On Fri, Nov 6, 2020 at 4:54 PM Thomas Munro wrote:
> On Fri, Nov 6, 2020 at 9:48 PM Andy Fan wrote:
> > I have 2 ideas about this. One is in the Read Committed level, we can
> advance xmin
> > aggressively. suppose it started at t1, and complete a query at t2. the
> xmin should
> > be t1 curr
On Fri, Nov 6, 2020 at 9:48 PM Andy Fan wrote:
> I have 2 ideas about this. One is in the Read Committed level, we can
> advance xmin
> aggressively. suppose it started at t1, and complete a query at t2. the xmin
> should
> be t1 currently. Can we advance the xmin to t2 since it is read comm
Since the impact of the long transaction is huge in postgresql, for example
a long transaction by incident, tables may become very huge and it can't
become small again even the transaction is completed unless a vacuum full
is used.
I have 2 ideas about this. One is in the Read Committed level,
