feat: create events table to store last 7 days of data (#26239)

posthog/clickhouse/kafka_engine.py

@@ -35,6 +35,8 @@
 , _partition UInt64
 """
 
+KAFKA_TIMESTAMP_MS_COLUMN = "_timestamp_ms DateTime64"
+
 
 def kafka_engine(topic: str, kafka_host: str | None = None, group="group1", serialization="JSONEachRow") -> str:
     if kafka_host is None:

posthog/clickhouse/migrations/0086_events_recent_table.py

@@ -0,0 +1,14 @@
+from posthog.clickhouse.client.migration_tools import run_sql_with_exceptions
+from posthog.models.event.sql import (
+    EVENTS_RECENT_TABLE_JSON_MV_SQL,
+    EVENTS_RECENT_TABLE_SQL,
+    KAFKA_EVENTS_RECENT_TABLE_JSON_SQL,
+    DISTRIBUTED_EVENTS_RECENT_TABLE_SQL,
+)
+
+operations = [
+    run_sql_with_exceptions(EVENTS_RECENT_TABLE_SQL()),
+    run_sql_with_exceptions(KAFKA_EVENTS_RECENT_TABLE_JSON_SQL()),
+    run_sql_with_exceptions(EVENTS_RECENT_TABLE_JSON_MV_SQL()),
+    run_sql_with_exceptions(DISTRIBUTED_EVENTS_RECENT_TABLE_SQL()),
+]

posthog/models/event/sql.py

@@ -4,6 +4,8 @@
 from posthog.clickhouse.indexes import index_by_kafka_timestamp
 from posthog.clickhouse.kafka_engine import (
     KAFKA_COLUMNS,
+    KAFKA_COLUMNS_WITH_PARTITION,
+    KAFKA_TIMESTAMP_MS_COLUMN,
     STORAGE_POLICY,
     kafka_engine,
     trim_quotes_expr,
@@ -18,7 +20,7 @@
 
 EVENTS_DATA_TABLE = lambda: "sharded_events"
 WRITABLE_EVENTS_DATA_TABLE = lambda: "writable_events"
-
+EVENTS_RECENT_DATA_TABLE = lambda: "events_recent"
 TRUNCATE_EVENTS_TABLE_SQL = (
     lambda: f"TRUNCATE TABLE IF EXISTS {EVENTS_DATA_TABLE()} ON CLUSTER '{settings.CLICKHOUSE_CLUSTER}'"
 )
@@ -185,6 +187,86 @@
     )
 )
 
+
+KAFKA_EVENTS_RECENT_TABLE_JSON_SQL = lambda: (
+    EVENTS_TABLE_BASE_SQL
+    + """
+    SETTINGS kafka_skip_broken_messages = 100
+"""
+).format(
+    table_name="kafka_events_recent_json",
+    cluster=settings.CLICKHOUSE_CLUSTER,
+    engine=kafka_engine(topic=KAFKA_EVENTS_JSON, group="group1_recent"),
+    extra_fields="",
+    materialized_columns="",
+    indexes="",
+)
+
+EVENTS_RECENT_TABLE_JSON_MV_SQL = (
+    lambda: """
+CREATE MATERIALIZED VIEW IF NOT EXISTS events_recent_json_mv ON CLUSTER '{cluster}'
+TO {database}.{target_table}
+AS SELECT
+uuid,
+event,
+properties,
+timestamp,
+team_id,
+distinct_id,
+elements_chain,
+created_at,
+person_id,
+person_created_at,
+person_properties,
+group0_properties,
+group1_properties,
+group2_properties,
+group3_properties,
+group4_properties,
+group0_created_at,
+group1_created_at,
+group2_created_at,
+group3_created_at,
+group4_created_at,
+person_mode,
+_timestamp,
+_timestamp_ms,
+_offset,
+_partition
+FROM {database}.kafka_events_recent_json
+""".format(
+        target_table=EVENTS_RECENT_DATA_TABLE(),
+        cluster=settings.CLICKHOUSE_CLUSTER,
+        database=settings.CLICKHOUSE_DATABASE,
+    )
+)
+
+EVENTS_RECENT_TABLE_SQL = lambda: (
+    EVENTS_TABLE_BASE_SQL
+    + """PARTITION BY toStartOfHour(_timestamp)
+ORDER BY (team_id, toStartOfHour(_timestamp), event, cityHash64(distinct_id), cityHash64(uuid))
+TTL _timestamp + INTERVAL 7 DAY
+{storage_policy}
+"""
+).format(
+    table_name=EVENTS_RECENT_DATA_TABLE(),
+    cluster=settings.CLICKHOUSE_CLUSTER,
+    engine=ReplacingMergeTree(EVENTS_RECENT_DATA_TABLE(), ver="_timestamp"),
+    extra_fields=KAFKA_COLUMNS_WITH_PARTITION + INSERTED_AT_COLUMN + f", {KAFKA_TIMESTAMP_MS_COLUMN}",
+    materialized_columns="",
+    indexes="",
+    storage_policy=STORAGE_POLICY(),
+)
+
+DISTRIBUTED_EVENTS_RECENT_TABLE_SQL = lambda: EVENTS_TABLE_BASE_SQL.format(
+    table_name="distributed_events_recent",
+    cluster=settings.CLICKHOUSE_CLUSTER,
+    engine=Distributed(data_table=EVENTS_RECENT_DATA_TABLE(), sharding_key="sipHash64(distinct_id)"),
+    extra_fields=KAFKA_COLUMNS_WITH_PARTITION + INSERTED_AT_COLUMN + f", {KAFKA_TIMESTAMP_MS_COLUMN}",
+    materialized_columns="",
+    indexes="",
+)
+
 # Distributed engine tables are only created if CLICKHOUSE_REPLICATED
 
 # This table is responsible for writing to sharded_events based on a sharding key.