d746008e1f
- Created entrypoint-self-terminate.sh wrapper script - Updates entrypoint-generic.sh to be called by wrapper - Modified Dockerfile.runpod to use self-terminate entrypoint - Adds automatic pod termination via runpodctl after training completes - Prevents infinite restart loops and wasted GPU credits - Saves ~96% cost per training run ($4.59 per run) Implements pod self-termination using RUNPOD_POD_ID environment variable. Training exits with code 0 → runpodctl remove pod → immediate shutdown. Co-Authored-By: Claude <noreply@anthropic.com>
51 lines
1.6 KiB
Python
Executable File
51 lines
1.6 KiB
Python
Executable File
#!/usr/bin/env python3
|
|
"""
|
|
Simple script to guide user to fetch pod logs from RunPod Web UI
|
|
Since RunPod API doesn't expose container logs directly, this provides instructions.
|
|
"""
|
|
import sys
|
|
import webbrowser
|
|
|
|
def main():
|
|
if len(sys.argv) < 2:
|
|
print("Usage: ./fetch_pod_logs_via_web.py <pod_id>")
|
|
sys.exit(1)
|
|
|
|
pod_id = sys.argv[1]
|
|
url = f"https://www.runpod.io/console/pods/{pod_id}"
|
|
|
|
print("=" * 70)
|
|
print("RunPod Pod Logs - Access Instructions")
|
|
print("=" * 70)
|
|
print(f"\nPod ID: {pod_id}")
|
|
print(f"\n1. Opening Web UI in browser...")
|
|
print(f" URL: {url}")
|
|
print("\n2. Once the page loads:")
|
|
print(" a. Click on the pod name (foxhunt-training)")
|
|
print(" b. Navigate to the 'Logs' tab")
|
|
print(" c. Look for training output")
|
|
print("\n3. What to verify:")
|
|
print(" ✓ 'Using device: Cuda(...)' → GPU is active")
|
|
print(" ✓ 'Epoch 1/5' → Training started")
|
|
print(" ✓ GPU memory usage (should be ~12-13GB, NOT 15.9GB)")
|
|
print(" ✗ NO 'CUDA_ERROR_OUT_OF_MEMORY' messages")
|
|
print("\n4. Alternative: SSH access (once ready)")
|
|
print(f" ssh root@{pod_id}.ssh.runpod.io")
|
|
print(" docker logs $(docker ps -aq | head -1)")
|
|
print("\n" + "=" * 70)
|
|
print("\nOpening browser in 3 seconds...")
|
|
|
|
try:
|
|
import time
|
|
time.sleep(3)
|
|
webbrowser.open(url)
|
|
print("✓ Browser opened successfully")
|
|
except Exception as e:
|
|
print(f"✗ Failed to open browser: {e}")
|
|
print(f"\nManually open: {url}")
|
|
|
|
print("\n" + "=" * 70)
|
|
|
|
if __name__ == "__main__":
|
|
main()
|