Bayes Rebuild not happening

David Lee t.d.lee at DURHAM.AC.UK
Fri Mar 26 16:10:49 GMT 2004


On Fri, 26 Mar 2004, Julian Field wrote:

> Wonderful! At long last we can actually get an error report. That extra
> eval{} was definitely worth it!
>
> Rather than having to get a patch from the SA guys, please can you try this
> patch to SA.pm
>
> -----SNIP-----
> --- SA.pm.old   2004-03-26 15:33:56.000000000 +0000
> +++ SA.pm       2004-03-26 15:35:24.000000000 +0000
> @@ -128,6 +128,7 @@
>           # Do the actual expiry run
>           MailScanner::Log::InfoLog('SpamAssassin Bayes database rebuild
> starting');
>           eval {
> +          $MailScanner::SA::SAspamtest->init(1);
>             $MailScanner::SA::SAspamtest->init_learner({
>                             force_expire => 1,
>                             learn_to_journal => 0,
> -----SNIP-----
>
> init() is supposed to be a non-public method, but I can't see any other way
> of doing it.
>
> Let me know if this fixes the problem. If it does, it will go in the next
> release.


Ouch!  OK it might temporarily dig us, the MailScanner subset of SA's user
community, out of a hole for the moment.  But doesn't it store up trouble
for the future?  At some point SA's internals, including the specifically
non-public "init()", might change, and drop us back into a hole.

Wouldn't it be much better, for all concerned, abd for both now and into
the future, to get this fixed within SA itself?  Anyone on the list in
regular contact with the SA folk, who could try to persuade them?

(Can this be a "both/and"?  Both Julian's MS patch for us, for the
interim, and trying to get SA itself patched as the proper, long-term
solution?)

--

:  David Lee                                I.T. Service          :
:  Systems Programmer                       Computer Centre       :
:                                           University of Durham  :
:  http://www.dur.ac.uk/t.d.lee/            South Road            :
:                                           Durham                :
:  Phone: +44 191 334 2752                  U.K.                  :



More information about the MailScanner mailing list