[MDEV-9329] 10.0 fails to build on official Ubuntu/s390x builders due to Spider test failures Created: 2015-12-28  Updated: 2021-08-14  Resolved: 2021-08-14

Status: Closed
Project: MariaDB Server
Component/s: Platform Debian, Storage Engine - Spider
Affects Version/s: 10.0.22, 10.0.23
Fix Version/s: N/A

Type: Bug Priority: Minor
Reporter: Otto Kekäläinen Assignee: VicenČ›iu Ciorbaru
Resolution: Fixed Votes: 0
Labels: None

Issue Links:
Relates
relates to MDEV-9730 Re-enable Spider tests in Debian git ... Stalled

 Description   

In Debian 10.0.23-1 builds OK on all official platforms, including s390x:
https://buildd.debian.org/status/fetch.php?pkg=mariadb-10.0&arch=s390x&ver=10.0.23-1&stamp=1450701833

When the same package is synced to Ubuntu and built there, it fails on s390x (only, other archs build OK):
https://launchpadlibrarian.net/231478614/buildlog_ubuntu-xenial-s390x.mariadb-10.0_10.0.23-1_BUILDING.txt.gz

spider/bg.spider3_fixes_part             w2 [ pass ]     96
 
spider/bg.basic_sql                      w1 [ pass ]    194
 
spider/bg.spider_fixes                   w2 [ pass ]    935
 
spider/bg.basic_sql_part                 w1 [ pass ]    102
 
spider/bg.spider_fixes_part              w2 [ pass ]    166
 
spider/bg.direct_aggregate               w1 [ pass ]     64
 
main.pool_of_threads                     w3 [ pass ]  12228
 
spider/bg.direct_aggregate_part          w1 [ pass ]     71
 
worker[3] mysql-test-run: WARNING: Process [mysqld.3.3 - pid: 19588, winpid: 19588, exit: 256] died after mysql-test-run waited 0.2 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/3/run/mysqld.3.3.pid to be created.
 
spider.basic_sql                         w3 [ fail ]
 
        Test ended at 2015-12-24 22:28:53
 
 
 
CURRENT_TEST: spider.basic_sql
 
 
 
 
 
Failed to start mysqld.3.3
 
mysqltest failed but provided no output
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/3/log/spider.basic_sql/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/spider.basic_sql/'
 
 
 
Retrying test spider.basic_sql, attempt(2/3)...
 
 
 
***Warnings generated in error logs during shutdown after running tests: spider.basic_sql
 
 
 
151224 22:28:53 [ERROR] Can't create interrupt-thread (error 11, errno: 11)
 
 
 
spider/bg.direct_update                  w1 [ pass ]    101
 
spider/bg.direct_update_part             w1 [ pass ]    101
 
spider/bg.function                       w1 [ pass ]    115
 
spider.basic_sql                         w3 [ retry-pass ]    230
 
 
 
Retrying test spider.basic_sql, attempt(3/3)...
 
 
 
worker[2] mysql-test-run: WARNING: Process [mysqld.1.1 - pid: 20359, winpid: 20359, exit: 256] died after mysql-test-run waited 0.3 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/2/run/mysqld.1.1.pid to be created.
 
spider.spider3_fixes                     w2 [ fail ]
 
        Test ended at 2015-12-24 22:28:55
 
 
 
CURRENT_TEST: spider.spider3_fixes
 
 
 
 
 
Failed to start mysqld.1.1
 
mysqltest failed but provided no output
 
The result from queries just before the failure was:
 
< snip >
 
5000
 
10000
 
 
 
deinit
 
DROP DATABASE IF EXISTS auto_test_local;
 
DROP DATABASE IF EXISTS auto_test_local;
 
DROP DATABASE IF EXISTS auto_test_remote;
 
DROP DATABASE IF EXISTS auto_test_remote2;
 
for slave1_1
 
for master_1
 
for child2
 
child2_1
 
child2_2
 
child2_3
 
for child3
 
child3_1
 
child3_2
 
child3_3
 
 
 
end of test
 
 
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/2/log/spider.spider3_fixes/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/spider.spider3_fixes/'
 
 
 
Retrying test spider.spider3_fixes, attempt(2/3)...
 
