v1.4.0

NVIDIA Fleet Intelligence Agent v1.4.0

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.

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

What's Changed

• fix: preserve counter metric types in OTLP by @jingxiang-z in #198
• docs: update supported platform docs by @jingxiang-z in #207
• feat: add validation for ng/cz input by @ambermingxin in #206
• chore: disable accelerator-nvidia-processes by default by @ambermingxin in #209
• ci: harden github actions workflows by @jingxiang-z in #205
• fix: allow user to only set one field for ng or cz by @ambermingxin in #211
• fix: remove warning message by @ambermingxin in #214
• chore: update go version to 1.26.4 by @jingxiang-z in #215
• fix: OTLP export silently skips when endpoints are configured but no auth token is present. by @miconeilaws in #199
• fix: ubuntu deb package upgrade restart the service by @jingxiang-z in #218
• fix: apt remove preserve service enable by @jingxiang-z in #221
• fix: make DCGM group name configurable and different for each process by @jingxiang-z in #220
• fix: degrade dcgm query timeouts by @jingxiang-z in #222
• fix: include metric scrape interval in agent config by @jingxiang-z in #223
• feat: mark persistence mode diable as warning level by @ambermingxin in #225
• fix: initialize node uuid during enrollment by @jingxiang-z in #224
• feat: Inband attestation should collect stderr from nvattest sdk by @ambermingxin in #228
• chore: Update OpenTelemetry dependencies to v1.44.0 by @rsampaio in #230
• chore: add readme for agent local db to be persistent by @ambermingxin in #232
• fix(inventory): make change hash order invariant by @ooolafhu in #233
• fix: export the component state error and suggested actions to backend by @ambermingxin in #234
• fix: bump grpc to v1.82.1 by @rsampaio in #237
• docs: add Fern scaffolding and publish agent docs by @ebalduf in #238
• fix: fern docs url by @ebalduf in #239
• feat: add authenticated OpenTelemetry gateway by @rsampaio in #231
• feat: enrich OTLP telemetry identity by @jingxiang-z in #240
• fix: bump the go crypto version to fix cve (#241) by @ambermingxin in #242

New Contributors

• @miconeilaws made their first contribution in #199
• @ooolafhu made their first contribution in #233

Full Changelog: v1.3.1...v1.4.0