John Scott <[EMAIL PROTECTED]> writes:
> but does retrying fix the problem? wouldn't the second try just often
> see the same oid again. does a failed attempt increment the oid????
Yes.
> by the way, have you written anything up on the txid wrapping problem?
> we talked about this at osdn last year in san jose. i proposed
> 64 bits, and you discussed a clever wrapping algorithm involving 48 bits and,
> if memory serves me, backward compatiablility of the disk format.
The wraparound idea was discussed on pghackers; people didn't like it
much, but I haven't heard a better near-term solution ...
regards, tom lane
---------------------------(end of broadcast)---------------------------
TIP 4: Don't 'kill -9' the postmaster
