[MDEV-28127] EXCHANGE PARTITION with non-matching vcol expression segfault Created: 2022-03-19  Updated: 2023-11-13  Resolved: 2023-11-10

Status: Closed
Project: MariaDB Server
Component/s: Virtual Columns
Affects Version/s: 10.2, 10.3, 10.4, 10.5, 10.6, 10.7, 10.8, 10.9, 10.10
Fix Version/s: 10.4.33, 10.5.24, 10.6.17, 10.11.7, 11.0.5, 11.1.4, 11.2.3

Type: Bug Priority: Critical
Reporter: Roel Van de Paar Assignee: Aleksey Midenkov
Resolution: Fixed Votes: 0
Labels: pushing


 Description   

SIGSEGV in row_mysql_store_col_in_innobase_format, and Assertion `pos < table->n_v_def' failed in dict_table_get_nth_v_col, and Assertion `table->s->db_options_in_use == part_table->s->db_options_in_use' failed in compare_table_with_partition

Even though it is present in all versions, this may be a new regression. Feel free to remove tag and lower prio if not.

SET sql_mode='';
 
CREATE TABLE t1 (a INT,KEY(a)) ENGINE=InnoDB PARTITION BY RANGE (a) (PARTITION p VALUES LESS THAN (1));
 
CREATE TABLE t (a INT GENERATED ALWAYS AS (1) VIRTUAL,KEY(a));
 
ALTER TABLE t1 EXCHANGE PARTITION p WITH TABLE t;
 
INSERT INTO t VALUES (1);

Leads to:

10.9.0 5be92887c2caacb45af87b1131db952ce627e83a (Optimized)

 
Core was generated by `/test/MD160322-mariadb-10.9.0-linux-x86_64-opt/bin/mysqld --no-defaults --core-'.
 
Program terminated with signal SIGSEGV, Segmentation fault.
 
#0  0x0000556de2e400eb in row_mysql_store_col_in_innobase_format (dfield=0x0, 
    buf=0x152df402ad11 "", row_format_col=1, mysql_data=0x152df40293e9 "\001", 
    col_len=4, comp=1) at /test/10.9_opt/storage/innobase/row/row0mysql.cc:363
 
363		type = dtype->mtype;
 
[Current thread is 1 (Thread 0x152e4c11a700 (LWP 4084564))]
 
(gdb) bt
 
#0  0x0000556de2e400eb in row_mysql_store_col_in_innobase_format (dfield=0x0, buf=0x152df402ad11 "", row_format_col=1, mysql_data=0x152df40293e9 "\001", col_len=4, comp=1) at /test/10.9_opt/storage/innobase/row/row0mysql.cc:363
 
#1  0x0000556de2e42bbc in row_mysql_convert_row_to_innobase (blob_heap=<synthetic pointer>, mysql_rec=0x152df40293e8 "\375\001", prebuilt=0x152df402a760, row=0x152df402ad18) at /test/10.9_opt/storage/innobase/include/dict0mem.h:1820
 
#2  row_insert_for_mysql (mysql_rec=mysql_rec@entry=0x152df40293e8 "\375\001", prebuilt=0x152df402a760, ins_mode=ROW_INS_NORMAL) at /test/10.9_opt/storage/innobase/row/row0mysql.cc:1290
 
#3  0x0000556de2d919da in ha_innobase::write_row (this=0x152df4029800, record=0x152df40293e8 "\375\001") at /test/10.9_opt/storage/innobase/handler/ha_innodb.cc:7800
 
#4  0x0000556de2a9af30 in handler::ha_write_row (this=0x152df4029800, buf=0x152df40293e8 "\375\001") at /test/10.9_opt/sql/handler.cc:7527
 
#5  0x0000556de280421d in write_record (thd=thd@entry=0x152df4000c58, table=table@entry=0x152df40177d8, info=info@entry=0x152e4c118c80, sink=sink@entry=0x0) at /test/10.9_opt/sql/sql_insert.cc:2156
 
