Update sync_data.sh

sync_data.sh

@@ -2,8 +2,8 @@
 
 # 检查环境变量
 if [[ -z "$HF_TOKEN" ]] || [[ -z "$DATASET_ID" ]]; then
-  echo "缺少 HF_TOKEN 或 DATASET_ID 环境变量，无法启用备份功能"
-  exit 0
+    echo "Starting without backup functionality - missing HF_TOKEN or DATASET_ID"
+    exit 0
 fi
 
 # 激活虚拟环境
@@ -11,29 +11,28 @@ source /opt/venv/bin/activate
 
 # 上传备份
 upload_backup() {
-  file_path="$1"
-  file_name="$2"
-  token="$HF_TOKEN"
-  repo_id="$DATASET_ID"
+    file_path="$1"
+    file_name="$2"
+    token="$HF_TOKEN"
+    repo_id="$DATASET_ID"
 
-  python3 -c "
+    python3 -c "
 from huggingface_hub import HfApi
 import sys
 import os
-
-def manage_backups(api, repo_id, max_files=50):
+def manage_backups(api, repo_id, max_files=30):
     files = api.list_repo_files(repo_id=repo_id, repo_type='dataset')
     backup_files = [f for f in files if f.startswith('qinglong_backup_') and f.endswith('.tar.gz')]
     backup_files.sort()
+    
     if len(backup_files) >= max_files:
         files_to_delete = backup_files[:(len(backup_files) - max_files + 1)]
         for file_to_delete in files_to_delete:
             try:
                 api.delete_file(path_in_repo=file_to_delete, repo_id=repo_id, repo_type='dataset')
-                print(f'已删除旧备份: {file_to_delete}')
+                print(f'Deleted old backup: {file_to_delete}')
             except Exception as e:
-                print(f'删除 {file_to_delete} 时出错: {str(e)}')
-
+                print(f'Error deleting {file_to_delete}: {str(e)}')
 api = HfApi(token='$token')
 try:
     api.upload_file(
@@ -42,34 +41,36 @@ try:
         repo_id='$repo_id',
         repo_type='dataset'
     )
-    print(f'成功上传 $file_name')
+    print(f'Successfully uploaded $file_name')
+    
     manage_backups(api, '$repo_id')
 except Exception as e:
-    print(f'上传文件时出错: {str(e)}')
+    print(f'Error uploading file: {str(e)}')
 "
 }
 
 # 下载最新备份
 download_latest_backup() {
-  token="$HF_TOKEN"
-  repo_id="$DATASET_ID"
+    token="$HF_TOKEN"
+    repo_id="$DATASET_ID"
 
-  python3 -c "
+    python3 -c "
 from huggingface_hub import HfApi
 import sys
 import os
 import tarfile
 import tempfile
-
 api = HfApi(token='$token')
 try:
     files = api.list_repo_files(repo_id='$repo_id', repo_type='dataset')
     backup_files = [f for f in files if f.startswith('qinglong_backup_') and f.endswith('.tar.gz')]
+    
     if not backup_files:
-        print('未找到备份文件')
+        print('No backup files found')
         sys.exit()
-    
+        
     latest_backup = sorted(backup_files)[-1]
+    
     with tempfile.TemporaryDirectory() as temp_dir:
         filepath = api.hf_hub_download(
             repo_id='$repo_id',
@@ -81,67 +82,41 @@ try:
         if filepath and os.path.exists(filepath):
             with tarfile.open(filepath, 'r:gz') as tar:
                 tar.extractall('/ql/data')
-            print(f'成功从 {latest_backup} 恢复备份')
+            print(f'Successfully restored backup from {latest_backup}')
+        
 except Exception as e:
-    print(f'下载备份时出错: {str(e)}')
+    print(f'Error downloading backup: {str(e)}')
 "
 }
 
-# 清理本地备份文件
-cleanup_local_backups() {
-  local_backup_dir="/tmp"
-  max_local_backups=5
-  
-  echo "正在清理本地备份文件..."
-  
-  # 列出本地备份文件并按时间排序
-  backup_files=$(find ${local_backup_dir} -name "qinglong_backup_*.tar.gz" -type f | sort)
-  
-  # 计算要删除的文件数量
-  file_count=$(echo "$backup_files" | wc -l)
-  files_to_delete=$((file_count - max_local_backups))
-  
-  # 如果需要删除文件
-  if [ $files_to_delete -gt 0 ]; then
-    echo "发现 $file_count 个备份文件，正在删除 $files_to_delete 个旧文件"
-    echo "$backup_files" | head -n $files_to_delete | xargs rm -f
-    echo "本地清理完成"
-  else
-    echo "无需删除本地备份文件"
-  fi
-}
-
 # 首次启动时下载最新备份
-echo "正在从 HuggingFace 下载最新备份..."
+echo "Downloading latest backup from HuggingFace..."
 download_latest_backup
 
 # 同步函数
 sync_data() {
-  while true; do
-    echo "开始同步进程，当前时间 $(date)"
-    if [ -d /ql/data ]; then
-      timestamp=$(date +%Y%m%d_%H%M%S)
-      backup_file="qinglong_backup_${timestamp}.tar.gz"
-      
-      # 压缩数据目录
-      tar -czf "/tmp/${backup_file}" -C /ql/data .
-      echo "正在上传备份到 HuggingFace..."
-      upload_backup "/tmp/${backup_file}" "${backup_file}"
-      
-      # 清理本地备份文件
-      cleanup_local_backups
-      
-      # 删除当前备份文件
-      rm -f "/tmp/${backup_file}"
-      echo "备份完成并清理了临时文件"
-    else
-      echo "数据目录尚不存在，等待下次同步..."
-    fi
-    
-    SYNC_INTERVAL=${SYNC_INTERVAL:-7200}
-    echo "下次同步将在 ${SYNC_INTERVAL} 秒后进行..."
-    sleep $SYNC_INTERVAL
-  done
+    while true; do
+        echo "Starting sync process at $(date)"
+        
+        if [ -d /ql/data ]; then
+            timestamp=$(date +%Y%m%d_%H%M%S)
+            backup_file="qinglong_backup_${timestamp}.tar.gz"
+            
+            # 压缩数据目录
+            tar -czf "/tmp/${backup_file}" -C /ql/data .
+            
+            echo "Uploading backup to HuggingFace..."
+            upload_backup "/tmp/${backup_file}" "${backup_file}"
+            
+            rm -f "/tmp/${backup_file}"
+        else
+            echo "Data directory does not exist yet, waiting for next sync..."
+        fi
+        
+        SYNC_INTERVAL=${SYNC_INTERVAL:-7200}
+        echo "Next sync in ${SYNC_INTERVAL} seconds..."
+        sleep $SYNC_INTERVAL
+    done
 }
 
 # 启动同步进程