Bug #47160 rpl_binlog_max_cache_size fails sporadically on PB2
Submitted: 6 Sep 2009 23:19 Modified: 2 Oct 2009 16:24
Reporter: Luis Soares Email Updates:
Status: Duplicate Impact on me:
None 
Category:MySQL Server: Replication Severity:S3 (Non-critical)
Version:5.1+ OS:Any
Assigned to: Assigned Account CPU Architecture:Any

[6 Sep 2009 23:19] Luis Soares
Description:
rpl.rpl_binlog_max_cache_size fails sporadically on PB2.

Symptom 1
=========

rpl.rpl_binlog_max_cache_size            w7 [ fail ]
        Test ended at 2009-09-04 23:29:11

CURRENT_TEST: rpl.rpl_binlog_max_cache_size
mysqltest: At line 65: query 'INSERT INTO t2 (a, data) VALUES (2,
CONCAT($data, $data, $data, $data, $data, $data))' failed with wrong errno 2013: 'Lost connection to MySQL server during query', instead of 1197...

Symptom 2
=========

rpl.rpl_binlog_max_cache_size            w5 [ fail ]
        Test ended at 2009-08-24 13:41:46

CURRENT_TEST: rpl.rpl_binlog_max_cache_size
Killed
mysqltest: In included file "./include/diff_master_slave.inc": At line 14: command "$MYSQL test -e "$diff_statement" > $MYSQLTEST_VARDIR/tmp/diff_master.out" failed

Output from before failure:
exec of '/export/home/pb2/test/sb_1-706458-1251113855.78/mysql-5.1.39-solaris10-sparc-test/bin/mysql --defaults-file=/export/home/pb2/test/sb_1-706458-1251113855.78/mysql-5.1.39-solaris10-sparc-test/mysql-test/var/5/my.cnf test -e "SELECT * FROM t1" > /export/home/pb2/test/sb_1-706458-1251113855.78/mysql-5.1.39-solaris10-sparc-test/mysql-test/var/5/tmp/diff_master.out' failed, error: 35072, status: 137, errno: 29

How to repeat:
Sample failure for Symptom 1:

http://pb2.norway.sun.com/web.py?template=mysql_show_test_failure&test_failure_id=2305801

Sample failure for Symptom 2:

http://pb2.norway.sun.com/web.py?template=mysql_show_test_failure&test_failure_id=2296743

Suggested fix:
n/a
[2 Oct 2009 15:06] Alfranio Junior
The errors reported seem to be related to environment problems, specifically storage space. There are other tests failing due to the same reasons. Please, check the links reported above.

However, we are going to try to minimize the problem by avoiding using
./include/diff_master_slave.inc as it dumps tables in order to the
comparison. In particular, this is not a good approach as the tables are
bigger than usual in this test to trigger the cache failures.
[2 Oct 2009 15:46] Bugs System
A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/85567

3109 Alfranio Correia	2009-10-02
      BUG#47160 rpl_binlog_max_cache_size fails sporadically on PB2
      
      The errors reported seem to be related to environment problems, specifically
      storage space. There are other tests failing due to the same reasons.
      
      In this patch, however, we are going to try to minimize the problem by avoiding
      using ./include/diff_master_slave.inc as it dumps tables in order to the
      comparison. In particular, this is not a good approach as the tables are bigger
      than usual to trigger the cache failures.
[2 Oct 2009 16:14] Alfranio Junior
Before both failures, one will find the following error message:

fork failed sleep 1 second and redo: Not enough space at lib/My/SafeProcess/Base.pm line 52.
[2 Oct 2009 16:17] Alfranio Junior
This is a duplicate of BUG#46530.
[2 Oct 2009 16:23] Alfranio Junior
We are not going to push the patch as it seems that the problem is not related to storage space but to the number of leftover processes.