--On 28 April 2009 14:16:17 +0400 Peter Volkov <volkov.peter@???>
wrote:
>
> Another problem is related to -odqs option. As I understand the manual
> this option tels exim to do only routing and *no* delivery. Then why I
> see the following messages in the log file:
You'll probably do better without using -odqs. You should avoid generating
a large queue if possible, since that's not Exim's most efficient feature.
Also, if your messages aren't personalised, make sure that they're fed to
Exim sorted by domain. It's much more efficient to deliver to 100 users in
a single domain than to 100 users in different domains. It probably doesn't
much matter how you order the domains, but if you do have lots of
recipients in a domain that you know you can deliver to quickly, then
prioritise that domain.
Oh, and because direct delivery is more efficient than queueing, try
feeding Exim more gradually (if you can).
Also, consider replicating your hardware, if nothing else helps.
Finally, how do you run your queue runners? Don't have too many running at
one time, and try using a two pass run. Think about using a virtual disk
for your queue. If it's a newsletter, it might not be a disaster to drop
the odd issue once in a while.
And, keep your recipient list up to date. There's no point wasting time
trying to deliver to addresses that don't want your newsletter. Especially
when the recipient address is not deliverable.
> 2009-04-28 11:34:09 1LyhpP-0004TO-HX no immediate delivery: load average
> 9.40
>
> ? Probably this is our performance bottleneck too?
>
>
> In any case thank you for any hints,
> --
> Peter.
--
Ian Eiloart
IT Services, University of Sussex
01273-873148 x3148
For new support requests, see
http://www.sussex.ac.uk/its/help/