 
 
***Warnings generated in error logs during shutdown after running tests: spider.spider3_fixes
 
 
 
InnoDB: Error: pthread_create returned 11
 
InnoDB: Error: pthread_create returned 11
 
InnoDB: Error: pthread_create returned 11
 
151224 22:28:55 [ERROR] Failed to create thread while initialising slave
 
151224 22:28:55 [ERROR] Aborting
 
 
 
spider.basic_sql_part                    w3 [ fail ]
 
        Test ended at 2015-12-24 22:28:55
 
 
 
CURRENT_TEST: spider.basic_sql_part
 
mysqltest: At line 549: query 'SELECT a, b, date_format(c, '%Y-%m-%d %H:%i:%s') FROM ta_l2 ORDER BY a' failed: 1429: Unable to connect to foreign data source: s_2_2
 
 
 
The result from queries just before the failure was:
 
< snip >
 
3	j	2009-03-03 03:03:03
 
4	i	2003-10-30 05:01:03
 
5	h	2001-10-31 23:59:59
 
 
 
delete partition pushdown
 
TRUNCATE TABLE ta_l2;
 
INSERT INTO ta_l2 SELECT a, b, c FROM tb_l;
 
DELETE FROM ta_l2 WHERE b = 'g';
 
SELECT a, b, date_format(c, '%Y-%m-%d %H:%i:%s') FROM ta_l2 ORDER BY a;
 
a	b	date_format(c, '%Y-%m-%d %H:%i:%s')
 
1	f	2008-07-01 10:21:39
 
3	j	2007-05-04 20:03:11
 
4	i	2003-10-30 05:01:03
 
5	h	2001-10-31 23:59:59
 
 
 
delete partition index pushdown
 
TRUNCATE TABLE ta_l2;
 
INSERT INTO ta_l2 SELECT a, b, c FROM tb_l;
 
DELETE FROM ta_l2 WHERE a > 0 AND b = 'g';
 
SELECT a, b, date_format(c, '%Y-%m-%d %H:%i:%s') FROM ta_l2 ORDER BY a;
 
 
 
More results from queries before failure can be found in /<<PKGBUILDDIR>>/builddir/mysql-test/var/3/log/basic_sql_part.log
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/3/log/spider.basic_sql_part/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/spider.basic_sql_part/'
 
 
 
Retrying test spider.basic_sql_part, attempt(2/3)...
 
 
 
spider/bg.ha                             w1 [ pass ]    587
 
worker[3] mysql-test-run: WARNING: Process [mysqld.3.1 - pid: 20810, winpid: 20810, exit: 256] died after mysql-test-run waited 0.8 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/3/run/mysqld.3.1.pid to be created.
 
spider/bg.ha_part                        w1 [ fail ]
 
        Test ended at 2015-12-24 22:28:57
 
 
 
CURRENT_TEST: spider/bg.ha_part
 
mysqltest: At line 228: query 'INSERT INTO ta_l2 (a, b, c) VALUES
 
(1, 'a', '2008-08-01 10:21:39'),
 
(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')' failed: 1429: Unable to connect to foreign data source: s_3_3
 
 
 
The result from queries just before the failure was:
 
< snip >
 
 
 
test select 1
 
SELECT 1;
 
1
 
1
 
 
 
create table with partition test
 
DROP TABLE IF EXISTS ta_l2;
 
CREATE TABLE ta_l2 (
 
a INT,
 
b CHAR(1),
 
c DATETIME,
 
PRIMARY KEY(a)
 
) MASTER_1_ENGINE MASTER_1_CHARSET MASTER_1_COMMENT_HA_P_2_1
 
INSERT INTO ta_l2 (a, b, c) VALUES
 
(1, 'a', '2008-08-01 10:21:39'),
 
(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');
 
 
 
More results from queries before failure can be found in /<<PKGBUILDDIR>>/builddir/mysql-test/var/1/log/ha_part.log
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/1/log/spider/bg.ha_part/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/bg.ha_part/'
 
 
 
Retrying test spider/bg.ha_part, attempt(2/3)...
 
