Details
-
Bug
-
Status: Open (View Workflow)
-
Minor
-
Resolution: Unresolved
-
10.11
-
None
-
None
-
None
Description
Note: armhf is not an officially supported platform by MariaDB. This bug report is filed mainly for tracking purposes and is unlikely to be fixed unless some new Raspberry PI / armhf enthusiast steps up and contributes fixes, or at least contributes debugging to make the issues obvious and easy to fix.
When running the full MariaDB 10.11 test suite (--big-test) on armhf on the following tests have failures (when run on Launchpad builders):
federated.federatedx_versioning
|
federated.federatedx_versioning
|
spider/bg.basic_sql
|
spider/bg.basic_sql
|
spider/bg.spider_fixes spider/bg.spider_fixes_part spider/bg.spider3_fixes spider/bg.spider3_fixes_part
|
spider/bg.spider_fixes_part
|
spider.auto_increment spider.basic_sql
|
spider.basic_sql_part spider.basic_sql
|
spider.spider3_fixes_part spider.spider3_fixes spider.spider_fixes_part spider.spider_fixes
|
spider.spider_fixes_part
|
versioning.trx_id
|
versioning.trx_id
|
versioning.online
|
versioning.online
|
versioning.partition
|
versioning.partition
|
period.versioning
|
This happens on Ubuntu 22.04 (Jammy), 23.10 and 24.04:
- https://launchpadlibrarian.net/712889019/buildlog_ubuntu-jammy-armhf.mariadb_1%3A10.11.6-3~bpo22.04.1~1707078933.a3220971295+feature.extend.mtr.to.big_BUILDING.txt.gz
- https://launchpadlibrarian.net/712875120/buildlog_ubuntu-mantic-armhf.mariadb_1%3A10.11.6-3~bpo23.10.1~1707078933.a3220971295+feature.extend.mtr.to.big_BUILDING.txt.gz
- https://launchpadlibrarian.net/712867854/buildlog_ubuntu-noble-armhf.mariadb_1%3A10.11.6-3~bpo24.04.1~1707078933.a3220971295+feature.extend.mtr.to.big_BUILDING.txt.gz
Summaries of failures from log:
federated.federatedx_versioning 'X,innodb,trx_id' w1 [ fail ]
|
Test ended at 2024-02-05 01:01:30
|
|
CURRENT_TEST: federated.federatedx_versioning
|
mysqltest: At line 18: query 'insert into t1 values (1)' failed: <Unknown> (2013): Lost connection to server during query
|
..
|
Query (0xf0a0e010): insert into t1 values (1)
|
spider/bg.basic_sql w3 [ fail ]
|
Test ended at 2024-02-05 01:16:35
|
|
CURRENT_TEST: spider/bg.basic_sql
|
mysqltest: At line 1875: query 'UPDATE ta_l a, tb_l b SET a.b = b.b, a.c = b.c WHERE a.a = b.a' failed: <Unknown> (2026): TLS/SSL error: unexpected eof while reading
|
|
The result from queries just before the failure was:
|
< snip >
|
tb_l);
|
connection master_1;
|
SELECT a, b, date_format(c, '%Y-%m-%d %H:%i:%s') FROM ta_l ORDER BY a;
|
a b date_format(c, '%Y-%m-%d %H:%i:%s')
|
1 g 2009-03-03 03:03:03
|
2 g 2009-03-03 03:03:03
|
|
update select a
|
connection master_1;
|
UPDATE ta_l SET b = 'h', c = '2010-04-04 04:04:04' WHERE a = (SELECT a FROM
|
tb_l ORDER BY a LIMIT 1);
|
connection master_1;
|
SELECT a, b, date_format(c, '%Y-%m-%d %H:%i:%s') FROM ta_l ORDER BY a;
|
a b date_format(c, '%Y-%m-%d %H:%i:%s')
|
1 h 2010-04-04 04:04:04
|
2 g 2009-03-03 03:03:03
|
|
update join
|
connection master_1;
|
UPDATE ta_l a, tb_l b SET a.b = b.b, a.c = b.c WHERE a.a = b.a;
|
spider/bg.spider_fixes_part w2 [ fail ]
|
Test ended at 2024-02-05 01:16:39
|
|
CURRENT_TEST: spider/bg.spider_fixes_part
|
mysqltest: At line 272: query 'UPDATE ta_l2 SET a = 4 WHERE a = 3' failed: <Unknown> (2026): TLS/SSL error: unexpected eof while reading
|
|
The result from queries just before the failure was:
|
< snip >
|
ORDER BY a;
|
a b date_format(c, '%Y-%m-%d %H:%i:%s')
|
2 b 2000-01-01 00:00:00
|
3 e 2007-06-04 20:03:11
