maria-discuss team mailing list archive

[Elena Stepanova <elenst@xxxxxxxxxxxxxxxx>]

Hi Reindl,

On 01/20/2017 06:03 PM, Reindl Harald wrote:
> shouldn't the testsuite at some point in time run without errors?

The testsuite should run without errors with some combinations of 
build/run server options which it's tuned for and which we normally run 
it with.

Unfortunately, with arbitrary non-default options in general MTR cannot 
be guaranteed to run successfully, it's just not that kind of tool. Some 
problems are easy to fix though, and we'll do it whenever we can. Please 
in future just file bug reports right away, no need to suffer for years. 
We can only fix something that we are aware of.


> "run STOP SLAVE '' first" - uhm did i start it? it's scripted by shipped
> scripts

It happens because of --mysqld=--binlog-format=mixed on the command 
line, more specifically due to a non-clean exit of a previous test which 
realizes too late that binlog format has been overridden from a command 
line, and skips the rest of execution, including the clean-up.

> [root@testserver:~]$ cat /usr/local/bin/mysql-test.sh
> #!/usr/bin/dash
> /usr/bin/su -c 'cd /usr/share/mysql-test; ./mysql-test-run.pl
> --parallel=1 --max-test-fail=0 --mysqld=--binlog-format=mixed --force'
> mysql
> rm -rf /usr/share/mysql-test/var/*
>
>
> rpl.rpl_binlog_index 'row'               [ fail ]
>         Test ended at 2017-01-20 16:56:16
> CURRENT_TEST: rpl.rpl_binlog_index
> mysqltest: In included file "./include/rpl_init.inc":
> included from ./include/master-slave.inc at line 38:
> included from /usr/share/mysql-test/suite/rpl/t/rpl_binlog_index.test at
> line 20:
> At line 165: query 'SET GLOBAL gtid_slave_pos= ""' failed: 1198: This
> operation cannot be performed as you have a running slave ''; run STOP
> SLAVE '' first
> The result from queries just before the failure was:
> include/master-slave.inc
>  - saving '/usr/share/mysql-test/var/log/rpl.rpl_binlog_index-row/' to
> '/usr/share/mysql-test/var/log/rpl.rpl_binlog_index-row/'
>
> rpl.sec_behind_master-5114 'stmt'        [ fail ]
>         Test ended at 2017-01-20 16:58:33
> CURRENT_TEST: rpl.sec_behind_master-5114
> mysqltest: In included file "./include/rpl_init.inc":
> included from ./include/master-slave.inc at line 38:
> included from
> /usr/share/mysql-test/suite/rpl/t/sec_behind_master-5114.test at line 1:
> At line 165: query 'SET GLOBAL gtid_slave_pos= ""' failed: 1198: This
> operation cannot be performed as you have a running slave ''; run STOP
> SLAVE '' first
> The result from queries just before the failure was:
> include/master-slave.inc
>
>
> rpl.rpl_semi_sync_event_after_sync 'mix,xtradb' [ fail ]
>         Test ended at 2017-01-20 16:59:24
> CURRENT_TEST: rpl.rpl_semi_sync_event_after_sync
> mysqltest: At line 1: query 'set global
> rpl_semi_sync_master_wait_point=AFTER_SYNC' failed: 1193: Unknown system
> variable 'rpl_semi_sync_master_wait_point'
> The result from queries just before the failure was:
> set global rpl_semi_sync_master_wait_point=AFTER_SYNC;
>  - saving
> '/usr/share/mysql-test/var/log/rpl.rpl_semi_sync_event_after_sync-mix,xtradb/'
> to
> '/usr/share/mysql-test/var/log/rpl.rpl_semi_sync_event_after_sync-mix,xtradb/'

it fails because there is no semisync plugin (you build without dynamic 
plugins, right?), but the test checks for that too late. A similar 
problem affects encryption tests that you mentioned in another email.


Regards,
Elena


> _______________________________________________
> Mailing list: https://launchpad.net/~maria-discuss
> Post to     : maria-discuss@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~maria-discuss
> More help   : https://help.launchpad.net/ListHelp