PG17 - Add Regression Test for Access Method Behavior on Partitioned Tables #7818
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## release-13.0 #7818 +/- ##
===============================================
Coverage ? 89.63%
===============================================
Files ? 274
Lines ? 59748
Branches ? 7452
===============================================
Hits ? 53556
Misses ? 4058
Partials ? 2134 |
src/test/regress/expected/pg17.out
Outdated
|
|
||
| (1 row) | ||
|
|
||
| -- Step 4: Verify that the table and partitions are created and distributed correctly |
There was a problem hiding this comment.
We have to verify step 4 in one of the worker nodes as well
|
|
||
| -- Step 5: Test ALTER TABLE ... SET ACCESS METHOD to a different method | ||
| ALTER TABLE test_partitioned_alter SET ACCESS METHOD columnar; | ||
| -- Step 6: Verify the change is applied to future partitions |
There was a problem hiding this comment.
similarly we have to verify step 5&6 in one of the worker nodes as well
There was a problem hiding this comment.
I was trying to verify -- Step 5: Test ALTER TABLE ... SET ACCESS METHOD to a different method in the worker nodes and I ran into a problem: the access method of the partitions is not updated! So, this PR needs code changes. The following output is from one of the worker nodes:
postgres=# \d+ test_partitioned_alter
Partitioned table "public.test_partitioned_alter"
...
Partitions: test_partition_1 FOR VALUES FROM (0) TO (100),
test_partition_2 FOR VALUES FROM (100) TO (200)
Access method: columnar
postgres=# \d+ test_partition_1
Table "public.test_partition_1"
...
Partition of: test_partitioned_alter FOR VALUES FROM (0) TO (100)
...
Access method: heap
There was a problem hiding this comment.
I just checked with the Postgres commit, and actually the existing partitions are not modified.
Specifying an AM for a partitioned table lets the value be used for all future partitions created under it, closely mirroring the behavior of the TABLESPACE option for partitioned
tables. Existing partitions are not modified.
Therefore, Citus is behaving correctly.
We need to modify the tests a little bit, I will leave some comments.
src/test/regress/sql/pg17.sql
Outdated
| -- Step 4: Verify that the table and partitions are created and distributed correctly | ||
| SELECT relname, relam | ||
| FROM pg_class | ||
| WHERE relname = 'test_partitioned_alter'; | ||
|
|
||
| -- Check the partitions' access methods | ||
| SELECT relname, relam | ||
| FROM pg_class | ||
| WHERE relname IN ('test_partition_1', 'test_partition_2') | ||
| ORDER BY relname; |
There was a problem hiding this comment.
We should repeat these steps in one of the worker nodes as well.
| -- Step 6: Verify the change is applied to future partitions | ||
| CREATE TABLE test_partition_3 PARTITION OF test_partitioned_alter | ||
| FOR VALUES FROM (200) TO (300); | ||
|
|
There was a problem hiding this comment.
Between step 5 and step 6 we should add another step 4 (in the coordinator) to verify that the access method in the distributed parent changed, but it didn't change in the existing partitions.
m3hm3t
force-pushed
the
m3hm3t/allow_access_part_table
branch
2 times, most recently
from
0709c77 to
97e4e1f
Compare
m3hm3t
force-pushed
the
m3hm3t/allow_access_part_table
branch
from
aaa940a to
947cd4d
Compare
m3hm3t
force-pushed
the
m3hm3t/allow_access_part_table
branch
from
947cd4d to
7d61038
Compare
update update fix whitespace update
m3hm3t
force-pushed
the
m3hm3t/allow_access_part_table
branch
from
7d61038 to
280456f
Compare
This PR adds a regression test to verify the behavior of access methods for partitioned and distributed tables, including: