Merge branch 'develop' into release-v0.27.0

20 files changed, 88 insertions, 47 deletions

diff --git a/CHANGES.rst b/CHANGES.rst
index 9f974e6875..df469a9930 100644
--- a/CHANGES.rst
+++ b/CHANGES.rst
@@ -105,7 +105,7 @@ Features:
 
 Changes:
 
-* Continue to factor out processing from main process and into worker processes. See updated `docs/workers.rst <docs/metrics-howto.rst>`_ (PR #2892 - #2904, #2913, #2920 - #2926, #2947, #2847, #2854, #2872, #2873, #2874, #2928, #2929, #2934, #2856, #2976 - #2984, #2987 - #2989, #2991 - #2993, #2995, #2784)
+* Continue to factor out processing from main process and into worker processes. See updated `docs/workers.rst <docs/workers.rst>`_ (PR #2892 - #2904, #2913, #2920 - #2926, #2947, #2847, #2854, #2872, #2873, #2874, #2928, #2929, #2934, #2856, #2976 - #2984, #2987 - #2989, #2991 - #2993, #2995, #2784)
 * Ensure state cache is used when persisting events (PR #2864, #2871, #2802, #2835, #2836, #2841, #2842, #2849)
 * Change the default config to bind on both IPv4 and IPv6 on all platforms (PR #2435) Thanks to @silkeh!
 * No longer require a specific version of saml2 (PR #2695) Thanks to @okurz!
diff --git a/synapse/app/appservice.py b/synapse/app/appservice.py
index c6fe4516d1..f2540023a7 100644
--- a/synapse/app/appservice.py
+++ b/synapse/app/appservice.py
@@ -36,7 +36,7 @@ from synapse.util.logcontext import LoggingContext, preserve_fn
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
 from twisted.internet import reactor
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 logger = logging.getLogger("synapse.app.appservice")
 
@@ -64,7 +64,7 @@ class AppserviceServer(HomeServer):
                 if name == "metrics":
                     resources[METRICS_PREFIX] = MetricsResource(self)
 
-        root_resource = create_resource_tree(resources, Resource())
+        root_resource = create_resource_tree(resources, NoResource())
 
         _base.listen_tcp(
             bind_addresses,
diff --git a/synapse/app/client_reader.py b/synapse/app/client_reader.py
index 0a8ce9bc66..267d34c881 100644
--- a/synapse/app/client_reader.py
+++ b/synapse/app/client_reader.py
@@ -44,7 +44,7 @@ from synapse.util.logcontext import LoggingContext
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
 from twisted.internet import reactor
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 logger = logging.getLogger("synapse.app.client_reader")
 
@@ -88,7 +88,7 @@ class ClientReaderServer(HomeServer):
                         "/_matrix/client/api/v1": resource,
                     })
 
-        root_resource = create_resource_tree(resources, Resource())
+        root_resource = create_resource_tree(resources, NoResource())
 
         _base.listen_tcp(
             bind_addresses,
diff --git a/synapse/app/event_creator.py b/synapse/app/event_creator.py
index 172e989b54..b915d12d53 100644
--- a/synapse/app/event_creator.py
+++ b/synapse/app/event_creator.py
@@ -52,7 +52,7 @@ from synapse.util.logcontext import LoggingContext
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
 from twisted.internet import reactor
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 logger = logging.getLogger("synapse.app.event_creator")
 
@@ -104,7 +104,7 @@ class EventCreatorServer(HomeServer):
                         "/_matrix/client/api/v1": resource,
                     })
 
-        root_resource = create_resource_tree(resources, Resource())
+        root_resource = create_resource_tree(resources, NoResource())
 
         _base.listen_tcp(
             bind_addresses,
diff --git a/synapse/app/federation_reader.py b/synapse/app/federation_reader.py
index 20d157911b..c1dc66dd17 100644
--- a/synapse/app/federation_reader.py
+++ b/synapse/app/federation_reader.py
@@ -41,7 +41,7 @@ from synapse.util.logcontext import LoggingContext
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
 from twisted.internet import reactor
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 logger = logging.getLogger("synapse.app.federation_reader")
 
@@ -77,7 +77,7 @@ class FederationReaderServer(HomeServer):
                         FEDERATION_PREFIX: TransportLayerServer(self),
                     })
 
