-- -- PG18 -- SHOW server_version \gset SELECT substring(:'server_version', '\d+')::int >= 18 AS server_version_ge_18 \gset -- test invalid statistics -- behavior is same among PG versions, error message differs -- relevant PG18 commit: 3eea4dc2c7, 38883916e CREATE STATISTICS tst ON a FROM (VALUES (x)) AS foo; CREATE FUNCTION tftest(int) returns table(a int, b int) as $$ SELECT $1, $1+i FROM generate_series(1,5) g(i); $$ LANGUAGE sql IMMUTABLE STRICT; CREATE STATISTICS alt_stat2 ON a FROM tftest(1); DROP FUNCTION tftest; \if :server_version_ge_18 \else \q \endif -- PG18-specific tests go here. -- -- Purpose: Verify PG18 behavior that NOT NULL constraints are materialized -- as pg_constraint rows with contype = 'n' on both coordinator and -- worker shards. Also confirm our helper view (table_checks) does -- NOT surface NOT NULL entries. -- https://github.com/postgres/postgres/commit/14e87ffa5c543b5f30ead7413084c25f7735039f CREATE SCHEMA pg18_nn; SET search_path TO pg18_nn; -- Local control table DROP TABLE IF EXISTS nn_local CASCADE; CREATE TABLE nn_local( a int NOT NULL, b int, c text NOT NULL ); -- Distributed table DROP TABLE IF EXISTS nn_dist CASCADE; CREATE TABLE nn_dist( a int NOT NULL, b int, c text NOT NULL ); SELECT create_distributed_table('nn_dist', 'a'); -- Coordinator: count NOT NULL constraint rows SELECT 'local_n_count' AS label, contype, count(*) FROM pg_constraint WHERE conrelid = 'pg18_nn.nn_local'::regclass GROUP BY contype ORDER BY contype; SELECT 'dist_n_count' AS label, contype, count(*) FROM pg_constraint WHERE conrelid = 'pg18_nn.nn_dist'::regclass GROUP BY contype ORDER BY contype; -- Our helper view should exclude NOT NULL SELECT 'table_checks_local_count' AS label, count(*) FROM public.table_checks WHERE relid = 'pg18_nn.nn_local'::regclass; SELECT 'table_checks_dist_count' AS label, count(*) FROM public.table_checks WHERE relid = 'pg18_nn.nn_dist'::regclass; -- Add a real CHECK to ensure table_checks still reports real checks ALTER TABLE nn_dist ADD CONSTRAINT nn_dist_check CHECK (b IS DISTINCT FROM 42); SELECT 'table_checks_dist_with_real_check' AS label, count(*) FROM public.table_checks WHERE relid = 'pg18_nn.nn_dist'::regclass; -- === Worker checks === \c - - - :worker_1_port SET client_min_messages TO WARNING; SET search_path TO pg18_nn; -- Pick one heap shard of nn_dist in our schema SELECT format('%I.%I', n.nspname, c.relname) AS shard_regclass FROM pg_class c JOIN pg_namespace n ON n.oid = c.relnamespace WHERE n.nspname = 'pg18_nn' AND c.relname LIKE 'nn_dist_%' AND c.relkind = 'r' ORDER BY c.relname LIMIT 1 \gset -- Expect: 2 NOT NULL rows (a,c) + 1 CHECK row on the shard SELECT 'worker_shard_n_count' AS label, contype, count(*) FROM pg_constraint WHERE conrelid = :'shard_regclass'::regclass GROUP BY contype ORDER BY contype; -- table_checks on shard should hide NOT NULL SELECT 'table_checks_worker_shard_count' AS label, count(*) FROM public.table_checks WHERE relid = :'shard_regclass'::regclass; -- Drop one NOT NULL on coordinator; verify propagation \c - - - :master_port SET search_path TO pg18_nn; ALTER TABLE nn_dist ALTER COLUMN c DROP NOT NULL; -- Re-check on worker: NOT NULL count should drop to 1 \c - - - :worker_1_port SET search_path TO pg18_nn; SELECT 'worker_shard_n_after_drop' AS label, contype, count(*) FROM pg_constraint WHERE conrelid = :'shard_regclass'::regclass GROUP BY contype ORDER BY contype; -- And on coordinator \c - - - :master_port SET search_path TO pg18_nn; SELECT 'dist_n_after_drop' AS label, contype, count(*) FROM pg_constraint WHERE conrelid = 'pg18_nn.nn_dist'::regclass GROUP BY contype ORDER BY contype; -- Purpose: test self join elimination for distributed, citus local and local tables. -- CREATE TABLE sje_d1 (id bigserial PRIMARY KEY, name text, created_at timestamptz DEFAULT now()); CREATE TABLE sje_d2 (id bigserial PRIMARY KEY, name text, created_at timestamptz DEFAULT now()); CREATE TABLE sje_local (id bigserial PRIMARY KEY, title text); SET citus.next_shard_id TO 4754000; SELECT create_distributed_table('sje_d1', 'id'); SELECT create_distributed_table('sje_d2', 'id'); INSERT INTO sje_d1 SELECT i, i::text, now() FROM generate_series(0,100)i; INSERT INTO sje_d2 SELECT i, i::text, now() FROM generate_series(0,100)i; INSERT INTO sje_local SELECT i, i::text FROM generate_series(0,100)i; -- Self-join elimination is applied when distributed tables are involved -- The query plan has only one join EXPLAIN (costs off) select count(1) from sje_d1 INNER JOIN sje_d2 u1 USING (id) INNER JOIN sje_d2 u2 USING (id) INNER JOIN sje_d2 u3 USING (id) INNER JOIN sje_d2 u4 USING (id) INNER JOIN sje_d2 u5 USING (id) INNER JOIN sje_d2 u6 USING (id); select count(1) from sje_d1 INNER JOIN sje_d2 u1 USING (id) INNER JOIN sje_d2 u2 USING (id) INNER JOIN sje_d2 u3 USING (id) INNER JOIN sje_d2 u4 USING (id) INNER JOIN sje_d2 u5 USING (id) INNER JOIN sje_d2 u6 USING (id); -- Self-join elimination applied to from list join EXPLAIN (costs off) SELECT count(1) from sje_d1 d1, sje_d2 u1, sje_d2 u2, sje_d2 u3 WHERE d1.id = u1.id and u1.id = u2.id and u3.id = d1.id; SELECT count(1) from sje_d1 d1, sje_d2 u1, sje_d2 u2, sje_d2 u3 WHERE d1.id = u1.id and u1.id = u2.id and u3.id = d1.id; -- Self-join elimination is not applied when a local table is involved -- This is a limitation that will be resolved in citus 14 EXPLAIN (costs off) select count(1) from sje_d1 INNER JOIN sje_local u1 USING (id) INNER JOIN sje_local u2 USING (id) INNER JOIN sje_local u3 USING (id) INNER JOIN sje_local u4 USING (id) INNER JOIN sje_local u5 USING (id) INNER JOIN sje_local u6 USING (id); select count(1) from sje_d1 INNER JOIN sje_local u1 USING (id) INNER JOIN sje_local u2 USING (id) INNER JOIN sje_local u3 USING (id) INNER JOIN sje_local u4 USING (id) INNER JOIN sje_local u5 USING (id) INNER JOIN sje_local u6 USING (id); -- to test USING vs ON equivalence EXPLAIN (costs off) SELECT count(1) FROM sje_d1 d JOIN sje_d2 u1 ON (d.id = u1.id) JOIN sje_d2 u2 ON (u1.id = u2.id); SELECT count(1) FROM sje_d1 d JOIN sje_d2 u1 ON (d.id = u1.id) JOIN sje_d2 u2 ON (u1.id = u2.id); -- Null-introducing join can have SJE EXPLAIN (costs off) SELECT count(*) FROM sje_d1 d LEFT JOIN sje_d2 u1 USING (id) LEFT JOIN sje_d2 u2 USING (id); SELECT count(*) FROM sje_d1 d LEFT JOIN sje_d2 u1 USING (id) LEFT JOIN sje_d2 u2 USING (id); -- prepared statement PREPARE sje_p(int,int) AS SELECT count(1) FROM sje_d1 d JOIN sje_d2 u1 USING (id) JOIN sje_d2 u2 USING (id) WHERE d.id BETWEEN $1 AND $2; EXPLAIN (costs off) EXECUTE sje_p(10,20); EXECUTE sje_p(10,20); -- cte EXPLAIN (costs off) WITH z AS (SELECT id FROM sje_d2 WHERE id % 2 = 0) SELECT count(1) FROM sje_d1 d JOIN z USING (id) JOIN sje_d2 u2 USING (id); WITH z AS (SELECT id FROM sje_d2 WHERE id % 2 = 0) SELECT count(1) FROM sje_d1 d JOIN z USING (id) JOIN sje_d2 u2 USING (id); -- PG18 Feature: JSON functionality - JSON_TABLE has COLUMNS clause for -- extracting multiple fields from JSON documents. -- PG18 commit: https://github.com/postgres/postgres/commit/bb766cd CREATE TABLE pg18_json_test (id serial PRIMARY KEY, data JSON); INSERT INTO pg18_json_test (data) VALUES ('{ "user": {"name": "Alice", "age": 30, "city": "San Diego"} }'), ('{ "user": {"name": "Bob", "age": 25, "city": "Los Angeles"} }'), ('{ "user": {"name": "Charlie", "age": 35, "city": "Los Angeles"} }'), ('{ "user": {"name": "Diana", "age": 28, "city": "Seattle"} } '), ('{ "user": {"name": "Evan", "age": 40, "city": "Portland"} } '), ('{ "user": {"name": "Ethan", "age": 32, "city": "Seattle"} } '), ('{ "user": {"name": "Fiona", "age": 27, "city": "Seattle"} } '), ('{ "user": {"name": "George", "age": 29, "city": "San Francisco"} } '), ('{ "user": {"name": "Hannah", "age": 33, "city": "Seattle"} } '), ('{ "user": {"name": "Ian", "age": 26, "city": "Portland"} } '), ('{ "user": {"name": "Jane", "age": 