Re: MultiXactID Wrap-Around

Thank you for the feed-back. In fact the risk seems neglibible, but it means that at least I'm not misunderstanding what that code does...

Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> ha scritto: paolo romano
writes:
> My doubts now concern MultixactID wrap-around management.
> Afaics, it is possible to spawn multixactids so quickly to have a
> wrap-around and to start overwriting the data stored in the offset
> slru (but analogous considerations apply to the member slru as
> well).

I looked into this when the multixact code was written. There is a
theoretical risk but I think it's entirely theoretical. MXIDs are
unlikely to be consumed faster than XIDs over the long term, and also
can be recycled sooner. So you'd run up against XID wraparound (which
we do defend against) first.

regards, tom lane

Chiacchiera con i tuoi amici in tempo reale!
http://it.yahoo.com/mail_it/foot/*http://it.messenger.yahoo.com