|
4 d 2003-11-30 05:01:03
|
5 c 2001-12-31 23:59:59
|
|
2.23
|
partition update with moving partition
|
connection master_1;
|
DROP TABLE IF EXISTS ta_l2;
|
connection master_1;
|
CREATE TABLE ta_l2 (
|
a INT,
|
b CHAR(1),
|
c DATETIME,
|
PRIMARY KEY(a)
|
) MASTER_1_ENGINE MASTER_1_COMMENT2_P_2_1
|
INSERT INTO ta_l2 (a, b, c) VALUES (3, 'B', '2010-09-26 00:00:00');
|
UPDATE ta_l2 SET a = 4 WHERE a = 3;
|
spider.basic_sql w2 [ fail ]
|
Test ended at 2024-02-05 01:17:23
|
|
CURRENT_TEST: spider.basic_sql
|
mysqltest: At line 1875: query 'UPDATE ta_l a, tb_l b SET a.b = b.b, a.c = b.c WHERE a.a = b.a' failed: <Unknown> (2026): TLS/SSL error: unexpected eof while reading
|
|
The result from queries just before the failure was:
|
< snip >
|
tb_l);
|
connection master_1;
|
SELECT a, b, date_format(c, '%Y-%m-%d %H:%i:%s') FROM ta_l ORDER BY a;
|
a b date_format(c, '%Y-%m-%d %H:%i:%s')
|
1 g 2009-03-03 03:03:03
|
2 g 2009-03-03 03:03:03
|
|
update select a
|
connection master_1;
|
UPDATE ta_l SET b = 'h', c = '2010-04-04 04:04:04' WHERE a = (SELECT a FROM
|
tb_l ORDER BY a LIMIT 1);
|
connection master_1;
|
SELECT a, b, date_format(c, '%Y-%m-%d %H:%i:%s') FROM ta_l ORDER BY a;
|
a b date_format(c, '%Y-%m-%d %H:%i:%s')
|
1 h 2010-04-04 04:04:04
|
2 g 2009-03-03 03:03:03
|
|
update join
|
connection master_1;
|
UPDATE ta_l a, tb_l b SET a.b = b.b, a.c = b.c WHERE a.a = b.a;
|
spider.spider_fixes_part w1 [ fail ]
|
Test ended at 2024-02-05 01:17:28
|
|
CURRENT_TEST: spider.spider_fixes_part
|
mysqltest: At line 272: query 'UPDATE ta_l2 SET a = 4 WHERE a = 3' failed: <Unknown> (2026): TLS/SSL error: unexpected eof while reading
|
|
The result from queries just before the failure was:
|
< snip >
|
ORDER BY a;
|
a b date_format(c, '%Y-%m-%d %H:%i:%s')
|
2 b 2000-01-01 00:00:00
|
3 e 2007-06-04 20:03:11
|
4 d 2003-11-30 05:01:03
|
5 c 2001-12-31 23:59:59
|
|
2.23
|
partition update with moving partition
|
connection master_1;
|
DROP TABLE IF EXISTS ta_l2;
|
connection master_1;
|
CREATE TABLE ta_l2 (
|
a INT,
|
b CHAR(1),
|
c DATETIME,
|
PRIMARY KEY(a)
|
) MASTER_1_ENGINE MASTER_1_COMMENT2_P_2_1
|
INSERT INTO ta_l2 (a, b, c) VALUES (3, 'B', '2010-09-26 00:00:00');
|
UPDATE ta_l2 SET a = 4 WHERE a = 3;
|
versioning.trx_id 'innodb' w2 [ fail ]
|
Test ended at 2024-02-05 01:19:05
|
|
CURRENT_TEST: versioning.trx_id
|
mysqltest: At line 29: query 'commit' failed: <Unknown> (2013): Lost connection to server during query
|
|
The result from queries just before the failure was:
|
set default_storage_engine= innodb;
|
create or replace table t1 (
|
x int,
|
sys_trx_start bigint(20) unsigned as row start invisible,
|
sys_trx_end bigint(20) unsigned as row end invisible,
|
period for system_time (sys_trx_start, sys_trx_end)
|
) with system versioning;
|
# No history inside the transaction
|
start transaction;
|
insert into t1 (x) values (1);
|
update t1 set x= x + 1;
|
update t1 set x= x + 1;
|
commit;
|
versioning.online 'innodb,release,trx_id' w4 [ fail ]
|
Test ended at 2024-02-05 01:19:55
|
|
CURRENT_TEST: versioning.online
|
mysqltest: At line 26: query 'insert into t values (1, 0)' failed: <Unknown> (2013): Lost connection to server during query
|
|
The result from queries just before the failure was:
|
< snip >
|
alter table t add system versioning, lock=none;
|
ERROR 0A000: LOCK=NONE is not supported. Reason: Not implemented for system-versioned timestamp tables. Try LOCK=SHARED
|
alter table t add system versioning, algorithm=inplace;
|
