HowardForums is actually smaller then FT with 13,708,041 posts, and city-data.com is just a little bit larger with 14,970,534 posts. (Flyertalk has 14,303,477 posts)

Now, let's think of Flyertalk in the context of our network. Flyertalk is the 5th or 6th largest forum we have here at the company. The largest being with http://www.honda-tech.com/ with 40,757,139 posts and http://www.corvetteforum.com/ with 34,056,868 posts. In total, we own over 100 forum sites of various sizes.

All of these forums send out a lot of emails, and when you send this many emails, there are a lot of problems that arise. One issue is simply processing that many emails. One larger problem is ISP's and email services noticing this large volume coming from our network. We occasionally get red-flagged by places like gmail for potentially sending out spam. This is simply due to the volume that we send.

When an issue arises, it can often be difficult to find the problem. It can be an issue with one of the servers, like it was the other day. It can be an issue with our network, which by necessity has its architecture modified continuously. It can be a problem with one of our dedicated outgoing mail servers. It can also be be a problem with ISP's temporary blocking us. Usually, it's a combination of two or more of these. For example, the other day VM1 couldn't send emails and 50k emails entered the delivery queue. When this was fixed, all of those emails were sent very quickly. This might have put a clog in one of the outgoing email servers. Maybe it didn't. I don't know. What I do know is that either the flood of emails from FT or a flood of emails from that mail server caused by the clog from FT caused a couple of ISP's to put a block on us. All of the emails that were blocked come back to us as failed and enter our differed queue. These sit in the queue and are periodically retried for a period of 72 hours (or something like that). This slowed the already clogged email server and caused additional issues.

From my understanding, the above issue was resolved shortly after the VM1 returned to sending emails. Also, from my understanding, another site of ours had a similar hiccup recently, which may or may not have affected FT.

I'm not completely sure if the current problems are because of this stuff, or because of a configuration change that was made on the servers to fix a different non-user facing email issue. Either way, it's apparent that there is still an issue, and we're addressing that issue.