Re: [Exim] exim's databases and sql

On Sat, Oct 04, 2003 at 11:12:27AM +0100, Philip Hazel wrote:
> On Fri, 3 Oct 2003, Avleen Vig wrote:
> > To work around file locking contention against the filesystem-based db.
> > This is a big problem on high-throughput relays.
>
> That is a concern, yes, but
[snip]
> (3) I can't see how using SQL can be *that* much better, in that it too
> must use some kind of locking feature in order to handle simultaneous
> accesses. Can it really do this in a more efficient way than the
> system's file locking? There also has to be a process switch (as far as
> I know) in order to interact with an SQL database.

Well, I only see the advantage in that writes only locks the row, not
the entire DB, but then we're way out in the extremes with a lot of
entries and a lot of writing, and the other options seem just as
well. But on the other hand, if you start spreading the load over a
lot of hosts, it seems like an interesting idea to share the data to
eliminate duplicate efforts and there's where an SQL database shows
promise. But I'm gagging in my own drool here thinking about the farms
you'd need for that...

Isn't the retry db and the spool files pretty much all you can't put
in from various DBs these days?

--
Thomas
beast@???