Jan Wieck JanWieck at Yahoo.com
Tue Aug 17 10:09:19 PDT 2010
On 8/17/2010 2:13 AM, sivakumar krishnamurthy wrote:
> 
> 
>     Triggers on the table may still need to fire; they certainly did, on the
>     origin, and they may also need to, on subscribers.
> 
> 
> Mine might be a naive question however why do we need a query like below 
> if both the old and new values are same in the UPDATE statement to be 
> executed on slaves.
> UPDATE tbl set key='x' where key='x'.
> This unnecessarily causes additional table and index bloat on the 
> subscribers.

Something on the origin did something similar. It updated a row and that 
update caused no single value in the row to change. As pointed out, the 
resulting update on the subscriber is done so that eventual triggers 
there will fire.

We can add a configuration option, suppressing those updates, to the 2.1 
todo list. In the meantime, if that happens that much that it causes 
problems for you, it may be better to fix your application. Because your 
origin must suffer the same bloat.


Jan

-- 
Anyone who trades liberty for security deserves neither
liberty nor security. -- Benjamin Franklin


More information about the Slony1-general mailing list