Re: BUG #9118: WAL Sender does not disconnect replication clients during shutdown

On 2014-03-19 02:53:18 +0900, Fujii Masao wrote:
> On Wed, Mar 19, 2014 at 2:20 AM, Andres Freund <andres(at)2ndquadrant(dot)com> wrote:
> > On 2014-03-19 02:16:00 +0900, Fujii Masao wrote:
> >> > Comments of WalSndWaitForWal would need an update as well in this
> >> > case. It is written on top of this function that it waits "until WAL <
> >> > loc is flushed to disk".
> >>
> >> I don't think that comment needs to be updated because it means that
> >> "until WAL < loc is flushed to the *master local* disk". It's not related to
> >> the flush location that the standby returns. Anyway, I attached the patch.
> >> Barring objection, I will commit this.
> >
> > I object on behalf of WalSndWaitForWal(). We really need the flush
> > location there because otherwise we can't remove old WAL.
>
> MyWalSnd->flush < sentPtr condition in WalSndWaitForWait() is always
> true if the flush location that the client returns is invalid. Is this OK?

Yes. A client doing so would retain WAL on the master
indefinitely. Which isn't a good idea...

Greetings,

Andres Freund

--
Andres Freund http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services