From: Wietse Venema (wietseporcupine.org)
Date: Tue Aug 29 2006 - 08:47:37 CDT

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Ramprasad:
> On Mon, 2006-08-28 at 11:27 -0500, Noel Jones wrote:
> > At 08:53 AM 8/28/2006, Ramprasad wrote:
> > >I have enabled Cached Remote Address verification on my
> > >postfix server
> > >which is used as a relay server
> > >
> > >I am confused about a few parameters
> > >What is the difference between
> > >address_verify_negative_expire_time and
> > >address_verify_negative_refresh_time
> >
> > The expire time is the time at which an entry is considered
> > no longer valid.
> > The refresh time is the time postfix will attempt to verify
> > the entry is still correct. If the entry is no longer
> > correct (ie. no answer or a different answer) postfix will
> > still wait *_expire_time before the status of the entry is
> > changed.
>
> Sorry, Didnt quiet get this. If postfix is anyway going to wait until
> _expire_time

The main reason for cache refresh-before-expire is that the SMTP
server does not have to wait for information to arrive in the cache.
It simple uses the information that isn't expired. The refresh probe
completes asynchronously at some later point in time.

        Wietse

> then why have the _refresh_time parameter
> Consider this
> 1) Mail arrives for id in negative cache
> 2) Cache Id is older than _refresh_time but not as old as _expire_time
>
> Will this mail get rejected ?

Mail is rejected when the cache says that it can't be delivered,
or when no cache entry exists within a few seconds from the start
of the lookup request.

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]