 
 
spider.basic_sql                         w3 [ retry-fail ]
 
        Test ended at 2015-12-24 22:28:57
 
 
 
CURRENT_TEST: spider.basic_sql
 
 
 
 
 
Failed to start mysqld.3.1
 
mysqltest failed but provided no output
 
The result from queries just before the failure was:
 
< snip >
 
truncate
 
TRUNCATE TABLE ta_l;
 
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')
 
 
 
deinit
 
DROP DATABASE IF EXISTS auto_test_local;
 
DROP DATABASE IF EXISTS auto_test_remote;
 
DROP DATABASE IF EXISTS auto_test_remote2;
 
for master_1
 
for child2
 
child2_1
 
child2_2
 
child2_3
 
for child3
 
child3_1
 
child3_2
 
child3_3
 
 
 
end of test
 
 
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/3/log/spider.basic_sql/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/spider.basic_sql/'
 
worker[3] mysql-test-run: WARNING: Process [mysqld.3.2 - pid: 21151, winpid: 21151, exit: 256] died after mysql-test-run waited 0.1 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/3/run/mysqld.3.2.pid to be created.
 
***Warnings generated in error logs during shutdown after running tests: spider.basic_sql
 
 
 
InnoDB: Error: pthread_create returned 11
 
InnoDB: Error: pthread_create returned 11
 
 
 
spider.basic_sql_part                    w3 [ retry-fail ]
 
        Test ended at 2015-12-24 22:29:01
 
 
 
CURRENT_TEST: spider.basic_sql
 
CURRENT_TEST: spider.basic_sql_part
 
 
 
 
 
Failed to start mysqld.3.2
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/3/log/spider.basic_sql_part/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/spider.basic_sql_part/'
 
 
 
Test spider.basic_sql_part has failed 2 times, no more retries!
 
 
 
***Warnings generated in error logs during shutdown after running tests: spider.basic_sql_part
 
 
 
151224 22:28:59 [ERROR] Can't create interrupt-thread (error 11, errno: 11)
 
InnoDB: Error: pthread_create returned 11
 
InnoDB: Error: pthread_create returned 11
 
 
 
worker[3] mysql-test-run: WARNING: Process [mysqld.3.2 - pid: 21486, winpid: 21486, exit: 256] died after mysql-test-run waited 0.2 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/3/run/mysqld.3.2.pid to be created.
 
rpl.rpl_heartbeat_basic 'mix'            w4 [ pass ]  55797
 
worker[2] mysql-test-run: WARNING: Process [mysqld.3.3 - pid: 20809, winpid: 20809, exit: 256] died after mysql-test-run waited 5.8 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/2/run/mysqld.3.3.pid to be created.
 
spider.spider3_fixes                     w2 [ retry-fail ]
 
        Test ended at 2015-12-24 22:29:02
 
 
 
CURRENT_TEST: spider.spider3_fixes
 
CURRENT_TEST: spider.spider3_fixes
 
 
 
 
 
Failed to start mysqld.3.3
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/2/log/spider.spider3_fixes/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/spider.spider3_fixes/'
 
 
 
Test spider.spider3_fixes has failed 2 times, no more retries!
 
 
 
***Warnings generated in error logs during shutdown after running tests: spider.spider3_fixes
 
 
 
151224 22:28:57 [ERROR] Aborting
 
 
 
spider.direct_aggregate                  w3 [ fail ]
 
        Test ended at 2015-12-24 22:29:02
 
 
 
CURRENT_TEST: spider.basic_sql
 
CURRENT_TEST: spider.basic_sql_part
 
CURRENT_TEST: spider.direct_aggregate
 
 
 
 
 
Failed to start mysqld.3.2
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/3/log/spider.direct_aggregate/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/spider.direct_aggregate/'
 
 
 
Retrying test spider.direct_aggregate, attempt(2/3)...
 