#6  0x0000556de280aa68 in mysql_insert (thd=thd@entry=0x152df4000c58, table_list=<optimized out>, fields=@0x152df4005d60: {<base_list> = {<Sql_alloc> = {<No data fields>}, first = 0x556de39d63b0 <end_of_list>, last = 0x152df4005d60, elements = 0}, <No data fields>}, values_list=@0x152df4005da8: {<base_list> = {<Sql_alloc> = {<No data fields>}, first = 0x152df40115a0, last = 0x152df40115a0, elements = 1}, <No data fields>}, update_fields=@0x152df4005d90: {<base_list> = {<Sql_alloc> = {<No data fields>}, first = 0x556de39d63b0 <end_of_list>, last = 0x152df4005d90, elements = 0}, <No data fields>}, update_values=@0x152df4005d78: {<base_list> = {<Sql_alloc> = {<No data fields>}, first = 0x556de39d63b0 <end_of_list>, last = 0x152df4005d78, elements = 0}, <No data fields>}, duplic=<optimized out>, ignore=<optimized out>, result=<optimized out>) at /test/10.9_opt/sql/sql_insert.cc:1127
 
#7  0x0000556de284596f in mysql_execute_command (thd=0x152df4000c58, is_called_from_prepared_stmt=<optimized out>) at /test/10.9_opt/sql/sql_parse.cc:4562
 
#8  0x0000556de28351c6 in mysql_parse (thd=0x152df4000c58, rawbuf=<optimized out>, length=<optimized out>, parser_state=<optimized out>) at /test/10.9_opt/sql/sql_parse.cc:8027
 
#9  0x0000556de2841375 in dispatch_command (command=COM_QUERY, thd=0x152df4000c58, packet=<optimized out>, packet_length=<optimized out>, blocking=<optimized out>) at /test/10.9_opt/sql/sql_class.h:1362
 
#10 0x0000556de2843567 in do_command (thd=0x152df4000c58, blocking=blocking@entry=true) at /test/10.9_opt/sql/sql_parse.cc:1402
 
#11 0x0000556de2962e97 in do_handle_one_connection (connect=<optimized out>, put_in_cache=true) at /test/10.9_opt/sql/sql_connect.cc:1418
 
#12 0x0000556de29631dd in handle_one_connection (arg=arg@entry=0x556de630d458) at /test/10.9_opt/sql/sql_connect.cc:1312
 
#13 0x0000556de2cdc8d1 in pfs_spawn_thread (arg=0x556de628cbc8) at /test/10.9_opt/storage/perfschema/pfs.cc:2201
 
#14 0x0000152e632d5609 in start_thread (arg=<optimized out>) at pthread_create.c:477
 
#15 0x0000152e62ec1163 in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95

10.9.0 5be92887c2caacb45af87b1131db952ce627e83a (Debug)

 
mysqld: /test/10.9_dbg/storage/innobase/include/dict0dict.inl:387: dict_v_col_t* dict_table_get_nth_v_col(const dict_table_t*, ulint): Assertion `pos < table->n_v_def' failed.

10.9.0 5be92887c2caacb45af87b1131db952ce627e83a (Debug)

 
Core was generated by `/test/MD160322-mariadb-10.9.0-linux-x86_64-dbg/bin/mysqld --no-defaults --core-'.
 
Program terminated with signal SIGABRT, Aborted.
 
#0  __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
 
[Current thread is 1 (Thread 0x149e1c1fe700 (LWP 4084799))]
 
(gdb) bt
 
#0  __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
 
#1  0x0000149e38873859 in __GI_abort () at abort.c:79
 
#2  0x0000149e38873729 in __assert_fail_base (fmt=0x149e38a09588 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=0x55872b6de646 "pos < table->n_v_def", file=0x55872b6e1278 "/test/10.9_dbg/storage/innobase/include/dict0dict.inl", line=387, function=<optimized out>) at assert.c:92
 
#3  0x0000149e38885006 in __GI___assert_fail (assertion=assertion@entry=0x55872b6de646 "pos < table->n_v_def", file=file@entry=0x55872b6e1278 "/test/10.9_dbg/storage/innobase/include/dict0dict.inl", line=line@entry=387, function=function@entry=0x55872b6e2d30 "dict_v_col_t* dict_table_get_nth_v_col(const dict_table_t*, ulint)") at assert.c:101
 
#4  0x000055872ae686b6 in dict_table_get_nth_v_col (pos=0, table=0x149ddc0242c8) at /test/10.9_dbg/storage/innobase/include/dict0dict.inl:387
 
