RE: btree split logic is fragile in the presence of lar ge index items

> -----Original Message-----
> From: Mikheev, Vadim
>
> > > > Do not add TID to key but store key anywhere in duplicate
> > > > chain and just read lefter child page while positioning index scan,
> > > > as we do right now for partial keys?
> > >
> > > > This will result in additional reads but I like it much more than
> > > > current "logic"...
> > >
> >
> > What about unique key insertions ?
>
> We'll have to find leftmost key in this case and do what we do now.
>

Currently the page contains the leftmost key is the target page of
insertion and is locked exclusively but it may be different in extra
TID implementation. There may be a very rare deadlock possibility.

Hiroshi Inoue
Inoue(at)tpf(dot)co(dot)jp