[SURBL-Discuss] RFC: consensus list?

Jeff Chan jeffc at surbl.org
Sun Nov 14 10:19:48 CET 2004


On Saturday, November 13, 2004, 7:45:51 PM, Frank Ellermann wrote:
> Jeff Chan wrote:

>> Setting up and tearing down separate test lists involves
>> some overhead.  And eventually multi will need to expand
>> beyond the last octet.

> I'll believe it when I see it (the expansion of "multi" ;-)

Well in addition to the combined lists we have another 3 we could
test or deploy:

1. DNS queries that match SBL
2. DNS queries that match XBL
3. trap data from Terry Sullivan,
    which is a bit faster than our other sources

> Don't do strange things with stuff used in production, you
> don't want to cause erroneous mail deletions.

Eventually we will need more than 7 bits.

How does anyone feel about:

1.  Propose a separate test list for trying new lists
but that should not be used for production servers.

2.  Ask application writers like SpamAssassin and others
to check or modify their code to be able to handle bits
in the other octets, i.e. 127.0.1.X, 127.64.X.X, etc.
Should we let them know ahead of time that their code
should be able to handle these cases without breaking?

Comments?

Jeff C.
--
"If it appears in hams, then don't list it."



More information about the Discuss mailing list