-        root_resource = create_resource_tree(resources, Resource())
+        root_resource = create_resource_tree(resources, NoResource())
 
         _base.listen_tcp(
             bind_addresses,
diff --git a/synapse/app/federation_sender.py b/synapse/app/federation_sender.py
index f760826d27..0cc3331519 100644
--- a/synapse/app/federation_sender.py
+++ b/synapse/app/federation_sender.py
@@ -42,7 +42,7 @@ from synapse.util.logcontext import LoggingContext, preserve_fn
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
 from twisted.internet import defer, reactor
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 logger = logging.getLogger("synapse.app.federation_sender")
 
@@ -91,7 +91,7 @@ class FederationSenderServer(HomeServer):
                 if name == "metrics":
                     resources[METRICS_PREFIX] = MetricsResource(self)
 
-        root_resource = create_resource_tree(resources, Resource())
+        root_resource = create_resource_tree(resources, NoResource())
 
         _base.listen_tcp(
             bind_addresses,
diff --git a/synapse/app/frontend_proxy.py b/synapse/app/frontend_proxy.py
index 816c080d18..de889357c3 100644
--- a/synapse/app/frontend_proxy.py
+++ b/synapse/app/frontend_proxy.py
@@ -44,7 +44,7 @@ from synapse.util.logcontext import LoggingContext
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
 from twisted.internet import defer, reactor
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 logger = logging.getLogger("synapse.app.frontend_proxy")
 
@@ -142,7 +142,7 @@ class FrontendProxyServer(HomeServer):
                         "/_matrix/client/api/v1": resource,
                     })
 
-        root_resource = create_resource_tree(resources, Resource())
+        root_resource = create_resource_tree(resources, NoResource())
 
         _base.listen_tcp(
             bind_addresses,
diff --git a/synapse/app/homeserver.py b/synapse/app/homeserver.py
index e477c7ced6..c00afbba28 100755
--- a/synapse/app/homeserver.py
+++ b/synapse/app/homeserver.py
@@ -56,7 +56,7 @@ from synapse.util.rlimit import change_resource_limit
 from synapse.util.versionstring import get_version_string
 from twisted.application import service
 from twisted.internet import defer, reactor
-from twisted.web.resource import EncodingResourceWrapper, Resource
+from twisted.web.resource import EncodingResourceWrapper, NoResource
 from twisted.web.server import GzipEncoderFactory
 from twisted.web.static import File
 
@@ -126,7 +126,7 @@ class SynapseHomeServer(HomeServer):
         if WEB_CLIENT_PREFIX in resources:
             root_resource = RootRedirect(WEB_CLIENT_PREFIX)
         else:
-            root_resource = Resource()
+            root_resource = NoResource()
 
         root_resource = create_resource_tree(resources, root_resource)
 
diff --git a/synapse/app/media_repository.py b/synapse/app/media_repository.py
index 84c5791b3b..fc8282bbc1 100644
--- a/synapse/app/media_repository.py
+++ b/synapse/app/media_repository.py
@@ -43,7 +43,7 @@ from synapse.util.logcontext import LoggingContext
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
 from twisted.internet import reactor
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 logger = logging.getLogger("synapse.app.media_repository")
 
@@ -84,7 +84,7 @@ class MediaRepositoryServer(HomeServer):
                         ),
                     })
 
-        root_resource = create_resource_tree(resources, Resource())
+        root_resource = create_resource_tree(resources, NoResource())
 
         _base.listen_tcp(
             bind_addresses,
diff --git a/synapse/app/pusher.py b/synapse/app/pusher.py
index 98a4a7c62c..d5c3a85195 100644
--- a/synapse/app/pusher.py
+++ b/synapse/app/pusher.py
@@ -37,7 +37,7 @@ from synapse.util.logcontext import LoggingContext, preserve_fn
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
 from twisted.internet import defer, reactor
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 logger = logging.getLogger("synapse.app.pusher")
 
@@ -94,7 +94,7 @@ class PusherServer(HomeServer):
                 if name == "metrics":
                     resources[METRICS_PREFIX] = MetricsResource(self)
 
-        root_resource = create_resource_tree(resources, Resource())
+        root_resource = create_resource_tree(resources, NoResource())
 
