Servers-known-about statistic (#5981)

Author: hawkowl

changelog.d/5981.feature

@@ -0,0 +1 @@
+Setting metrics_flags.known_servers to True in the configuration will publish the synapse_federation_known_servers metric over Prometheus. This represents the total number of servers your server knows about (i.e. is in rooms with), including itself.

docs/sample_config.yaml

@@ -958,6 +958,16 @@ account_threepid_delegates:
 #sentry:
 #    dsn: "..."
 
+# Flags to enable Prometheus metrics which are not suitable to be
+# enabled by default, either for performance reasons or limited use.
+#
+metrics_flags:
+    # Publish synapse_federation_known_servers, a g auge of the number of
+    # servers this homeserver knows about, including itself. May cause
+    # performance problems on large homeservers.
+    #
+    #known_servers: true
+
 # Whether or not to report anonymized homeserver usage statistics.
 # report_stats: true|false
 

synapse/config/metrics.py

@@ -1,5 +1,6 @@
 # -*- coding: utf-8 -*-
 # Copyright 2015, 2016 OpenMarket Ltd
+# Copyright 2019 The Matrix.org Foundation C.I.C.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,20 +14,40 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import attr
+
 from ._base import Config, ConfigError
 
 MISSING_SENTRY = """Missing sentry-sdk library. This is required to enable sentry
     integration.
     """
 
 
+@attr.s
+class MetricsFlags(object):
+    known_servers = attr.ib(default=False, validator=attr.validators.instance_of(bool))
+
+    @classmethod
+    def all_off(cls):
+        """
+        Instantiate the flags with all options set to off.
+        """
+        return cls(**{x.name: False for x in attr.fields(cls)})
+
+
 class MetricsConfig(Config):
     def read_config(self, config, **kwargs):
         self.enable_metrics = config.get("enable_metrics", False)
         self.report_stats = config.get("report_stats", None)
         self.metrics_port = config.get("metrics_port")
         self.metrics_bind_host = config.get("metrics_bind_host", "127.0.0.1")
 
+        if self.enable_metrics:
+            _metrics_config = config.get("metrics_flags") or {}
+            self.metrics_flags = MetricsFlags(**_metrics_config)
+        else:
+            self.metrics_flags = MetricsFlags.all_off()
+
         self.sentry_enabled = "sentry" in config
         if self.sentry_enabled:
             try:
@@ -58,6 +79,16 @@ def generate_config_section(self, report_stats=None, **kwargs):
         #sentry:
         #    dsn: "..."
 
+        # Flags to enable Prometheus metrics which are not suitable to be
+        # enabled by default, either for performance reasons or limited use.
+        #
+        metrics_flags:
+            # Publish synapse_federation_known_servers, a g auge of the number of
+            # servers this homeserver knows about, including itself. May cause
+            # performance problems on large homeservers.
+            #
+            #known_servers: true
+
         # Whether or not to report anonymized homeserver usage statistics.
         """
 

synapse/storage/roommember.py

@@ -24,8 +24,10 @@
 from twisted.internet import defer
 
 from synapse.api.constants import EventTypes, Membership
+from synapse.metrics import LaterGauge
 from synapse.metrics.background_process_metrics import run_as_background_process
 from synapse.storage._base import LoggingTransaction
+from synapse.storage.engines import Sqlite3Engine
 from synapse.storage.events_worker import EventsWorkerStore
 from synapse.types import get_domain_from_id
 from synapse.util.async_helpers import Linearizer
@@ -74,6 +76,63 @@ def __init__(self, db_conn, hs):
         self._check_safe_current_state_events_membership_updated_txn(txn)
         txn.close()
 
