Improve database configuration docs (#6988)

Attempts to clarify the sample config for databases, and add some stuff about
tcp keepalives to `postgres.md`.

changelog.d/6988.doc

@@ -0,0 +1 @@
+Improve the documentation for database configuration.

docs/postgres.md

@@ -105,7 +105,8 @@ of free memory the database host has available.
 When you are ready to start using PostgreSQL, edit the `database`
 section in your config file to match the following lines:
 
-    database:
+```yaml
+database:
   name: psycopg2
   args:
     user: <user>
@@ -114,10 +115,31 @@ section in your config file to match the following lines:
     host: <host>
     cp_min: 5
     cp_max: 10
+```
 
 All key, values in `args` are passed to the `psycopg2.connect(..)`
 function, except keys beginning with `cp_`, which are consumed by the
-twisted adbapi connection pool.
+twisted adbapi connection pool. See the [libpq
+documentation](https://www.postgresql.org/docs/current/libpq-connect.html#LIBPQ-PARAMKEYWORDS)
+for a list of options which can be passed.
+
+You should consider tuning the `args.keepalives_*` options if there is any danger of
+the connection between your homeserver and database dropping, otherwise Synapse
+may block for an extended period while it waits for a response from the
+database server. Example values might be:
+
+```yaml
+# seconds of inactivity after which TCP should send a keepalive message to the server
+keepalives_idle: 10
+
+# the number of seconds after which a TCP keepalive message that is not
+# acknowledged by the server should be retransmitted
+keepalives_interval: 10
+
+# the number of TCP keepalives that can be lost before the client's connection
+# to the server is considered dead
+keepalives_count: 3
+```
 
 ## Porting from SQLite
 

docs/sample_config.yaml

@@ -578,13 +578,46 @@ acme:
 
 ## Database ##
 
+# The 'database' setting defines the database that synapse uses to store all of
+# its data.
+#
+# 'name' gives the database engine to use: either 'sqlite3' (for SQLite) or
+# 'psycopg2' (for PostgreSQL).
+#
+# 'args' gives options which are passed through to the database engine,
+# except for options starting 'cp_', which are used to configure the Twisted
+# connection pool. For a reference to valid arguments, see:
+#   * for sqlite: https://docs.python.org/3/library/sqlite3.html#sqlite3.connect
+#   * for postgres: https://www.postgresql.org/docs/current/libpq-connect.html#LIBPQ-PARAMKEYWORDS
+#   * for the connection pool: https://twistedmatrix.com/documents/current/api/twisted.enterprise.adbapi.ConnectionPool.html#__init__
+#
+#
+# Example SQLite configuration:
+#
+#database:
+#  name: sqlite3
+#  args:
+#    database: /path/to/homeserver.db
+#
+#
+# Example Postgres configuration:
+#
+#database:
+#  name: psycopg2
+#  args:
+#    user: synapse
+#    password: secretpassword
+#    database: synapse
+#    host: localhost
+#    cp_min: 5
+#    cp_max: 10
+#
+# For more information on using Synapse with Postgres, see `docs/postgres.md`.
+#
 database:
-  # The database engine name
-  name: "sqlite3"
-  # Arguments to pass to the engine
+  name: sqlite3
   args:
-    # Path to the database
-    database: "DATADIR/homeserver.db"
+    database: DATADIR/homeserver.db
 
 # Number of events to cache in memory.
 #

synapse/config/_base.py

@@ -294,7 +294,6 @@ class RootConfig(object):
         report_stats=None,
         open_private_ports=False,
         listeners=None,
-        database_conf=None,
         tls_certificate_path=None,
         tls_private_key_path=None,
         acme_domain=None,
@@ -367,7 +366,6 @@ class RootConfig(object):
                 report_stats=report_stats,
                 open_private_ports=open_private_ports,
                 listeners=listeners,
-                database_conf=database_conf,
                 tls_certificate_path=tls_certificate_path,
                 tls_private_key_path=tls_private_key_path,
                 acme_domain=acme_domain,

synapse/config/database.py

@@ -1,5 +1,6 @@
 # -*- coding: utf-8 -*-
 # Copyright 2014-2016 OpenMarket Ltd
+# Copyright 2020 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.
@@ -14,14 +15,60 @@
 # limitations under the License.
 import logging
 import os
-from textwrap import indent
-
-import yaml
 
 from synapse.config._base import Config, ConfigError
 
 logger = logging.getLogger(__name__)
 
