fix(dashboard): follow HAMi metric rename so GPU panels resolve - #606
Merged
ZhangEnYao merged 2 commits intoJul 30, 2026
Merged
Conversation
HAMi renamed its scheduler and device-plugin metrics (GPUDeviceMemoryLimit, Device_*_desc_of_container, vGPU_device_memory_*_in_bytes) to a hami_* scheme, and relabelled deviceuuid → device_uuid and podnamespace/podname → exported_namespace/exported_pod. The dashboard still queried the old names, so every GPU query returned an empty vector while Prometheus reported success — the cluster GPU cards and the node table's GPU columns showed "no data" on a cluster with eight healthy RTX 4090s. Point the queries and label lookups at the new names. GPU Utilization now averages DCGM_FI_DEV_GPU_UTIL instead of HAMi's per-container ratio: the card is titled "average across all GPU cards", DCGM reports per physical card, and the card count beside it already came from DCGM.
ZhangEnYao
approved these changes
Jul 30, 2026
ZhangEnYao
deleted the
605-gpu-panels-show-no-data-after-the-hami-metric-rename
branch
July 30, 2026 06:58
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
HAMi renamed its scheduler and device-plugin metrics (GPUDeviceMemoryLimit, Device_desc_of_container, vGPU_device_memoryin_bytes) to a hami* scheme, and relabelled deviceuuid → device_uuid and podnamespace/podname → exported_namespace/exported_pod. The dashboard still queried the old names, so every GPU query returned an empty vector while Prometheus reported success — the cluster GPU cards and the node table's GPU columns showed "no data" on a cluster with eight healthy RTX 4090s.
Point the queries and label lookups at the new names. GPU Utilization now averages DCGM_FI_DEV_GPU_UTIL instead of HAMi's per-container ratio: the card is titled "average across all GPU cards", DCGM reports per physical card, and the card count beside it already came from DCGM.