ERROR 0A000: ALGORITHM=INPLACE is not supported. Reason: Not implemented for system-versioned timestamp tables. Try ALGORITHM=COPY
|
alter table t add system versioning, lock=shared;
|
alter table t add column b int, change column a a int without system versioning, lock=none;
|
alter table t drop system versioning, lock=none;
|
ERROR 0A000: LOCK=NONE is not supported. Reason: Not implemented for system-versioned operations. Try LOCK=SHARED
|
alter table t drop system versioning, algorithm=inplace;
|
create or replace table t (
|
a int, b int,
|
row_start SYS_DATATYPE as row start invisible,
|
row_end SYS_DATATYPE as row end invisible,
|
period for system_time (row_start, row_end)
|
) with system versioning;
|
insert into t values (1, 0);
|
versioning.partition 'innodb,trx_id' w4 [ fail ]
|
Test ended at 2024-02-05 01:20:00
|
|
CURRENT_TEST: versioning.partition
|
mysqltest: At line 28: query 'insert into t1 values (3), (300)' failed: <Unknown> (2013): Lost connection to server during query
|
|
The result from queries just before the failure was:
|
set @save_persistent=@@global.innodb_stats_persistent;
|
set global innodb_stats_persistent= 0;
|
call mtr.add_suppression("need more HISTORY partitions");
|
set system_versioning_alter_history=keep;
|
# Check conventional partitioning on temporal tables
|
create or replace table t1 (
|
x int,
|
row_start SYS_DATATYPE as row start invisible,
|
row_end SYS_DATATYPE as row end invisible,
|
period for system_time(row_start, row_end))
|
with system versioning
|
partition by range columns (x) (
|
partition p0 values less than (100),
|
partition p1 values less than (1000));
|
insert into t1 values (3), (300);
|
period.versioning 'innodb,trx_id' w1 [ fail ]
|
Test ended at 2024-02-05 01:20:12
|
|
CURRENT_TEST: period.versioning
|
mysqltest: At line 12: query 'insert into t values('1999-01-01', '2018-12-12'),
|
('1999-01-01', '1999-12-12')' failed: <Unknown> (2013): Lost connection to server during query
|
|
The result from queries just before the failure was:
|
# DELETE
|
create table t (
|
s date, e date,
|
row_start SYS_TYPE as row start invisible,
|
row_end SYS_TYPE as row end invisible,
|
period for apptime(s, e),
|
period for system_time (row_start, row_end)) with system versioning;
|
insert into t values('1999-01-01', '2018-12-12'),
|
('1999-01-01', '1999-12-12');
|
versioning.autoinc 'innodb,trx_id' w1 [ fail ]
|
Test ended at 2024-02-05 01:20:12
|
|
CURRENT_TEST: versioning.autoinc
|
mysqltest: At line 19: query 'insert into t1(x, y) values(1, 11)' failed: <Unknown> (2013): Lost connection to server during query
|
|
The result from queries just before the failure was:
|
create table t1(
|
id int unsigned auto_increment primary key,
|
x int unsigned,
|
y int unsigned,
|
sys_start SYS_DATATYPE as row start invisible,
|
sys_end SYS_DATATYPE as row end invisible,
|
period for system_time (sys_start, sys_end))
|
with system versioning;
|
create table t2(
|
id int unsigned auto_increment primary key,
|
x int unsigned,
|
y int unsigned);
|
insert into t1(x, y) values(1, 11);
|
Many of these seem to be related to time. Perhaps this armhf as 32-bit platform has some year 2038 adaptations which MariaDB isn't compatible - but this is just pure speculation. I will however re-test this once MDEV-32188 is done to see if it made a difference.
Attachments
Issue Links
- relates to
-
MDBF-155 Test running (almost) all tests on all Debian/Ubuntu archs
- Closed
-
MDEV-32188 make TIMESTAMP use whole 32-bit unsigned range
- Closed