List:Cluster« Previous MessageNext Message »
From:Tomas Ulin Date:October 7 2004 9:49am
Subject:Re: Testing the 6oct04 build
View as plain text  
Joseph,

thank you for putting our attention to this.  Attmepted to fix all the 
order by.

regarding "error":

select a,b from t1 limit 1;


Our sql experts say the result is undefined.  Hence we've changed the 
test to:

select a,b from t1 order by a limit 1;


BR,

Tomas

Joseph E. Sacco, Ph.D. wrote:

>System:
>* PowerMac[silver] with dual G4 533-MHz CPU's, 1GB RAM, SCSI drives
>* YDL-3.0.1
>* gcc-3.2.2
>* 6oct04 BK code
>==========================================================================
>Pulled, built and tested the latest source [up through change set
>1.2062] from the BK tree:
>
>        ./mysql-test/mysql-test-run --force --with-ndbcluster
>
>Results:
>
>Failed 2/289 tests, 99.30% successful.
>
>ndb_subquery                   [ fail ]
>ps_7ndb                        [ fail ]
>
>Discussion:
>
>The ndb_subquery test did not really fail.  The select statements
>require "SORT BY" clauses to handle output differences between
>big-endian and little-endian machines.
>
>All but one of the reported failures in the ps_7ndb test are also output
>ordering differences that can be removed by adding "SORT BY" clauses. 
>There appears to be one real error:
>
>*** 444,450 ****
>  limit 1 ';
>  execute stmt1 ;
>  a     b
>! 2     two
>  prepare stmt1 from ' select a,b from t1
>  limit ? ';
>  ERROR 42000: You have an error in your SQL syntax; check the manual
>that corresponds to your MySQL server version for the right syntax to
>use near '?' at line 2
>--- 444,450 ----
>  limit 1 ';
>  execute stmt1 ;
>  a     b
>! 3     three
>  prepare stmt1 from ' select a,b from t1
>  limit ? ';
>  ERROR 42000: You have an error in your SQL syntax; check the manual
>that corresponds to your MySQL server version for the right syntax to
>use near '?' at line 2
>
>
>NOTE:
>I also tried building the code using gcc-3.4.2. Everything built, but
>several of the ndb tests failed. GCC-3.4.2 is "stricter" than gcc-3.2.2.
>Whether or not that is the issue is unclear.  It is always possible that
>the compiler I built is broken in some obscure way.
>
>QUESTION:
>Should the code build and run properly under gcc-3.4.2?
>
>-Joseph
>
>
>  
>

Thread
Testing the 6oct04 buildPh.D. Joseph E. Sacco7 Oct
  • Re: Testing the 6oct04 buildTomas Ulin7 Oct