can't send email

[markgroen]

I can't send email even with it set to port 26 as suggested. This just started today and it is with any domain that I have as a reseller on cadillac.

Any other suggestions?

[esl]

Is your SPF record setup correctly?

[markgroen]

It's the same as it has been for the past couple years and hostgator has not told me to change it.

Any other suggestions? Accounts with hosts other than hostgator do not have this problem.

[quietFinn]

What does "I can't send email" mean?
Do you get an error message?
Do the mails bounce back?

when I try to connect to the smtp server in cadillac I get this:

bambi:~# telnet cadillac.websitewelcome.com 26
Trying 67.19.28.98...
Connected to cadillac.websitewelcome.com.
Escape character is '^]'.
220-cadillac.websitewelcome.com ESMTP Exim 4.52 #1 Mon, 08 May 2006 05:55:47 -0500
220-We do not authorize the use of this system to transport unsolicited,
220 and/or bulk e-mail.

so at least the smtp server is running...

[markgroen]

The CP shows that it is running as well, any suggestions as to why sending email doesn't work with this web host and does with others?

Has hostgator got so large they are a target and it's time to find another host?

[gtgeorge]

All our domains that we host on GMC quite sending or receiving email on the same day they upgraded the kernel. We received email bounce errors for sending and receiving though.

[markgroen]

What does "I can't send email" mean?
I can receive email but not send. The FAQ suggested specifying port 26 in case it's an ISP thing, which I did to no avail.

Do you get an error message?
Yes, sent from me at mydomain.com: RCPT TO <markgroen@gmail.com> failed: Requested action not taken: mailbox unavailable.

Do the mails bounce back?
They don't get sent. Exim is running, get a 200 response etc. tho'....

My DNS report isn't as it was the last time I checked, have a ticket in asking about that and hoping some kernel upgrade borked things so they can just reboot to previous and start combing logs for what happened.

It's Monday morning here and will know soon if clients start seeing this behavior too, just started happening Sunday morning for about 4 domains out of 30 or so, weirdness that I was hoping would go away after the weekend but it's still happening.

[markgroen]

Apparently this is not going to get fixed, Hostgator insists that nothing is wrong, but has no suggestions as to what to do to fix it except for what is contained on the FAQ, which didn't work. <strike>Exim was restarted and failed four times this morning and that's the last I heard from them.</strike> Post edited: that was another cpanel I was in the midst of setting up somewhere else.

This happens using any email program I have here, on both Windows and Linux boxes.

I have accounts with other web site hosting providers, looks like they are going to get a few more domains and hostgator gets less as everything works fine over there for other domains.

Hostgator has become too large and it's too easy for them to just forget about someone like me that only has 30 or so domains with them, oh well, nice while it lasted but I need my email and using webmail all the time just wil not do.

Mark,

Please be aware that my latest update to your ticket did not insinuate that it was resolved: The ticket is still open and one of our level 2 technicians is currently investigating it as we speak.

We'll ensure your issue is resolved as promptly as possible and we do appreciate your patience throughout the ordeal.

Thank you for your time.

[markgroen]

txs, we're on the second day now of no email going out and heading into the third so it's not unusual that I should feel nothing is going to be fixed any time soon.

[chaloupe]

Does your issue has been resove with your email?

If not,

If you use your ISP SMTP mail server, you have to run on it on port 25.
If you use your server SMTP mail server, you have to use it on port 26.

Make sure you set My Outgoing server (SMTP) requires authentication.

Next, when you are trying to send, what error do you have? Have you tried with the webmail interface, outlook, thunderbird?

If we have more information we can maybe try to help you much more because you are going in the third day.

Best regards,

[markgroen]

With Evolution:
RCPT TO <anyaddress@anydomain.net> failed: Requested action not taken:
mailbox unavailable.

With Outlook Express:
An unknown error occurred. Subject 'test Win98', Account:
'test@markgroen.com', Server: 'mail.markgroen.com', Protocol: SMTP,
Port: 25, Secure(SSL): No, Error Number 0x800CCC08

The Windows error says that we are using the correct spelling for the
test account, the server is correct, the port is standard, No ssl
involved.

I have taken the router and firewalls out of the picture hoping that
there was a blockage there maybe, but to no avail. Also tried setting
the port to 26, and three different computers and operating systems -
WinXP, Win98, and Fedora.

Talking directly through CLI to the server returns correct 220 response etc. Using my ISP for a mail server is not an option.

[chaloupe]

With Outlook Express:
An unknown error occurred. Subject 'test Win98', Account:
'test@markgroen.com', Server: 'mail.markgroen.com', Protocol: SMTP,
Port: 25, Secure(SSL): No, Error Number 0x800CCC08

When you use your mail server('mail.markgroen.com'), you need to use port 26.

The Error Number 0x800CCC08 is an odd one. There is no description except INVALID_STATE.

If you use port 25 without the SMTP server of your ISP, that's why they can block you. For me I had to put my ISP SMTP server and change smtp port from 26 to 25 and it work.

Here's and exemple:
My domain is jbwseries.com

mail.jbwseries.com for my server smtp on port 26
relai.videotron.ca for my ISP smtp on port 25

If you say that the Webmail is working, it is really what you are doing for settings on your machine. If your ISP also block port 26, you will need to use their port 25 with your ISP smtp server. There is no way around it and you are not the first one that happen to. If you continu to say no and will not test with your ISP smtp, you will never know why it is not working. Try it you have nothing to loose at this point.

Best regards,

[markgroen]

Finally...

The last entry in our mail logs shows an email being sent from you to your gmail account,
2006-05-09 11:26:13 H=s0106000f3d01a09e.vn.shawcable.net ([192.168.0.100]) [24.82.141.xxx]:61467 I=[67.19.28.xx]:25 F=mark@xxxxxxxxx.com rejected RCPT me@gmail.com: s0106000f3d01a09e.vn.shawcable.net ([192.168.0.100]) [24.82.141.xxx]:12345 is currently not permitted to relay through this server. Perhaps you have not logged into the pop/imap server in the last 30 minutes or do not have SMTP Authentication turned on in your email client.

This is caused from not having SMTP authentication enabled through your email client. Please enable this and try sending again. Let us know if you are still having problems sending after making this change.

The rest of the answer appears to be - Sunday morning/Saturday night sometime, this started happening but only to some people using this account's mailserver.

Apparently, my ISP may or may not require authentication and there is no way from a user point of view, except for the cryptic error messages from the email software - as to whether or not SMTP authentication must be specified in an email client. If at some point the ISP decides to change this, in my case at least, they don't send out emails telling you to modify your email software settings.

The mail logs also plainly show what the problem is. Having that sent to me this morning, I was able to go to the machines that are using that IP, change the setting, and all is working once more.

Thanks for trying to help; the biggest help to me was being able to see the email log. Is there some way for us to see them and self-diagnose, or is this just available to support?

Update from support:

email logs are restricted to the root user only, through shell access. Resellers and shared clients cannot view this. Only if you have a dedicated server.

[markgroen]

Originally Posted by esl

Is your SPF record setup correctly?

Just a note for the archive, every once in a while go to dnsreport.com and check your report. Software and kernels get upgraded, no way around that and sometimes settings will change and it's a handy tool for determining if anything is out of order. Although at one time my report was clean, this incident made me check it again and indeed something was different from the last time I checked.

A ticket submitted to support resulted in a quick fix from hostgator for the offending test result from DNSreport.com, many thanks for that

[Serra]

Originally Posted by markgroen

Thanks for trying to help; the biggest help to me was being able to see the email log. Is there some way for us to see them and self-diagnose, or is this just available to support?

The mail logs are part of the system that support can only see because of their location. There is no real way to safely give users access to this info.

However, if you move to a dedicated server, you will have access to the logs and much more. I had a mail problem a while back and if I wasn't on a dedicated box, I would have not even noticed the problem for at least 24 hours. I found it 60 minutes after it started and I was able to make sure that there was a problem, not just a slow mail period.

Beyond problems, I have clients who call me and say things like, "A few days ago, I had an email that ________." (didn't work, was lost, bounced, had a virus). I have a system on my server that lets me track down every email sent or received in the last 60 days! Very helpful when looking for email problems. I can actually see every incoming and outgoing email!

Move to a dedicated server... it is much better.

[JAC]

Originally Posted by markgroen

With Outlook Express:
An unknown error occurred. Subject 'test Win98', Account:
'test@markgroen.com', Server: 'mail.markgroen.com', Protocol: SMTP,
Port: 25, Secure(SSL): No, Error Number 0x800CCC08

I had a client call me today that is getting this exact same error message. This started happening a few days ago for them all of a sudden. They have not made any changes to their outlook or any hardware changes, just one day all was fine now they get that message when trying to send emails. I have gone over their settings and I do not see anything configured incorrectly.

Any resolution to this would be very much appreciated.

[crxfisher]

Same error for me and some clients on the Tiburon server. When I try to manually connect to port 25 or 26, it fails.

[markgroen]

toggle between "server requires authentication" worked for me, hope that helps.

[chaloupe]

The error is cause by Outlook Express and it is not fixeable,.,,,.

http://www.modemhelp.net/oeerrors/oe.shtml

It is not a server error, it is an Outlook express error. So if you see error 0X800CCC08, it is a bug from outlook and not the server. changing port can maybe work but it is maybe because it will get rid of the error of outlook for some times.

Best Regards,

[gtgeorge]

I too had clients getting this error on gmc suddenly and reported it as an Outlook error only to find that smtp authentication was not enabled on their end and had not been for almost a year for some. Server changes seem to bee requiring authentication now whereas users were getting by without it.