38, "city": "San Francisco"} } '); SELECT jt.name, jt.age FROM pg18_json_test, JSON_TABLE( data, '$.user' COLUMNS ( age INT PATH '$.age', name TEXT PATH '$.name' ) ) AS jt WHERE jt.age between 25 and 35 ORDER BY jt.age, jt.name; SELECT jt.city, count(1) FROM pg18_json_test, JSON_TABLE( data, '$.user' COLUMNS ( city TEXT PATH '$.city' ) ) AS jt GROUP BY jt.city ORDER BY count(1) DESC; -- Make it distributed and repeat the queries SELECT create_distributed_table('pg18_json_test', 'id'); SELECT jt.name, jt.age FROM pg18_json_test, JSON_TABLE( data, '$.user' COLUMNS ( age INT PATH '$.age', name TEXT PATH '$.name' ) ) AS jt WHERE jt.age between 25 and 35 ORDER BY jt.age, jt.name; SELECT jt.city, count(1) FROM pg18_json_test, JSON_TABLE( data, '$.user' COLUMNS ( city TEXT PATH '$.city' ) ) AS jt GROUP BY jt.city ORDER BY count(1) DESC; -- PG18 Feature: WITHOUT OVERLAPS can appear in PRIMARY KEY and UNIQUE constraints. -- PG18 commit: https://github.com/postgres/postgres/commit/fc0438b4e CREATE TABLE temporal_rng ( -- Since we can't depend on having btree_gist here, -- use an int4range instead of an int. -- (The rangetypes regression test uses the same trick.) id int4range, valid_at daterange, CONSTRAINT temporal_rng_pk PRIMARY KEY (id, valid_at WITHOUT OVERLAPS) ); SELECT create_distributed_table('temporal_rng', 'id'); -- okay: INSERT INTO temporal_rng (id, valid_at) VALUES ('[1,2)', daterange('2018-01-02', '2018-02-03')); INSERT INTO temporal_rng (id, valid_at) VALUES ('[1,2)', daterange('2018-03-03', '2018-04-04')); INSERT INTO temporal_rng (id, valid_at) VALUES ('[2,3)', daterange('2018-01-01', '2018-01-05')); INSERT INTO temporal_rng (id, valid_at) VALUES ('[3,4)', daterange('2018-01-01', NULL)); -- should fail: INSERT INTO temporal_rng (id, valid_at) VALUES ('[1,2)', daterange('2018-01-01', '2018-01-05')); -- NULLs are not allowed in the shard key: INSERT INTO temporal_rng (id, valid_at) VALUES (NULL, daterange('2018-01-01', '2018-01-05')); INSERT INTO temporal_rng (id, valid_at) VALUES ('[3,4)', NULL); -- rejects empty: INSERT INTO temporal_rng (id, valid_at) VALUES ('[3,4)', 'empty'); SELECT * FROM temporal_rng ORDER BY id, valid_at; -- Repeat with UNIQUE constraint CREATE TABLE temporal_rng_uq ( -- Since we can't depend on having btree_gist here, -- use an int4range instead of an int. id int4range, valid_at daterange, CONSTRAINT temporal_rng_uq_uk UNIQUE (id, valid_at WITHOUT OVERLAPS) ); SELECT create_distributed_table('temporal_rng_uq', 'id'); -- okay: INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[1,2)', daterange('2018-01-02', '2018-02-03')); INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[1,2)', daterange('2018-03-03', '2018-04-04')); INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[2,3)', daterange('2018-01-01', '2018-01-05')); INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[3,4)', daterange('2018-01-01', NULL)); -- should fail: INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[1,2)', daterange('2018-01-01', '2018-01-05')); -- NULLs are not allowed in the shard key: INSERT INTO temporal_rng_uq (id, valid_at) VALUES (NULL, daterange('2018-01-01', '2018-01-05')); INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[3,4)', NULL); -- rejects empty: INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[3,4)', 'empty'); SELECT * FROM temporal_rng_uq ORDER BY id, valid_at; DROP TABLE temporal_rng CASCADE; DROP TABLE temporal_rng_uq CASCADE; -- Repeat the tests with the PRIMARY KEY and UNIQUE constraints added -- after the table is created and distributed. INSERTs produce the -- same results as before. CREATE TABLE temporal_rng ( -- Since we can't depend on having btree_gist here, -- use an int4range instead of an int. -- (The rangetypes regression test uses the same trick.) id int4range, valid_at daterange ); SELECT create_distributed_table('temporal_rng', 'id'); -- okay: INSERT INTO temporal_rng (id, valid_at) VALUES ('[1,2)', daterange('2018-01-02', '2018-02-03')); INSERT INTO temporal_rng (id, valid_at) VALUES ('[1,2)', daterange('2018-03-03', '2018-04-04')); INSERT INTO temporal_rng (id, valid_at) VALUES ('[2,3)', daterange('2018-01-01', '2018-01-05')); INSERT INTO temporal_rng (id, valid_at) VALUES ('[3,4)', daterange('2018-01-01', NULL)); ALTER TABLE temporal_rng ADD CONSTRAINT temporal_rng_pk PRIMARY KEY (id, valid_at WITHOUT OVERLAPS); -- should fail: INSERT INTO temporal_rng (id, valid_at) VALUES ('[1,2)', daterange('2018-01-01', '2018-01-05')); -- NULLs are not allowed in the shard key: INSERT INTO temporal_rng (id, valid_at) VALUES (NULL, daterange('2018-01-01', '2018-01-05')); INSERT INTO temporal_rng (id, valid_at) VALUES ('[3,4)', NULL); -- rejects empty: INSERT INTO temporal_rng (id, valid_at) VALUES ('[3,4)', 'empty'); SELECT * FROM temporal_rng ORDER BY id, valid_at; -- Repeat with UNIQUE constraint CREATE TABLE temporal_rng_uq ( -- Since we can't depend on having btree_gist here, -- use an int4range instead of an int. id int4range, valid_at daterange ); SELECT create_distributed_table('temporal_rng_uq', 'id'); -- okay: INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[1,2)', daterange('2018-01-02', '2018-02-03')); INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[1,2)', daterange('2018-03-03', '2018-04-04')); INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[2,3)', daterange('2018-01-01', '2018-01-05')); INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[3,4)', daterange('2018-01-01', NULL)); ALTER TABLE temporal_rng_uq ADD CONSTRAINT temporal_rng_uq_uk UNIQUE (id, valid_at WITHOUT OVERLAPS); -- should fail: INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[1,2)', daterange('2018-01-01', '2018-01-05')); -- NULLs are not allowed in the shard key: INSERT INTO temporal_rng_uq (id, valid_at) VALUES (NULL, daterange('2018-01-01', '2018-01-05')); INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[3,4)', NULL); -- rejects empty: INSERT INTO temporal_rng_uq (id, valid_at) VALUES ('[3,4)', 'empty'); SELECT * FROM temporal_rng_uq ORDER BY id, valid_at; -- PG18 Feature: RETURNING old and new values in DML statements -- PG18 commit: https://github.com/postgres/postgres/commit/80feb727c CREATE TABLE users (id SERIAL PRIMARY KEY, email text, category int); INSERT INTO users (email, category) SELECT 'xxx@foo.com', i % 10 from generate_series (1,100) t(i); SELECT create_distributed_table('users','id'); UPDATE users SET email = 'colm@planet.com' WHERE id = 1 RETURNING OLD.email AS previous_email, NEW.email AS current_email; SELECT * FROM users WHERE id = 1 ORDER BY id; UPDATE users SET email = 'tim@arctic.net' WHERE id = 22 RETURNING OLD.email AS previous_email, NEW.email AS current_email; UPDATE users SET email = 'john@farm.ie' WHERE id = 33 RETURNING OLD.email AS previous_email, NEW.email AS current_email; SELECT * FROM users WHERE id = 22 ORDER BY id; SELECT * FROM users WHERE email not like 'xxx@%' ORDER BY id; -- NULL values creep into the email column.. INSERT INTO users (email, category) VALUES (null, 5) RETURNING OLD.email AS previous_email, NEW.email AS current_email; UPDATE users SET email = NULL WHERE id = 79 RETURNING OLD.email AS previous_email, NEW.email AS current_email; -- Now add a NOT NULL constraint on email, but do -- not apply it to existing rows yet. ALTER TABLE users ADD CONSTRAINT users_email_not_null CHECK (email IS NOT NULL) NOT VALID; UPDATE users SET email = NULL WHERE id = 50 RETURNING OLD.email AS previous_email, NEW.email AS current_email; -- Validation should fail due to existing NULLs ALTER TABLE users VALIDATE CONSTRAINT users_email_not_null; -- Fix NULL emails to a default value UPDATE users SET email = 'xxx@foo.com' WHERE email IS NULL RETURNING OLD.email AS previous_email, NEW.email AS current_email; -- Validation should now succeed ALTER TABLE users VALIDATE CONSTRAINT