#5  build_template_field (prebuilt=0x149ddc02bfc8, clust_index=clust_index@entry=0x149ddc026ae8, index=index@entry=0x149ddc026ae8, table=<optimized out>, field=field@entry=0x149ddc03a090, i=<optimized out>, i@entry=0, v_no=<optimized out>) at /test/10.9_dbg/storage/innobase/handler/ha_innodb.cc:7175
 
#6  0x000055872ae6d909 in ha_innobase::build_template (this=this@entry=0x149ddc03a4d0, whole_row=<optimized out>, whole_row@entry=true) at /test/10.9_dbg/storage/innobase/handler/ha_innodb.cc:7587
 
#7  0x000055872ae6de5d in ha_innobase::write_row (this=0x149ddc03a4d0, record=0x149ddc03a048 "\375\001") at /test/10.9_dbg/storage/innobase/handler/ha_innodb.cc:7793
 
#8  0x000055872aac7601 in handler::ha_write_row (this=0x149ddc03a4d0, buf=0x149ddc03a048 "\375\001") at /test/10.9_dbg/sql/handler.cc:7527
 
#9  0x000055872a75d494 in write_record (thd=thd@entry=0x149ddc000db8, table=table@entry=0x149ddc0358a8, info=info@entry=0x149e1c1fccd0, sink=sink@entry=0x0) at /test/10.9_dbg/sql/sql_insert.cc:2156
 
#10 0x000055872a7691ea in mysql_insert (thd=thd@entry=0x149ddc000db8, table_list=0x149ddc013eb8, fields=@0x149ddc006080: {<base_list> = {<Sql_alloc> = {<No data fields>}, first = 0x55872be009a0 <end_of_list>, last = 0x149ddc006080, elements = 0}, <No data fields>}, values_list=@0x149ddc0060c8: {<base_list> = {<Sql_alloc> = {<No data fields>}, first = 0x149ddc014ac0, last = 0x149ddc014ac0, elements = 1}, <No data fields>}, update_fields=@0x149ddc0060b0: {<base_list> = {<Sql_alloc> = {<No data fields>}, first = 0x55872be009a0 <end_of_list>, last = 0x149ddc0060b0, elements = 0}, <No data fields>}, update_values=@0x149ddc006098: {<base_list> = {<Sql_alloc> = {<No data fields>}, first = 0x55872be009a0 <end_of_list>, last = 0x149ddc006098, elements = 0}, <No data fields>}, duplic=DUP_ERROR, ignore=false, result=0x0) at /test/10.9_dbg/sql/sql_insert.cc:1127
 
#11 0x000055872a7afa4c in mysql_execute_command (thd=thd@entry=0x149ddc000db8, is_called_from_prepared_stmt=is_called_from_prepared_stmt@entry=false) at /test/10.9_dbg/sql/sql_parse.cc:4562
 
#12 0x000055872a79a437 in mysql_parse (thd=thd@entry=0x149ddc000db8, rawbuf=<optimized out>, length=<optimized out>, parser_state=parser_state@entry=0x149e1c1fd400) at /test/10.9_dbg/sql/sql_parse.cc:8027
 
#13 0x000055872a7a90d3 in dispatch_command (command=command@entry=COM_QUERY, thd=thd@entry=0x149ddc000db8, packet=packet@entry=0x149ddc00b889 "INSERT INTO t VALUES (1)", packet_length=packet_length@entry=24, blocking=blocking@entry=true) at /test/10.9_dbg/sql/sql_class.h:1362
 
#14 0x000055872a7ac51a in do_command (thd=0x149ddc000db8, blocking=blocking@entry=true) at /test/10.9_dbg/sql/sql_parse.cc:1402
 
#15 0x000055872a927228 in do_handle_one_connection (connect=<optimized out>, connect@entry=0x55872e275a48, put_in_cache=put_in_cache@entry=true) at /test/10.9_dbg/sql/sql_connect.cc:1418
 
#16 0x000055872a92782d in handle_one_connection (arg=arg@entry=0x55872e275a48) at /test/10.9_dbg/sql/sql_connect.cc:1312
 
#17 0x000055872adadf23 in pfs_spawn_thread (arg=0x55872e188fd8) at /test/10.9_dbg/storage/perfschema/pfs.cc:2201
 
#18 0x0000149e38d84609 in start_thread (arg=<optimized out>) at pthread_create.c:477
 
