Skip to content

gpu memory check - #295

Open
jaspals3123 wants to merge 1 commit into
developmentfrom
jaspal_gpuchecks
Open

jaspals3123 wants to merge 1 commit into
developmentfrom
jaspal_gpuchecks

Conversation

@jaspals3123

Copy link
Copy Markdown
Collaborator

added new analyzer arg minimum_available_percent
Plugin config:

{
  "name": "GPU memory validation",
  "plugins": {
    "AmdSmiPlugin": {
      "collection_args": {},
      "analysis_args": {
        "gpu_memory": {
          "minimum_available_percent": 95
        }
      }
    }
  }
}

cmd:
node-scraper --plugin-configs=/home/jaspals/node-scraper/plugin.json

if total_vram is None or free_vram is None:
self._log_event(
category=EventCategory.PLATFORM,
description=f"GPU {metric.gpu} VRAM availability is not available",

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this sounds odd "availability is not available" :)

except (TypeError, ValueError):
self._log_event(
category=EventCategory.PLATFORM,
description=f"GPU {metric.gpu} VRAM availability is invalid",

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

make this more descriptive, like "was not able to parse available blabla form blabla"

)
continue

if total_value <= 0:

@alexandraBara alexandraBara Sep 23, 2026 •

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

how is this ever going to happen? this seems like something we should check for before we try and compute it in such a way that it gives negative values

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants