I’d make sure to validate all assumptions, preferably with ONE message from ONE dedicated IP for the test, and make sure that everything does in fact line up the way you want it to. I don’t really know anything about that particular system, but … I’d want to make sure, as you do, that everything the report tells you is what actually happened.
Aloha, Michael. -- Michael J Wise Microsoft Corporation| Spam Analysis "Your Spam Specimen Has Been Processed." Open a ticket for Hotmail<http://go.microsoft.com/fwlink/?LinkID=614866> ? From: mailop <mailop-boun...@mailop.org> On Behalf Of Fernando MM via mailop Sent: Wednesday, February 24, 2021 4:54 AM To: mailop@mailop.org Subject: [EXTERNAL] [mailop] SNDS trap message period accuracy Hi, I'm trying to use the "Trap message period" information that is provided by SNDS in order to detect and remove users that might be sending spam. The IPs are dynamically assigned and some of them are assigned to multiple users at the same day. The issue that I'm experiencing is that, although the documentation says that this info is accurate to the minute, sometimes it fails because no users were using that IP at that minute. If I increase the search time range to 5 minutes, it usually works but sometimes I also find out that 2+ customers were assigned within that time range. I already checked everything that I could think of on our side: our IP assignment logs are correct/reliable, server's times are accurate, logs time zone is the same as SNDS etc. Does anyone have any experience with this? How accurate is the information provided in SNDS? We do check other factors but I'm worried that we could end up cancelling an user's account that isn't doing anything wrong based on unreliable data. Best regards, Fernando Marcelo
_______________________________________________ mailop mailing list mailop@mailop.org https://list.mailop.org/listinfo/mailop