#19 0x0000149e38970163 in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95

Bug confirmed present in:
MariaDB: 10.2.44 (dbg), 10.2.44 (opt), 10.3.35 (dbg), 10.3.35 (opt), 10.4.25 (dbg), 10.4.25 (opt), 10.5.16 (dbg), 10.5.16 (opt), 10.6.8 (dbg), 10.6.8 (opt), 10.7.4 (dbg), 10.7.4 (opt), 10.8.3 (dbg), 10.8.3 (opt), 10.9.0 (dbg), 10.9.0 (opt)

Bug (or feature/syntax) confirmed not present in:
MySQL: 5.5.62 (dbg), 5.5.62 (opt), 5.6.51 (dbg), 5.6.51 (opt), 5.7.37 (dbg), 5.7.37 (opt), 8.0.28 (dbg), 8.0.28 (opt)

 Comments   
Comment by Marko Mäkelä [ 2022-03-19 ]

I think that EXCHANGE PARTITION is implemented entirely outside InnoDB. The operation should be rejected, because a normal column and a virtual column (a constant one at that) are not really interchangeable.

Comment by Kiyoshi Takeda [ 2022-05-27 ]

In my local environment, this error is not reproducible.
On the other hand, different assertion error occurs.

The test code is as below:

--source include/have_partition.inc
 
SET sql_mode='';
 
CREATE TABLE t1 (a INT,KEY(a)) ENGINE=InnoDB PARTITION BY RANGE (a) (PARTITION p VALUES LESS THAN (1));
 
CREATE TABLE t (a INT GENERATED ALWAYS AS (1) VIRTUAL,KEY(a));
 
ALTER TABLE t1 EXCHANGE PARTITION p WITH TABLE t;
 
INSERT INTO t VALUES (1);

in branch 10.2 this test code leads to

mysqld: /path/to/server/sql/sql_partition_admin.cc:248: bool compare_table_with_partition(THD*, TABLE*, TABLE*, partition_element*, uint): Assertion `table->s->db_options_in_use == part_table->s->db_options_in_use' failed.
 
220527 11:36:00 [ERROR] mysqld got signal 6 ;

I appreciate it if someone run this test code and see if the same error occurs.

If so, I need to seek advice on whether to create a new issue or proceed with debugging as this issue.

Comment by Nayuta Yanagisawa (Inactive) [ 2022-05-27 ]

Hi, Takekiyo! Thanks for your interest.

Looks the same to me. I cannot reproduce the exact assertion failure on 10.3 HEAD and 10.9 HEAD. The server crashes by Assertion `table->s->db_options_in_use == part_table->s->db_options_in_use'. I've also checked 10.3.35 but have got the same result.

Roel Could you take another look at the present issue to see whether the assertion error shown in the issue description is still reproducible in your environment?

Comment by Roel Van de Paar [ 2022-05-27 ]

Takekiyo tip: use {noformat}...{noformat} or {code:sql}...{code} tags above/below excerpts to pretify comments. Thank you for your interest +1.

Comment by Roel Van de Paar [ 2022-05-27 ]

The difference in asserts (nice find!) is related to MTR.

I retested the original testcase on 10.3 and 10.9, at the CLI (command line) and the assert is always:

10.9.0 0b14dbd45b5a1c02616d611876158d44b92b77bf (Debug)

 
mysqld: /test/10.9_dbg/storage/innobase/include/dict0dict.inl:387: dict_v_col_t* dict_table_get_nth_v_col(const dict_table_t*, ulint): Assertion `pos < table->n_v_def' failed.

At the CLI, by default, table creations are InnoDB. In MTR, by default they are MyISAM. The second CREATE TABLE does not have an ENGINE specified so that may be related, as well as the sql_mode setting.

Changing the testcase to be (at the CLI):

CREATE TABLE t1 (a INT,KEY(a)) ENGINE=MyISAM PARTITION BY RANGE (a) (PARTITION p VALUES LESS THAN (1));
 
CREATE TABLE t (a INT GENERATED ALWAYS AS (1) VIRTUAL,KEY(a)) ENGINE=MyISAM;
 
ALTER TABLE t1 EXCHANGE PARTITION p WITH TABLE t;
 
INSERT INTO t VALUES (1);

Leads to the assert (at the CLI) observed in MTR in the comments above:

10.3.35 6a2d88c132221ea07dd322060089c85ff5e469b5 (Debug)

 
mysqld: /test/10.3_dbg/sql/sql_partition_admin.cc:254: bool compare_table_with_partition(THD*, TABLE*, TABLE*, partition_element*, uint): Assertion `table->s->db_options_in_use == part_table->s->db_options_in_use' failed.

The issue is likely the same (and as such it is likely that both outcomes need to be addressed in this bug), whereas the resulting codepath/assert hit is not. Adding full analysis of modified testcase in next comment. Original assert can likely be reproduced in MTR by specifying ENGINE=InnoDB for the second create table and/or removing the sql_mode.

Comment by Roel Van de Paar [ 2022-05-27 ]

Note that 10.2 is not being tested anymore (EOL). Report below starts with 10.3+ even if issue is present in 10.2 as well. Any fixes should be in 10.3.

Comment by Roel Van de Paar [ 2022-05-27 ] 

CREATE TABLE t1 (a INT,KEY(a)) ENGINE=MyISAM PARTITION BY RANGE (a) (PARTITION p VALUES LESS THAN (1));
 
CREATE TABLE t (a INT GENERATED ALWAYS AS (1) VIRTUAL,KEY(a)) ENGINE=MyISAM;
 
ALTER TABLE t1 EXCHANGE PARTITION p WITH TABLE t;
 
INSERT INTO t VALUES (1);

Leads to:

10.3.35 6a2d88c132221ea07dd322060089c85ff5e469b5 (Debug)

 
mysqld: /test/10.3_dbg/sql/sql_partition_admin.cc:254: bool compare_table_with_partition(THD*, TABLE*, TABLE*, partition_element*, uint): Assertion `table->s->db_options_in_use == part_table->s->db_options_in_use' failed.

10.3.35 6a2d88c132221ea07dd322060089c85ff5e469b5 (Debug)

 
Core was generated by `/test/MD160322-mariadb-10.3.35-linux-x86_64-dbg/bin/mysqld --no-defaults --core'.
 
Program terminated with signal SIGABRT, Aborted.
 
#0  __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
 
[Current thread is 1 (Thread 0x149144138700 (LWP 1280459))]
 
(gdb) bt
 
#0  __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
 
#1  0x00001491591d2859 in __GI_abort () at abort.c:79
 
#2  0x00001491591d2729 in __assert_fail_base (fmt=0x149159368588 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=0x558bb9c50930 "table->s->db_options_in_use == part_table->s->db_options_in_use", file=0x558bb9c505d0 "/test/10.3_dbg/sql/sql_partition_admin.cc", line=254, function=<optimized out>) at assert.c:92
 
#3  0x00001491591e3fd6 in __GI___assert_fail (assertion=assertion@entry=0x558bb9c50930 "table->s->db_options_in_use == part_table->s->db_options_in_use", file=file@entry=0x558bb9c505d0 "/test/10.3_dbg/sql/sql_partition_admin.cc", line=line@entry=254, function=function@entry=0x558bb9c508d8 "bool compare_table_with_partition(THD*, TABLE*, TABLE*, partition_element*, uint)") at assert.c:101
 
#4  0x0000558bb93b5b6b in compare_table_with_partition (part_id=0, part_elem=0x1490f8024518, part_table=0x1490f8021fa0, table=0x1490f802bc70, thd=0x1490f8000d90) at /test/10.3_dbg/sql/sql_partition_admin.cc:254
 
#5  Sql_cmd_alter_table_exchange_partition::exchange_partition (this=this@entry=0x1490f80123a8, thd=thd@entry=0x1490f8000d90, table_list=<optimized out>, table_list@entry=0x1490f8011680, alter_info=alter_info@entry=0x149144135bd0) at /test/10.3_dbg/sql/sql_partition_admin.cc:593
 
#6  0x0000558bb93b60d4 in Sql_cmd_alter_table_exchange_partition::execute (this=0x1490f80123a8, thd=0x1490f8000d90) at /test/10.3_dbg/sql/sql_partition_admin.cc:99
 
#7  0x0000558bb8f41cec in mysql_execute_command (thd=thd@entry=0x1490f8000d90) at /test/10.3_dbg/sql/sql_parse.cc:6075
 
