More email delays?

[#1]

@tdadyslexia
Looking at the headers your email has been delayed by 7 minutes between the cgirelays and the mxcores. Have you been experiencing delays longer than this?

[#2]

Yes

[#3]

Do you have the headers of the emails so we can look into it?

[#4]

I'd agree with Barry on this one. Looks like this again. I mentioned it in the last Service Status update and we're looking into what we can do to combat it. We're inclined to think DNS is playing a part as it sometimes happens when some of the mail servers have plenty of capacity to handle new connections

[#5]

Quote from: Chris

Do you have the headers of the emails so we can look into it?

No sorry been deleted. 

[#6]

I wonder if mail is being routed directly to pih-sunmxcorexx.plus.net, instead of using the mx.core.plus.net.  I'm assuming that the mx.core address is the load balancers.
For whatever reason, maybe mail is targetting specific mail servers, causing them to be more heavily loaded than others?
Either that, or the load balancers forgot what the word 'balance' means
B.

[#7]

Quote from: Barry

Either that, or the load balancers forgot what the word 'balance' means

I think that the load balancers think balancing eMail is to put it all in one eMail Server. 

[#8]

Quote from: Barry

I wonder if mail is being routed directly to pih-sunmxcorexx.plus.net, instead of using the mx.core.plus.net.  I'm assuming that the mx.core address is the load balancers.

That's always going to happen to a lesser or greater extent.
Heads up BTW that the rDNS blocking has been rolled back again. I've posted more detail to Service Status but there's a library on Solaris that the code in the script calls that acts strangely and has been causing forward/reverse DNS mismatches to be deferred again 
It may well be that we shift towards concentrating on the longer term stuff now assuming the other upgrades conducted this week can keep us above water.

[#9]

There seems a fair amount of technical expertise here which goes straight over my head. A delay of 10 minutes - 30 minutes even, I can live with but 10 hours - 24 hours even and the service isn't worthy of the name.
What problems can cause an e-mail to sit with JTN for almost 24 hours as the example below? There are many others!
Envelope-to: xxxxxxx@metronet.co.uk
Delivery-date: Fri, 12 Oct 2007 11:24:06 +0100
Received: from just-the-name.co.uk ([213.162.97.161] helo=mail.just-the-name.co.uk)
    by fhw-sunmxcore01.plus.net with esmtp (PlusNet MXCore v2.00) id 1IgHgc-0003Tr-FX
    for xxxxxx@xxxxx.co.uk; Fri, 12 Oct 2007 11:24:06 +0100
Received: from main.xxxxxxx.com (xxxxxxx.demon.co.uk
  by mail.just-the-name.co.uk (Postfix) with ESMTP id 28510405DCA
  for <xxxxxxx@xxxxxx.co.uk>; Thu, 11 Oct 2007 12:14:03 +0100 (BST)
Subject: Order
MIME-Version: 1.0
Content-Type: multipart/alternative;
  boundary="----_=_NextPart_001_01C80BF7.CC737348"
content-class: urn:content-classes:message
X-MimeOLE: Produced By Microsoft Exchange V6.0.6619.12
Date: Thu, 11 Oct 2007 12:13:49 +0100
Message-ID: <E75DD9B64D1F214292F8E675836DC5B7271150@main.xxxxxxx.com>
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
Thread-Topic: Order
Thread-Index: AcgL98xsyqVwGMq3QXOFw2t7IaJ/8g==
X-Priority: 1
Priority: Urgent
Importance: high

[#10]

Quote from: Amos

What problems can cause an e-mail to sit with JTN for almost 24 hours as the example below? There are many others!

For 'JTN' read 'uk2.net'. This has been happening for some weeks now and while some mail arrives as it should, large swathes are delayed for 24 hours and more, while in other cases delivery fails.
What is going on?
Thank you.

[#11]

Earlier this week one of my clients could not send any email out through relay.plus.net. The sent emails did not arrive and no errors reported.  The Broadband connection seemed fine with no problems.
After restarting the router after a 5min shutdown, emails could then be sent and arrived as expected.
Could this be a problem caused by one of your gateways as presumably the router logon connected us through another gateway.

[#12]

I think the router restart's a bit of a red herring. It's possible to send via relay irrespective of the gateway you're on.
Not sure what could have been causing this issue - It's a bit of a peculiar one. I'm assuming the emails can be seen in the 'Sent' folder of the customer's mail client?

