instructions, see the project README.

Changelog

Bug Fixes

• c88bd44 fix: pass chart env to enroll init container

Verification

All release artifacts are checksummed. Verify downloads using:

sha256sum -c checksums.txt

Support

For issues and questions, please visit: https://github.com/NVIDIA/fleet-intelligence-agent/issues

NVIDIA Fleet Intelligence Agent v1.0.1-rc.2

This release includes a host agent for GPU telemetry collection and attestation.

Available Packages

• DEB packages: amd64 and arm64 for Debian/Ubuntu systems
• RPM packages: x86_64 and aarch64 for RHEL/CentOS/Fedora systems
• Binary archives: Standalone binaries for both architectures

Installation

Download the appropriate package for your system from the assets below. For installation instructions, see the project README.

Changelog

New Features

• da8306d [GPUHEALTH-1520] feat: add gpu index to machine info (#132)
• 81a88dd feat: add enrollment precheck command
• d27258e feat: add helm enroll force option

Bug Fixes

• 3b61269 fix: enforce minimum driver version in precheck
• 6806019 fix: remove retired page metrics

Documentation Updates

• 2767124 [GPUHEALTH-1563] docs: fix the xid component name

Other Changes

• 313b24e [GPUHEALTH-1556] Change to Prod URL and remove deployEnv
• aed2cb6 add header to missing file

Verification

All release artifacts are checksummed. Verify downloads using:

sha256sum -c checksums.txt

Support

For issues and questions, please visit: https://github.com/NVIDIA/fleet-intelligence-agent/issues

NVIDIA Fleet Intelligence Agent v1.0.0

This release includes a host agent for GPU telemetry collection and attestation.

Available Packages

• DEB packages: amd64 and arm64 for Debian/Ubuntu systems
• RPM packages: x86_64 and aarch64 for RHEL/CentOS/Fedora systems
• Binary archives: Standalone binaries for both architectures

Installation

Download the appropriate package for your system from the assets below. For installation instructions, see the project README.

Changelog

Bug Fixes

• f2b4b8e fix: remove retired page metrics

Documentation Updates

• 2767124 [GPUHEALTH-1563] docs: fix the xid component name

Other Changes

• 313b24e [GPUHEALTH-1556] Change to Prod URL and remove deployEnv
• aed2cb6 add header to missing file

Verification

All release artifacts are checksummed. Verify downloads using:

sha256sum -c checksums.txt

Support

For issues and questions, please visit: https://github.com/NVIDIA/fleet-intelligence-agent/issues

NVIDIA Fleet Intelligence Agent v1.0.0-rc.6

This release includes a host agent for GPU telemetry collection and attestation.

Available Packages

• DEB packages: amd64 and arm64 for Debian/Ubuntu systems
• RPM packages: x86_64 and aarch64 for RHEL/CentOS/Fedora systems
• Binary archives: Standalone binaries for both architectures

Installation

Download the appropriate package for your system from the assets below. For installation instructions, see the project README.

Changelog

New Features

• e23fd1c [GPUHEALTH-1522] feat: add DCGM ECC and NVLink RX metrics

Documentation Updates

• 47c3d48 [GPUHEALTH-1535] docs: restore component list to configuration docs

Verification

All release artifacts are checksummed. Verify downloads using:

sha256sum -c checksums.txt

Support

For issues and questions, please visit: https://github.com/NVIDIA/fleet-intelligence-agent/issues

NVIDIA Fleet Intelligence Agent v1.0.0-rc.5

This release includes a host agent for GPU telemetry collection and attestation.

Available Packages

• DEB packages: amd64 and arm64 for Debian/Ubuntu systems
• RPM packages: x86_64 and aarch64 for RHEL/CentOS/Fedora systems
• Binary archives: Standalone binaries for both architectures

Installation

Download the appropriate package for your system from the assets below. For installation instructions, see the project README.

Changelog

New Features

• 1dd7613 [GPUHEALTH-1522] feat: add approved DCGM metrics

Verification

All release artifacts are checksummed. Verify downloads using:

sha256sum -c checksums.txt

Support

For issues and questions, please visit: https://github.com/NVIDIA/fleet-intelligence-agent/issues

NVIDIA Fleet Intelligence Agent v1.0.0-rc.4

This release includes a host agent for GPU telemetry collection and attestation.

Available Packages

• DEB packages: amd64 and arm64 for Debian/Ubuntu systems
• RPM packages: x86_64 and aarch64 for RHEL/CentOS/Fedora systems
• Binary archives: Standalone binaries for both architectures

Installation

Download the appropriate package for your system from the assets below. For installation instructions, see the project README.

Changelog

Bug Fixes

• 14512c9 [GPUHEALTH-1331] fix: export structured event extra_info

Documentation Updates

• 7cd59e1 [GPUHEALTH-1491] docs: update install prerequisites and supported GPU platforms

Other Changes

• a5676b0 [GPUHEALTH-1331] chore: bump Go toolchain to 1.26.1

Verification

All release artifacts are checksummed. Verify downloads using:

sha256sum -c checksums.txt

Support

For issues and questions, please visit: https://github.com/NVIDIA/fleet-intelligence-agent/issues

NVIDIA Fleet Intelligence Agent v1.0.0-rc.3

This release includes a host agent for GPU telemetry collection and attestation.