#8  0x0000558bb8f43d24 in mysql_parse (thd=thd@entry=0x1490f8000d90, rawbuf=<optimized out>, length=<optimized out>, parser_state=parser_state@entry=0x149144137530, is_com_multi=is_com_multi@entry=false, is_next_command=is_next_command@entry=false) at /test/10.3_dbg/sql/sql_parse.cc:7870
 
#9  0x0000558bb8f46495 in dispatch_command (command=command@entry=COM_QUERY, thd=thd@entry=0x1490f8000d90, packet=packet@entry=0x1490f80198f1 "ALTER TABLE t1 EXCHANGE PARTITION p WITH TABLE t", packet_length=packet_length@entry=48, is_com_multi=is_com_multi@entry=false, is_next_command=is_next_command@entry=false) at /test/10.3_dbg/sql/sql_class.h:1152
 
#10 0x0000558bb8f49625 in do_command (thd=0x1490f8000d90) at /test/10.3_dbg/sql/sql_parse.cc:1398
 
#11 0x0000558bb907185f in do_handle_one_connection (connect=connect@entry=0x558bbc2247e0) at /test/10.3_dbg/sql/sql_connect.cc:1403
 
#12 0x0000558bb9071a96 in handle_one_connection (arg=0x558bbc2247e0) at /test/10.3_dbg/sql/sql_connect.cc:1308
 
#13 0x00001491593aa609 in start_thread (arg=<optimized out>) at pthread_create.c:477
 
#14 0x00001491592cf133 in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95

Bug confirmed present in:
MariaDB: 10.3.35 (dbg), 10.4.25 (dbg), 10.5.16 (dbg), 10.6.8 (dbg), 10.7.4 (dbg), 10.8.3 (dbg), 10.9.0 (dbg), 10.10.0 (dbg)

Bug (or feature/syntax) confirmed not present in:
MariaDB: 10.3.35 (opt), 10.4.25 (opt), 10.5.16 (opt), 10.6.8 (opt), 10.7.4 (opt), 10.8.3 (opt), 10.9.0 (opt), 10.10.0 (opt)
MySQL: 5.5.62 (dbg), 5.5.62 (opt), 5.6.51 (dbg), 5.6.51 (opt), 5.7.37 (dbg), 5.7.37 (opt), 8.0.28 (dbg), 8.0.28 (opt)

Comment by Roel Van de Paar [ 2022-05-28 ]

Observe that the secondary testcase is debug only whereas the original (which should also be reproducible in MTR with a few modifications) is present in optimized+debug builds.

Comment by Roel Van de Paar [ 2022-05-28 ]

UniqueID's seen so far for this bug (one per line)

SIGSEGV|row_mysql_store_col_in_innobase_format|row_mysql_convert_row_to_innobase|row_insert_for_mysql|ha_innobase::write_row
 
pos < table->n_v_def|SIGABRT|dict_table_get_nth_v_col|build_template_field|ha_innobase::build_template|ha_innobase::write_row
 
table->s->db_options_in_use == part_table->s->db_options_in_use|SIGABRT|compare_table_with_partition|Sql_cmd_alter_table_exchange_partition::exchange_partition|Sql_cmd_alter_table_exchange_partition::execute|mysql_execute_command

Comment by Nikita Malyavin [ 2022-06-08 ]

https://github.com/MariaDB/server/pull/2149

Comment by Nikita Malyavin [ 2022-11-23 ]

The last PR was closed as the proposed fix was incorrect and no further progress was made on it, so this task is free to take now.
Please also consider the following test case to fix in scope of this task:

create table t1 (b int, a int, key(a), c smallint as (1)) 
                      partition by range (a) (partition p values less than (1));
 
create table t2 (b int, a int as (1),key(a), c smallint);
 
ALTER TABLE t1 EXCHANGE PARTITION p WITH TABLE t2;
 
insert into t1(b, a) values (0, -1);
 
insert into t2(b, c) values (2, 4);

Comment by Aleksey Midenkov [ 2023-09-13 ]

Please review bb-10.4-midenok

Comment by Nikita Malyavin [ 2023-09-13 ]

https://github.com/MariaDB/server/commit/1ae447360f10c7f94830c7bd9afe9bb6464d90ba#r127273469

Comment by Sergei Golubchik [ 2023-10-17 ]

8a16079ee78a is ok to push

Generated at Thu Feb 08 09:58:16 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.