[#13]

Well, I'm getting some pretty serious e-mail delays at the moment. (I always seem to be saying that... sigh.) Here's the situation:
On late Thursday afternoon, I ordered some music scores from Boosey & Hawkes. Unusually for online ordering companies, they asked for not only my e-mail address, but the e-mail address of the person the order was being delivered to - in this case my mother. As a result, a copy of the order confirmation e-mail was sent to both me and her. However, we both received them at vastly different times. My mother received her copy at 00:02 on Friday, whereas mine took until 20:00 - over 26 hours after it was sent - as you can see from the headers from my copy below:

Quote

From:  musicshop@boosey.com Subject: B&H Music Shop Order Confirmation - Order No: 65375 Date: 11 October 2007 17:40:40 BDT To:  <removed>, <removed> X-Daemon-Classification: INNOCENT Envelope-To: <removed> Delivery-Date: Fri, 12 Oct 2007 21:03:45 +0100 Received: from [212.134.91.200] (helo=www.booseywww.com) by fhw-sunmxcore07.plus.net with esmtp (PlusNet MXCore v2.00) id 1IgQjZ-0003QA-2J  for <removed>; Fri, 12 Oct 2007 21:03:45 +0100 Received: from mail pickup service by www.booseywww.com with Microsoft SMTPSVC; Thu, 11 Oct 2007 17:40:40 +0100 Thread-Index: AcgMJXWWpx6mJS3RQ0mX4xa5Of9WGQ== Thread-Topic: B&H Music Shop Order Confirmation - Order No: 65375 Message-Id: <019F3338129E4C3E9CAB031F2E60B220@booseywww.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit X-Mailer: Microsoft CDO for Windows 2000 Content-Class: urn:content-classes:message Importance: normal Priority: normal X-Mimeole: Produced By Microsoft MimeOLE V6.00.3790.4073 X-Originalarrivaltime: 11 Oct 2007 16:40:40.0627 (UTC) FILETIME=[7597E030:01C80C25] X-Pn-Virusfiltered: by PlusNet MXCore (v4.00) X-Dspam-Result: Innocent X-Dspam-Processed: Fri Oct 12 21:03:46 2007 X-Dspam-Confidence: 0.7555 X-Dspam-Improbability: 1 in 310 chance of being spam X-Dspam-Probability: 0.0000 X-Dspam-Factors: 27, assistance+or, 0.00124, Delivery-date*12+Oct, 0.00398, Delivery-date*Fri+12, 0.00595, X-OriginalArrivalTime*11+Oct, 0.00649, Product+Qty, 0.01000, Subject*Confirmation+Order, 0.01000, 4778, 0.01000, Thread-Topic*Order+Confirmation, 0.01000, X-MimeOLE*MimeOLE+V6.00.3790.4073, 0.99000, only)+or, 0.01000, X-MimeOLE*V6.00.3790.4073, 0.99000, Order+Details, 0.01094, queries+regarding, 0.01114, 50+Total, 0.01490, quote+this, 0.01944, any+queries, 0.01975, Qty, 0.02028, X-OriginalArrivalTime*11, 0.02193, (UK+only), 0.02339, order+number, 0.02524, to+cancel, 0.02528, (0)870, 0.02616, 44+(0)870, 0.02690, order+Order, 0.02794, Subject*Order+Confirmation, 0.02829, queries, 0.02867, X-Mailer*Windows+2000, 0.02907

But the problem doesn't end there. Wondering why there had been such a delay, I thought I'd forward a copy of my mum's e-mail to my account, complete with full headers, for further analysis. I did this at precisely 20:30 last night, and as of writing this at 13:22 on Saturday, that e-mail has not yet arrived in my PlusNet inbox. Bear in mind that both my and my mother's e-mail accounts are with PlusNet, so the forwarded e-mail from her to me should never have had to leave the PlusNet network.
What is going on?

[#14]

Quote

Received: from mail pickup service by www.booseywww.com with Microsoft SMTPSVC; Thu, 11 Oct 2007 17:40:40 +0100

Looks like that's where the email was sat. There's a possibility this is related to the problem with the mail exchangers being busy but I wouldn't expect stuff to be delayed to the extent you're reporting and it's very concerning that the email you forwarded hasn't turned up!
Edit: BTW I modified your post and changed the code tags to quote tags as it was breaking the forum layout.

[#15]

Exactly - whilst I would have originally put it down to the B&H servers being slow, the fact that the forwarded e-mail is also being seriously delayed is what made me think there's more to this than meets the eye. I'll report here when/if the forwarded e-mail turns up...
(And sorry about the code tags - everything looked fine in Safari; I'll bear in mind not to use code tags again.)