users_email_not_null; -- And prevent future NULLs INSERT INTO users (email, category) VALUES (null, 10) RETURNING OLD.email AS previous_email, NEW.email AS current_email; -- PG18 Feature: support for LIKE in CREATE FOREIGN TABLE -- PG18 commit: https://github.com/postgres/postgres/commit/302cf1575 SET citus.use_citus_managed_tables TO ON; CREATE EXTENSION postgres_fdw; CREATE SERVER foreign_server FOREIGN DATA WRAPPER postgres_fdw OPTIONS (host 'localhost', port :'master_port', dbname 'regression'); CREATE USER MAPPING FOR CURRENT_USER SERVER foreign_server OPTIONS (user 'postgres'); CREATE TABLE ctl_table(a int PRIMARY KEY, b varchar COMPRESSION pglz, c int GENERATED ALWAYS AS (a * 2) STORED, d bigint GENERATED ALWAYS AS IDENTITY, e int DEFAULT 1); CREATE INDEX ctl_table_ab_key ON ctl_table(a, b); COMMENT ON COLUMN ctl_table.b IS 'Column b'; CREATE STATISTICS ctl_table_stat ON a,b FROM ctl_table; INSERT INTO ctl_table VALUES (1, 'first'), (2, 'second'), (3, 'third'), (4, 'fourth'); -- Test EXCLUDING ALL CREATE FOREIGN TABLE ctl_ft1(LIKE ctl_table EXCLUDING ALL) SERVER foreign_server OPTIONS (schema_name 'pg18_nn', table_name 'ctl_table'); -- Test INCLUDING ALL CREATE FOREIGN TABLE ctl_ft2(LIKE ctl_table INCLUDING ALL) SERVER foreign_server OPTIONS (schema_name 'pg18_nn', table_name 'ctl_table'); -- check that the foreign tables are citus local table SELECT partmethod, repmodel FROM pg_dist_partition WHERE logicalrelid IN ('ctl_ft1'::regclass, 'ctl_ft2'::regclass) ORDER BY logicalrelid; -- we can query the foreign tables EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ctl_ft1 ORDER BY a; SELECT * FROM ctl_ft1 ORDER BY a; EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ctl_ft2 ORDER BY a; SELECT * FROM ctl_ft2 ORDER BY a; -- Clean up foreign table test RESET citus.use_citus_managed_tables; SELECT undistribute_table('ctl_ft1'); SELECT undistribute_table('ctl_ft2'); DROP SERVER foreign_server CASCADE; -- PG18 Feature: PERIOD clause in foreign key constraint definitions. -- PG18 commit: https://github.com/postgres/postgres/commit/89f908a6d -- This test verifies that the PG18 tests apply to Citus tables CREATE EXTENSION btree_gist; -- needed for range type indexing CREATE TABLE temporal_test ( id integer, valid_at daterange, CONSTRAINT temporal_test_pk PRIMARY KEY (id, valid_at WITHOUT OVERLAPS) ); SET citus.shard_count TO 4; SELECT create_reference_table( 'temporal_test'); INSERT INTO temporal_test VALUES (1, '[2000-01-01,2001-01-01)'); -- same key, doesn't overlap: INSERT INTO temporal_test VALUES (1, '[2001-01-01,2002-01-01)'); -- overlaps but different key: INSERT INTO temporal_test VALUES (2, '[2000-01-01,2001-01-01)'); -- should fail: INSERT INTO temporal_test VALUES (1, '[2000-06-01,2001-01-01)'); -- Required for foreign key constraint on distributed table SET citus.shard_replication_factor TO 1; -- Create and distribute a table with temporal foreign key constraints CREATE TABLE temporal_fk_rng2rng ( id integer, valid_at daterange, parent_id integer, CONSTRAINT temporal_fk_rng2rng_pk PRIMARY KEY (id, valid_at WITHOUT OVERLAPS) ); SELECT create_distributed_table( 'temporal_fk_rng2rng', 'id'); -- -- Add foreign key constraint with PERIOD clause -- This is propagated to worker shards ALTER TABLE temporal_fk_rng2rng ADD CONSTRAINT temporal_fk_rng2rng_fk FOREIGN KEY (parent_id, PERIOD valid_at) REFERENCES temporal_test (id, PERIOD valid_at); INSERT INTO temporal_fk_rng2rng VALUES (1, '[2000-01-01,2001-01-01)', 1); -- okay spanning two parent records: INSERT INTO temporal_fk_rng2rng VALUES (2, '[2000-01-01,2002-01-01)', 1); -- key is missing INSERT INTO temporal_fk_rng2rng VALUES (3, '[2000-01-01,2001-01-01)', 3); -- key exist but is outside range INSERT INTO temporal_fk_rng2rng VALUES (4, '[2001-01-01,2002-01-01)', 2); -- key exist but is partly outside range INSERT INTO temporal_fk_rng2rng VALUES (5, '[2000-01-01,2002-01-01)', 2); -- PG18 Feature: REJECT_LIMIT option for COPY errors -- PG18 commit: https://github.com/postgres/postgres/commit/4ac2a9bec -- Citus does not support COPY with ON_ERROR so just need to -- ensure the appropriate error is returned. CREATE TABLE check_ign_err (n int, m int[], k int); SELECT create_distributed_table('check_ign_err', 'n'); COPY check_ign_err FROM STDIN WITH (on_error stop, reject_limit 5); COPY check_ign_err FROM STDIN WITH (ON_ERROR ignore, REJECT_LIMIT 100); COPY check_ign_err FROM STDIN WITH (on_error ignore, log_verbosity verbose, reject_limit 50); COPY check_ign_err FROM STDIN WITH (reject_limt 77, log_verbosity verbose, on_error ignore); -- PG requires on_error when reject_limit is specified COPY check_ign_err FROM STDIN WITH (reject_limit 100); -- PG18 Feature: COPY TABLE TO on a materialized view -- PG18 commit: https://github.com/postgres/postgres/commit/534874fac -- This does not work in Citus as a materialized view cannot be distributed. -- So just verify that the appropriate error is raised. CREATE MATERIALIZED VIEW copytest_mv AS SELECT i as id, md5(i::text) as hashval FROM generate_series(1,100) i; -- Attempting to make it distributed should fail with appropriate error as -- Citus does not yet support materialized views. SELECT create_distributed_table('copytest_mv', 'id'); -- After that, any command on the materialized view is outside Citus support. -- PG18: verify publish_generated_columns is preserved for distributed tables -- https://github.com/postgres/postgres/commit/7054186c4 \c - - - :master_port CREATE SCHEMA pg18_publication; SET search_path TO pg18_publication; -- table with a stored generated column CREATE TABLE gen_pub_tab ( id int primary key, a int, b int GENERATED ALWAYS AS (a * 10) STORED ); -- make it distributed so CREATE PUBLICATION goes through Citus metadata/DDL path SELECT create_distributed_table('gen_pub_tab', 'id', colocate_with := 'none'); -- publication using the new PG18 option: stored CREATE PUBLICATION pub_gen_cols_stored FOR TABLE gen_pub_tab WITH (publish = 'insert, update', publish_generated_columns = stored); -- second publication explicitly using "none" for completeness CREATE PUBLICATION pub_gen_cols_none FOR TABLE gen_pub_tab WITH (publish = 'insert, update', publish_generated_columns = none); -- On coordinator: pubgencols must be 's' and 'n' respectively SELECT pubname, pubgencols FROM pg_publication WHERE pubname IN ('pub_gen_cols_stored', 'pub_gen_cols_none') ORDER BY pubname; -- On worker 1: both publications must exist and keep pubgencols in sync \c - - - :worker_1_port SET search_path TO pg18_publication; SELECT pubname, pubgencols FROM pg_publication WHERE pubname IN ('pub_gen_cols_stored', 'pub_gen_cols_none') ORDER BY pubname; -- On worker 2: same check \c - - - :worker_2_port SET search_path TO pg18_publication; SELECT pubname, pubgencols FROM pg_publication WHERE pubname IN ('pub_gen_cols_stored', 'pub_gen_cols_none') ORDER BY pubname; -- Now verify ALTER PUBLICATION .. SET (publish_generated_columns = none) -- propagates to workers as well. \c - - - :master_port SET search_path TO pg18_publication; ALTER PUBLICATION pub_gen_cols_stored SET (publish_generated_columns = none); -- coordinator: both publications should now have pubgencols = 'n' SELECT pubname, pubgencols FROM pg_publication WHERE pubname IN ('pub_gen_cols_stored', 'pub_gen_cols_none') ORDER BY pubname; -- worker 1: pubgencols must match coordinator \c - - - :worker_1_port SET search_path TO pg18_publication; SELECT pubname, pubgencols FROM pg_publication WHERE pubname IN ('pub_gen_cols_stored', 'pub_gen_cols_none') ORDER BY pubname; -- worker 2: same check \c - - - :worker_2_port SET search_path TO pg18_publication; SELECT pubname, pubgencols FROM pg_publication WHERE pubname IN ('pub_gen_cols_stored', 'pub_gen_cols_none') ORDER BY pubname; -- Column list precedence test: Citus must preserve both prattrs and pubgencols \c - - - :master_port SET search_path TO pg18_publication; -- Case 1: column list explicitly includes the generated column, flag = none CREATE PUBLICATION pub_gen_cols_list_includes_b FOR TABLE gen_pub_tab (id, a, b) WITH (publish_generated_columns = none); -- Case 2: column list excludes the generated column, flag = stored CREATE PUBLICATION pub_gen_cols_list_excludes_b FOR TABLE gen_pub_tab (id, a) WITH (publish_generated_columns = stored); -- Helper: show pubname, pubgencols, and column list (prattrs) for gen_pub_tab SELECT p.pubname, p.pubgencols, r.prattrs FROM pg_publication p JOIN pg_publication_rel r ON p.oid = r.prpubid JOIN pg_class c ON c.oid = r.prrelid WHERE p.pubname IN ('pub_gen_cols_list_includes_b', 'pub_gen_cols_list_excludes_b') AND c.relname = 'gen_pub_tab' ORDER BY p.pubname; -- worker 1: must see the same pubgencols + prattrs \c - - - :worker_1_port SET search_path TO pg18_publication; SELECT p.pubname, p.pubgencols, r.prattrs FROM pg_publication p JOIN pg_publication_rel r ON p.oid = r.prpubid JOIN pg_class c ON c.oid = r.prrelid WHERE p.pubname IN ('pub_gen_cols_list_includes_b', 'pub_gen_cols_list_excludes_b') AND c.relname = 'gen_pub_tab' ORDER BY p.pubname; -- worker 2: same check \c - - - :worker_2_port SET search_path TO pg18_publication; SELECT p.pubname, p.pubgencols, r.prattrs FROM pg_publication p JOIN pg_publication_rel r ON p.oid = r.prpubid JOIN pg_class c ON c.oid = r.prrelid WHERE p.pubname IN ('pub_gen_cols_list_includes_b', 'pub_gen_cols_list_excludes_b') AND c.relname = 'gen_pub_tab' ORDER BY p.pubname; -- back to coordinator for subsequent tests / cleanup \c - - - :master_port SET search_path TO pg18_publication; DROP PUBLICATION pub_gen_cols_stored; DROP PUBLICATION pub_gen_cols_none; DROP PUBLICATION pub_gen_cols_list_includes_b; DROP PUBLICATION pub_gen_cols_list_excludes_b; -- ============================================================ -- PG18: replicate stored generated columns when included in publication column list -- Upstream: 745217a05 -- Publisher: worker1 | Subscriber: worker2 -- We validate initial COPY + streaming INSERT/UPDATE semantics. -- ============================================================ CREATE OR REPLACE FUNCTION wait_for_expected_rowcount_at_table(tableName text, expectedCount integer) RETURNS void AS $$ DECLARE actualCount integer; i int; BEGIN FOR i IN 1..600 LOOP EXECUTE FORMAT('SELECT COUNT(*) FROM %s', tableName) INTO actualCount; EXIT WHEN actualCount = expectedCount; PERFORM pg_sleep(0.05); END LOOP; IF actualCount IS DISTINCT FROM expectedCount THEN RAISE EXCEPTION 'timeout waiting for % rows in %, got %', expectedCount, tableName, actualCount; END IF; END$$ LANGUAGE plpgsql; CREATE OR REPLACE FUNCTION wait_for_expected_rowcount_at_query(query text, expectedCount integer) RETURNS void AS $$ DECLARE actualCount integer; i int; BEGIN FOR i IN 1..600 LOOP EXECUTE query INTO actualCount; EXIT WHEN actualCount = expectedCount; PERFORM pg_sleep(0.05); END LOOP; IF actualCount IS DISTINCT FROM expectedCount THEN RAISE EXCEPTION 'timeout waiting for query [%] to return %, got %', query, expectedCount, actualCount; END IF; END$$ LANGUAGE plpgsql; -- Build conninfo safely (psql vars do NOT expand inside single quotes) \set conninfo_w1 '\'user=postgres host=localhost port=' :worker_1_port ' dbname=regression\'' -- -------------------------- -- Case A: column list includes generated column b -- Expectation: subscriber receives b values (publisher b = a*10) because b is published in the column list. -- -------------------------- \c - - - :worker_1_port SET search_path TO pg18_publication; DROP PUBLICATION IF EXISTS pub_gen_repl_a; DROP TABLE IF EXISTS gen_pub_repl_a; CREATE TABLE gen_pub_repl_a ( id int PRIMARY KEY, a int, b int GENERATED ALWAYS AS (a * 10) STORED ); INSERT INTO gen_pub_repl_a (id, a) VALUES (1, 2), (2, 7); SET citus.enable_ddl_propagation TO off; CREATE PUBLICATION pub_gen_repl_a FOR TABLE gen_pub_repl_a (id, a, b) WITH (publish = 'insert, update', publish_generated_columns = none); RESET citus.enable_ddl_propagation; \c - - - :worker_2_port SET search_path TO pg18_publication; DROP SUBSCRIPTION IF EXISTS sub_gen_repl_a; DROP TABLE IF EXISTS gen_pub_repl_a; -- IMPORTANT: subscriber b is NOT generated. This is the PG18 “replicate generated values” use-case. CREATE TABLE gen_pub_repl_a ( id int PRIMARY KEY, a int, b int ); CREATE SUBSCRIPTION sub_gen_repl_a CONNECTION :conninfo_w1 PUBLICATION pub_gen_repl_a WITH (copy_data = true); SELECT wait_for_expected_rowcount_at_table('gen_pub_repl_a', 2); SELECT * FROM gen_pub_repl_a ORDER BY id; -- expect b=20,70 \c - - - :worker_1_port SET search_path TO pg18_publication; INSERT INTO gen_pub_repl_a (id, a) VALUES (3, 9); UPDATE gen_pub_repl_a SET a = 5 WHERE id = 1; \c - - - :worker_2_port SET search_path TO pg18_publication; SELECT wait_for_expected_rowcount_at_table('gen_pub_repl_a', 3); SELECT wait_for_expected_rowcount_at_query( $$SELECT COUNT(*) FROM gen_pub_repl_a WHERE id=1 AND a=5 AND b=50$$, 1); SELECT * FROM gen_pub_repl_a ORDER BY id; -- expect (1,5,50) (2,7,70) (3,9,90) -- Cleanup Case A DROP SUBSCRIPTION sub_gen_repl_a; DROP TABLE gen_pub_repl_a; \c - - - :worker_1_port SET search_path TO pg18_publication; DROP PUBLICATION pub_gen_repl_a; DROP TABLE gen_pub_repl_a; -- -------------------------- -- Case B: column list excludes generated column b but publish_generated_columns=stored -- Expectation (precedence): b is NOT replicated because column list excludes it, so subscriber b stays NULL. -- -------------------------- \c - - - :worker_1_port SET search_path TO pg18_publication; DROP PUBLICATION IF EXISTS pub_gen_repl_b; DROP TABLE IF EXISTS gen_pub_repl_b; CREATE TABLE gen_pub_repl_b ( id int PRIMARY KEY, a int, b int GENERATED ALWAYS AS (a * 10) STORED ); INSERT INTO gen_pub_repl_b (id, a) VALUES (1, 2), (2, 7); SET citus.enable_ddl_propagation TO off; CREATE PUBLICATION pub_gen_repl_b FOR TABLE gen_pub_repl_b (id, a) WITH (publish = 'insert, update', publish_generated_columns = stored); RESET citus.enable_ddl_propagation; \c - - - :worker_2_port SET search_path TO pg18_publication; DROP SUBSCRIPTION IF EXISTS sub_gen_repl_b; DROP TABLE IF EXISTS gen_pub_repl_b; CREATE TABLE gen_pub_repl_b ( id int PRIMARY KEY, a int, b int ); CREATE SUBSCRIPTION sub_gen_repl_b CONNECTION :conninfo_w1 PUBLICATION pub_gen_repl_b WITH (copy_data = true); SELECT wait_for_expected_rowcount_at_table('gen_pub_repl_b', 2); SELECT * FROM gen_pub_repl_b ORDER BY id; -- expect b is NULL \c - - - :worker_1_port SET search_path TO pg18_publication; INSERT INTO gen_pub_repl_b (id, a) VALUES (3, 9); UPDATE gen_pub_repl_b SET a = 5 WHERE id = 1; \c - - - :worker_2_port SET search_path TO pg18_publication; SELECT wait_for_expected_rowcount_at_table('gen_pub_repl_b', 3); SELECT wait_for_expected_rowcount_at_query( $$SELECT COUNT(*) FROM gen_pub_repl_b WHERE id=1 AND a=5 AND b IS NULL$$, 1); SELECT * FROM gen_pub_repl_b ORDER BY id; -- Cleanup B DROP SUBSCRIPTION sub_gen_repl_b; DROP TABLE gen_pub_repl_b; \c - - - :worker_1_port SET search_path TO pg18_publication; DROP PUBLICATION pub_gen_repl_b; DROP TABLE gen_pub_repl_b; \c - - - :master_port SET search_path TO pg18_publication; SET client_min_messages TO ERROR; DROP SCHEMA pg18_publication CASCADE; RESET client_min_messages; SET search_path TO pg18_nn; -- END: PG18: verify publish_generated_columns is preserved for distributed tables -- PG18 Feature: FOREIGN KEY constraints can be specified as NOT ENFORCED -- PG18 commit: https://github.com/postgres/postgres/commit/eec0040c4 CREATE TABLE customers( customer_id INT GENERATED ALWAYS AS IDENTITY, customer_name VARCHAR(255) NOT NULL, PRIMARY KEY(customer_id) ); SET citus.shard_replication_factor TO 1; SELECT create_distributed_table('customers', 'customer_id'); CREATE TABLE contacts( contact_id INT