Email Propagation

[chernove]

I know that propagation can/does happen at different rates, but I am curious as to whether pop can propagate at a different rate than mail (i.e., smtp)? I am working on a site recently transfered to FQ and while email can be received, there is difficulty with the client in sending mail. I have doubled-checked all the usual suspects.

Any thoughts?

Thanks,
Eric

[sheila]

Each "host" will propagate independently.

At FutureQuest, here are the list of default "hosts" that are set up for your domain ( on full hosting packages ):

example.com
www.example.com
pop.example.com
mail.example.com
lists.example.com
mlm.example.com (for backwards compatibility)
ftp.example.com

and if you activate MySQL, then also
mysql.example.com

Each of the above propagates independently of the others, and so it is entirely possible that pop works whereas smtp (the mail. host) doesn't yet work.

The easiest way to check if it is propagation, is to open a command line window on your computer and type:

ping mail.example.com

See if it resolves to your IP address or not. If it doesn't, then propagation is not yet complete. If it does, then it is something else.

In the case of sending email, "something else" is often Port 25 blocking by the ISP. If propagation is complete, and the email client is configured with correct host, username, and password, then I would suggest trying Port 1025 as the outbound mail server (SMTP) and see if that resolves the issue.

[chernove]

Thanks for the speedy reply, Sheila!
I'll run the ping, as you suggest.

-Eric

PS: I checked the 1025 vs 25 issue as one of the usual suspects already. Not the issue here. I suspect it's basic propagation stuff . . . wish I had time for that fishing trip! :smile:

[chernove]

Oy. The mail ping DID resolve correctly. The OE error, FYI, is a socket error 10060 which, if I recall correctly, is a timeout error.

Any other thoughts? :dunno:

Thanks,
Eric

[sheila]

Ummm
Norton Internet Security?
Antivirus program?
Router?

To establish whether it is independent of the email program, in a command line window type:

telnet mail.example.com 25

Which will try to open a telnet connection to the mail server on the usual SMTP port.

If you are able to successfully do this, then the problem is somehow specific to the email program. If that also bombs, then it is your local network or ISP connection (most likely).

You could also try the above command with port 1025 to see if you get different results.

telnet mail.example.com 1025

[chernove]

Tried; both of them bombed. However, thinking earlier that this might have been the case, I made sure to check my own FQ-hosted email from my office as well (where I am right now), and it worked fine.

Another thought occured to me after my last post. I only checked the mail.domain.tld when pinging, but it occured to me that you gave me a whole list of "hosts." Among them was smtp.domain.tld which did NOT resolve (neither did several others on your list). So, is it possible that even though "mail" resolved and it is "mail.domain.tld" that is entered in OE that this is still a propagation issue?

Thanks,
Eric

[sheila]

Another thought occured to me after my last post. I only checked the mail.domain.tld when pinging, but it occured to me that you gave me a whole list of "hosts." Among them was smtp.domain.tld which did NOT resolve (neither did several others on your list). So, is it possible that even though "mail" resolved and it is "mail.domain.tld" that is entered in OE that this is still a propagation issue?
Ugg. The smtp.example.com host was a mistake. I have removed it from my post above and replaced it with the correct entry, which is mail.example.com.

So, smtp.example.com should not and is not expected to resolve.

Technically, the IP address for mail.example.com should either resolve or not. If you can ping it, and it resolves for you, then the same DNS server that you are using to do the ping should resolve it when you try to telnet or use your mail app. However, the fact that several of the other hosts are not resolving does sound like the domain is still in propagation.

Sometimes when a host is propagating it will work intermittently for a while. Sometimes work, short time later not work. Lather, rinse, repeat. So, I guess chalk this up to propagation, wait an additional 12 hours without visiting the domain in the meantime (as that can cause the old results ... or non-results, as the case may be... to be cached, delaying the completion of propagation for you locally).

I really have no explanation for this at this time, other than propagation oddities. Recommendation: Forget about it for 12 hours and do not even go peek or look, and see if after 12 hours it resolves itself.

[chernove]

:fishing:
Doctor's orders!
Okay, thanks Sheila (as always!).

Will let you know what happens.

-Eric