Re: huge table occupation after updates

2016-12-10 18:33 GMT+01:00 Francisco Olarte <folarte(at)peoplecall(dot)com>:

> Tom:
>
> On Sat, Dec 10, 2016 at 6:01 PM, Tom DalPozzo <t(dot)dalpozzo(at)gmail(dot)com>
> wrote:
> > As for crash proof, I meant that once my client app is told that her
> update
> > request was committed, it mustn't get lost (hdd failure apart of course).
> > And I can't wait to flush the cache before telling to the app
> :"committed".
> > I can replicate also the cache on the standby PC of course.
>
> You are making inconsistent requests. When the server tells your app
> it's commited, it has flush the transaction log cache. If your
> assertion about is real, you cannot wait for commit, so your
> requirements are imposible to satisfy ( of course, you could run with
> scissors, but that will loose data without hdd failure ).
>
> Francisco Olarte.
>

​Hi, perhaps I was not clear. The cache I mentioned is a possible cache in
my app, outside postgresql server.
I answered to Rob's question with more details regarding my app.

Regards
Pupillo