         _base.listen_tcp(
             bind_addresses,
diff --git a/synapse/app/synchrotron.py b/synapse/app/synchrotron.py
index abe91dcfbd..508b66613d 100644
--- a/synapse/app/synchrotron.py
+++ b/synapse/app/synchrotron.py
@@ -56,7 +56,7 @@ from synapse.util.manhole import manhole
 from synapse.util.stringutils import random_string
 from synapse.util.versionstring import get_version_string
 from twisted.internet import defer, reactor
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 logger = logging.getLogger("synapse.app.synchrotron")
 
@@ -269,7 +269,7 @@ class SynchrotronServer(HomeServer):
                         "/_matrix/client/api/v1": resource,
                     })
 
-        root_resource = create_resource_tree(resources, Resource())
+        root_resource = create_resource_tree(resources, NoResource())
 
         _base.listen_tcp(
             bind_addresses,
diff --git a/synapse/app/user_dir.py b/synapse/app/user_dir.py
index 494ccb702c..5f845e80d1 100644
--- a/synapse/app/user_dir.py
+++ b/synapse/app/user_dir.py
@@ -43,7 +43,7 @@ from synapse.util.logcontext import LoggingContext, preserve_fn
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
 from twisted.internet import reactor
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 logger = logging.getLogger("synapse.app.user_dir")
 
@@ -116,7 +116,7 @@ class UserDirectoryServer(HomeServer):
                         "/_matrix/client/api/v1": resource,
                     })
 
-        root_resource = create_resource_tree(resources, Resource())
+        root_resource = create_resource_tree(resources, NoResource())
 
         _base.listen_tcp(
             bind_addresses,
diff --git a/synapse/handlers/register.py b/synapse/handlers/register.py
index ed5939880a..dd03705279 100644
--- a/synapse/handlers/register.py
+++ b/synapse/handlers/register.py
@@ -24,7 +24,7 @@ from synapse.api.errors import (
 from synapse.http.client import CaptchaServerHttpClient
 from synapse import types
 from synapse.types import UserID
-from synapse.util.async import run_on_reactor
+from synapse.util.async import run_on_reactor, Linearizer
 from synapse.util.threepids import check_3pid_allowed
 from ._base import BaseHandler
 
@@ -46,6 +46,10 @@ class RegistrationHandler(BaseHandler):
 
         self.macaroon_gen = hs.get_macaroon_generator()
 
+        self._generate_user_id_linearizer = Linearizer(
+            name="_generate_user_id_linearizer",
+        )
+
     @defer.inlineCallbacks
     def check_username(self, localpart, guest_access_token=None,
                        assigned_user_id=None):
@@ -345,9 +349,11 @@ class RegistrationHandler(BaseHandler):
     @defer.inlineCallbacks
     def _generate_user_id(self, reseed=False):
         if reseed or self._next_generated_user_id is None:
-            self._next_generated_user_id = (
-                yield self.store.find_next_generated_user_id_localpart()
-            )
+            with (yield self._generate_user_id_linearizer.queue(())):
+                if reseed or self._next_generated_user_id is None:
+                    self._next_generated_user_id = (
+                        yield self.store.find_next_generated_user_id_localpart()
+                    )
 
         id = self._next_generated_user_id
         self._next_generated_user_id += 1
diff --git a/synapse/http/server.py b/synapse/http/server.py
index 1551db239d..f19c068ef6 100644
--- a/synapse/http/server.py
+++ b/synapse/http/server.py
@@ -488,6 +488,7 @@ def respond_with_json_bytes(request, code, json_bytes, send_cors=False,
     request.setHeader(b"Content-Type", b"application/json")
     request.setHeader(b"Server", version_string)
     request.setHeader(b"Content-Length", b"%d" % (len(json_bytes),))
+    request.setHeader(b"Cache-Control", b"no-cache, no-store, must-revalidate")
 
     if send_cors:
         set_cors_headers(request)
diff --git a/synapse/python_dependencies.py b/synapse/python_dependencies.py
index 91179ce532..40eedb63cb 100644
--- a/synapse/python_dependencies.py
+++ b/synapse/python_dependencies.py
@@ -34,7 +34,6 @@ REQUIREMENTS = {
     "bcrypt": ["bcrypt>=3.1.0"],
     "pillow": ["PIL"],
     "pydenticon": ["pydenticon"],
-    "ujson": ["ujson"],
     "blist": ["blist"],
     "pysaml2>=3.0.0": ["saml2>=3.0.0"],
     "pymacaroons-pynacl": ["pymacaroons"],
diff --git a/synapse/state.py b/synapse/state.py
index 932f602508..a7f20350f1 100644
--- a/synapse/state.py
+++ b/synapse/state.py
@@ -132,7 +132,7 @@ class StateHandler(object):
 
