Albert,

Perhaps you can submit an enhancement request so that constraints are unloaded 
such that the order is:
PRIMARY KEY
UNIQUE INDEX/KEY
FOREIGN KEY
regular INDEXES 

 
Jim Bentley
American Celiac Society
[email protected]
tel: 1-504-737-3293


>________________________________
> From: Albert Berry <[email protected]>
>To: RBASE-L Mailing List <[email protected]> 
>Sent: Thursday, May 3, 2012 10:23 AM
>Subject: [RBASE-L] - RE: Constraint troubleshooting?
> 
>Javier, I am using this routine to rebuild a database. I discovered one 
>thing that must be edited before the constraints definitions are run. 
>The output does not unload the constraints in the correct order. I 
>discovered this when the error showed up.
>
>-ERROR- The referenced table Races does not have a compatible Primary 
>Key. (2769)
>
>The table Races primary key definition was the last entry in the file. I 
>cut and pasted all the primary keys first, followed by the unique keys 
>and lastly the foreign keys.
>
>Albert
>
>On 02/05/2012 2:58 PM, Javier Valencia wrote:
>>
>> Buddy,
>>
>> R:Base has provided us a better way of doing this:
>>
>> OUTPUT MyDB.STR
>>
>> UNLOAD ALL FOR TABLEDEF
>>
>> OUTPUT MyDB.VIE
>>
>> UNLOAD ALL FOR VIEWS
>>
>> OUTPUT MyDB.CST
>>
>> UNLOAD ALL FOR CONSTRAINTS
>>
>> OUTPUT MyDB.IND
>>
>> UNLOAD ALL FOR INDEXES
>>
>> OUTPUT MyDB.RUL
>>
>> UNLOAD ALL FOR RULES
>>
>> OUTPUT MyDB.ACC
>>
>> UNLOAD ALL FOR ACCESS
>>
>> OUTPUT MyDB.AUT
>>
>> UNLOAD ALL FOR AUTONUM
>>
>> OUTPUT MyDB.COM
>>
>> UNLOAD ALL FOR COMMENTS
>>
>> OUTPUT MyDB.TRG
>>
>> UNLOAD ALL FOR TRIGGERS
>>
>> OUTPUT MyDB.DAT
>>
>> UNLOAD DATA
>>
>> OUTPUT SCREEN
>>
>> And now you have all components separated and can load them as needed; 
>> using the TRACE feature lets you know exactly where the problem happens.
>>
>> Javier,
>>
>> Javier Valencia, PE
>>
>> O: 913-829-0888
>>
>> H: 913-397-9605
>>
>> C: 913-915-3137
>>
>> *From:*[email protected] [mailto:[email protected]] *On Behalf Of 
>> *Walker, Buddy
>> *Sent:* Wednesday, May 02, 2012 12:41 PM
>> *To:* RBASE-L Mailing List
>> *Subject:* [RBASE-L] - RE: Constraint troubleshooting?
>>
>> Karen
>>
>>    What I have done in the past was this
>>
>>   Make sure messages and error messages on are
>>
>>    OUT whatever.ALL
>>
>>                 UNLOAD ALL
>>
>>   OUT TERM
>>
>>    In the whatever.all file I cut out all of the create views, create 
>> index and alter table
>>
>>    I put each one of them in their own file.
>>
>>   RENAME the old database
>>
>>   Now using 9.1(64)
>>
>>    Run whatever.all
>>
>>    If this is really big database I would make a copy of it at this time.
>>
>>   Run the create views file
>>
>>   Run the create index file
>>
>> Since you are having problems with constraints I would
>>
>>   Trace the alter table file. It should stop on the error.
>>
>> Buddy
>>
>> *From:*[email protected] <mailto:[email protected]> 
>> [mailto:[email protected]] <mailto:[mailto:[email protected]]> *On 
>> Behalf Of *[email protected] <mailto:[email protected]>
>> *Sent:* Wednesday, May 02, 2012 12:47 PM
>> *To:* RBASE-L Mailing List
>> *Subject:* [RBASE-L] - Constraint troubleshooting?
>>
>> We used "unload all" to recreate the database for a recent upgrade 
>> from 7.6 to 9.1 64-bit.   An autochk on the database gives us 6 
>> instances of an error "A foreign key references a table not known to 
>> be referenced" on various tables.
>>
>> First I used some existing code I had to create a cursor thru all FK 
>> columns and write out what PK column/table it references per the 
>> sys_indexes table, and also displayed the "referenced" flag of that 
>> PK.   Everything looks great.
>>
>> So then I did an "unload schema" and looked at the bottom at all the 
>> FKs (luckily only 25 of them) and all of them reference good tables & 
>> columns.  If I list each PK table, each one shows "PK referenced".
>>
>> Anyone have a clue what that message means and how I can find out what 
>> the offending entries are?    If we reload the database, the reloaded 
>> copy shows the same errors.
>>
>> Karen
>>
>> No virus found in this message.
>> Checked by AVG - www.avg.com <http://www.avg.com>
>> Version: 2012.0.2169 / Virus Database: 2411/4973 - Release Date: 05/02/12
>>
>
>
>
>
>

Reply via email to