 
 
***Warnings generated in error logs during shutdown after running tests: spider.direct_aggregate
 
 
 
151224 22:29:01 [ERROR] Can't create interrupt-thread (error 11, errno: 11)
 
 
 
fork failed sleep 1 second and redo: Resource temporarily unavailable at lib/My/SafeProcess/Base.pm line 53.
 
fork failed sleep 1 second and redo: Resource temporarily unavailable at lib/My/SafeProcess/Base.pm line 53.
 
worker[3] mysql-test-run: WARNING: Process [mysqld.3.3 - pid: 21850, winpid: 21850, exit: 256] died after mysql-test-run waited 0.2 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/3/run/mysqld.3.3.pid to be created.
 
spider.direct_aggregate_part             w3 [ fail ]
 
        Test ended at 2015-12-24 22:29:03
 
 
 
CURRENT_TEST: spider.basic_sql
 
CURRENT_TEST: spider.basic_sql_part
 
CURRENT_TEST: spider.direct_aggregate
 
CURRENT_TEST: spider.direct_aggregate_part
 
 
 
 
 
Failed to start mysqld.3.3
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/3/log/spider.direct_aggregate_part/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/spider.direct_aggregate_part/'
 
 
 
Retrying test spider.direct_aggregate_part, attempt(2/3)...
 
 
 
maria.small_blocksize                    w4 [ pass ]      8
 
***Warnings generated in error logs during shutdown after running tests: spider.direct_aggregate_part
 
 
 
InnoDB: Error: pthread_create returned 11
 
 
 
spider/bg.ha_part                        w1 [ retry-pass ]   2313
 
 
 
Retrying test spider/bg.ha_part, attempt(3/3)...
 
 
 
worker[3] mysql-test-run: WARNING: Process [mysqld.3.2 - pid: 22091, winpid: 22091, exit: 256] died after mysql-test-run waited 0.1 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/3/run/mysqld.3.2.pid to be created.
 
worker[4] mysql-test-run: WARNING: Process [mysqld.1 - pid: 22049, winpid: 22049, exit: 256] died after mysql-test-run waited 2 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/4/run/mysqld.1.pid to be created.
 
sys_vars.autocommit_func3 'innodb_plugin' w4 [ fail ]
 
        Test ended at 2015-12-24 22:29:05
 
 
 
CURRENT_TEST: sys_vars.autocommit_func3
 
 
 
 
 
Failed to start mysqld.1
 
mysqltest failed but provided no output
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/4/log/sys_vars.autocommit_func3-innodb_plugin/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/sys_vars.autocommit_func3-innodb_plugin/'
 
 
 
Retrying test sys_vars.autocommit_func3, attempt(2/3)...
 
 
 
***Warnings generated in error logs during shutdown after running tests: sys_vars.autocommit_func3
 
 
 
InnoDB: Error: pthread_create returned 11
 
 
 
spider.direct_update                     w3 [ fail ]
 
        Test ended at 2015-12-24 22:29:06
 
 
 
CURRENT_TEST: spider.basic_sql
 
CURRENT_TEST: spider.basic_sql_part
 
CURRENT_TEST: spider.direct_aggregate
 
CURRENT_TEST: spider.direct_aggregate_part
 
CURRENT_TEST: spider.direct_update
 
 
 
 
 
Failed to start mysqld.3.2
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/3/log/spider.direct_update/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/spider.direct_update/'
 
 
 
Retrying test spider.direct_update, attempt(2/3)...
 
 
 
***Warnings generated in error logs during shutdown after running tests: spider.direct_update
 
 
 
InnoDB: Error: pthread_create returned 11
 
InnoDB: Error: pthread_create returned 11
 
 
 
spider.spider3_fixes_part                w2 [ pass ]   1344
 
fork failed sleep 1 second and redo: Resource temporarily unavailable at lib/My/SafeProcess/Base.pm line 53.
 
fork failed sleep 1 second and redo: Resource temporarily unavailable at lib/My/SafeProcess/Base.pm line 53.
 
sys_vars.autocommit_func3 'innodb_plugin' w4 [ retry-pass ]    547
 
 
 
Retrying test sys_vars.autocommit_func3, attempt(3/3)...
 
