Change os.uname()[1] and socket.gethostname() to the portable and faster platform.node_ip() (#8839)

Co-authored-by: Mehrdad <noreply@github.com>

python/ray/dashboard/dashboard.py

@@ -12,7 +12,7 @@ import errno
 import json
 import logging
 import os
-import socket
+import platform
 import threading
 import time
 import traceback
@@ -965,7 +965,7 @@ if __name__ == "__main__":
             args.redis_address, password=args.redis_password)
         traceback_str = ray.utils.format_error_message(traceback.format_exc())
         message = ("The dashboard on node {} failed with the following "
-                   "error:\n{}".format(socket.gethostname(), traceback_str))
+                   "error:\n{}".format(platform.node(), traceback_str))
         ray.utils.push_error_to_driver_through_redis(
             redis_client, ray_constants.DASHBOARD_DIED_ERROR, message)
         if isinstance(e, OSError) and e.errno == errno.ENOENT:

python/ray/log_monitor.py

@@ -4,8 +4,8 @@ import glob
 import json
 import logging
 import os
+import platform
 import shutil
-import socket
 import time
 import traceback
 
@@ -302,7 +302,7 @@ if __name__ == "__main__":
             args.redis_address, password=args.redis_password)
         traceback_str = ray.utils.format_error_message(traceback.format_exc())
         message = ("The log monitor on node {} failed with the following "
-                   "error:\n{}".format(socket.gethostname(), traceback_str))
+                   "error:\n{}".format(platform.node(), traceback_str))
         ray.utils.push_error_to_driver_through_redis(
             redis_client, ray_constants.LOG_MONITOR_DIED_ERROR, message)
         raise e

python/ray/memory_monitor.py

@@ -1,5 +1,6 @@
 import logging
 import os
+import platform
 import sys
 import time
 
@@ -47,7 +48,7 @@ class RayOutOfMemoryError(Exception):
                 " ".join(cmdline)[:100].strip())
         return ("More than {}% of the memory on ".format(int(
             100 * threshold)) + "node {} is used ({} / {} GB). ".format(
-                os.uname()[1], round(used_gb, 2), round(total_gb, 2)) +
+                platform.node(), round(used_gb, 2), round(total_gb, 2)) +
                 "The top 10 memory consumers are:\n\n{}".format(proc_str) +
                 "\n\nIn addition, up to {} GiB of shared memory is ".format(
                     round(get_shared(psutil.virtual_memory()) / (1024**3), 2))

python/ray/reporter.py

@@ -6,7 +6,7 @@ import traceback
 import time
 import datetime
 import grpc
-import socket
+import platform
 import subprocess
 import sys
 from concurrent import futures
@@ -92,7 +92,7 @@ class Reporter:
         """Initialize the reporter object."""
         self.cpu_counts = (psutil.cpu_count(), psutil.cpu_count(logical=False))
         self.ip = ray.services.get_node_ip_address()
-        self.hostname = socket.gethostname()
+        self.hostname = platform.node()
 
         _ = psutil.cpu_percent()  # For initialization
 
@@ -252,7 +252,7 @@ if __name__ == "__main__":
             args.redis_address, password=args.redis_password)
         traceback_str = ray.utils.format_error_message(traceback.format_exc())
         message = ("The reporter on node {} failed with the following "
-                   "error:\n{}".format(socket.gethostname(), traceback_str))
+                   "error:\n{}".format(platform.node(), traceback_str))
         ray.utils.push_error_to_driver_through_redis(
             redis_client, ray_constants.REPORTER_DIED_ERROR, message)
         raise e

python/ray/tune/trainable.py

@@ -6,6 +6,7 @@ import logging
 import glob
 import os
 import pickle
+import platform
 import pandas as pd
 from six import string_types
 import shutil
@@ -308,7 +309,7 @@ class Trainable:
             time_this_iter_s=time_this_iter,
             time_total_s=self._time_total,
             pid=os.getpid(),
-            hostname=os.uname()[1],
+            hostname=platform.node(),
             node_ip=self._local_ip,
             config=self.config,
             time_since_restore=self._time_since_restore,

python/ray/tune/trial.py

@@ -3,6 +3,7 @@ from collections import deque
 import copy
 from datetime import datetime
 import logging
+import platform
 import shutil
 import uuid
 import time
@@ -42,7 +43,7 @@ class Location:
     def __str__(self):
         if not self.pid:
             return ""
-        elif self.hostname == os.uname()[1]:
+        elif self.hostname == platform.node():
             return "pid={}".format(self.pid)
         else:
             return "{}:{}".format(self.hostname, self.pid)