v1.3.1-rc.1

NVIDIA Fleet Intelligence Agent v1.3.1-rc.1

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

• 457a29b: feat: mark persistence mode diable as warning level (#225) (@ambermingxin)

Bug Fixes

• 446d2c4: fix: OTLP export silently skips when endpoints are configured but no auth token is present. (#199) (@miconeilaws)
• eb518eb: fix: apt remove preserve service enable (#221) (@jingxiang-z)
• e17a58d: fix: degrade dcgm query timeouts (#222) (@jingxiang-z)
• 1593ca1: fix: include metric scrape interval in agent config (#223) (@jingxiang-z)
• ea870bb: fix: initialize node uuid during enrollment (#224) (@jingxiang-z)
• b2af283: fix: make DCGM group name configurable and different for each process (#220) (@jingxiang-z)
• d575c33: fix: preserve counter metric types in OTLP (#198) (@jingxiang-z)
• 87f1a58: fix: remove warning message (#214) (@ambermingxin)
• eb7aa39: fix: ubuntu deb package upgrade restart the service (#218) (@jingxiang-z)

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