 
 
worker[3] mysql-test-run: WARNING: Process [mysqld.3.2 - pid: 22506, winpid: 22506, exit: 256] died after mysql-test-run waited 0.1 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/3/run/mysqld.3.2.pid to be created.
 
spider.direct_update_part                w3 [ fail ]
 
        Test ended at 2015-12-24 22:29:07
 
 
 
CURRENT_TEST: spider.basic_sql
 
CURRENT_TEST: spider.basic_sql_part
 
CURRENT_TEST: spider.direct_aggregate
 
CURRENT_TEST: spider.direct_aggregate_part
 
CURRENT_TEST: spider.direct_update
 
CURRENT_TEST: spider.direct_update_part
 
 
 
 
 
Failed to start mysqld.3.2
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/3/log/spider.direct_update_part/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/spider.direct_update_part/'
 
 
 
Retrying test spider.direct_update_part, attempt(2/3)...
 
 
 
***Warnings generated in error logs during shutdown after running tests: spider.direct_update_part
 
 
 
151224 22:29:06 [ERROR] Can't create interrupt-thread (error 11, errno: 11)
 
 
 
spider/bg.ha_part                        w1 [ retry-fail ]
 
        Test ended at 2015-12-24 22:29:07
 
 
 
CURRENT_TEST: spider/bg.ha_part
 
 
 
 
 
Could not execute 'check-warnings' for testcase 'spider/bg.ha_part' (res: 1) server: 'mysqld.1.1:
 
mysqltest: Could not open connection 'default': 1135 Can't create a new thread (errno 11 "Resource temporarily unavailable"); if you are not out of available memory, you can consult the manual for a possible OS-dependent bug
 
not ok
 
mysqltest failed but provided no output
 
The result from queries just before the failure was:
 
< snip >
 
for master_1
 
for child2
 
child2_1
 
child2_2
 
child2_3
 
for child3
 
child3_1
 
child3_2
 
child3_3
 
for master_1
 
for child2
 
child2_1
 
child2_2
 
child2_3
 
for child3
 
child3_1
 
child3_2
 
child3_3
 
 
 
end of test
 
 
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/1/log/spider/bg.ha_part/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/bg.ha_part/'
 
worker[4] mysql-test-run: WARNING: Process [mysqld.1 - pid: 22954, winpid: 22954, exit: 256] died after mysql-test-run waited 1.1 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/4/run/mysqld.1.pid to be created.
 
worker[1] mysql-test-run: WARNING: Process [mysqld.2.2 - pid: 22871, winpid: 22871, exit: 256] died after mysql-test-run waited 0.2 seconds for /<<PKGBUILDDIR>>/builddir/mysql-test/var/1/run/mysqld.2.2.pid to be created.
 
sys_vars.autocommit_func2 'innodb_plugin' w4 [ fail ]
 
        Test ended at 2015-12-24 22:29:10
 
 
 
CURRENT_TEST: sys_vars.autocommit_func2
 
 
 
 
 
Failed to start mysqld.1
 
mysqltest failed but provided no output
 
 
 
 
 
 - saving '/<<PKGBUILDDIR>>/builddir/mysql-test/var/4/log/sys_vars.autocommit_func2-innodb_plugin/' to '/<<PKGBUILDDIR>>/builddir/mysql-test/var/log/sys_vars.autocommit_func2-innodb_plugin/'
 
Too many tests(10) failed! Terminating...
 
 
 
Only  1419  of 4314 completed.
 
--------------------------------------------------------------------------
 
The servers were restarted 511 times
 
Spent 2142.256 of 858 seconds executing testcases
 
 
 
Too many failed: Failed 5/921 tests, 99.46% were successful.
 
 
 
Failing test(s): spider.basic_sql spider.basic_sql_part spider.spider3_fixes spider/bg.ha_part sys_vars.autocommit_func2

Ubuntu builds overview visible at https://launchpad.net/ubuntu/+source/mariadb-10.0/10.0.23-1

 Comments   
Comment by Otto Kekäläinen [ 2015-12-28 ]

This seems to be sporadic: exactly the same source version failed in 10.0.22-3 (https://launchpadlibrarian.net/229927811/buildlog_ubuntu-xenial-s390x.mariadb-10.0_10.0.22-3_BUILDING.txt.gz) but succeeded in 10.0.22-4 (https://launchpadlibrarian.net/230250486/buildlog_ubuntu-xenial-s390x.mariadb-10.0_10.0.22-4_BUILDING.txt.gz).