         state_map = yield self.store.get_events(state.values(), get_prev_content=False)
         state = {
-            key: state_map[e_id] for key, e_id in state.items() if e_id in state_map
+            key: state_map[e_id] for key, e_id in state.iteritems() if e_id in state_map
         }
 
         defer.returnValue(state)
@@ -378,7 +378,7 @@ class StateHandler(object):
             new_state = resolve_events_with_state_map(state_set_ids, state_map)
 
         new_state = {
-            key: state_map[ev_id] for key, ev_id in new_state.items()
+            key: state_map[ev_id] for key, ev_id in new_state.iteritems()
         }
 
         return new_state
@@ -458,15 +458,15 @@ class StateResolutionHandler(object):
             # build a map from state key to the event_ids which set that state.
             # dict[(str, str), set[str])
             state = {}
-            for st in state_groups_ids.values():
-                for key, e_id in st.items():
+            for st in state_groups_ids.itervalues():
+                for key, e_id in st.iteritems():
                     state.setdefault(key, set()).add(e_id)
 
             # build a map from state key to the event_ids which set that state,
             # including only those where there are state keys in conflict.
             conflicted_state = {
                 k: list(v)
-                for k, v in state.items()
+                for k, v in state.iteritems()
                 if len(v) > 1
             }
 
@@ -480,7 +480,7 @@ class StateResolutionHandler(object):
                     )
             else:
                 new_state = {
-                    key: e_ids.pop() for key, e_ids in state.items()
+                    key: e_ids.pop() for key, e_ids in state.iteritems()
                 }
 
             # if the new state matches any of the input state groups, we can
@@ -488,8 +488,8 @@ class StateResolutionHandler(object):
             # which will be used as a cache key for future resolutions, but
             # not get persisted.
             state_group = None
-            new_state_event_ids = frozenset(new_state.values())
-            for sg, events in state_groups_ids.items():
+            new_state_event_ids = frozenset(new_state.itervalues())
+            for sg, events in state_groups_ids.iteritems():
                 if new_state_event_ids == frozenset(e_id for e_id in events):
                     state_group = sg
                     break
@@ -702,7 +702,7 @@ def _resolve_with_state(unconflicted_state_ids, conflicted_state_ds, auth_event_
 
     auth_events = {
         key: state_map[ev_id]
-        for key, ev_id in auth_event_ids.items()
+        for key, ev_id in auth_event_ids.iteritems()
         if ev_id in state_map
     }
 
@@ -740,7 +740,7 @@ def _resolve_state_events(conflicted_state, auth_events):
 
     auth_events.update(resolved_state)
 
-    for key, events in conflicted_state.items():
+    for key, events in conflicted_state.iteritems():
         if key[0] == EventTypes.JoinRules:
             logger.debug("Resolving conflicted join rules %r", events)
             resolved_state[key] = _resolve_auth_events(
@@ -750,7 +750,7 @@ def _resolve_state_events(conflicted_state, auth_events):
 
     auth_events.update(resolved_state)
 
-    for key, events in conflicted_state.items():
+    for key, events in conflicted_state.iteritems():
         if key[0] == EventTypes.Member:
             logger.debug("Resolving conflicted member lists %r", events)
             resolved_state[key] = _resolve_auth_events(
@@ -760,7 +760,7 @@ def _resolve_state_events(conflicted_state, auth_events):
 
     auth_events.update(resolved_state)
 
-    for key, events in conflicted_state.items():
+    for key, events in conflicted_state.iteritems():
         if key not in resolved_state:
             logger.debug("Resolving conflicted state %r:%r", key, events)
             resolved_state[key] = _resolve_normal_events(
diff --git a/synapse/storage/events.py b/synapse/storage/events.py
index 85ce6bea1a..f3d65f4338 100644
--- a/synapse/storage/events.py
+++ b/synapse/storage/events.py
@@ -53,6 +53,22 @@ event_counter = metrics.register_counter(
     "persisted_events_sep", labels=["type", "origin_type", "origin_entity"]
 )
 
