MatrixSynapse/synapse/storage/controllers/stats.py

113 lines
4.0 KiB
Python
Raw Permalink Normal View History

# Copyright 2023 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.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
from typing import TYPE_CHECKING, Collection, Counter, List, Tuple
from synapse.api.errors import SynapseError
from synapse.storage.database import LoggingTransaction
from synapse.storage.databases import Databases
from synapse.storage.engines import PostgresEngine
if TYPE_CHECKING:
from synapse.server import HomeServer
logger = logging.getLogger(__name__)
class StatsController:
"""High level interface for getting statistics."""
def __init__(self, hs: "HomeServer", stores: Databases):
self.stores = stores
async def get_room_db_size_estimate(self) -> List[Tuple[str, int]]:
"""Get an estimate of the largest rooms and how much database space they
use, in bytes.
Only works against PostgreSQL.
Note: this uses the postgres statistics so is a very rough estimate.
"""
# Note: We look at both tables on the main and state databases.
if not isinstance(self.stores.main.database_engine, PostgresEngine):
raise SynapseError(400, "Endpoint requires using PostgreSQL")
if not isinstance(self.stores.state.database_engine, PostgresEngine):
raise SynapseError(400, "Endpoint requires using PostgreSQL")
# For each "large" table, we go through and get the largest rooms
# and an estimate of how much space they take. We can then sum the
# results and return the top 10.
#
# This isn't the most accurate, but given all of these are estimates
# anyway its good enough.
room_estimates: Counter[str] = Counter()
# Return size of the table on disk, including indexes and TOAST.
table_sql = """
SELECT pg_total_relation_size(?)
"""
# Get an estimate for the largest rooms and their frequency.
#
# Note: the cast here is a hack to cast from `anyarray` to an actual
# type. This ensures that psycopg2 passes us a back a a Python list.
column_sql = """
SELECT
most_common_vals::TEXT::TEXT[], most_common_freqs::TEXT::NUMERIC[]
FROM pg_stats
WHERE tablename = ? and attname = 'room_id'
"""
def get_room_db_size_estimate_txn(
txn: LoggingTransaction,
tables: Collection[str],
) -> None:
for table in tables:
txn.execute(table_sql, (table,))
row = txn.fetchone()
assert row is not None
(table_size,) = row
txn.execute(column_sql, (table,))
row = txn.fetchone()
assert row is not None
vals, freqs = row
for room_id, freq in zip(vals, freqs):
room_estimates[room_id] += int(freq * table_size)
await self.stores.main.db_pool.runInteraction(
"get_room_db_size_estimate_main",
get_room_db_size_estimate_txn,
(
"event_json",
"events",
"event_search",
"event_edges",
"event_push_actions",
"stream_ordering_to_exterm",
),
)
await self.stores.state.db_pool.runInteraction(
"get_room_db_size_estimate_state",
get_room_db_size_estimate_txn,
("state_groups_state",),
)
return room_estimates.most_common(10)