|
#!/bin/bash |
|
|
|
if [ -z "$HF_TOKEN" ] || [ -z "$DATASET_ID" ]; then |
|
echo "Starting without backup functionality - missing HF_TOKEN or DATASET_ID" |
|
exec uvicorn app.main:app --host 0.0.0.0 --port 7860 |
|
exit 0 |
|
fi |
|
|
|
export HUGGING_FACE_HUB_TOKEN=$HF_TOKEN |
|
|
|
restore_latest() { |
|
echo "Restoring latest backup..." |
|
python3 -c " |
|
from huggingface_hub import HfApi |
|
import os |
|
api = HfApi() |
|
files = api.list_repo_files(repo_id='${DATASET_ID}', repo_type='dataset') |
|
backup_files = sorted([f for f in files if f.startswith('backup_') and f.endswith('.tar.gz')]) |
|
if backup_files: |
|
latest = backup_files[-1] |
|
api.hf_hub_download(repo_id='${DATASET_ID}', filename=latest, repo_type='dataset', local_dir='/tmp') |
|
os.system(f'tar -xzf /tmp/{latest} -C /app/app') |
|
os.remove(f'/tmp/{latest}') |
|
print(f'Restored from {latest}') |
|
else: |
|
print('No backup found') |
|
" |
|
} |
|
|
|
sync_data() { |
|
while true; do |
|
echo "Starting backup at $(date)" |
|
cd /app |
|
timestamp=$(date +%Y%m%d_%H%M%S) |
|
backup_file="backup_${timestamp}.tar.gz" |
|
if [ -d "app" ] && [ "$(ls -A app 2>/dev/null)" ]; then |
|
tar -czf "/tmp/${backup_file}" app/ |
|
python3 -c " |
|
from huggingface_hub import HfApi |
|
api = HfApi() |
|
api.upload_file(path_or_fileobj='/tmp/${backup_file}', path_in_repo='${backup_file}', repo_id='${DATASET_ID}', repo_type='dataset') |
|
print('Backup uploaded successfully') |
|
backup_files = sorted([f for f in api.list_repo_files('${DATASET_ID}', repo_type='dataset') if f.startswith('backup_')]) |
|
for old_backup in backup_files[:-50]: |
|
api.delete_file(path_in_repo=old_backup, repo_id='${DATASET_ID}', repo_type='dataset') |
|
print(f'Deleted old backup: {old_backup}') |
|
" |
|
rm -f "/tmp/${backup_file}" |
|
else |
|
echo "No data to backup" |
|
fi |
|
sleep ${SYNC_INTERVAL:-7200} |
|
done |
|
} |
|
|
|
( |
|
restore_latest |
|
sync_data & |
|
exec uvicorn app.main:app --host 0.0.0.0 --port 7860 |
|
) 2>&1 | tee -a /app/data/backup.log |
|
|