Another potential cause of this behavior is the built-in maximum connection limit in non-server versions of Windows. Typically we recommend deploying SmarterTools products in Server 2008 R2 at least, but recommend Windows 2012 Server. There may be a way to disable the concurrent connection limits in Windows 10 Professional, but I'd recommend testing these instructions in a non-production environment first.

For my understanding the connection limit of a Windows client OS is regarding incomming connections, isn't it? And the POP3 retrieval is an outgoing connection to my mail provider.

For more clarification: I don't use any mail client software to work with SmarterMail. I'm only using the web UI, which is really great. In my home network there aren't many concurrent connections.

While the POP3 retrieval isn't working all other connections are available: SMTP from internal devices to SmarterMail and from SmarterMail to my mail Provider and the access to the web UI.

I haven't found a detailed POP log. There's only a log file regarding POP when the SmarterMail service is restarted. So I'm not able to troubleshot this issue.

 

After restarting the SmarterMail service yesterday in the evening, POP3 retrieval was working again until today in the morning. This time the working period was very short.

Now I found the setting for a detailed log of SmarterMail. And I will monitor the TCP connections when POP3 retrieval won't work anymore. So hopefully there will be a hint for further troubleshooting.

 

Today I saw that the POP3s connection (TCP port 995) was still active. It didn't terminate. But after closing this connection SmarterMail still wasn't able to do POP3 retrievals. I had to restart the SmarterMail Service and then the retrieval was working again.

 

I have to say that the Windows 10 machine is only running SmarterMail and FreeFileSync for backups. So there are no other incoming connections than web mail access and SMTP for local devices.

Hello all,

 

today I saw that the message retrieval wasn't working again. The POP retrieval log showed only "Starting processing for ...".

 

Looking at the opened TCP ports I could see that SmarterMail didn't close a POP3s connection to the mail provider (in the last both cases it was Microsoft). In the POP retrieval log you can see:

 

18:37:32 [...] [...] Connection to 65.55.162.199 succeeded
18:37:32 [...] [...] CMD: USER <...>
18:37:33 [...] [...] CMD: PASS XXXX
18:37:37 [...] [...] CMD: STAT

 

So there's no QUIT entry in the log. After closing this TCP connection SmarterMail still isn't able to do POP3 retrievals. I have to restart the SmarterMail service and then the retrieval is working again.

 

I don't believe there's an issue related to Windows 10 because the TCP connection is established. But it wouldn't be closed sometimes. I can also say that there's no relationship to the hourly backup because this happens to full hour times, i. e. 17:00, 18:00 and so on.

 

Who should close the connection? Should SmarterMail close it? Why this doesn't happen sometimes?

 

Thank you.

Hello all,

 

for me it looks like SmarterMail isn't sometimes sending the QUIT command for closing the POP3(s) connection. Shouldn't there be a timeout for a open session which then will be closed by a QUIT command?

 

This seems to happen with the POP3s connection to Microsoft only. With GMX as my second mail provider I don't have this issue.

 

The protocol settings in my SmarterMail installation are the default settings so there is a timeout activated for the POP protocol. But does this timeout also belong to POP retrieval of SmarterMail itself (when it's retrieving mails from another mail server) or is this setting only related to POP retrievals from mail clients connecting to the SmarterMail server?

 

Thank You.

Hello all,

 

does nobody know this issue or can help troubleshooting?

 

Kind Regards.

Wish I could be of more assistance but here's what I think: it seems to be either that connection limit problem or a bug. Oh yes, and make sure your Windows firewall is turned off, and, at least as a test, completely uninstall all anti-virus programs (don't just disable them!).

 

Next comment: you can download a trial of Windows server, and it will run for 6 months, after which time you can extend the trial ("rearm" command - look it up). You can also buy fully licensed, genuine sealed OEM versions of Windows server on places like ebay for under $ 200.00.

Hi,

 

I user SmarterMail (latest 15 version) on Server 2008 (not R2) for several years, and since 2 weeks I have the same issue. The pop retrieval process didn't close the connection for only one user. After theis happens, the Mailserver process uses 100% cpu. If I stop and restart the pop retrival service inside smartermail, nothing changes.Only workaround is the restart of the whole SmarterMail windows service.

Even lower the pop3 timeount (for the incomming connections?) did't change the issue.

 

Hint: The failing user has more than 1.000 mail in his inbox. Thats the only difference compared to the oter users. Pop retrieval puts the mails in the inbox ... provider is the same for all users. Uses Unix/Linux based mailserver for millions of accounts.

 

Hope this helps to find the issue.

 

Best regaards

 

Lars

 

 

 

"Mail Server," in your case it sounds as if SM is choking on that large mailbox. One fix for this might be to move all, or most of, the already-retrieved emails to another mailbox (under that same user) in SM. That would make SM not need to "traverse" all those other emails when it goes to retrieve.

 

If all those emails are on the mail server, because all clients for that user "leave mail on server," and that is causing the problem, then you probably need to clean out or move the emails on the mail server. And yes, in either event it sounds like bug in SM, which should not choke on these.

 

Confusing! Possible glitch in SM user account? What happens if you create another SM user and try the same POP retrieval process from the original user into the new SM account?

Perplexing still. Perhaps there is some glitch in the user's settings (or a bad file) on the mail server you're retrieving from (creating an incompatibility with the SM process?). At this point it's the only other thing I can think of.

 

Dear Paul,

 

since my 2008 server go end of support, I setup a new one 2016 Standard. Copy the xml config files out of the programm tree and robocopy the mail store to the new server.

 

Same behaviour.

 

Only possibility to stay online is Service Protector ,,,

 

Do you have any hint how to debug? Log is set to Detailed for Pop-Retrieval.

 

Regards

 

Lars