Plusnet Email servers giving random SMTP 550 errors for inbound emails

[#1]

---

@bobpullen wrote:
@cerij101, can you private message me the address you're sending from and to please when these failures occur?
Edit: can you also provide me with the date and time of a recent bounce so I can try and locate it in our mail logs?

---

Done, thanks.

[#2]

---

@spraxyt wrote:

Turning Spam checking off will have changed mail server connections from the avasin (mx.avasasin.plus.net) to the core (mx.core.plus.net) servers; do your logs show any error 452 instances to the core servers since that Spam setting was changed?

---

The Duocircle logs are per message rather than a flat file but I have been checking occasional messages since changing the PN spam setting and haven't seen any errors at all.

Am a bit reluctant to turn spam checking back on again to see if the problem reoccurs!!

[#3]

Since turning the Spam back on is likely to result in some inbound emails being rejected I can understand your reluctance to do that.

I have a collection of 291 "unknown sender domain" reject messages over period 11/02/2015 (when I started saving them) to 17/01/2017 (nearly 2 years). At this point the problem stopped until one instance on 21/09/2017 and two instances on 29.09.2017. None since, including from 7 mailings in Noverber 2017.These are service status notices which are not sent through relay.plus.net. Mostly they are to Plusnet/Force9/Free-Online addresses, but at least one was to a 123-reg hosted mailbox.

The irony is that the bounce messages are sent using SMTP to the mailbox on the domain the message says doesn't exist. Of course it's not the failing connection that sends these.

@bobpullen These messages are on the nosenderdomain folder.

[#4]

---

@cerij101 wrote:

---

@bobpullen wrote:
@cerij101, can you private message me the address you're sending from and to please when these failures occur?
Edit: can you also provide me with the date and time of a recent bounce so I can try and locate it in our mail logs?

---

Done, thanks.

---

Thanks.

I've had a look and I can see the corresponding logs at our side.

I'll admit, I was hoping the errors were all being generated by a particular group of servers, but alas that's not the case:

Whilst I've redacted the email address below, it's not malformed, so that's not where the problem lies:

"20171211 21:10:17.043 messages id=OVL1enCF1D9NPOVL3etJOy action=REJECT wf=p_smtp:0000:4:5 smtp=MAILFROM:550 ip=209.85.216.179 from=""[redacted]"" to="""" host = avasin18
"20171211 07:44:12.616 messages id=OIkxe7iBTVb5JOIkzekavC action=REJECT wf=p_smtp:0000:4:5 smtp=MAILFROM:550 ip=209.85.216.180 from=""[redacted]"" to="""" host = avasin12
"20171210 09:12:34.902 messages id=NxevekHTpD9NPNxexerSm5 action=REJECT wf=p_smtp:0000:4:5 smtp=MAILFROM:550 ip=209.85.216.179 from=""[redacted]"" to="""" host = avasin18

There are similar messages from pretty much all of the other inbound servers too, across both data centres.

From what I can establish, DNS for your domain checks out fine too.

If I trawl the logs back to the start of October for all instances of these error (regardless of sender address), I don't see any unusual trends, other than the usual weekly fluctuations (mid-week always sees a lot of them).

I probably need to defer to one of our network engineers at this point, although I can see the issue being a bit of a bugger to troubleshoot

Edit: @cerij101, forgot to mention, the earliest example I can see going back to the start of October is from the 21/10/17.

[#5]

---

Thanks.

I've had a look and I can see the corresponding logs at our side.

I'll admit, I was hoping the errors were all being generated by a particular group of servers, but alas that's not the case:

Whilst I've redacted the email address below, it's not malformed, so that's not where the problem lies:

"20171211 21:10:17.043 messages id=OVL1enCF1D9NPOVL3etJOy action=REJECT wf=p_smtp:0000:4:5 smtp=MAILFROM:550 ip=209.85.216.179 from=""[redacted]"" to="""" host = avasin18
"20171211 07:44:12.616 messages id=OIkxe7iBTVb5JOIkzekavC action=REJECT wf=p_smtp:0000:4:5 smtp=MAILFROM:550 ip=209.85.216.180 from=""[redacted]"" to="""" host = avasin12
"20171210 09:12:34.902 messages id=NxevekHTpD9NPNxexerSm5 action=REJECT wf=p_smtp:0000:4:5 smtp=MAILFROM:550 ip=209.85.216.179 from=""[redacted]"" to="""" host = avasin18

There are similar messages from pretty much all of the other inbound servers too, across both data centres.

From what I can establish, DNS for your domain checks out fine too.

If I trawl the logs back to the start of October for all instances of these error (regardless of sender address), I don't see any unusual trends, other than the usual weekly fluctuations (mid-week always sees a lot of them).

I probably need to defer to one of our network engineers at this point, although I can see the issue being a bit of a [-Censored-] to troubleshoot

Edit: @cerij101, forgot to mention, the earliest example I can see going back to the start of October is from the 21/10/17.

---

Hi Bob, many thanks for this, just checked my end and the 21st was indeed the first failure as far as I can tell. Shame it's not an easy one to solve, but appreciate any help you can give.

[#6]

@bobpullen To expand on message 18 above:

---

@spraxyt wrote:

I have a collection of 291 "unknown sender domain" reject messages over period 11/02/2015 (when I started saving them) to 17/01/2017 (nearly 2 years). At this point the problem stopped until one instance on 21/09/2017 and two instances on 29.09.2017. None since, including from 7 mailings in November 2017.These are service status notices which are not sent through relay.plus.net. Mostly they are to Plusnet/Force9/Free-Online addresses, but at least one was to a 123-reg hosted mailbox.