+        if self.hs.config.metrics_flags.known_servers:
+            self._known_servers_count = 1
+            self.hs.get_clock().looping_call(
+                run_as_background_process,
+                60 * 1000,
+                "_count_known_servers",
+                self._count_known_servers,
+            )
+            self.hs.get_clock().call_later(
+                1000,
+                run_as_background_process,
+                "_count_known_servers",
+                self._count_known_servers,
+            )
+            LaterGauge(
+                "synapse_federation_known_servers",
+                "",
+                [],
+                lambda: self._known_servers_count,
+            )
+
+    @defer.inlineCallbacks
+    def _count_known_servers(self):
+        """
+        Count the servers that this server knows about.
+
+        The statistic is stored on the class for the
+        `synapse_federation_known_servers` LaterGauge to collect.
+        """
+
+        def _transact(txn):
+            if isinstance(self.database_engine, Sqlite3Engine):
+                query = """
+                    SELECT COUNT(DISTINCT substr(out.user_id, pos+1))
+                    FROM (
+                        SELECT rm.user_id as user_id, instr(rm.user_id, ':')
+                            AS pos FROM room_memberships as rm
+                        INNER JOIN current_state_events as c ON rm.event_id = c.event_id
+                        WHERE c.type = 'm.room.member'
+                    ) as out
+                """
+            else:
+                query = """
+                    SELECT COUNT(DISTINCT split_part(state_key, ':', 2))
+                    FROM current_state_events
+                    WHERE type = 'm.room.member' AND membership = 'join';
+                """
+            txn.execute(query)
+            return list(txn)[0][0]
+
+        count = yield self.runInteraction("get_known_servers", _transact)
+
+        # We always know about ourselves, even if we have nothing in
+        # room_memberships (for example, the server is new).
+        self._known_servers_count = max([count, 1])
+        return self._known_servers_count
+
     def _check_safe_current_state_events_membership_updated_txn(self, txn):
         """Checks if it is safe to assume the new current_state_events
         membership column is up to date

tests/config/test_generate.py

@@ -17,6 +17,8 @@
 import re
 import shutil
 import tempfile
+from contextlib import redirect_stdout
+from io import StringIO
 
 from synapse.config.homeserver import HomeServerConfig
 
@@ -32,17 +34,18 @@ def tearDown(self):
         shutil.rmtree(self.dir)
 
     def test_generate_config_generates_files(self):
-        HomeServerConfig.load_or_generate_config(
-            "",
-            [
-                "--generate-config",
-                "-c",
-                self.file,
-                "--report-stats=yes",
-                "-H",
-                "lemurs.win",
-            ],
-        )
+        with redirect_stdout(StringIO()):
+            HomeServerConfig.load_or_generate_config(
+                "",
+                [
+                    "--generate-config",
+                    "-c",
+                    self.file,
+                    "--report-stats=yes",
+                    "-H",
+                    "lemurs.win",
+                ],
+            )
 
         self.assertSetEqual(
             set(["homeserver.yaml", "lemurs.win.log.config", "lemurs.win.signing.key"]),

tests/config/test_load.py

@@ -15,6 +15,8 @@
 import os.path
 import shutil
 import tempfile
+from contextlib import redirect_stdout
+from io import StringIO
 
 import yaml
 
@@ -26,7 +28,6 @@
 class ConfigLoadingTestCase(unittest.TestCase):
     def setUp(self):
         self.dir = tempfile.mkdtemp()
-        print(self.dir)
         self.file = os.path.join(self.dir, "homeserver.yaml")
 
     def tearDown(self):
@@ -94,18 +95,27 @@ def test_disable_registration(self):
         )
         self.assertTrue(config.enable_registration)
 
+    def test_stats_enabled(self):
+        self.generate_config_and_remove_lines_containing("enable_metrics")
+        self.add_lines_to_config(["enable_metrics: true"])
+
+        # The default Metrics Flags are off by default.
+        config = HomeServerConfig.load_config("", ["-c", self.file])
+        self.assertFalse(config.metrics_flags.known_servers)
+
     def generate_config(self):
-        HomeServerConfig.load_or_generate_config(
-            "",
-            [
-                "--generate-config",
-                "-c",
-                self.file,
-                "--report-stats=yes",
-                "-H",
-                "lemurs.win",
-            ],
-        )
+        with redirect_stdout(StringIO()):
+            HomeServerConfig.load_or_generate_config(
+                "",
+                [
+                    "--generate-config",
+                    "-c",
+                    self.file,
+                    "--report-stats=yes",
+                    "-H",
+                    "lemurs.win",
+                ],
+            )
 
     def generate_config_and_remove_lines_containing(self, needle):
         self.generate_config()