X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/6e76e3322d66f609dabcd34c98cba34bd739e089..446e64dc143148e52a126fa502bf63299c94197e:/sdk/python/arvados/api.py

diff --git a/sdk/python/arvados/api.py b/sdk/python/arvados/api.py
index 5ec5ac2e8a..e69f1a112d 100644
--- a/sdk/python/arvados/api.py
+++ b/sdk/python/arvados/api.py
@@ -1,21 +1,37 @@
+# Copyright (C) The Arvados Authors. All rights reserved.
+#
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import absolute_import
+from future import standard_library
+standard_library.install_aliases()
+from builtins import range
 import collections
+import http.client
 import httplib2
 import json
 import logging
 import os
 import re
 import socket
+import time
 import types
 
 import apiclient
 from apiclient import discovery as apiclient_discovery
 from apiclient import errors as apiclient_errors
-import config
-import errors
-import util
+from . import config
+from . import errors
+from . import util
+from . import cache
 
 _logger = logging.getLogger('arvados.api')
 
+MAX_IDLE_CONNECTION_DURATION = 30
+RETRY_DELAY_INITIAL = 2
+RETRY_DELAY_BACKOFF = 2
+RETRY_COUNT = 2
+
 class OrderedJsonModel(apiclient.model.JsonModel):
     """Model class for JSON that preserves the contents' order.
 
@@ -36,9 +52,7 @@ class OrderedJsonModel(apiclient.model.JsonModel):
         return body
 
 
-def _intercept_http_request(self, uri, **kwargs):
-    from httplib import BadStatusLine
-
+def _intercept_http_request(self, uri, method="GET", **kwargs):
     if (self.max_request_size and
         kwargs.get('body') and
         self.max_request_size < len(kwargs['body'])):
@@ -51,32 +65,54 @@ def _intercept_http_request(self, uri, **kwargs):
         kwargs['headers']['X-External-Client'] = '1'
 
     kwargs['headers']['Authorization'] = 'OAuth2 %s' % self.arvados_api_token
-    try:
-        return self.orig_http_request(uri, **kwargs)
-    except BadStatusLine:
-        # This is how httplib tells us that it tried to reuse an
-        # existing connection but it was already closed by the
-        # server. In that case, yes, we would like to retry.
-        # Unfortunately, we are not absolutely certain that the
-        # previous call did not succeed, so this is slightly
-        # risky.
-        return self.orig_http_request(uri, **kwargs)
-    except socket.error:
-        # This is the one case where httplib2 doesn't close the
-        # underlying connection first.  Close all open connections,
-        # expecting this object only has the one connection to the API
-        # server.  This is safe because httplib2 reopens connections when
-        # needed.
-        _logger.debug("Retrying API request after socket error", exc_info=True)
-        for conn in self.connections.itervalues():
+
+    retryable = method in [
+        'DELETE', 'GET', 'HEAD', 'OPTIONS', 'PUT']
+    retry_count = self._retry_count if retryable else 0
+
+    if (not retryable and
+        time.time() - self._last_request_time > self._max_keepalive_idle):
+        # High probability of failure due to connection atrophy. Make
+        # sure this request [re]opens a new connection by closing and
+        # forgetting all cached connections first.
+        for conn in self.connections.values():
             conn.close()
-        return self.orig_http_request(uri, **kwargs)
+        self.connections.clear()
+
+    delay = self._retry_delay_initial
+    for _ in range(retry_count):
+        self._last_request_time = time.time()
+        try:
+            return self.orig_http_request(uri, method, **kwargs)
+        except http.client.HTTPException:
+            _logger.debug("Retrying API request in %d s after HTTP error",
+                          delay, exc_info=True)
+        except socket.error:
+            # This is the one case where httplib2 doesn't close the
+            # underlying connection first.  Close all open
+            # connections, expecting this object only has the one
+            # connection to the API server.  This is safe because
+            # httplib2 reopens connections when needed.
+            _logger.debug("Retrying API request in %d s after socket error",
+                          delay, exc_info=True)
+            for conn in self.connections.values():
+                conn.close()
+        time.sleep(delay)
+        delay = delay * self._retry_delay_backoff
+
+    self._last_request_time = time.time()
+    return self.orig_http_request(uri, method, **kwargs)
 
 def _patch_http_request(http, api_token):
     http.arvados_api_token = api_token
     http.max_request_size = 0
     http.orig_http_request = http.request
     http.request = types.MethodType(_intercept_http_request, http)
+    http._last_request_time = 0
+    http._max_keepalive_idle = MAX_IDLE_CONNECTION_DURATION
+    http._retry_delay_initial = RETRY_DELAY_INITIAL
+    http._retry_delay_backoff = RETRY_DELAY_BACKOFF
+    http._retry_count = RETRY_COUNT
     return http
 
 # Monkey patch discovery._cast() so objects and arrays get serialized
@@ -85,6 +121,7 @@ _cast_orig = apiclient_discovery._cast
 def _cast_objects_too(value, schema_type):
     global _cast_orig
     if (type(value) != type('') and
+        type(value) != type(b'') and
         (schema_type == 'object' or schema_type == 'array')):
         return json.dumps(value)
     else:
@@ -108,8 +145,8 @@ def http_cache(data_type):
     try:
         util.mkdir_dash_p(path)
     except OSError:
-        path = None
-    return path
+        return None
+    return cache.SafeHTTPCache(path, max_age=60*60*24*2)
 
 def api(version=None, cache=True, host=None, token=None, insecure=False, **kwargs):
     """Return an apiclient Resources object for an Arvados instance.
@@ -178,8 +215,9 @@ def api(version=None, cache=True, host=None, token=None, insecure=False, **kwarg
 
     kwargs['http'] = _patch_http_request(kwargs['http'], token)
 
-    svc = apiclient_discovery.build('arvados', version, **kwargs)
+    svc = apiclient_discovery.build('arvados', version, cache_discovery=False, **kwargs)
     svc.api_token = token
+    svc.insecure = insecure
     kwargs['http'].max_request_size = svc._rootDesc.get('maxRequestSize', 0)
     kwargs['http'].cache = None
     return svc