I know there's been many posts on the Pix firewall's SMTP fixup command. However, the vast majority seem to be dealing with issues related to Exchange servers. We use Netscape Messaging Server (3.6 in production, and I'm testing 4.15).
Right now, our mail server is outside the firewall. On my testing box, which is inside the firewall, I realized that the SMTP fixup command is preventing the use of SMTP's AUTH, since it restricts the session to straight SMTP, and not ESMTP. (And, no, this isn't one of those port 113 IDENT things.)
Our mail server requires that you either pass the AUTH test or match a predefined list of IP addresses that can relay. (My testing box is the same way, even though it's implemented differently. That's because Netscape Messaging Server 4.15 can do it at the protocol level, rather than at the post-SMTP receive stage. Much nicer!)
Here's my overall question -- What is the danger in turning off the SMTP fixup in our situation? I have two mail servers (both Netscape, not including my testing box) behind our firewall already, and I don't want to open them up to potential problems. I've seen several mentions of "not-so-nice" SMTP commands that would be permitted, but I'd like to know what impact they might have. Any thoughts would be greatly appreciated.
Thanks!
Right now, our mail server is outside the firewall. On my testing box, which is inside the firewall, I realized that the SMTP fixup command is preventing the use of SMTP's AUTH, since it restricts the session to straight SMTP, and not ESMTP. (And, no, this isn't one of those port 113 IDENT things.)
Our mail server requires that you either pass the AUTH test or match a predefined list of IP addresses that can relay. (My testing box is the same way, even though it's implemented differently. That's because Netscape Messaging Server 4.15 can do it at the protocol level, rather than at the post-SMTP receive stage. Much nicer!)
Here's my overall question -- What is the danger in turning off the SMTP fixup in our situation? I have two mail servers (both Netscape, not including my testing box) behind our firewall already, and I don't want to open them up to potential problems. I've seen several mentions of "not-so-nice" SMTP commands that would be permitted, but I'd like to know what impact they might have. Any thoughts would be greatly appreciated.
Thanks!