Re: [Gusdev-gusdev] ga problem

[Steve F. <sfi...@pc...>]

jonathan-

which workaround?  the original one you put in place is this, at the end 
of DbiDbHandle.pm:


# JC 9/6/2002
#
# Subclassing not set up correctly according to DBI manpage;
# this appears to fix it, although the structure is still
# a little off.  The fact that we're reblessing the return
# value from the DBI connect() method is making everything
# work, albeit in a roundabout way.
#
package GUS::ObjRelP::DbiDbHandle::db;
use vars qw(@ISA);
@ISA = qw( DBI::db );

Thats been in place for ages.

is there another one?  

i think that there may be something else going on here, having to do 
with version of perl,  or something.

steve

Jonathan Crabtree wrote:

>Hi Sam-
>
>On Sat, 4 Oct 2003, sam wang wrote:
>  
>
>>DBI subclasses 'GUS::ObjRelP::DbiDbHandle::db' and ::st are not setup, 
>>RootClass ignored at 
>>    
>>
>
>The above error is annoying, but should not affect the correct operation 
>of the plugin.  It has to do with the way that the DBI classes are subclassed 
>in the GUS code, and a workaround is available (could someone commit this 
>to CVS?)
>
>  
>
>>DBD::Oracle::st execute failed: ORA-02291: integrity constraint 
>>(CORE.ALGORITHMINVOCATION_FK04) violated - parent key not found (DBD ERROR: 
>>OCIStmtExecute) [for statement ``
>>    INSERT INTO Core.AlgorithmInvocation (  group_read, user_read, 
>>algorithm_implementation_id, modification_date, end_time,
>>other_write, row_group_id, machine_id, user_write, other_read, 
>>group_write, cpus_used, start_time, row_user_id, row_alg_invocation_id, 
>>row_project_id, algorithm_invocation_id, result )
>>    VALUES   (  ?, ?, ?, SYSDATE, SYSDATE, ?, ?, ?, ?, ?, ?, ?, 
>>SYSDATE, ?, ?, ?, ?, ? ) '' with params: :p5='1', :p12=1, :p8=1, :p14='2', 
>>:p10=1, :p13='1', :p2=1, :p6=0, :p3='2', :p15='meta', :p7=1, :p1=1, :p4=0, 
>>:p9=1, :p11='1']) at 
>>/opt/gus/gus_home/lib/perl/GUS/ObjRelP/DbiDbHandle.pm line 144.
>> 
>> SQL ERROR!! involving
>>  
>>    INSERT INTO Core.AlgorithmInvocation (  group_read, user_read, 
>>algorithm_implementation_id, modification_date, end_time,
>>other_write, row_group_id, machine_id, user_write, other_read, 
>>group_write, cpus_used, start_time, row_user_id, row_alg_invocation_id, 
>>row_project_id, algorithm_invocation_id, result )
>>    VALUES   (  ?, ?, ?, SYSDATE, SYSDATE, ?, ?, ?, ?, ?, ?, ?, 
>>SYSDATE, ?, ?, ?, ?, ? )
>> Values: 1, 1, 2, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 2, meta at 
>>/opt/gus/gus_home/lib/perl/GUS/ObjRelP/DbiDbHandle.pm line 184
>>    
>>
>
>Now this is the real error; a failed INSERT statement.  Notice that the 
>error mentions a violation of the integrity constraint 
>"CORE.ALGORITHMINVOCATION_FK04".  You can find the definition of this 
>constraint in the file GUS/Model/schema/oracle/core-constraints.sql (from 
>the GUS CVS repository):
>
>alter table @oracle_core@.ALGORITHMINVOCATION add constraint 
>ALGORITHMINVOCATION_FK04 foreign key (MACHINE_ID) references 
>@oracle_core@.MACHINE (MACHINE_ID);
>
>So the problem is that your INSERT statement is referencing a nonexistent 
>row in core.Machine (namely the one with machine_id = 0).  Insert this 
>missing row into the core.Machine table (using SQL*PLUS, since you won't 
>be able to do it with 'ga' until the problem is fixed) and it should 
>allow the insert to succeed.
>
>Jonathan
>
>
>
>
>-------------------------------------------------------
>This sf.net email is sponsored by:ThinkGeek
>Welcome to geek heaven.
>http://thinkgeek.com/sf
>_______________________________________________
>Gusdev-gusdev mailing list
>Gus...@li...
>https://lists.sourceforge.net/lists/listinfo/gusdev-gusdev
>  
>