+DEFAULT_CONFIG = """\
+## Database ##
+
+# The 'database' setting defines the database that synapse uses to store all of
+# its data.
+#
+# 'name' gives the database engine to use: either 'sqlite3' (for SQLite) or
+# 'psycopg2' (for PostgreSQL).
+#
+# 'args' gives options which are passed through to the database engine,
+# except for options starting 'cp_', which are used to configure the Twisted
+# connection pool. For a reference to valid arguments, see:
+#   * for sqlite: https://docs.python.org/3/library/sqlite3.html#sqlite3.connect
+#   * for postgres: https://www.postgresql.org/docs/current/libpq-connect.html#LIBPQ-PARAMKEYWORDS
+#   * for the connection pool: https://twistedmatrix.com/documents/current/api/twisted.enterprise.adbapi.ConnectionPool.html#__init__
+#
+#
+# Example SQLite configuration:
+#
+#database:
+#  name: sqlite3
+#  args:
+#    database: /path/to/homeserver.db
+#
+#
+# Example Postgres configuration:
+#
+#database:
+#  name: psycopg2
+#  args:
+#    user: synapse
+#    password: secretpassword
+#    database: synapse
+#    host: localhost
+#    cp_min: 5
+#    cp_max: 10
+#
+# For more information on using Synapse with Postgres, see `docs/postgres.md`.
+#
+database:
+  name: sqlite3
+  args:
+    database: %(database_path)s
+
+# Number of events to cache in memory.
+#
+#event_cache_size: 10K
+"""
+
 
 class DatabaseConnectionConfig:
     """Contains the connection config for a particular database.
@@ -36,10 +83,12 @@ class DatabaseConnectionConfig:
     """
 
     def __init__(self, name: str, db_config: dict):
-        if db_config["name"] not in ("sqlite3", "psycopg2"):
-            raise ConfigError("Unsupported database type %r" % (db_config["name"],))
+        db_engine = db_config.get("name", "sqlite3")
 
-        if db_config["name"] == "sqlite3":
+        if db_engine not in ("sqlite3", "psycopg2"):
+            raise ConfigError("Unsupported database type %r" % (db_engine,))
+
+        if db_engine == "sqlite3":
             db_config.setdefault("args", {}).update(
                 {"cp_min": 1, "cp_max": 1, "check_same_thread": False}
             )
@@ -97,34 +146,10 @@ class DatabaseConfig(Config):
 
             self.set_databasepath(config.get("database_path"))
 
-    def generate_config_section(self, data_dir_path, database_conf, **kwargs):
-        if not database_conf:
-            database_path = os.path.join(data_dir_path, "homeserver.db")
-            database_conf = (
-                """# The database engine name
-          name: "sqlite3"
-          # Arguments to pass to the engine
-          args:
-            # Path to the database
-            database: "%(database_path)s"
-            """
-                % locals()
-            )
-        else:
-            database_conf = indent(yaml.dump(database_conf), " " * 10).lstrip()
-
-        return (
-            """\
-        ## Database ##
-
-        database:
-          %(database_conf)s
-        # Number of events to cache in memory.
-        #
-        #event_cache_size: 10K
-        """
-            % locals()
-        )
+    def generate_config_section(self, data_dir_path, **kwargs):
+        return DEFAULT_CONFIG % {
+            "database_path": os.path.join(data_dir_path, "homeserver.db")
+        }
 
     def read_arguments(self, args):
         self.set_databasepath(args.database_path)

tests/config/test_database.py

@@ -21,9 +21,9 @@ from tests import unittest
 
 
 class DatabaseConfigTestCase(unittest.TestCase):
-    def test_database_configured_correctly_no_database_conf_param(self):
+    def test_database_configured_correctly(self):
         conf = yaml.safe_load(
-            DatabaseConfig().generate_config_section("/data_dir_path", None)
+            DatabaseConfig().generate_config_section(data_dir_path="/data_dir_path")
         )
 
         expected_database_conf = {
@@ -32,21 +32,3 @@ class DatabaseConfigTestCase(unittest.TestCase):
         }
 
         self.assertEqual(conf["database"], expected_database_conf)
-
-    def test_database_configured_correctly_database_conf_param(self):
-
-        database_conf = {
-            "name": "my super fast datastore",
-            "args": {
-                "user": "matrix",
-                "password": "synapse_database_password",
-                "host": "synapse_database_host",
-                "database": "matrix",
-            },
-        }
-
-        conf = yaml.safe_load(
-            DatabaseConfig().generate_config_section("/data_dir_path", database_conf)
-        )
-
-        self.assertEqual(conf["database"], database_conf)