mirror of
https://github.com/scylladb/scylladb.git
synced 2026-06-03 13:37:04 +00:00
cc034f84c5
Tables created before the GROUP0_SCHEMA_VERSIONING feature was enabled have committed_by_group0 = null in system_schema.scylla_tables. This causes maybe_delete_schema_version() to delete their version cell, forcing the legacy hash-based schema version computation path. Add ensure_committed_by_group0() which runs on boot and fixes up any non-system tables where committed_by_group0 is not true (null or false): 1. Queries system_schema.scylla_tables for rows where committed_by_group0 is null or false, skipping system keyspaces (system, system_schema). 2. Takes a group0 guard 3. Re-checks after the raft barrier in case another node already fixed it. 4. For each table needing fixup, creates a mutation writing the version cell (from the in-memory schema). The committed_by_group0 = true flag is stamped by add_committed_by_group0_flag() inside announce(). 5. Announces via raft group0. 6. Retries with a small random delay on group0_concurrent_modification. On other nodes, schema_applier will detect these as "altered" tables (scylla_tables mutation changed), but since the actual table definition is unchanged, update_column_family is effectively a no-op. This is a prerequisite for eventually removing the legacy hash-based schema versioning code path. Closes scylladb/scylladb#29911
53 lines
2.1 KiB
Python
53 lines
2.1 KiB
Python
#
|
|
# Copyright (C) 2026-present ScyllaDB
|
|
#
|
|
# SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.1
|
|
#
|
|
"""
|
|
Test that ensure_committed_by_group0() fixes tables missing the flag on boot.
|
|
"""
|
|
import pytest
|
|
import logging
|
|
from test.pylib.manager_client import ManagerClient
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_ensure_committed_by_group0(manager: ManagerClient):
|
|
"""Tables with committed_by_group0 = null or false get fixed on restart."""
|
|
servers = await manager.servers_add(1)
|
|
(cql, _) = await manager.get_ready_cql(servers)
|
|
|
|
await cql.run_async("CREATE KEYSPACE ks WITH replication = "
|
|
"{'class': 'NetworkTopologyStrategy', 'replication_factor': 1}")
|
|
await cql.run_async("CREATE TABLE ks.tbl_null (pk int PRIMARY KEY)")
|
|
await cql.run_async("CREATE TABLE ks.tbl_false (pk int PRIMARY KEY)")
|
|
|
|
# Verify both have committed_by_group0 = true initially
|
|
for tbl in ['tbl_null', 'tbl_false']:
|
|
rows = await cql.run_async(
|
|
f"SELECT committed_by_group0 FROM system_schema.scylla_tables "
|
|
f"WHERE keyspace_name = 'ks' AND table_name = '{tbl}'")
|
|
assert rows[0].committed_by_group0 == True
|
|
|
|
# Simulate pre-group0 table (null) and recovery-mode table (false)
|
|
await cql.run_async(
|
|
"DELETE committed_by_group0 FROM system_schema.scylla_tables "
|
|
"WHERE keyspace_name = 'ks' AND table_name = 'tbl_null'")
|
|
await cql.run_async(
|
|
"UPDATE system_schema.scylla_tables SET committed_by_group0 = false "
|
|
"WHERE keyspace_name = 'ks' AND table_name = 'tbl_false'")
|
|
|
|
# Restart — ensure_committed_by_group0() should fix both on boot
|
|
await manager.server_restart(servers[0].server_id)
|
|
(cql, _) = await manager.get_ready_cql(servers)
|
|
|
|
# Verify fixup happened for both tables
|
|
for tbl in ['tbl_null', 'tbl_false']:
|
|
rows = await cql.run_async(
|
|
f"SELECT committed_by_group0 FROM system_schema.scylla_tables "
|
|
f"WHERE keyspace_name = 'ks' AND table_name = '{tbl}'")
|
|
assert rows[0].committed_by_group0 == True, \
|
|
f"committed_by_group0 not fixed for {tbl}"
|