http://bugzilla.spamassassin.org/show_bug.cgi?id=4188
------- Additional Comments From [EMAIL PROTECTED] 2005-07-15 11:03 ------- Subject: Re: RCVD_HELO_IP_MISMATCH should check address literals On 7/15/2005 1:12 PM, [EMAIL PROTECTED] wrote: >>Your assumption is correct. The rule only fires against n.n.n.n in the >>HELO greeting, and doesn't fire against [n.n.n.n] (I don't know if this is >>still current, but I doubt this has changed since the report was filed) > > In many cases in SpamAssassin, this is by design btw -- [n.n.n.n] is the > RFC-approved way to do this, while n.n.n.n is more common in spam. I know, and it's sometimes useful to ignore it (as is the case when a client behind a NAT doesn't know it's public address, and which is obviously going to be different from the local private address). But the point remains that there is some value to the test, and spammers can easily bypass the current rule just by wrapping the addr in brackets. Probably the best thing to do here actually is to have a separate rule that is scored independently, so that it can be disabled if needed. ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.
