Before this patch KILL MUTATION marks mutation as canceled just by name
(and part numbers) so if you have multiple tables with the same part
name, then killing mutation for one table, will mark it as killed for
another too.
Fix this by comparing StorageID too (it is better to use StorageID over
database/table to avoid ambiguity by using UUIDs for comparing).
Here is a failure of the 01414_freeze_does_not_prevent_alters on CI [1].
[1]: https://clickhouse-test-reports.s3.yandex.net/24069/9fb69dcf98c71a939d200cad3c8491bf43a44622/functional_stateless_tests_(ubsan).html#fail1
Will fix possible issues in CI [1].
2021-06-07 03:42:56 01520_client_print_query_id: [ FAIL ] 8.24 sec. - return code 1
2021-06-07 03:42:56 , result:
And in server logs:
2021.06.07 03:42:48.732289 [ 18471 ] {7df001e8-f9bb-488f-9fe1-6dc538f24808} <Debug> executeQuery: (from [::1]:45680, using production parser) (comment: /usr/share/clickhouse-test/queries/0_stateless/01520_client_print_query_id.expect) CREATE DATABASE test
...
2021.06.07 03:42:52.048064 [ 18703 ] {8526892a-f99a-4949-9f7d-3c8e82969439} <Debug> executeQuery: (from [::1]:45702, using production parser) SELECT DISTINCT arrayJoin(extractAll(name, '[\\w_]{2,}')) AS res FROM (SELECT name FROM system.functions UNION ALL SELECT name FROM system.table_engines UNION ALL SELECT name FROM system.formats UNION ALL SELECT name FROM system.table_functions UNION ALL SELECT name FROM system.data_type_families UNION ALL SELECT name FROM system.merge_tree_settings UNION ALL SELECT name FROM system.settings UNION ALL SELECT cluster FROM system.clusters UNION ALL SELECT macro FROM system.macros UNION ALL SELECT policy_name FROM system.storage_policies UNION ALL SELECT concat(func.name, comb.name) FROM system.functions AS func CROSS JOIN system.aggregate_function_combinators AS comb WHERE is_aggregate UNION ALL SELECT name FROM system.databases LIMIT 10000 UNION ALL SELECT DISTINCT name FROM system.tables LIMIT 10000 UNION ALL SELECT DISTINCT name FROM system.dictionaries LIMIT 10000 UNION ALL SELECT DISTINCT name FROM system.columns LIMIT 10000) WHERE notEmpty(res)
...
2021.06.07 03:42:54.523976 [ 5575 ] {d97df847-0feb-4e8d-92bd-3236e1f7f711} <Debug> executeQuery: (from [::1]:45710, using production parser) (comment: /usr/share/clickhouse-test/queries/0_stateless/01520_client_print_query_id.expect) DROP DATABASE test_3bce13
...
2021.06.07 03:42:54.722391 [ 18703 ] {8526892a-f99a-4949-9f7d-3c8e82969439} <Error> executeQuery: Code: 210, e.displayText() = DB::NetException: I/O error: Broken pipe, while writing to socket ([::1]:45702) (version 21.7.1.7071) (from [::1]:45702) (in query: SELECT DISTINCT arrayJoin(extractAll(name, '[\\w_]{2,}')) AS res FROM (SELECT name FROM system.functions UNION ALL SELECT name FROM system.table_engines UNION ALL SELECT name FROM system.formats UNION ALL SELECT name FROM system.table_functions UNION ALL SELECT name FROM system.data_type_families UNION ALL SELECT name FROM system.merge_tree_settings UNION ALL SELECT name FROM system.settings UNION ALL SELECT cluster FROM system.clusters UNION ALL SELECT macro FROM system.macros UNION ALL SELECT policy_name FROM system.storage_policies UNION ALL SELECT concat(func.name, comb.name) FROM system.functions AS func CROSS JOIN system.aggregate_function_combinators AS comb WHERE is_aggregate UNION ALL SELECT name FROM system.databases LIMIT 10000 UNION ALL SELECT DISTINCT name FROM system.tables LIMIT 10000 UNION ALL SELECT DISTINCT name FROM system.dictionaries LIMIT 10000 UNION ALL SELECT DISTINCT name FROM system.columns LIMIT 10000) WHERE notEmpty(res)), Stack trace (when copying this message, always include the lines below):
[1]: https://clickhouse-test-reports.s3.yandex.net/24069/263c06137cda16213c14ca60a89cfdd7fb511c81/functional_stateless_tests_(thread).html
After #24483 had been merged the only place where the allocation may
fail is the insert into PODArray in DB::OwnSplitChannel::log, but this
MR ignores the errors in this function, so to check new behaviour
separate server is required.
