In this case the AWL gives 4.21 points, that makes the spamfilter block the message.
Any idea why? In my opinion when an address is whitelisted it should give a lower score?
This question came out of a few messages that are being blocked.
I saw the AWL giving the mail a higher score.
The issue is that some messages are being blocked that we don’t want to be blocked.
Those are mostly messages from services as Sendgrid.
I have address the IP address in the whitelist as XXX.XX. but still the messages are being blocked.
I have three Message IDs, is it possible to take a look and explain why those are blocked?
has received a score of 5.99 from AWL, which would be enough to trigger the SPAM flag. However it has also been caught by 4 RBL, with an extra score of 18, which means that the sender is blacklisted in public registers.
Message is almost the same
Message has received a negative score from AWL (non-spam), but has been caught by 3 RBL, which have added a score of 13.50. So, this one has not been blocked by AWL.
We may create an AWL config interface in the Protector to alter the behavior of the AWL filter, but the main issue here seems to be the RBL.
But doesnt it help that we’ve whitelisted the IP range to avoid the blacklist?
The IP address added to the white/black list is checked against the sender's IP address, but there's no check for partial IPs. I have removed the "XXX.XX." from the white list and added 4 complete IPs. So, the answer to this issue is that the IP address must be complete.
As an improvement, we may look at a more recent version of the check function that also checks for partial IP addresses. Or another version that allows white/blacklisting per TLD
However this improvement should also be done with the AWL management, so that the user is able to watch the content of the AWL and remove email addresses from it.
Ok perfect. I’ll keep an eye to see how it goes now.