>
> On 10 Jun 2011, at 16:09, Quanah Gibson-Mount wrote:
>
>> --On Friday, June 10, 2011 1:35 PM -0400 Nigel Horne <njh_at_bandsman.co.uk> wrote:
>>
>>>> <http://marc.info/?l=amavis-user&m=130772228532647&w=2>
>>>
>>> Which version of Mail::OpenDKIM was this against? A performance
>>> improvement was noticed and fixed in the most recent version.
>>
>>
>> Mail-OpenDKIM-3230 with an opendkim-2.3.1 library, Perl 5.14.0.
>>
>> If you read the updated thread, Mark has included his test program if you want to try it out.
>
I've looked into this further.
Mark's example hardly uses the Mail::OpenDKIM module, by which I mean the data goes straight through the Mail::OpenDKIM layer with little processing; most of the time he measures is spent within the OpenDKIM library. It is therefore better for one of the OpenDKIM library maintainers to talk about benchmark issues than me.
His comment is true, you get more performance out of the OpenDKIM library if you store the data in large chunk then call dkim_chunk() on a large amount of data, than call dkim_chunk lots of times on small amounts of data. There is a memory hit using that method, but not so much that would matter. Again the question of the best way to use OpenDKIM is a question for the maintainers of that library than for Mail::OpenDKIM's maintainer.
I know that the above sounds like I'm saying "not my fault" as though I'm copping out (something which I *hate* when others do!). I did look into the issue, and will continue so to do in case I've missed something, but at the moment there is nothing more that I can do because I don't believe that the fault lies within Mail::OpenDKIM, it is for the others I mentioned above to help you and I suggest that you ask them for help.
Regards,
-Nigel
Received on Tue Jun 14 2011 - 16:29:24 PST