GENERATED ALWAYS AS IDENTITY, customer_id INT, contact_name VARCHAR(255) NOT NULL, phone VARCHAR(15), email VARCHAR(100), CONSTRAINT fk_customer FOREIGN KEY(customer_id) REFERENCES customers(customer_id) ON DELETE CASCADE NOT ENFORCED ); -- The foreign key constraint is propagated to worker nodes. SELECT create_distributed_table('contacts', 'customer_id'); SELECT pg_get_constraintdef(oid, true) AS "Definition" FROM pg_constraint WHERE conrelid = 'contacts'::regclass AND conname = 'fk_customer'; INSERT INTO customers(customer_name) VALUES('BlueBird Inc'), ('Dolphin LLC'); INSERT INTO contacts(customer_id, contact_name, phone, email) VALUES(1,'John Doe','(408)-111-1234','john.doe@example.com'), (1,'Jane Doe','(408)-111-1235','jane.doe@example.com'), (2,'David Wright','(408)-222-1234','david.wright@example.com'); DELETE FROM customers WHERE customer_name = 'Dolphin LLC'; -- After deleting 'Dolphin LLC' from customers, the corresponding contact -- 'David Wright' is not deleted from contacts due to the NOT ENFORCED. SELECT * FROM contacts ORDER BY contact_id; -- Test that ALTER TABLE .. ADD CONSTRAINT .. FOREIGN KEY .. NOT ENFORCED -- is propagated to worker nodes. First drop the foreign key: ALTER TABLE contacts DROP CONSTRAINT fk_customer; SELECT pg_get_constraintdef(oid, true) AS "Definition" FROM pg_constraint WHERE conrelid = 'contacts'::regclass AND conname = 'fk_customer'; -- Now add the foreign key constraint back with NOT ENFORCED. ALTER TABLE contacts ADD CONSTRAINT fk_customer FOREIGN KEY(customer_id) REFERENCES customers(customer_id) ON DELETE CASCADE NOT ENFORCED; -- The foreign key is propagated to worker nodes. SELECT pg_get_constraintdef(oid, true) AS "Definition" FROM pg_constraint WHERE conrelid = 'contacts'::regclass AND conname = 'fk_customer'; DELETE FROM customers WHERE customer_name = 'BlueBird Inc'; -- The customers table is now empty but the contacts table still has -- the contacts due to the NOT ENFORCED foreign key. SELECT * FROM customers ORDER BY customer_id; SELECT * FROM contacts ORDER BY contact_id; -- ALTER TABLE .. ALTER CONSTRAINT is not supported in Citus, -- so the following command should fail ALTER TABLE contacts ALTER CONSTRAINT fk_customer ENFORCED; -- PG18 Feature: ENFORCED / NOT ENFORCED check constraints -- PG18 commit: https://github.com/postgres/postgres/commit/ca87c415e -- In Citus, CHECK constraints are propagated on promoting a postgres table -- to a citus table, on adding a new CHECK constraint to a citus table, and -- on adding a node to a citus cluster. Postgres does not support altering a -- check constraint's enforcement status, so Citus does not either. CREATE TABLE NE_CHECK_TBL (x int, y int, CONSTRAINT CHECK_X CHECK (x > 3) NOT ENFORCED, CONSTRAINT CHECK_Y CHECK (y < 20) ENFORCED ); SET citus.next_shard_id TO 4754044; SELECT create_distributed_table('ne_check_tbl', 'x'); -- CHECK_X is NOT ENFORCED, so these inserts should succeed INSERT INTO NE_CHECK_TBL (x) VALUES (5), (4), (3), (2), (6), (1); SELECT x FROM NE_CHECK_TBL ORDER BY x; -- CHECK_Y is ENFORCED, so this insert should fail INSERT INTO NE_CHECK_TBL (x, y) VALUES (1, 15), (2, 25), (3, 10), (4, 30); -- Test adding new constraints with enforcement status ALTER TABLE NE_CHECK_TBL ADD CONSTRAINT CHECK_Y2 CHECK (y > 10) NOT ENFORCED; -- CHECK_Y2 is NOT ENFORCED, so these inserts should succeed INSERT INTO NE_CHECK_TBL (x, y) VALUES (1, 8), (2, 9), (3, 10), (4, 11); SELECT x, y FROM NE_CHECK_TBL ORDER BY x, y; ALTER TABLE NE_CHECK_TBL ADD CONSTRAINT CHECK_X2 CHECK (x < 10) ENFORCED; -- CHECK_X2 is ENFORCED, so these inserts should fail INSERT INTO NE_CHECK_TBL (x) VALUES (5), (15), (8), (12); -- PG18 Feature: dropping of constraints ONLY on partitioned tables -- PG18 commit: https://github.com/postgres/postgres/commit/4dea33ce7 -- Here we verify that dropping constraints ONLY on partitioned tables -- works correctly in Citus. This is done by repeating the tests of the -- PG commit (4dea33ce7) on a table that is a distributed table in Citus, -- in addition to a Postgres partitioned table. CREATE TABLE partitioned_table ( a int, b char(3) ) PARTITION BY LIST (a); SELECT create_distributed_table('partitioned_table', 'a'); -- check that violating rows are correctly reported CREATE TABLE part_2 (LIKE partitioned_table); INSERT INTO part_2 VALUES (3, 'aaa'); ALTER TABLE partitioned_table ATTACH PARTITION part_2 FOR VALUES IN (2); -- should be ok after deleting the bad row DELETE FROM part_2; ALTER TABLE partitioned_table ATTACH PARTITION part_2 FOR VALUES IN (2); -- PG18's "cannot add NOT NULL or check constraints to *only* the parent, when -- partitions exist" applies to Citus distributed tables as well. ALTER TABLE ONLY partitioned_table ALTER b SET NOT NULL; ALTER TABLE ONLY partitioned_table ADD CONSTRAINT check_b CHECK (b <> 'zzz'); -- Dropping constraints from parent should be ok ALTER TABLE partitioned_table ALTER b SET NOT NULL; ALTER TABLE ONLY partitioned_table ALTER b DROP NOT NULL; ALTER TABLE partitioned_table ADD CONSTRAINT check_b CHECK (b <> 'zzz'); ALTER TABLE ONLY partitioned_table DROP CONSTRAINT check_b; -- ... and the partitions still have the NOT NULL constraint: select relname, attname, attnotnull from pg_class inner join pg_attribute on (oid=attrelid) where relname = 'part_2' and attname = 'b' ; -- ... and the check_b constraint: select relname, conname, pg_get_expr(conbin, conrelid, true) from pg_class inner join pg_constraint on (pg_class.oid=conrelid) where relname = 'part_2' and conname = 'check_b' ; -- PG18 Feature: partitioned tables can have NOT VALID foreign keys -- PG18 commit: https://github.com/postgres/postgres/commit/b663b9436 -- As with dropping constraints only on the partitioned tables, for -- NOT VALID foreign keys, we verify that foreign key declarations -- that use NOT VALID work correctly in Citus by repeating the tests -- of the PG commit (b663b9436) on a table that is a distributed -- table in Citus, in addition to a Postgres partitioned table. CREATE TABLE fk_notpartitioned_pk (a int, b int, PRIMARY KEY (a, b), c int); CREATE TABLE fk_partitioned_fk (b int, a int) PARTITION BY RANGE (a, b); SELECT create_reference_table('fk_notpartitioned_pk'); SELECT create_distributed_table('fk_partitioned_fk', 'a'); ALTER TABLE fk_partitioned_fk ADD FOREIGN KEY (a, b) REFERENCES fk_notpartitioned_pk NOT VALID; -- Attaching a child table with the same valid foreign key constraint. CREATE TABLE fk_partitioned_fk_1 (a int, b int); ALTER TABLE fk_partitioned_fk_1 ADD FOREIGN KEY (a, b) REFERENCES fk_notpartitioned_pk; ALTER TABLE fk_partitioned_fk ATTACH PARTITION fk_partitioned_fk_1 FOR VALUES FROM (0,0) TO (1000,1000); -- Child constraint will remain valid. SELECT conname, convalidated, conrelid::regclass FROM pg_constraint WHERE conrelid::regclass::text like 'fk_partitioned_fk%' ORDER BY oid; -- Validate the constraint ALTER TABLE fk_partitioned_fk VALIDATE CONSTRAINT fk_partitioned_fk_a_b_fkey; -- All constraints are now valid. SELECT conname, convalidated, conrelid::regclass FROM pg_constraint WHERE conrelid::regclass::text like 'fk_partitioned_fk%' ORDER BY oid; -- Attaching a child with a NOT VALID constraint. CREATE TABLE fk_partitioned_fk_2 (a int, b int); INSERT INTO fk_partitioned_fk_2 VALUES(1000, 1000); -- doesn't exist in referenced table ALTER TABLE fk_partitioned_fk_2 ADD FOREIGN KEY (a, b) REFERENCES fk_notpartitioned_pk NOT VALID; -- It will fail because the attach operation implicitly validates the data. ALTER TABLE fk_partitioned_fk ATTACH PARTITION fk_partitioned_fk_2 FOR VALUES FROM (1000,1000) TO (2000,2000); -- Remove the invalid data and try again. TRUNCATE fk_partitioned_fk_2; ALTER TABLE fk_partitioned_fk ATTACH PARTITION fk_partitioned_fk_2 FOR VALUES FROM (1000,1000) TO (2000,2000); -- The child constraint will also be valid. SELECT