Rate Limits


We enforce rate limits on mail; those limits are applied to the following metrics:

  • Recipients per message (100 maximum)
  • Recipients per hour
  • Recipients per connection
  • Simultaneous or conncurrent connections

Note our use of the word recipients, not messages, when referencing rate limits. One message to one hundred recipients or one hundred messages each to one recipient will both count as one hundred recipients toward the sending IP's limit.

Other than a system-wide limit on recipients per message, the rates we impose on an IP will vary based on several criteria, including:

  • Past sending history - IPs with a history of sending wanted mail will enjoy higher rate limits than will either IPs with a history of sending mail that draws complaints and/or triggers our spam filter or IPs that have no recent history of sending any mail to our customers.
  • Full Circle Reverse DNS (FCrDNS) - IPs that do not have FCrDNS (that is, the IP resolves to a name that does not resolve back to the IP) will be severely limited in their ability to get mail to Road Runner customers.
  • FBL Enrollment - IPs enrolled in our FBL will typically enjoy more generous rate limits than those that are not enrolled in our FBL, although such IPs can still end up blocked here for bad behavior.

As discussed above, we impose rate limits on inbound mail, and servers that exceed those rate limits will receive error messages in response to their attempts to send mail inbound to our customers. We try to make the messages self-explanatory to match the rate limit condition that's being violated:

  • 421 4.7.1 - Too many recipients for this message
  • 452 4.7.1 - ERROR: Mail refused - IP.ADD.RE.SS- Too many recipients received this hour.
  • 452 4.7.1 - ERROR: Mail refused - IP.ADD.RE.SS - Exceeded limit for recipients per connection
  • 421 4.7.1 - Connection refused - IP.ADD.RE.SS - Too many concurrent connections from source IP