FAQ

I can't sign in! I get the message "You don't have an active account on the system".
I forgot my password. What can I do?
I see a message in my Marvin report that I want to read. How do I get it?
I have configured Marvin, but I am still getting spam and viruses. Why?
Since Marvin has anti-virus protection, do I still need anti-virus software on my computer?
What changes need to be made for a domain to be processed by Marvin?
These MX records look strange, how does Marvin use them?
If the first two MX records are unresolvable, doesn't that cause problems for people sending mail to our domain?
How do spammers find e-mail addresses?
How does Marvin identify spam?
What happens to messages that are identified as spam?
What is the purpose of the "delay" feature?

Q. I can't sign in! I get the message "You don't have an active account on the system".
A. Before you can configure Marvin, you need to register your e-mail address with Marvin. Click on the First Time User link under the sign in boxes (or here) and enter in the e-mail address you want Marvin to protect. A password will be sent to the address you entered. Use that password to sign into Marvin.

Another possibility is that your domain administrator has not allowed any other people to sign in. Contact your domain administrator to turn it on or confirm this is the case.

If your domain has not been registered with Marvin, you will get this error. If you have sent in the e-mail listing the domain, domain administrator and e-mail addresses, we will send you a confirmation that everything has been set up and is ready to go. If you have not gotten this message, then you will not be able to sign into Marvin yet.

Back to Questions

Q. I forgot my password. What can I do?
A. Click on the Forgot Password under the sign in boxes (or here) and enter in your e-mail address that you have Marvin protecting. A new password will be e-mailed to your inbox.

Back to Questions

Q. I see a message in my Marvin report that I want to read. How do I get it?
A. First sign into Marvin. Click on the Retrieve link. Messages from your last report are listed. Find the message you want and check the recall box and then click the recall selected messages button. The message is sent directly to your inbox.

Remember that if the message is on the domain level, click the retrieve link under the domain name to find the correct message.

Back to Questions

Q. I have configured Marvin, but I am still getting spam and viruses. Why?
A. If you have multiple e-mail addresses going to the same mailbox, you will need to configure each one or remove the forwarding of those e-mail addresses. Let's say you receive mail addresses to you.name@company.com. You also get e-mail for sales@company.com. If you just configured you.name@company.com in Marvin, spam sent to sales@company.com would still go through since there are no configuration settings for that e-mail address. You would need to configure Marvin for that e-mail address also.

Back to Questions

Q. Since Marvin has anti-virus protection, do I still need anti-virus software on my computer?
A. YES. The biggest reason is because e-mail isn't the only way viruses spread. You can get a virus by going to a webpage, from a floppy or CD-ROM, downloading software from the Internet, or over your corporate network. By having your own anti-virus software installed, you can prevent these other types of infection, and it gives you a second layer of protection from e-mail viruses.

Back to Questions

Q. What changes need to be made for a domain to be processed by Marvin?
A. The MX records have to be changed. In BIND, they would look like this:

IN MX 10 yourdomain.com.mx.marvin.sihope.com.
IN MX 20 queue.mx.marvin.sihope.com.
IN MX 30 marvin.sihope.com.
IN MX 40 actual-mail-server.yourdomain.com.
IN MX 50 backup-mail-servers.yourdomain.com.

Back to Questions

Q. These MX records look strange, how does Marvin use them?
A. The first two addresses listed (yourdomain.com.mx.marvin.sihope.com and
queue.mx.marvin.sihope.com) can only be resolved to IP addresses by the Marvin servers. When another mail server on the Internet tries to send mail to your domain, it will not be able to reach the first or second server, so it will fall back to the third (marvin.sihope.com), which can be resolved, and mail will be sent there.

After Marvin processes the mail, it looks up the MX records for your domain to see where the mail should be delivered. When you sign up with Marvin, you provide the IP addresses of your mail servers, so Marvin knows where to deliver your mail. Since the Marvin servers can resolve the first two addresses, they will send the mail to your mailserver. If your mailserver is unreachable for some reason, Marvin will queue the mail and try to redeliver the mail at another time.

Back to Questions

Q. If the first two MX records are unresolvable, doesn't that cause problems for people trying to send mail to our domain?
A. Typically not. The SMTP specification in RFC 2821 says, "To provide reliable mail transmission, the SMTP client MUST be able to try (and retry) each of the relevant addresses in this list in order, until a delivery attempt succeeds." Since much of the software spammers use is poorly written, and is designed for speed and not reliable mail delivery, this MX configuration prevents a lot of spam from even getting to the Marvin servers. Most modern SMTP implementations correctly adhere to RFC 2821 and have no problems delivering mail to Marvin-enabled domains.

Back to Questions

Q. How do spammers find e-mail addresses?
A. The most common ways spammers get lists of addresses is through a process known as harvesting. Spammers employ software that gathers addresses by scanning websites, mailing list archives, news groups, domain records, and any other public sources. Another technique that is becoming more prevalent is known as a dictionary attack. In a dictionary attack, the spammer's software sends the messages to e-mail addresses that it "guesses" on the fly, sometimes listing 50 or more recipients at a time.

This doesn't require any lists of addresses, and your e-mail address doesn't have to appear anywhere on the Internet to be found in this manner. These kinds of attacks can last for hours, and can cause excessive load, or even crash a mail server.

Back to Questions

Q. How does Marvin identify spam?
A. Marvin is really a framework made up of different modules. Marvin itself handles passing messages between modules, logging, and other details.

The modules are what actually filter the e-mail. Some of the modules are public, third-party anti-spam tools such as Spam Assassin and Vipul's Razor. Other modules, like the Whitelist and Blacklist, are custom made.

This modular design allows new modules to be added in a very short amount of time (sometimes in less than an hour). This flexibility means that Marvin can continue to be effective, even as spammers find ways around the current systems.

Back to Questions

Q. What happens to messages that are identified as spam?
A. They go to e-mail purgatory. Actually, messages that are identified as spam are added to your rejected e-mail report, and put into a holding area. If a message that you want is rejected, you can go to the Marvin website and retrieve it up to 60 days later.

Back to Questions

Q. What is the purpose of the "delay" feature?
A. By delaying your messages by a few minutes (30 minutes is the smallest interval that can be effective), you increase the chance that someone else will receive the spam and report it to one of the spam tools that Marvin uses. This is most effective with the DNS-based blacklists, since they rely on recipients to report spam. Also note that because your whitelist is processed before the delay, any messages matching your whitelist are delivered immediately, so the delay has no effect on them.

Back to Questions