How come CBL blocks my NON reply masking OUTGOING messages?

[lwc]

I rather fake my "from" address than slowing down the sending process by using reply masking (I do turn it on for any case, though).

Anyway, until today I happily faked lots of "from" address. But today I sent 2 messages - one of them was to a fellow Spamgourmet user - and got this:

After sending a message to a non Spamgourmet user:

----- Transcript of session follows -----
... while talking to *non_spamgourmet_server*.:
>>> DATA
<<< 554 5.7.1 Service unavailable; Client host [*myIP*] blocked using cbl.abuseat.org; Blocked - see http://cbl.abuseat.org/lookup.cgi?ip=*myIP*
554 5.0.0 Service unavailable
<<< 554 5.5.1 Error: no valid recipients

After sending a message to a fellow Spamgourmet user:

----- Transcript of session follows -----
... while talking to gourmet.spamgourmet.com.:
>>> MAIL From:<>
<<< 553 5.3.0 Message from *myIP* rejected **** see http://cbl.abuseat.org/lookup.cgi?ip= *myIP*
501 5.6.0 Data format error

Now, *myIP* happens to be a shared IP (well, it saves me $1 a month) so I don't think it's my fault.

BTW, what's with the space in "ip= *myIP*"?

Anyway, here's the catch - in the "What's New" section you admitted you started using this CBL service, but my messages are NON reply masking OUTGOING messages, so how come Spamgourmet was involved in both messages (it should have only been involved with the message to the fellow user)?! Is there a chance the other server also uses CBL? What are the odds to send messages for years without crossing paths with CBL and then cross paths with it twice in the span of 5 minutes...?

[SysKoll]

OK, that's really weird. You need to send me the CBL response with the complete headers (you can remove the quoted body of your message, but I want to see the CBL-related text).

Please PM these to user info or to me.

[josh]

Without knowing the details, it looks like your IP address was listed in CBL for some reason, and then both our server and another server picked it up in an update -- that is, the two messages you posted are from two different servers (the second of which is ours) that both use CBL. Not much of a coincidence.

As for the space, that's a sendmail config problem, and I can't figure out how to get rid of it. Not at the top of my list, but I do work on it from time to time.

When you say "shared" IP address, what do you mean? The only reason I ask is that we only use the CBL list that claims to include virus infected computers - not general spam-sending computers, and so somebody or something somwhere thought that a computer using that IP address had a virus, and so you might look into it to make sure your stuff is safe.

[lwc]

Well, when you buy hosting you must decide if you want a shared IP address or a constant ("dedicated") IP. In my host, a shared IP address is "free" (just part of the package), while a dedicated IP address is a special add-on that costs $1 a month for each site you want that feature in. And, unless you're a resller that has enough patience to use its ability to remove the whole site and re-create it with the other option, once you pay for it, you can't return to having a shared IP address.

So let's say we both bought hosting from the same host and chose to have a shared IP address. Then luck has it we happen to share the same IP address (although keep in mind it's dynamic, so partners are dynamic too). The result is that if someone pings either of our sites, they get the same IP address (and if they directly ping that IP address, they get the host's hostname).

So like everying that involves sharing, if one of us messes up (knowingly or not), the other suffers for it too. Partners are dynamic, but it's always hundreds of partners so it just takes one who messes up (in their current IP address).

In Spamgourmet terms, it's like users who use it for evil purposes and ruin things for the rest of us (like getting Spamgourmet reported to Spamcop).

[josh]

In Spamgourmet terms, it's like users who use it for evil purposes and ruin things for the rest of us (like getting Spamgourmet reported to Spamcop).

True, but for the record, it's not evil purposes that wind up getting spamgourmet reported to spamcop, but rather overzealous spam reporters (who false report without knowing).

[lwc]

Then I guess I mean those who run automated codes.

But Spamcop reports contain the full headers, so why would Spamgourmet be blamed? Its RECEIVED headers are never the first headers in line. I report to Spamcop myself, so how do I know I don't do something wrong?

[SysKoll]

I believe we were whitelisted with Spamcop, weren't we?