Christopher Browne <cbbrowne at ca.afilias.info> writes:
> I'm discovering some problems with the new log shipping handling; note
> in the following that there are sequence numbers duplicated.  

It strikes me that perhaps there's a race condition here...  My first
thought is "perhaps a lock should be taken out on sl_archive_counter
at the start of each archive log..."

That seems like not a bad idea, but it doesn't seem to be the key...
-- 
(reverse (concatenate 'string "gro.mca" "@" "enworbbc"))
http://cbbrowne.com/info/spreadsheets.html
Stop the world! I want to get off!!