+# The number of times we are recalculating the current state
+state_delta_counter = metrics.register_counter(
+    "state_delta",
+)
+# The number of times we are recalculating state when there is only a
+# single forward extremity
+state_delta_single_event_counter = metrics.register_counter(
+    "state_delta_single_event",
+)
+# The number of times we are reculating state when we could have resonably
+# calculated the delta when we calculated the state for an event we were
+# persisting.
+state_delta_reuse_delta_counter = metrics.register_counter(
+    "state_delta_reuse_delta",
+)
+
 
 def encode_json(json_object):
     if USE_FROZEN_DICTS:
@@ -368,7 +384,8 @@ class EventsStore(EventsWorkerStore):
                                 room_id, ev_ctx_rm, latest_event_ids
                             )
 
-                            if new_latest_event_ids == set(latest_event_ids):
+                            latest_event_ids = set(latest_event_ids)
+                            if new_latest_event_ids == latest_event_ids:
                                 # No change in extremities, so no change in state
                                 continue
 
@@ -389,6 +406,26 @@ class EventsStore(EventsWorkerStore):
                                 if all_single_prev_not_state:
                                     continue
 
+                            state_delta_counter.inc()
+                            if len(new_latest_event_ids) == 1:
+                                state_delta_single_event_counter.inc()
+
+                                # This is a fairly handwavey check to see if we could
+                                # have guessed what the delta would have been when
+                                # processing one of these events.
+                                # What we're interested in is if the latest extremities
+                                # were the same when we created the event as they are
+                                # now. When this server creates a new event (as opposed
+                                # to receiving it over federation) it will use the
+                                # forward extremities as the prev_events, so we can
+                                # guess this by looking at the prev_events and checking
+                                # if they match the current forward extremities.
+                                for ev, _ in ev_ctx_rm:
+                                    prev_event_ids = set(e for e, _ in ev.prev_events)
+                                    if latest_event_ids == prev_event_ids:
+                                        state_delta_reuse_delta_counter.inc()
+                                        break
+
                             logger.info(
                                 "Calculating state delta for room %s", room_id,
                             )
diff --git a/synapse/storage/group_server.py b/synapse/storage/group_server.py
index 8fde1aab8e..d03858234b 100644
--- a/synapse/storage/group_server.py
+++ b/synapse/storage/group_server.py
@@ -19,7 +19,7 @@ from synapse.api.errors import SynapseError
 
 from ._base import SQLBaseStore
 
-import ujson as json
+import simplejson as json
 
 
 # The category ID for the "default" category. We don't store as null in the
diff --git a/synapse/storage/registration.py b/synapse/storage/registration.py
index d809b2ba46..6b557ca0cf 100644
--- a/synapse/storage/registration.py
+++ b/synapse/storage/registration.py
@@ -460,14 +460,12 @@ class RegistrationStore(RegistrationWorkerStore,
         """
         def _find_next_generated_user_id(txn):
             txn.execute("SELECT name FROM users")
-            rows = self.cursor_to_dict(txn)
 
             regex = re.compile("^@(\d+):")
 
             found = set()
 
-            for r in rows:
-                user_id = r["name"]
+            for user_id, in txn:
                 match = regex.search(user_id)
                 if match:
                     found.add(int(match.group(1)))
diff --git a/synapse/util/httpresourcetree.py b/synapse/util/httpresourcetree.py
index 45be47159a..d747849553 100644
--- a/synapse/util/httpresourcetree.py
+++ b/synapse/util/httpresourcetree.py
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from twisted.web.resource import Resource
+from twisted.web.resource import NoResource
 
 import logging
 
@@ -45,7 +45,7 @@ def create_resource_tree(desired_tree, root_resource):
         for path_seg in full_path.split('/')[1:-1]:
             if path_seg not in last_resource.listNames():
                 # resource doesn't exist, so make a "dummy resource"
-                child_resource = Resource()
+                child_resource = NoResource()
                 last_resource.putChild(path_seg, child_resource)
                 res_id = _resource_id(last_resource, path_seg)
                 resource_mappings[res_id] = child_resource