Andrew Dunstan <[EMAIL PROTECTED]> writes:
> Jan Wieck wrote:
>> Then we better make sure that 8.0 -> 8.1 does not require dump&reload.
> There was some mention of an upgrade tool which would avoid the need for
> a dump/restore - did that idea die?
No, but I don't see anyone volunteering to work on it now --- much less
to make it robust and reliable in the next two months, which is what
would have to happen to make it a useful answer in the timeframe we need.
At the moment I think that the most sane way to proceed is to back-patch
one of the 2Q variants I posted into 8.0.*, so as to get out of the
patent issue in that branch with minimum effort, and then proceed with a
"normal" development cycle for 8.1. The discussions currently going on
about the bufmgr are focusing on abandoning LRU/ARC/2Q entirely in favor
of something that requires only local state updates, so it seems a bit
pointless to expend a major amount of work on that line of code.
regards, tom lane
---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
