Outgoing email "time warp" problem

[intrigue]

our company's OUTgoing emails have been experiencing a strange problem. at random, some of the email we send out isn't getting delivered immediately, like they usually are. it doesn't happen every day, but when it does, several messages get lost in a "time warp". it can take HOURS for them to get through. no bounces and no errors, just an extreme delay. there's usually no way to know that there is even a problem until someone calls us to complain that they didn't get our mail. Inbound messages are ok. also, sending mail from one co-worker to another is also near instant.

also, whenever this "time warp" happens, all messages that get sent out at that time are affected. After the "time warp" ends, all NEW messages DO get delivered immediately. So what ends up happening is that NEW messages get received before OLD messages!

Like I said, the problem comes and goes. Hard to tell what the duration of the "time warp" is but it seems to start and stop on a whim. I've contacted support twice. Both times I've been told it's not a bluehost server problem. One theory was a bulk emailer had tied up the outgoing mail. another is that it's the receiving server causing the delay. I have a hard time believing either theory since mail is sent to several different servers but I have no technical way to dispute it. I checked the headers and I can't tell anything from it.

Anybody else experience this situation?
:confused:

[Early Out]

We've all come to expect email to be almost instantaneous, simply because it usually works that way. In effect, we've gotten spoiled. Depending upon how the mail servers are set up, however, there is nothing remarkable about a delay of several hours in the delivery of an email message.

If the sending server encounters any problem contacting the receiving server, regardless of which end the problem is on, the waiting messages just stay in the queue, and the sending server tries to make the connection again later. Again, depending upon how things are set up, the next connection attempt may not occur for a few hours. A final delivery "failure" may not occur for a day or two, after repeated connection attempts. As email has gotten more reliable, many server administrators have chosen to reduce the retry time.

[intrigue]

thanks for the insight. I guess it just comes down to a simple "communication" problem. it's just frustrating when this happens because it can happen without anyone's knowledge. For example: Imagine having a quick 3 or 4 email correspondence with a client. The first bunch goes through immediately. The last message gets "delayed". As the sender, you won't know it's not yet received. The receiver doesn't know it has already been sent. Because the first group went through quickly, neither has any reason to believe there's a server communication issue.

As an example, I sent a message out today at 12:55 pm and another at 2:04pm. The 2:04 message was received at 2:04. The 12:55 message came in at 2:42. So, clearly, the old message got stuck somewhere. What I'm still not understanding is, how did the 2:04 come in at 2:04 and not AFTER the 12:55 message's arrival at 2:42. Does the server use multiple queues?

[Early Out]

There are often multiple servers on each end of a conversation, so it shouldn't come as a surprise when messages sometimes arrive out of sequence. SenderA may have trouble connecting to ReceiverA, but not to ReceiverB, or SenderB may have trouble connecting to ReceiverB, but not to ReceiverA, and so on.

Mail I send to my Comcast account from my Bluehost account can go out through one of several different BH sending mail servers, and can come in on any one of many, many different Comcast mail servers. There's probably some load-balancing going on, but the net effect is that it appears to be random.

[intrigue]

definitely makes sense to me. I hear exactly what you are saying. And I totally understand that mail can sometimes get delayed. however, the part that I still don't get is: For a period of 1 or 2 hours, why did the mail not get delivered to aol.com AND optonline.net, among a handful of services we sent to? bluehost says their server sent it out and that the problem is elsewhere. OK, fine. but did ALL the other mail host services decide to stop talking to bluehost simultaneously? it just seems too coincidental that ALL the receivers we sent mail to failed to communicate in the exact SAME window of time. that's why I'm still suspecting the issue stems from the outgoing server. even if it is, and even if it could be verified (and I'm not sure it can) how could that be fixed? all I can say is, I'm glad the mail eventually does go through!