feat: add homelab deployment configs, documentation, and automation scripts
This commit is contained in:
12
scripts/test_dcgm_metrics.py
Normal file
12
scripts/test_dcgm_metrics.py
Normal file
@@ -0,0 +1,12 @@
|
||||
import requests
|
||||
|
||||
for ip in ["140.44.4.81", "140.44.4.82", "140.44.4.83", "140.44.4.84"]:
|
||||
try:
|
||||
r = requests.get(f"http://{ip}:9400/metrics", timeout=3)
|
||||
print(f"=== {ip}:9400 -> HTTP {r.status_code}, len={len(r.text)} ===")
|
||||
gpu_metrics = [line for line in r.text.split('\n') if 'DCGM_FI_DEV_GPU_UTIL' in line or 'DCGM_FI_DEV_FB_USED' in line or 'DCGM_FI_DEV_GPU_TEMP' in line]
|
||||
for m in gpu_metrics[:3]:
|
||||
print(" ", m)
|
||||
except Exception as e:
|
||||
print(f"{ip}:9400 -> error: {e}")
|
||||
|
||||
Reference in New Issue
Block a user