conname, convalidated FROM pg_constraint WHERE conrelid = 'fk_partitioned_fk_2'::regclass; -- Test case where the child constraint is invalid, the grandchild constraint -- is valid, and the validation for the grandchild should be skipped when a -- valid constraint is applied to the top parent. CREATE TABLE fk_partitioned_fk_3 (a int, b int) PARTITION BY RANGE (a, b); ALTER TABLE fk_partitioned_fk_3 ADD FOREIGN KEY (a, b) REFERENCES fk_notpartitioned_pk NOT VALID; SELECT create_distributed_table('fk_partitioned_fk_3', 'a'); CREATE TABLE fk_partitioned_fk_3_1 (a int, b int); ALTER TABLE fk_partitioned_fk_3_1 ADD FOREIGN KEY (a, b) REFERENCES fk_notpartitioned_pk; SELECT create_distributed_table('fk_partitioned_fk_3_1', 'a'); ALTER TABLE fk_partitioned_fk_3 ATTACH PARTITION fk_partitioned_fk_3_1 FOR VALUES FROM (2000,2000) TO (3000,3000); -- Fails because Citus does not support multi-level (grandchild) partitions ALTER TABLE fk_partitioned_fk ATTACH PARTITION fk_partitioned_fk_3 FOR VALUES FROM (2000,2000) TO (3000,3000); -- All constraints are now valid, except for fk_partitioned_fk_3 -- because the attach failed because of Citus not yet supporting -- multi-level partitions. SELECT conname, convalidated, conrelid::regclass FROM pg_constraint WHERE conrelid::regclass::text like 'fk_partitioned_fk%' ORDER BY oid; DROP TABLE fk_partitioned_fk, fk_notpartitioned_pk CASCADE; -- NOT VALID foreign key on a non-partitioned table referencing a partitioned table CREATE TABLE fk_partitioned_pk (a int, b int, PRIMARY KEY (a, b)) PARTITION BY RANGE (a, b); SELECT create_distributed_table('fk_partitioned_pk', 'a'); CREATE TABLE fk_partitioned_pk_1 PARTITION OF fk_partitioned_pk FOR VALUES FROM (0,0) TO (1000,1000); CREATE TABLE fk_notpartitioned_fk (b int, a int); SELECT create_distributed_table('fk_notpartitioned_fk', 'a'); ALTER TABLE fk_notpartitioned_fk ADD FOREIGN KEY (a, b) REFERENCES fk_partitioned_pk NOT VALID; -- Constraint will be invalid. SELECT conname, convalidated FROM pg_constraint WHERE conrelid = 'fk_notpartitioned_fk'::regclass; ALTER TABLE fk_notpartitioned_fk VALIDATE CONSTRAINT fk_notpartitioned_fk_a_b_fkey; -- All constraints are now valid. SELECT conname, convalidated FROM pg_constraint WHERE conrelid = 'fk_notpartitioned_fk'::regclass; DROP TABLE fk_notpartitioned_fk, fk_partitioned_pk; -- PG18 Feature: Generated Virtual Columns -- PG18 commit: https://github.com/postgres/postgres/commit/83ea6c540 -- Verify that generated virtual columns are supported on distributed tables. CREATE TABLE v_reading ( celsius DECIMAL(5,2), farenheit DECIMAL(6, 2) GENERATED ALWAYS AS (celsius * 9/5 + 32) VIRTUAL, created_at TIMESTAMPTZ DEFAULT now(), device_id INT ); -- Cannot distribute on a generated column (#4616) applies -- to VIRTUAL columns. SELECT create_distributed_table('v_reading', 'farenheit'); SELECT create_distributed_table('v_reading', 'device_id'); INSERT INTO v_reading (celsius, device_id) VALUES (0, 1), (100, 1), (37.5, 2), (25, 2), (-40, 3); SELECT device_id, celsius, farenheit FROM v_reading ORDER BY device_id; ALTER TABLE v_reading ADD COLUMN kelvin DECIMAL(6, 2) GENERATED ALWAYS AS (celsius + 273.15) VIRTUAL; SELECT device_id, celsius, kelvin FROM v_reading ORDER BY device_id, celsius; -- Show all columns that are generated SELECT s.relname, a.attname, a.attgenerated FROM pg_class s JOIN pg_attribute a ON a.attrelid=s.oid WHERE s.relname LIKE 'v_reading%' and attgenerated::int != 0 ORDER BY 1,2; -- Generated columns are virtual by default - repeat the test without VIRTUAL keyword CREATE TABLE d_reading ( celsius DECIMAL(5,2), farenheit DECIMAL(6, 2) GENERATED ALWAYS AS (celsius * 9/5 + 32), created_at TIMESTAMPTZ DEFAULT now(), device_id INT ); SELECT create_distributed_table('d_reading', 'farenheit'); SELECT create_distributed_table('d_reading', 'device_id'); INSERT INTO d_reading (celsius, device_id) VALUES (0, 1), (100, 1), (37.5, 2), (25, 2), (-40, 3); SELECT device_id, celsius, farenheit FROM d_reading ORDER BY device_id; ALTER TABLE d_reading ADD COLUMN kelvin DECIMAL(6, 2) GENERATED ALWAYS AS (celsius + 273.15) VIRTUAL; SELECT device_id, celsius, kelvin FROM d_reading ORDER BY device_id, celsius; -- Show all columns that are generated SELECT s.relname, a.attname, a.attgenerated FROM pg_class s JOIN pg_attribute a ON a.attrelid=s.oid WHERE s.relname LIKE 'd_reading%' and attgenerated::int != 0 ORDER BY 1,2; -- COPY implementation needs to handle GENERATED ALWAYS AS (...) VIRTUAL columns. \COPY d_reading FROM STDIN WITH DELIMITER ',' 3.00,2025-11-24 09:46:17.390872+00,1 6.00,2025-11-24 09:46:17.390872+00,5 2.00,2025-11-24 09:46:17.390872+00,1 22.00,2025-11-24 09:46:17.390872+00,5 15.00,2025-11-24 09:46:17.390872+00,1 13.00,2025-11-24 09:46:17.390872+00,5 27.00,2025-11-24 09:46:17.390872+00,1 14.00,2025-11-24 09:46:17.390872+00,5 2.00,2025-11-24 09:46:17.390872+00,1 23.00,2025-11-24 09:46:17.390872+00,5 22.00,2025-11-24 09:46:17.390872+00,1 3.00,2025-11-24 09:46:17.390872+00,5 2.00,2025-11-24 09:46:17.390872+00,1 7.00,2025-11-24 09:46:17.390872+00,5 6.00,2025-11-24 09:46:17.390872+00,1 21.00,2025-11-24 09:46:17.390872+00,5 30.00,2025-11-24 09:46:17.390872+00,1 1.00,2025-11-24 09:46:17.390872+00,5 31.00,2025-11-24 09:46:17.390872+00,1 22.00,2025-11-24 09:46:17.390872+00,5 \. SELECT device_id, count(device_id) as count, round(avg(celsius), 2) as avg, min(farenheit), max(farenheit) FROM d_reading GROUP BY device_id ORDER BY count DESC; -- Test GROUP BY on tables with generated virtual columns - this requires -- special case handling in distributed planning. Test it out on some -- some queries involving joins and set operations. SELECT device_id, max(kelvin) as Kel FROM v_reading WHERE (device_id, celsius) NOT IN (SELECT device_id, max(celsius) FROM v_reading GROUP BY device_id) GROUP BY device_id ORDER BY device_id ASC; SELECT device_id, round(AVG( (d_farenheit + v_farenheit) / 2), 2) as Avg_Far FROM (SELECT * FROM (SELECT device_id, round(AVG(farenheit),2) as d_farenheit FROM d_reading GROUP BY device_id) AS subq RIGHT JOIN (SELECT device_id, MAX(farenheit) AS v_farenheit FROM d_reading GROUP BY device_id) AS subq2 USING (device_id) ) AS finalq GROUP BY device_id ORDER BY device_id ASC; SELECT device_id, MAX(farenheit) as farenheit FROM ((SELECT device_id, round(AVG(farenheit),2) as farenheit FROM d_reading GROUP BY device_id) UNION ALL (SELECT device_id, MAX(farenheit) AS farenheit FROM d_reading GROUP BY device_id) ) AS unioned GROUP BY device_id ORDER BY device_id ASC; SELECT device_id, MAX(farenheit) as farenheit FROM ((SELECT device_id, round(AVG(farenheit),2) as farenheit FROM d_reading GROUP BY device_id) INTERSECT (SELECT device_id, MAX(farenheit) AS farenheit FROM d_reading GROUP BY device_id) ) AS intersected GROUP BY device_id ORDER BY device_id ASC; SELECT device_id, MAX(farenheit) as farenheit FROM ((SELECT device_id, round(AVG(farenheit),2) as farenheit FROM d_reading GROUP BY device_id) EXCEPT (SELECT device_id, MAX(farenheit) AS farenheit FROM d_reading GROUP BY device_id) ) AS excepted GROUP BY device_id ORDER BY device_id ASC; -- Ensure that UDFs such as alter_distributed_table, undistribute_table -- and add_local_table_to_metadata work fine with VIRTUAL columns. For -- this, PR #4616 changes are modified to handle VIRTUAL columns in -- addition to STORED columns. CREATE TABLE generated_stored_dist ( col_1 int, "col\'_2" text, col_3 text generated always as (UPPER("col\'_2")) virtual ); SELECT create_distributed_table ('generated_stored_dist', 'col_1'); INSERT INTO generated_stored_dist VALUES (1, 'text_1'), (2, 'text_2'); SELECT * FROM generated_stored_dist ORDER BY 1,2,3; INSERT INTO generated_stored_dist VALUES (1, 'text_1'), (2, 'text_2'); SELECT alter_distributed_table('generated_stored_dist', shard_count := 5, cascade_to_colocated := false); SELECT * FROM generated_stored_dist