On 9/21/2010 7:54 PM, Aleksey Tsalolikhin wrote:
> On Tue, Sep 21, 2010 at 2:04 PM, Aleksey Tsalolikhin
> <atsaloli.tech at gmail.com>  wrote:
>>  Dear Steve,
>>
>>    You are right, I saw these errors on the Slony master.  The master
>>  was NOT quiesced for the Slony drop/recreate; data was rolling in.
>>
>>    So does this mean 12,000 queries failed?  I don't see anything in
>>  the database log like "query cancelled" or anything like that.
>
> I just found in my application logs that I had 35 queries fail earlier in the
> Slony cluster drop/recreate timeline than the 12000 ERROR's in the
> database log.
>
> I guess that means I have to put my application into "maintenance mode"
> and quisce it before doing a Slony drop/recreate.  Is that what people
> do, commonly?

What you should do is to restart the application between the Slony-I 
uninstall and the reinstall. That will ensure that all prepared query 
plans are recreated.

The message says "cache lookup" because it is coming from the internal 
query executor, that searches for the object (in this case a data type) 
in the system catalog cache and it isn't there.


Jan

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