Available Packages

• DEB packages: amd64 and arm64 for Debian/Ubuntu systems
• RPM packages: x86_64 and aarch64 for RHEL/CentOS/Fedora systems
• Binary archives: Standalone binaries for both architectures

Installation

Download the appropriate package for your system from the assets below. For installation instructions, see the project README.

Changelog

Other Changes

• f82c671 [GPUHEALTH-1408] chore: finalize fleet intelligence naming in release and packaging text

Verification

All release artifacts are checksummed. Verify downloads using:

sha256sum -c checksums.txt

Support

For issues and questions, please visit: https://github.com/NVIDIA/fleet-intelligence-agent/issues

NVIDIA Fleet Intelligence Agent v1.0.0-rc.2

This release includes GPU health monitoring and reporting capabilities for NVIDIA GPU infrastructure.

Available Packages

• DEB packages: amd64 and arm64 for Debian/Ubuntu systems
• RPM packages: x86_64 and aarch64 for RHEL/CentOS/Fedora systems
• Binary archives: Standalone binaries for both architectures

Installation

Download the appropriate package for your system from the assets below. For installation instructions, see the project README.

Changelog

Verification

All release artifacts are checksummed. Verify downloads using:

sha256sum -c checksums.txt

Support

For issues and questions, please visit: https://github.com/NVIDIA/fleet-intelligence-agent/issues

NVIDIA Fleet Intelligence Agent v1.0.0-rc.1

This release includes GPU health monitoring and reporting capabilities for NVIDIA GPU infrastructure.

Available Packages

• DEB packages: amd64 and arm64 for Debian/Ubuntu systems
• RPM packages: x86_64 and aarch64 for RHEL/CentOS/Fedora systems
• Binary archives: Standalone binaries for both architectures

Installation

Download the appropriate package for your system from the assets below. For installation instructions, see the project README.

Changelog

New Features

• 1e45823 feat: Agent status should report configured endpoint

Bug Fixes

• 4b8e899 [GPUHEALTH-1393] fix: use DCGM NVLink counters and align help text
• 2015b4c [GPUHEALTH-1417] fix: use default GPU group for profiling watch

Other Changes

• 9f0b094 Add missing dependencies for vendor sdk tests and adjust dockerfile for local tests
• 0a01d9e Exclude third_party from codecoverage check by sonar
• a27624f [GPUHEALTH-1405] refactor: remove deprecated gpud components from registry
• f4ac109 [GPUHEALTH-1417] chore: ignore local worktree directory
• d4e41bd [GPUHEALTH-1422] GPUHealth changes to vendor dependency libraries from GPUd for opensourcing
• c198d42 exclude third_party/* from sonar coverage check

Verification

All release artifacts are checksummed. Verify downloads using:

sha256sum -c checksums.txt

Support

For issues and questions, please visit: https://github.com/NVIDIA/fleet-intelligence-agent/issues

NVIDIA Fleet Intelligence Agent v0.3.4

This release includes GPU health monitoring and reporting capabilities for NVIDIA GPU infrastructure.

Available Packages

• DEB packages: amd64 and arm64 for Debian/Ubuntu systems
• RPM packages: x86_64 and aarch64 for RHEL/CentOS/Fedora systems
• Binary archives: Standalone binaries for both architectures

Installation

Download the appropriate package for your system from the assets below. For installation instructions, see the project README.

Changelog

New Features

• f70a5ea [GPUHEALTH-1408] feat: strict rename to fleet intelligence agent naming

Bug Fixes

• 318214d [GPUHEALTH-1408] fix: allow goreleaser to replace existing release assets
• 9ab4cbb [GPUHEALTH-1408] fix: apply gofmt after naming updates
• 5b46ffa [GPUHEALTH-1408] fix: keep gpuhealth artifactory repo root
• b38cd36 [GPUHEALTH-1408] fix: make lint resilient to inherited go.work
• 426ec1b [GPUHEALTH-1408] fix: release to fleet-intelligence-agent repo
• 7486269 [GPUHEALTH-1408] fix: rename machine-info version label to Fleetint
• 02ec84c [GPUHEALTH-1408] fix: update CI smoke build path after fleetint rename

Documentation Updates

• dc62ae1 [GPUHEALTH-1408] docs: finalize naming updates in top-level and helm install docs

Other Changes

• 5d7f4cf [GPUHEALTH-1408] chore: align dev image tag with public image name
• 4c1fd07 [GPUHEALTH-1408] refactor: remove remaining GPU Health references in code and docs

Verification

All release artifacts are checksummed. Verify downloads using:

sha256sum -c checksums.txt

Support

For issues and questions, please visit: https://github.com/NVIDIA/fleet-intelligence-agent/issues