我在HDFS文件夹中有15 k个不同日期的文件。我正在尝试删除超过10天的文件，其中我有10 k个文件要从HDFS中删除。当我试图通过传递到列表来一次删除这些10 k个文件时，我遇到了错误。您能帮助我解决这个问题吗？我尝试使用find和xargs，但它接受单个文件，而不是多个文件。

delete_cmd = 'hadoop fs -find /test/folder/* -name ' +filepath+ ' | xargs hadoop fs -rm -r -skipTrash'

OSError：[Errno 7]参数列表太长：'/bin/sh'

folder = '/test/folder/*'
no_of_days = 10
now_dt = datetime.now()

def list_files():
  hdfs_cmd = "hdfs dfs -ls -r " + folder  + " | grep / " 
  hdfs_output = subprocess.getoutput(hdfs_cmd)
  return hdfs_output

def delete_cmd(filepath):
  delete_cmd = 'hdfs dfs -rm -r -skipTrash '  + filepath 
  delete_output = subprocess.getoutput(delete_cmd)
  print(delete_output)

def delete_olderfiles():
  hdfs_outputfun = list_files()
  delete_list = []
  filepath_list = [line.split(' ')[-1] for line in hdfs_outputfun.splitlines()]
  for filepath in filepath_list:
        filename = filepath.split('/')[-1]
        filename_dt = datetime.strptime(filename, '%Y%m%d.csv')
    diff_days = (now_dt - filename_dt).days
    if diff_days > no_of_days:
        delete_list.append(filepath)
 string = " " 
 delete_folders = string.join(delete_list)
 delete_cmd(delete_folders)

delete_olderfiles()

删除文件的输出：hdfs dfs -rm -r -跳过垃圾文件20210901.csv文件20210801.csv文件20210903.csv文件...10k个文件