ExchangeDefender vs. Hacked Exim/cPanel servers
ExchangeDefender has been seeing an elevated amount of malware originating from hacked Exim mail servers. While we tend to score those messages higher by default to keep our clients protected, one of our clients discovered a scenario in which a user could get a dangerous payload through our scanners (requires multiple manual steps and a sophisticated recipient with imaging software willing to go through multiple hoops). Which this is exceptionally unlikely, we wanted to address a few of the topics anyhow.
1. CIOs, MSPs, and Domain Administrators can manage attachment policies
If you go to https://admin.exchangedefender.com and login as the Domain Administrator, you can manage attachment policies under Configuration > Attachments. You can find more about ExchangeDefender configuration at https://www.exchangedefender.com/docs/domain#configuration
2. We do not deep-scan file system images (.iso/.img)
As a policy we do not deep scan .iso or .img file system images. The files themselves are scanned for both malware, viruses, and other parameters (for example, if someone renames a .exe to .img, or embeds malware in one we will still filter it out) but we will not mount file system images and go through each file inside. This is not a popular attack vector (requires multiple actions by the user and most will require Administrative access and specialized software) but it is technically possible.
3. You should implicitly distrust anyone on hacked Exim servers
ExchangeDefender cannot globally block Exim servers (because there is always going to be that one “business case scenario!!! we cannot block our $2 cPanel VPS!”) but if you can possibly block them – by all means do. While this is generally not necessary (ExchangeDefender maintains a proprietary list of pwn3d Exim servers and routinely moves them to SPAM or SureSPAM), it’s a good idea not to accept any mail from these servers at all.
4. You should implicitly junk SPF failures
Same as #3, it’s a really good idea if you have the luxury of not dealing with people that shouldn’t be running an email server. ExchangeDefender tracks SPF failures and notes them in the headers that can be used to aggressively filter out messages sent out from invalid ranges. Just look for a “Received-SPF: softfail” in the message headers.
Received-SPF: softfail (inboundXX.exchangedefender.com: domain of transitioning postmaster@gmail.com does not designate 67.82.55.11 as permitted sender)
What this means is that the organization has designated an IP range that legitimately relays messages, and this message came from an IP address outside of that range. 99.99999% of the time it’s a spammer. 0.00001% of the time it’s just a poorly configured server. It’s your choice to assess the risk and implement this if possible and we recommend it.
Finally, if you are actively monitoring security and communicating with your clients, we do manage a NOC site that logs major issues at https://www.anythingdown.com. If you’re one of our MSP or enterprise clients, you also have a branded version of this software free of charge at https://www.xdnoc.com that you can attach your domain name to and offer these alerts to your clients without copying and pasting around.We hope this helps and we appreciate your trust in keeping you safe online.
For more information, please see our ExchangeDefender Guide for Domain.
