Darcy Buskermolen wrote:

>On Tuesday 08 March 2005 02:42, Chris Newland wrote:
>  
>
>>Thanks to all for the info.
>>
>>I'll keep an eye out for announcements as I'd like to test log shipping
>>in Slony 1.1 in my dev environment.
>>    
>>
>
>You can always track the cvs -HEAD if you want a preview of what's up and 
>coming in 1.1  (as well as providing feedback on anything that may not work)
>  
>
I checked all of the log shipping functionality into CVS HEAD last 
night, so those that are "risk tolerant/interested in testing" would 
find it valuable to take a peek.

There's a new function in the cleanup thread, static unsigned long 
get_earliest_xid (PGconn *dbconn), where I'm suspicious of its behaviour 
with large XIDs; that one's very worthy of "taking a peek." 

Actually, I expect the ideal person to take a poke at it would be Rod 
Taylor, who I suspect is, serendipitously, one of the most likely 
beneficiaries of its benefits, namely that Slony-I 1.1 won't bother with 
(guaranteed to be futile) vacuums if the earliest XID hasn't changed 
since the last cleanup run...