Retry non-streaming gRPC calls

appengine/swarming/swarming_bot/bot_code/remote_client_grpc.py

Index: appengine/swarming/swarming_bot/bot_code/remote_client_grpc.py
diff --git a/appengine/swarming/swarming_bot/bot_code/remote_client_grpc.py b/appengine/swarming/swarming_bot/bot_code/remote_client_grpc.py
index 3ef861d0017d73dc99af579aa4cbc7c56ac65bf7..4bcfb736690b9350ec0f37b3318deb4dff9c7548 100644
--- a/appengine/swarming/swarming_bot/bot_code/remote_client_grpc.py
+++ b/appengine/swarming/swarming_bot/bot_code/remote_client_grpc.py
@@ -7,6 +7,9 @@
 
 import json
 import logging
+import math
+import random
+import time
 
 import grpc
 import google.protobuf.json_format
@@ -19,6 +22,14 @@ from remote_client_errors import InternalError
 NET_CONNECTION_TIMEOUT_SEC = 5*60
 
 
+# How many times to retry a gRPC call
+MAX_GRPC_ATTEMPTS = 30
+
+
+# Longest time to sleep between gRPC calls
+MAX_GRPC_SLEEP = 10.
+
+
 class RemoteClientGrpc(object):
   """RemoteClientGrpc knows how to make calls via gRPC.
   """
@@ -67,8 +78,7 @@ class RemoteClientGrpc(object):
     google.protobuf.json_format.ParseDict(params, request)
 
     # Perform update
-    response = self._stub.TaskUpdate(request,
-                                     timeout=NET_CONNECTION_TIMEOUT_SEC)
+    response = call_grpc(self._stub.TaskUpdate, request)
     logging.debug('post_task_update() = %s', request)
     if response.error:
       raise InternalError(response.error)
@@ -81,7 +91,7 @@ class RemoteClientGrpc(object):
     request.msg = message
     logging.error('post_task_error() = %s', request)
 
-    response = self._stub.TaskError(request, timeout=NET_CONNECTION_TIMEOUT_SEC)
+    response = call_grpc(self._stub.TaskError, request)
     return response.ok
 
   def _attributes_json_to_proto(self, json_attr, msg):
@@ -95,7 +105,7 @@ class RemoteClientGrpc(object):
   def do_handshake(self, attributes):
     request = swarming_bot_pb2.HandshakeRequest()
     self._attributes_json_to_proto(attributes, request.attributes)
-    response = self._stub.Handshake(request, timeout=NET_CONNECTION_TIMEOUT_SEC)
+    response = call_grpc(self._stub.Handshake, request)
     resp = {
         'server_version': response.server_version,
         'bot_version': response.bot_version,
@@ -113,7 +123,7 @@ class RemoteClientGrpc(object):
     request = swarming_bot_pb2.PollRequest()
     self._attributes_json_to_proto(attributes, request.attributes)
     # TODO(aludwin): gRPC-specific exception handling
-    response = self._stub.Poll(request, timeout=NET_CONNECTION_TIMEOUT_SEC)
+    response = call_grpc(self._stub.Poll, request)
 
     if response.cmd == swarming_bot_pb2.PollResponse.UPDATE:
       return 'update', response.version
@@ -173,7 +183,7 @@ class RemoteClientGrpc(object):
     logging.info('Updating to version: %s', bot_version)
     request = swarming_bot_pb2.BotUpdateRequest()
     request.bot_version = bot_version
-    response = self._stub.BotUpdate(request, timeout=NET_CONNECTION_TIMEOUT_SEC)
+    response = call_grpc(self._stub.BotUpdate, request)
     with open(new_zip_fn, 'wb') as f:
       f.write(response.bot_code)
 
@@ -221,3 +231,26 @@ def insert_dict_as_submessage(message, keyname, value):
   """
   sub_msg = getattr(message, keyname)
   google.protobuf.json_format.Parse(json.dumps(value), sub_msg)
+
+
+def call_grpc(method, request):
+  """Retries a command a set number of times"""
+  num_attempts = 0
+  while True:
+    try:
+      num_attempts += 1
+      return method(request, timeout=NET_CONNECTION_TIMEOUT_SEC)
+    except grpc.RpcError as rpc_error:
+      logging.warning('call_grpc - gRPC error: %s', str(rpc_error))
+      if (rpc_error.code() is grpc.StatusCode.UNAVAILABLE
+          and num_attempts < MAX_GRPC_ATTEMPTS):
+        logging.warning('Swallowing UNAVAILABLE error (attempt %d/%d)',
+                        num_attempts, MAX_GRPC_ATTEMPTS)
+        # random.random() returns [0.0, 1.0). Starts with relatively short

[M-A Ruel, 2016/12/20 22:05:55]

Please call net.calculate_sleep_before_retry(num_attempts, 10.) here so we don't
duplicate this code and if we want to update the retry mechanism we can update
all locations at once.

+        # waiting time by starting with 1.5/2+1.5^-1 median offset.
+        duration = (random.random() * 1.5) + math.pow(1.5, (num_attempts - 1))
+        assert duration > 0.1
+        duration = min(MAX_GRPC_SLEEP, duration)
+        time.sleep(duration)
+      else:
+        raise