Vincent Lefevre wrote: > Nicolas George wrote: > > Vincent Lefevre a écrit : > > > /^Message-ID:\s+(<\S+>)( \(added by .*\))?$/i or next; > > > > I have never seen this "added by" in my mails, but assuming it is > > necessary for you, > > Yes, obviously. This came from some MTA's when the MUA didn't generate > a Message-ID. This lasted at least until 2005.
Messages arriving at border mail relays often arrive without a Message-Id header. When the message comes from a local submitter this is normal for mail user agents that don't add it themselves. It isn't good for messages arriving from the outside world and is almost always a spam sign that the message came from spam injection software. The spammer is trying to attack the system. Not having a message id is one of the symptoms. But from valid local submitters it is possibly normal. Unfortunately the Message-Id is an optional header field. One can't say by the RFCs that a message without the Message-Id is invalid and rejected solely upon that basis. The client may not have a good unique name or a good unique clock time and therefore making a globally unique id might be hard for leaf node clients that validly send mail. They leave it to the next mail relay to add. In practice mail from the outside Internet without a Message-Id header is very likely spam. But not always because some mail clients don't create one. Most mail transfer agents such as Postfix and others add a valid Message-Id if one is not present. I use this as a spam sign to add to SpamAssassin scoring. If the message comes in from the outside world and doesn't have a Message-Id then for me on my system's unique environments that is guarenteed to be spam. Can't say that for everyone else though. It depends. This is still the way things are both in 2005 and 2015 and as far as that goes for 1995 too that matter. Bob
signature.asc
Description: Digital signature
