Peter D. Gray wrote:
> As I say, by solaris mail hubs are not the problem. But when the mail
> gets to the stores, they kind of get slow.
Your stores are running exim too? Then implement the limits there. The
Hubs will queue the mails, you may be interested in Michael Haardts
"randomized retrytimes":
http://exim.inode.at/exim-html-4.60/doc/html/spec.html/ch32.html#id2651626
(Retry algorithm "H", available since exim 4.60 or so.)
> It would be nice if the smtp system on the mail stores knew its own
> limits and only accepted email at a rate it can handle, but that is
> not the case.
I remember this has been discussed on the -devel list some time ago, but
I don't think we came to a final solution, besides the above mentioned.
> No matter how much machinery I throw at the hubs or the stores, I
> suspect I can always have performance problems with high enough email
> rates.
Well, no matter how big your cluster is, there's always a bigger one out
there. ;o)
lg,
daniel
