[MDEV-7241] rpl.rpl_parallel2 fails sporadically in buildbot Created: 2014-12-01  Updated: 2014-12-02  Resolved: 2014-12-02

Status: Closed
Project: MariaDB Server
Component/s: Tests
Affects Version/s: 10.0
Fix Version/s: 10.0.16

Type: Bug Priority: Major
Reporter: Elena Stepanova Assignee: Kristian Nielsen
Resolution: Fixed Votes: 0
Labels: buildbot, tests

Issue Links:
Relates
relates to MDEV-7069 Fix buildbot failures in main server ... Stalled

 Description   

http://buildbot.askmonty.org/buildbot/builders/kvm-bintar-centos5-x86/builds/1915/steps/test/logs/stdio

rpl.rpl_parallel2 'stmt'                 w2 [ fail ]
        Test ended at 2014-11-29 04:50:39
 
CURRENT_TEST: rpl.rpl_parallel2
--- /usr/local/mariadb-10.0.15-linux-i686/mysql-test/suite/rpl/r/rpl_parallel2.result	2014-11-28 15:20:23.000000000 +0100
+++ /usr/local/mariadb-10.0.15-linux-i686/mysql-test/suite/rpl/r/rpl_parallel2.reject	2014-11-29 04:50:38.000000000 +0100
@@ -10,7 +10,7 @@
 Warnings:
 Note	1592	Unsafe statement written to the binary log using statement format since BINLOG_FORMAT = STATEMENT. Statement is unsafe because it uses a system function that may return a different value on the slave.
 Seconds_Behind_Master should be zero here because the slave is fully caught up and idle.
-Seconds_Behind_Master = '0'
+Seconds_Behind_Master = '5'
 include/stop_slave.inc
 SET GLOBAL slave_parallel_threads=@old_parallel_threads;
 include/start_slave.inc
 
mysqltest: Result content mismatch

Strangely, I can also see the failure on 5.5 tree http://buildbot.askmonty.org/buildbot/builders/kvm-deb-sid-amd64/builds/475/steps/test_4/logs/stdio , but can't see the actual test (and can't understand how it could have been there, if it's parallel replication...)



 Comments   
Comment by Kristian Nielsen [ 2014-12-02 ]

The reason we see the failure in the 5.5 build is the following. This is
Debian unstable (I think). In Debian unstable, there are now native MariaDB 10
packages. The Buildbot job does:

apt-get install mariadb-server mariadb-test

The Buildbot job has setup a repository with mariadb-server version 5.5, but
the native repository has a newer version 10.0, so the newer version is pulled
in and tested.

I would suggest that, since Debian sid now has 10.0 native, it does not make
much sense to build 5.5 for it.

Comment by Kristian Nielsen [ 2014-12-02 ]

Failure can be reproduced reliably by inserting a sleep:

=== modified file 'sql/rpl_rli.cc'
--- sql/rpl_rli.cc	2014-12-01 12:53:57 +0000
+++ sql/rpl_rli.cc	2014-12-02 08:10:32 +0000
@@ -1055,6 +1055,7 @@ void Relay_log_info::inc_group_relay_log
   mysql_cond_broadcast(&data_cond);
   if (!skip_lock)
     mysql_mutex_unlock(&data_lock);
+my_sleep(100000);
   DBUG_VOID_RETURN;
 }
 
 

Comment by Kristian Nielsen [ 2014-12-02 ]

http://lists.askmonty.org/pipermail/commits/2014-December/007128.html

Comment by Kristian Nielsen [ 2014-12-02 ]

Pushed to 10.0.16

Generated at Thu Feb 08 07:18:05 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.