The irony is that the bounce messages are sent using SMTP to the mailbox on the domain the message says doesn't exist. Of course it's not the failing connection that sends these.  

---

In fact there were 16 of these messages rejected by various IPs in the mx0.123-reg.co.uk cluster over the period 10/07/2015 to 09/12/2016. The last date is shortly before the regular stream of rejects stopped in January 2017. The spread of dates is illustrated in the following image.

The reject message from the final instance on 09/12/2016 is

The main point here is that

• the Mail Sender Agent is the Usertools server, not the Plusnet relays
• the MX rejecting the messages because of a non-existent sender domain is not a Plusnet avasin

However the sender address is a valid Plusnet email address which has an MX record. In fact the Non Delivery Report is sent successfully via SMTP to that same address.

I think the topic https://community.plus.net/t5/Email/False-rejection-of-valid-sender-domain/m-p/1492200#M21693 is also relevant to this problem. That topic shows the relay.plus.net cluster can also be mislead.

[#7]

I am seeing a similar problem with a mail forwarder list that I maintain (not hosted by Plusnet). One sender on the list (not on Plusnet) is getting these bounce-back messages from one receiver who is on Plusnet. The problem doesn't occur with other senders and other receivers. Is this a problem with mx.avasin.plus.net or is the mail forwarder list doing something wrong?

Here is the bounce-back message (email details redacted):

- - - - - - - - - - - - - - - - - - - - -

From: Mail Delivery System <MAILER-DAEMON@eforward3b.registrar-servers.com>
To: SRS0+hXDn=EA=btinternet.com=********@eforward3b.registrar-servers.com
Sent: Friday, 5 January 2018, 16:24
Subject: Undelivered Mail Returned to Sender

This is the mail system at host eforward3b.registrar-servers.com.

I'm sorry to have to inform you that your message could not

be delivered to one or more recipients. It's attached below.

For further assistance, please send mail to postmaster.

If you do so, please include this problem report. You can

delete your own text from the attached returned message.

                  The mail system

<******@******.plus.com> (expanded from <*****@********.com>): host

    mx.avasin.plus.net[212.159.9.200] said: 550 #5.1.8 Domain of sender address

    <srs0+hxdn=ea=btinternet.com=********@eforward3b.registrar-servers.com>

    does not exist (in reply to MAIL FROM command)

[#8]

I am having the exact same problem with one of my contact who is with Totalise.co.uk it started a few month ago and didn't pay too much attention as it was sporadic but now most email sent seems to fail!

I understand that totalise.co.uk belongs to plusnet now? could it be related?

Edit: here is the error I get! I have redacted the emails

Reporting-MTA: dns; googlemail.com
Received-From-MTA: dns; xxx@p p m t.org
Arrival-Date: Sat, 13 Jan 2018 10:50:00 -0800 (PST)
X-Original-Message-ID: <56284e46-46c9-aae7-48f6-13c8da36b51a@ppmt.org>

Final-Recipient: rfc822; xxx@totalise.co.uk
Action: failed
Status: 5.0.0
Remote-MTA: dns; mx-ironport.core.plus.net. (212.159.9.200, the server for the
 domain totalise.co.uk.)
Diagnostic-Code: smtp; 550 #5.1.8 Domain of sender address <pxxx@ p p m t.org> does not exist
Last-Attempt-Date: Sat, 13 Jan 2018 10:50:07 -0800 (PST)

[#9]

@bobpullen
It seems to me that enough people have experienced this issue to justify some sort of detailed investigation - do you agree?

[#10]

Over the past two months I have seen regular examples of this occuring for messages sent to club members with plusnet email addresses from a site that we manage. Despite the domain having fully validated DNS the messages are rejected with 550 'Domain does not exist' messages. This does not appear to happen to every message as some are delivered, it is a lottery. Below is another example from today:

The mail system

<************@kornweg.plus.com>: host mx.avasin.plus.net[212.159.8.200] said: 550

    #5.1.8 Domain of sender address <************@theheatingoilclub.co.uk> does not

    exist (in reply to MAIL FROM command)

I can dig out further examples if they are useful.

This is causing a lot of frustration amongst club members with plusnet email addresses as they are not receiving confirmations of orders place, etc. and I would have expected the problem to have been identified and fixed when first reported.

[#11]

Your description implies this problem occurs randomly with messages sent to a single member, but not always to the same one. Does it also occur when a single message (eg a newsletter) is sent to several members?

If the latter applies does the message get through to any Plusnet addresses, or is there a full house of rejections?

[#12]

Hi spraxyt

You are correct that the problem occurs inconsistently with messages to any single member and has occured for multiple members with plus.net, plus.com, waitrose.com, madasafish.com email addresses since mid November 2017. This has happened both with individual responses, automated replies and newsletters. However there have only been very few 550 rejections for newsletters and NOT a full house when newsletters have been sent out. I don't have any hard evidence to support this but get the impression that the problem is intermittent at the Plusnet end and it may depend upon the timing of the emails rather than the individual addresses.

I hope this helps.

Keith

[#13]

You can add totalise.co.uk to the list.

It has happened again tonight to another totalise.co.uk email.

It is frustrating because you never know when it will fail and you end up loosing confidence. I end up sending a sms to confirm that they have received the email!

Is Plusnet doing anything to fix it? can they share any progress?

I am a plusnet customer but don't use their email...this definitely doesn't make me to want to!

[#14]

I'm amazed a company like Plusnet could let something as fundamental as this go unresolved.

I did menton @bobpullen but no response so far.

[#15]

Thanks for contacting us via Twitter @cerij101

We've raised this with our incident management team to investigate.

EDIT: For internal use, the reference number is: 105238