Looking at the failures, even when the Spider failures occur, the failing tests sometimes includes and sometimes not: spider/bg.ha, spider.spider3_fixes, sys_vars.autocommit_func2.

As the tests are for now unstable and unsuitable for production builds, I've added them to the Debian build unstable list in https://github.com/ottok/mariadb-10.0/commit/7c5bca8810b2d0357c0939cdfd7d5bd481312938

Comment by Otto Kekäläinen [ 2015-12-28 ]

Looking at older builds on it seems the s390x builds have been enabled for Ubuntu only very recently. For example https://launchpad.net/ubuntu/+source/mariadb-10.0/10.0.20-0ubuntu0.15.04.1 did not have any s390x builds.

Comment by Elena Stepanova [ 2015-12-29 ]

otto,

Doesn't the reason that the server suggests sound more likely than a mystical test failure which only happens on s390x?

if you are not out of available memory, you can consult the manual for a possible OS-dependent bug

How much memory does the machine have?

Comment by Otto Kekäläinen [ 2016-01-01 ]

I filed the memory limit issue at Launchpad: https://bugs.launchpad.net/launchpad/+bug/1530482
Let's see what they reply.

Comment by Otto Kekäläinen [ 2016-01-02 ]

Ubuntu reply:

They've got 8GiB of RAM and 6GiB of swap. It's deeply unlikely that
available memory on the builders is in fact the problem here, not least
because other architectures have rather less memory, but also because it
really seems quite implausible that MariaDB would in fact need a total
of more than 14GiB to build. I don't think adding more memory would be
a wise resolution to this; it would probably just mean that the build
would fail in much the same way but take longer to do so. It's much
more likely that there's some bug that causes something during the build
to allocate an indefinite amount of memory or other resources; please
investigate that possibility instead.

Comment by Otto Kekäläinen [ 2016-02-25 ]

More replies from Matthias Klose / Ubuntu:

here's the 10.0.24 story:

  • 10.0.24-1: the synced package failed to build on arm64, ppc64el, s390x.
    So all 64bit archs.
  • 10.0.24-1ubuntu1: Run tests with --parallel=2 on these above, arm64
    and ppc64el succeed. s390x still fails.
    A local s390x build using this configuration succeeded with these memory
    constraints.
  • 10.0.24-1ubuntu2: Run tests sequentially on s390x, still running.

**

failed, down to 5 test failures:

Completed: Failed 5/3689 tests, 99.86% were successful.

Looks like the testsuite is not safe to run in parallel. s390x default
to -fPIE, so maybe mariadb has issues with option.

Comment by Otto Kekäläinen [ 2016-02-25 ]

The issue is clearly random to some degree, compare:
https://launchpad.net/ubuntu/+source/mariadb-10.0/10.0.23-1
https://launchpad.net/ubuntu/+source/mariadb-10.0/10.0.23-2
https://launchpad.net/ubuntu/+source/mariadb-10.0/10.0.23-3
https://launchpad.net/ubuntu/+source/mariadb-10.0/10.0.24-1

Comment by Otto Kekäläinen [ 2016-02-26 ]

svoj bar There is a bug about this on Launchpad too: https://bugs.launchpad.net/ubuntu/+source/mariadb-10.0/+bug/1530482
Matthias Klose from Ubuntu has been trying to fix it. Please collaborate. I suggest everybody would discuss in the comments of this upstream bug report to find the final solution.

Comment by Otto Kekäläinen [ 2016-11-25 ]

These are still disabled for s390x:

spider.basic_sql                         [ disabled ]  https://jira.mariadb.org/browse/MDEV-9329
 
spider.basic_sql_part                    [ disabled ]  MDEV-9329

https://launchpadlibrarian.net/291808701/buildlog_ubuntu-zesty-s390x.mariadb-10.0_10.0.28-1_BUILDING.txt.gz

Comment by Nayuta Yanagisawa (Inactive) [ 2021-08-13 ]

Hi cvicentiu! Do you think is this still a problem?

Generated at Thu Feb 08 07:33:51 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.