Bug #5556 Test failure: 'ps' with 4.1.5
Submitted: 14 Sep 2004 5:53 Modified: 16 Sep 2004 18:16
Reporter: Matt Wagner Email Updates:
Status: Closed Impact on me:
None 
Category:MySQL Server: Compiling Severity:S1 (Critical)
Version:4.1.5-gamma OS:Other (QNX)
Assigned to: Konstantin Osipov CPU Architecture:Any

[14 Sep 2004 5:53] Matt Wagner
Description:
On 'buildqnx', the test 'ps' fails with this result:

Below are the diffs between actual and expected results:
-------------------------------------------------------
*** r/ps.result Mon Sep 13 08:19:05 2004
--- r/ps.reject Tue Sep 14 02:44:39 2004
***************
*** 213,222 ****
  prepare stmt1 from ' show table status from test like ''t1%'' ';
  execute stmt1;
  Name  Engine  Version Row_format      Rows    Avg_row_length  Data_length     
Max_data_length Index_length    Data_free       Auto_increment  Create_time     
Update_time     Check_time      Collation       Checksum        Create_options  
Comment
! t1    MyISAM  9       Dynamic 0       0       0       4294967295      1024    
0       NULL    #       #       #       latin1_swedish_ci       NULL            
  show table status from test like 't1%' ;
  Name  Engine  Version Row_format      Rows    Avg_row_length  Data_length     
Max_data_length Index_length    Data_free       Auto_increment  Create_time     
Update_time     Check_time      Collation       Checksum        Create_options  
Comment
! t1    MyISAM  9       Dynamic 0       0       0       4294967295      1024    
0       NULL    #       #       #       latin1_swedish_ci       NULL            
  deallocate prepare stmt1 ;
  drop table t1;
  create table t1(a varchar(2), b varchar(3));
--- 213,222 ----
  prepare stmt1 from ' show table status from test like ''t1%'' ';
  execute stmt1;
  Name  Engine  Version Row_format      Rows    Avg_row_length  Data_length     
Max_data_length Index_length    Data_free       Auto_increment  Create_time     
Update_time     Check_time      Collation       Checksum        Create_options  
Comment
! t1    MyISAM  9       Dynamic 0       0       0       2147483647      1024    
0       NULL    #       #       #       latin1_swedish_ci       NULL            
  show table status from test like 't1%' ;
  Name  Engine  Version Row_format      Rows    Avg_row_length  Data_length     
Max_data_length Index_length    Data_free       Auto_increment  Create_time     
Update_time     Check_time      Collation       Checksum        Create_options  
Comment
! t1    MyISAM  9       Dynamic 0       0       0       2147483647      1024    
0       NULL    #       #       #       latin1_swedish_ci       NULL            
  deallocate prepare stmt1 ;
  drop table t1;
  create table t1(a varchar(2), b varchar(3));
-------------------------------------------------------

How to repeat:
CFLAGS="-O1" CXXFLAGS="-O1"  ./configure --prefix=/usr/local/mysql --localstatedir=/usr/local/
mysql/data --libexecdir=/usr/local/mysql/bin --with-comment="Official MySQL-debug binary" --
with-extra-charsets=complex --with-server-suffix="-debug" --enable-thread-safe-client --enable-
local-infile  --disable-shared --with-zlib-dir=bundled --with-debug --with-readline --with-
embedded-server --with-innodb

./mysql-test-run  --tmpdir=/tmp/my_build-buildqnx --master_port=3334 --slave_port=3350 --
ndbcluster_port=9350 --manager-port=3335 --no-manager --sleep=10
[15 Sep 2004 14:58] Konstantin Osipov
This bug is not prepared-statements specific, and has always been there.
The difference is in test results, 'Max_data_length' column.
The value of this column is taken from handler::max_data_file_length, which in this particular case is calculated in mi_open. Apparently this calculation is platform-specific.
We need to investigate the reason why qnx is different, but meanwhile I would just suppress
this test result.
[15 Sep 2004 17:22] Lenz Grimmer
Thanks for the analysis!
[16 Sep 2004 18:16] Konstantin Osipov
Fixed:
 bk commit - 4.1 tree (konstantin:1.2025) BUG#5556