Duplicated emails in Inbox (POP).

[#1]

No you reject b4 I get that option

[#2]

@bobpullen 

Is there a global filter on @free.fr before considering a user's whitelist choices?

[#3]

Zen tech support say:

>There's nothing to say the email is being sent 2/3 times, there's only the one delivered message.
>Can plus net check to see how many received messages they are getting?
>Also I can try re-setting up the forwarding to see if that negates it.

David

[#4]

Oh dear, this is not great!

There might be / will only be one delivered message (it was successful) ... but how many delivery attempt records are there (I tried, but did not get a delivered message).

This has previously been seen to be the heart of the multiple receipt of forwarded emails - delivery attempts are successful, but not recorded as such ... AIUI due to a misunderstanding of the message transfer termination status.

The help you ask for requires a staff member.  In the meantime I will keep a closer watch on a message submission failure I see from time to time, but have no necessity to resolve.

[#5]

OK I pressed on with Zen support and today they sent this reply which confirms the fault is with your servers

Thanks for reaching out to us, I'm sorry to hear you are having trouble with duplicate emails. From what I can see on our own system, our mailcluster is sending these duplicates but this is due to the poor termination occurring from Plus net servers. I can provide some of our logs to show this. 

Mar 16, 2023 @ 10:01:37.844    development-bounces@qt-project.org    dp@perdrix52.plus.com    delivered    bastion02a.mail.zen.net.uk    1pckBb-00En2T-RX    1pckBb-00En2T-RX => dp@perdrix52.plus.com <david.partridge@perdrix.co.uk> F=<development-bounces@qt-project.org> R=dnslookup T=remote_smtp S=36523 H=mx.core.plus.net [212.159.9.1] K C="250- 30547 byte chunk, total 36523\\n250 OK id=1pckQT-000DB2-6G" DT=0s


Mar 16, 2023 @ 09:59:33.441     -      -     deferred    bastion04d.mail.zen.net.uk    1pckOS-004PZv-It    1pckOS-004PZv-It == dp@perdrix52.plus.com <david.partridge@perdrix.co.uk> R=dnslookup T=remote_smtp defer (-18) H=mx.core.plus.net [212.159.9.1] DT=0s: Remote host closed connection in response to end of data

The email above was deferred 5 times before successful receiving. Each time previous the same message occurred 'Remote host closed connection in response to end of data'

This is essentially the Plusnet server just disconnecting when we send data over. Our server is configured to attempt to send again if a connection is interrupted, from my knowledge this tends to be the standard configuration for most servers, if a sending does not complete satisfactorily then it will send again. Realistically the fault lies with Plusnet and how the servers they are using are terminating connections. This will need to be addressed by them.

---

So will you please escalate this to the relevant people to fix your servers ASAP - this is driving me nuts.

Regards

David

[#6]

I am a customer the same as yourself ... so they are not my servers.

Zen's response is a little economic with the truth - the transmission did not fail, it was successful, because if it was a failure, you would not have received ALL of their transmission attempts.

Closing the connection after the end of data does not seem to be an unreasonable action ... but given that eventually the transmission is successful, one has to conclude that one of the following must be true...

1. A different sending SMTP server at Zen is content with a (seems reasonably legitimate) close of the connection on end of data and deems the transmission to be successful ... as it evidently is because you get the message multiple times
2. A different receiving server within the Plusnet space hangs around holding the connection open after the end of data has been received successfully just hoping that the sending client does something useful - I have yet to find clear affirmation that such is required

@bobpullen - is this something you can assist with please?

[#7]

The "your" was of course addressed to plusnet.

The problem is dropping the connection early (before the end of communication protocol) isn't correct

D.

[#8]

@bobpullen Any chance you can get this issue fixed or a fix scheduled.

This is driving me up the wall to the extent I am seriously contemplating asking for a MAC - and saying don't even think about charging me an early termination fee

David

[#9]