ORDER BY 1,2,3; CREATE TABLE generated_stored_local ( col_1 int, "col\'_2" text, col_3 text generated always as (UPPER("col\'_2")) stored ); SELECT citus_add_local_table_to_metadata('generated_stored_local'); INSERT INTO generated_stored_local VALUES (1, 'text_1'), (2, 'text_2'); SELECT * FROM generated_stored_local ORDER BY 1,2,3; SELECT create_distributed_table ('generated_stored_local', 'col_1'); INSERT INTO generated_stored_local VALUES (1, 'text_1'), (2, 'text_2'); SELECT * FROM generated_stored_local ORDER BY 1,2,3; CREATE TABLE generated_stored_ref ( col_1 int, col_2 int, col_3 int generated always as (col_1+col_2) virtual, col_4 int, col_5 int generated always as (col_4*2-col_1) virtual ); SELECT create_reference_table ('generated_stored_ref'); INSERT INTO generated_stored_ref (col_1, col_4) VALUES (1,2), (11,12); INSERT INTO generated_stored_ref (col_1, col_2, col_4) VALUES (100,101,102), (200,201,202); SELECT * FROM generated_stored_ref ORDER BY 1,2,3,4,5; BEGIN; SELECT undistribute_table('generated_stored_ref'); INSERT INTO generated_stored_ref (col_1, col_4) VALUES (11,12), (21,22); INSERT INTO generated_stored_ref (col_1, col_2, col_4) VALUES (200,201,202), (300,301,302); SELECT * FROM generated_stored_ref ORDER BY 1,2,3,4,5; ROLLBACK; BEGIN; -- drop some of the columns not having "generated always as virtual" expressions SET client_min_messages TO WARNING; ALTER TABLE generated_stored_ref DROP COLUMN col_1 CASCADE; RESET client_min_messages; ALTER TABLE generated_stored_ref DROP COLUMN col_4; -- show that undistribute_table works fine SELECT undistribute_table('generated_stored_ref'); INSERT INTO generated_stored_ref VALUES (5); SELECT * FROM generated_stored_REF ORDER BY 1; ROLLBACK; BEGIN; -- now drop all columns ALTER TABLE generated_stored_ref DROP COLUMN col_3; ALTER TABLE generated_stored_ref DROP COLUMN col_5; ALTER TABLE generated_stored_ref DROP COLUMN col_1; ALTER TABLE generated_stored_ref DROP COLUMN col_2; ALTER TABLE generated_stored_ref DROP COLUMN col_4; -- show that undistribute_table works fine SELECT undistribute_table('generated_stored_ref'); SELECT * FROM generated_stored_ref; ROLLBACK; -- PG18 Feature: VACUUM/ANALYZE support ONLY to limit processing to the parent. -- For Citus, ensure ONLY does not trigger shard propagation. -- PG18 commit: https://github.com/postgres/postgres/commit/62ddf7ee9 CREATE SCHEMA pg18_vacuum_part; SET search_path TO pg18_vacuum_part; CREATE TABLE vac_analyze_only (a int); SELECT create_distributed_table('vac_analyze_only', 'a'); INSERT INTO vac_analyze_only VALUES (1), (2), (3); -- ANALYZE (no ONLY) should recurse into shard placements ANALYZE vac_analyze_only; \c - - - :worker_1_port SET search_path TO pg18_vacuum_part; SELECT coalesce(max(last_analyze), 'epoch'::timestamptz) AS analyze_before_only FROM pg_stat_user_tables WHERE schemaname = 'pg18_vacuum_part' AND relname LIKE 'vac_analyze_only_%' \gset \c - - - :master_port SET search_path TO pg18_vacuum_part; -- ANALYZE ONLY should not recurse into shard placements ANALYZE ONLY vac_analyze_only; \c - - - :worker_1_port SET search_path TO pg18_vacuum_part; SELECT max(last_analyze) = :'analyze_before_only'::timestamptz AS analyze_only_skipped FROM pg_stat_user_tables WHERE schemaname = 'pg18_vacuum_part' AND relname LIKE 'vac_analyze_only_%'; \c - - - :master_port SET search_path TO pg18_vacuum_part; -- VACUUM (no ONLY) should recurse into shard placements VACUUM vac_analyze_only; \c - - - :worker_1_port SET search_path TO pg18_vacuum_part; SELECT coalesce(max(last_vacuum), 'epoch'::timestamptz) AS vacuum_before_only FROM pg_stat_user_tables WHERE schemaname = 'pg18_vacuum_part' AND relname LIKE 'vac_analyze_only_%' \gset \c - - - :master_port SET search_path TO pg18_vacuum_part; -- VACUUM ONLY should not recurse into shard placements VACUUM ONLY vac_analyze_only; \c - - - :worker_1_port SET search_path TO pg18_vacuum_part; SELECT max(last_vacuum) = :'vacuum_before_only'::timestamptz AS vacuum_only_skipped FROM pg_stat_user_tables WHERE schemaname = 'pg18_vacuum_part' AND relname LIKE 'vac_analyze_only_%'; \c - - - :master_port SET search_path TO pg18_vacuum_part; DROP SCHEMA pg18_vacuum_part CASCADE; SET search_path TO pg18_nn; -- END PG18 Feature: VACUUM/ANALYZE support ONLY to limit processing to the parent -- PG18 Feature: VACUUM/ANALYZE ONLY on a partitioned distributed table -- Ensure Citus does not recurse into shard placements when ONLY is used -- on the partitioned parent. -- PG18 commit: https://github.com/postgres/postgres/commit/62ddf7ee9 CREATE SCHEMA pg18_vacuum_part_dist; SET search_path TO pg18_vacuum_part_dist; SET citus.shard_count = 2; SET citus.shard_replication_factor = 1; CREATE TABLE part_dist (id int, v int) PARTITION BY RANGE (id); CREATE TABLE part_dist_1 PARTITION OF part_dist FOR VALUES FROM (1) TO (100); CREATE TABLE part_dist_2 PARTITION OF part_dist FOR VALUES FROM (100) TO (200); SELECT create_distributed_table('part_dist', 'id'); INSERT INTO part_dist SELECT g, g FROM generate_series(1, 199) g; -- ANALYZE (no ONLY) should recurse into partitions and shard placements ANALYZE part_dist; \c - - - :worker_1_port SET search_path TO pg18_vacuum_part_dist; SELECT coalesce(max(last_analyze), 'epoch'::timestamptz) AS analyze_before_only FROM pg_stat_user_tables WHERE schemaname = 'pg18_vacuum_part_dist' AND relname LIKE 'part_dist_%' \gset \c - - - :master_port SET search_path TO pg18_vacuum_part_dist; -- ANALYZE ONLY should not recurse into shard placements ANALYZE ONLY part_dist; \c - - - :worker_1_port SET search_path TO pg18_vacuum_part_dist; SELECT max(last_analyze) = :'analyze_before_only'::timestamptz AS analyze_only_partitioned_skipped FROM pg_stat_user_tables WHERE schemaname = 'pg18_vacuum_part_dist' AND relname LIKE 'part_dist_%'; \c - - - :master_port SET search_path TO pg18_vacuum_part_dist; -- VACUUM (no ONLY) should recurse into partitions and shard placements VACUUM part_dist; \c - - - :worker_1_port SET search_path TO pg18_vacuum_part_dist; SELECT coalesce(max(last_vacuum), 'epoch'::timestamptz) AS vacuum_before_only FROM pg_stat_user_tables WHERE schemaname = 'pg18_vacuum_part_dist' AND relname LIKE 'part_dist_%' \gset \c - - - :master_port SET search_path TO pg18_vacuum_part_dist; -- VACUUM ONLY parent: core warns and does no work; Citus must not -- propagate to shard placements. VACUUM ONLY part_dist; \c - - - :worker_1_port SET search_path TO pg18_vacuum_part_dist; SELECT max(last_vacuum) = :'vacuum_before_only'::timestamptz AS vacuum_only_partitioned_skipped FROM pg_stat_user_tables WHERE schemaname = 'pg18_vacuum_part_dist' AND relname LIKE 'part_dist_%'; \c - - - :master_port SET search_path TO pg18_vacuum_part_dist; DROP SCHEMA pg18_vacuum_part_dist CASCADE; SET search_path TO pg18_nn; -- END PG18 Feature: VACUUM/ANALYZE ONLY on partitioned distributed table -- PG18 Feature: text search with nondeterministic collations -- PG18 commit: https://github.com/postgres/postgres/commit/329304c90 -- This test verifies that the PG18 tests apply to Citus tables; Citus -- just passes through the collation info and text search queries to -- worker shards. CREATE COLLATION ignore_accents (provider = icu, locale = '@colStrength=primary;colCaseLevel=yes', deterministic = false); -- nondeterministic collations CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true); CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false); CREATE TABLE strtest1 (a int, b text); SELECT create_distributed_table('strtest1', 'a'); INSERT INTO strtest1 VALUES (1, U&'zy\00E4bc'); INSERT INTO strtest1 VALUES (2, U&'zy\0061\0308bc'); INSERT INTO strtest1 VALUES (3, U&'ab\00E4cd'); INSERT INTO strtest1 VALUES (4, U&'ab\0061\0308cd'); INSERT INTO strtest1 VALUES (5, U&'ab\00E4cd'); INSERT INTO strtest1 VALUES (6, U&'ab\0061\0308cd'); INSERT INTO strtest1 VALUES (7, U&'ab\00E4cd'); SELECT * FROM strtest1 WHERE b = 'zyäbc' COLLATE ctest_det ORDER BY a; SELECT * FROM strtest1 WHERE b = 'zyäbc' COLLATE ctest_nondet ORDER BY a; SELECT strpos(b COLLATE ctest_det, 'bc') FROM strtest1 ORDER BY a; SELECT strpos(b COLLATE ctest_nondet, 'bc') FROM strtest1 ORDER BY a; SELECT replace(b COLLATE ctest_det, U&'\00E4b', 'X') FROM strtest1 ORDER BY a; SELECT replace(b COLLATE ctest_nondet, U&'\00E4b', 'X') FROM strtest1 ORDER BY a; SELECT a, split_part(b COLLATE ctest_det, U&'\00E4b', 2) FROM strtest1 ORDER BY a; SELECT a, split_part(b COLLATE ctest_nondet, U&'\00E4b', 2) FROM strtest1 ORDER BY a; SELECT a, split_part(b COLLATE ctest_det, U&'\00E4b', -1) FROM strtest1 ORDER BY a; SELECT a, split_part(b COLLATE ctest_nondet, U&'\00E4b', -1) FROM strtest1 ORDER BY a; SELECT a, string_to_array(b COLLATE ctest_det, U&'\00E4b') FROM strtest1 ORDER BY a; SELECT a, string_to_array(b COLLATE ctest_nondet, U&'\00E4b') FROM strtest1 ORDER BY a; SELECT * FROM strtest1 WHERE b LIKE 'zyäbc' COLLATE ctest_det ORDER BY a; SELECT * FROM strtest1 WHERE b LIKE 'zyäbc' COLLATE ctest_nondet ORDER BY a; CREATE TABLE strtest2 (a int, b text); SELECT create_distributed_table('strtest2', 'a'); INSERT INTO strtest2 VALUES (1, 'cote'), (2, 'côte'), (3, 'coté'), (4, 'côté'); CREATE TABLE strtest2nfd (a int, b text); SELECT create_distributed_table('strtest2nfd', 'a'); INSERT INTO strtest2nfd VALUES (1, 'cote'), (2, 'côte'), (3, 'coté'), (4, 'côté'); UPDATE strtest2nfd SET b = normalize(b, nfd); -- This shows why replace should be greedy. Otherwise, in the NFD -- case, the match would stop before the decomposed accents, which -- would leave the accents in the results. SELECT a, b, replace(b COLLATE ignore_accents, 'co', 'ma') FROM strtest2 ORDER BY a, b; SELECT a, b, replace(b COLLATE ignore_accents, 'co', 'ma') FROM strtest2nfd ORDER BY a, b; -- PG18 Feature: LIKE support for non-deterministic collations -- PG18 commit: https://github.com/postgres/postgres/commit/85b7efa1c -- As with non-deterministic collation text search, we verify that -- LIKE with non-deterministic collation is passed through by Citus -- and expected results are returned by the queries. INSERT INTO strtest1 VALUES (8, U&'abc'); INSERT INTO strtest1 VALUES (9, 'abc'); SELECT a, b FROM strtest1 WHERE b LIKE 'abc' COLLATE ctest_det ORDER BY a; SELECT a, b FROM strtest1 WHERE b LIKE 'a\bc' COLLATE ctest_det ORDER BY a; SELECT a, b FROM strtest1 WHERE b LIKE 'abc' COLLATE ctest_nondet ORDER BY a; SELECT a, b FROM strtest1 WHERE b LIKE 'a\bc' COLLATE ctest_nondet ORDER BY a; CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false); SELECT a, b FROM strtest1 WHERE b LIKE 'ABC' COLLATE case_insensitive ORDER BY a; SELECT a, b FROM strtest1 WHERE b LIKE 'ABC%' COLLATE case_insensitive ORDER BY a; INSERT INTO strtest1 VALUES (10, U&'\00E4bc'); INSERT INTO strtest1 VALUES (12, U&'\0061\0308bc'); SELECT * FROM strtest1 WHERE b LIKE 'äbc' COLLATE ctest_det ORDER BY a; SELECT * FROM strtest1 WHERE b LIKE 'äbc' COLLATE ctest_nondet ORDER BY a; -- Tests with ignore_accents collation. Taken from -- PG18 regress tests and applied to a Citus table. INSERT INTO strtest1 VALUES (10, U&'\0061\0308bc'); INSERT INTO strtest1 VALUES (11, U&'\00E4bc'); INSERT INTO strtest1 VALUES (12, U&'cb\0061\0308'); INSERT INTO strtest1 VALUES (13, U&'\0308bc'); INSERT INTO strtest1 VALUES (14, 'foox'); SELECT a, b FROM strtest1 WHERE b LIKE U&'\00E4_c' COLLATE ignore_accents ORDER BY a, b; -- and in reverse: SELECT a, b FROM strtest1 WHERE b LIKE U&'\0061\0308_c' COLLATE ignore_accents ORDER BY a, b; -- inner % matches b: SELECT a, b FROM strtest1 WHERE b LIKE U&'\00E4%c' COLLATE ignore_accents ORDER BY a, b; -- inner %% matches b then zero: SELECT a, b FROM strtest1 WHERE b LIKE U&'\00E4%%c' COLLATE ignore_accents ORDER BY a, b; -- inner %% matches b then zero: SELECT a, b FROM strtest1 WHERE b LIKE U&'c%%\00E4' COLLATE ignore_accents ORDER BY a, b; -- trailing _ matches two codepoints that form one grapheme: SELECT a, b FROM strtest1 WHERE b LIKE U&'cb_' COLLATE ignore_accents ORDER BY a, b; -- trailing __ matches two codepoints that form one grapheme: SELECT a, b FROM strtest1 WHERE b LIKE U&'cb__' COLLATE ignore_accents ORDER BY a, b; -- leading % matches zero: SELECT a, b FROM strtest1 WHERE b LIKE U&'%\00E4bc' COLLATE ignore_accents ORDER BY a; -- leading % matches zero (with later %): SELECT a, b FROM strtest1 WHERE b LIKE U&'%\00E4%c' COLLATE ignore_accents ORDER BY a, b; -- trailing % matches zero: SELECT a, b FROM strtest1 WHERE b LIKE U&'\00E4bc%' COLLATE ignore_accents ORDER BY a, b; -- trailing % matches zero (with previous %): SELECT a, b FROM strtest1 WHERE b LIKE U&'\00E4%c%' COLLATE ignore_accents ORDER BY a, b; -- _ versus two codepoints that form one grapheme: SELECT a, b FROM strtest1 WHERE b LIKE U&'_bc' COLLATE ignore_accents ORDER BY a, b; -- (actually this matches because) SELECT a, b FROM strtest1 WHERE b = 'bc' COLLATE ignore_accents ORDER BY a, b; -- __ matches two codepoints that form one grapheme: SELECT a, b FROM strtest1 WHERE b LIKE U&'__bc' COLLATE ignore_accents ORDER BY a, b; -- _ matches one codepoint that forms half a grapheme: SELECT a, b FROM strtest1 WHERE b LIKE U&'_\0308bc' COLLATE ignore_accents ORDER BY a, b; -- doesn't match because \00e4 doesn't match only \0308 SELECT a, b FROM strtest1 WHERE b LIKE U&'_\00e4bc' COLLATE ignore_accents ORDER BY a, b; -- escape character at end of pattern SELECT a, b FROM strtest1 WHERE b LIKE 'foo\' COLLATE ignore_accents ORDER BY a, b; DROP TABLE strtest1; DROP COLLATION ignore_accents; DROP COLLATION ctest_det; DROP COLLATION ctest_nondet; DROP COLLATION case_insensitive; -- PG18 Feature: GUC for CREATE DATABASE file copy method -- PG18 commit: https://github.com/postgres/postgres/commit/f78ca6f3e -- Citus supports the wal_log strategy only for CREATE DATABASE. -- Here we show that the expected error (from PR #7249) occurs -- when the file_copy strategy is attempted. SET citus.enable_create_database_propagation=on; SHOW file_copy_method; -- Error output is expected here CREATE DATABASE copied_db WITH strategy file_copy; SET file_copy_method TO clone; -- Also errors out, per #7249 CREATE DATABASE cloned_db WITH strategy file_copy; RESET file_copy_method; -- This is okay CREATE DATABASE copied_db WITH strategy wal_log; -- Show that file_copy works for ALTER DATABASE ... SET TABLESPACE \set alter_db_tablespace :abs_srcdir '/tmp_check/ts3' CREATE TABLESPACE alter_db_tablespace LOCATION :'alter_db_tablespace'; \c - - - :worker_1_port \set alter_db_tablespace :abs_srcdir '/tmp_check/ts4' CREATE TABLESPACE alter_db_tablespace LOCATION :'alter_db_tablespace'; \c - - - :worker_2_port \set alter_db_tablespace :abs_srcdir '/tmp_check/ts5' CREATE TABLESPACE alter_db_tablespace LOCATION :'alter_db_tablespace'; \c - - - :master_port SET citus.enable_create_database_propagation TO on; SET file_copy_method TO clone; SET citus.log_remote_commands TO true; SELECT datname, spcname FROM pg_database d, pg_tablespace t WHERE d.dattablespace = t.oid AND d.datname = 'copied_db'; ALTER DATABASE copied_db SET TABLESPACE alter_db_tablespace; SELECT datname, spcname FROM pg_database d, pg_tablespace t WHERE d.dattablespace = t.oid AND d.datname = 'copied_db'; RESET file_copy_method; RESET citus.log_remote_commands; -- Enable alter_db_tablespace to be dropped ALTER DATABASE copied_db SET TABLESPACE pg_default; DROP DATABASE copied_db; -- Done with DATABASE commands RESET citus.enable_create_database_propagation; SELECT result FROM run_command_on_all_nodes( $$ DROP TABLESPACE "alter_db_tablespace" $$ ); -- cleanup with minimum verbosity SET client_min_messages TO ERROR; RESET search_path; RESET citus.shard_count; RESET citus.shard_replication_factor; DROP SCHEMA pg18_nn CASCADE; RESET client_min_messages;