June 10, 2009 at 10:36 pm
Hi I have two databases on same server and both databases.These do databases are same (we can say replica) .There is a table name 'Transactions' on both databases.Now I have to create a job which runs on daily basis and check whether the data in Table 'Transaction' in both the databases is same or not (i.e to check whether table in both database is in sync) .How can I do this.Please suggest.
June 10, 2009 at 11:06 pm
jain_abhishek (6/10/2009)
Hi I have two databases on same server and both databases.These do databases are same (we can say replica) .There is a table name 'Transactions' on both databases.Now I have to create a job which runs on daily basis and check whether the data in Table 'Transaction' in both the databases is same or not (i.e to check whether table in both database is in sync) .How can I do this.Please suggest.
you can create a checksum for the values in the table (see checksum http://msdn.microsoft.com/en-us/library/ms189788(SQL.90).aspx and cheksum_agg (http://msdn.microsoft.com/en-us/library/ms188920(SQL.90).aspx) and then compare the checksums. You can do this on a per row level (you may want to do this via a linked server), or the whole table. The fewer checksums you compare the less accurate the result will be (i.e. if you calculate a checksum for the whole table you may encounter cases where the checksums match, but the tables are different. You should also consider calculating the checksums based on some inside knowledge (e.g. do the checksums on only a few columns if that is sufficient to determine a difference). The checksum calculation may be expensive, so do some initial tests (like row counts). If you need to compare all the data without checksums, use either linked servers or there are third party tools too.
Regards,
Andras
June 10, 2009 at 11:18 pm
use tablediff utility : C:\Program Files\Microsoft SQL Server\90\COM
give source and destination [servername , tablename , databasename]..
simple to use (consult BOL).
make sure that the table has unique/primary key (clustered index)
HTH
Abhay Chaudhary
Sr.DBA (MCITP/MCTS :SQL Server 2005/2008 ,OCP 9i)
June 10, 2009 at 11:19 pm
but both the tables are in different database.how can i compare
June 10, 2009 at 11:24 pm
One more thing If both table are not having same data then how will I make both tablles in sync ie same data in both table
June 11, 2009 at 12:07 am
its possible ........databases can be different ...
for your second query : its very much possible..
Very busy right now ....wil reply later ...till then do soem research 🙂
Abhay Chaudhary
Sr.DBA (MCITP/MCTS :SQL Server 2005/2008 ,OCP 9i)
June 11, 2009 at 5:37 am
Since databases are different, use fully qualified name of the databases i.e. prefix the table name with server IP.database.user.table
To comapre the results, u can use EXCEPT operator.
loke
select * from d1..table1
except
select * from d2..table2
June 12, 2009 at 7:13 am
After you get the first compare done, you could add an update_dt field and have it default to the current getdate() value. Then, if there are dates since the last comparison, then those are the ones that are different.
There's also merge replication, once you got them in a spot where they are the same.
The EXCEPT is the way I would make that first compare, too.
Randy
Viewing 8 posts - 1 through 7 (of 7 total)
You must be logged in to reply to this topic. Login to reply