More email delays?

[#1]

keerto1 and rogerloxton,
Neither of your emails were delayed on our platform.
Everybody else, thanks for the examples, I'll have them looked into immediately. I know the work referenced in the most recent Service Status is due to take place today so I'm wondering if we'll see anything after that?
We sent around 90,000 emails ourselves yesterday to customers but I doubt that would have been phased and it doesn't coincide with the peak on the graphs.
I wouldn't normally expect significant widespread delays unless the queues across all the cores exceeded 12k or so, certainly not 6hrs+
More news as I have it.
Edit: Checking the graphs again the same thing happened the night before last. In fact the rise does coincide almost exactly with when the ACL blocking was backed out. Granted the queues weren't that high before implementation but we've mentioned elsewhere that the volumes of spam have started to ramp up recently.

[#2]

Quote from: Heloman

I received a mail,  posted at 2239 on the 24th, at approx. 2000 on the 25th. A 22hour delay.

Heloman, that email doesn't look delayed to me.

[#3]

Here's one - a  PlusNet support notification
Just an hour and a half....

Quote

Envelope-to: xxxx@xxxxx.f9.co.uk Delivery-date: Thu, 27 Sep 2007 11:50:31 +0100 Received: from ptb-relay03.plus.net ([212.159.14.214])   by pih-sunmxcore10.plus.net with esmtp (PlusNet MXCore v2.00) id 1IapVY-0001Oh-7G   for xxxx@xxxxx.f9.co.uk; Thu, 27 Sep 2007 10:18:08 +0100

[#4]

A couple of Community site notifications delayed yesterday.

X-Daemon-Classification: INNOCENT
Envelope-to: ********@********
Delivery-date: Thu, 27 Sep 2007 01:36:53 +0100       10 hours
Received: from fhw-relay07.plus.net ([212.159.14.215])
	  by pih-sunmxcore11.plus.net with esmtp (PlusNet MXCore v2.00) id 1IaYCX-00039L-Al 
	  for ********@********; Wed, 26 Sep 2007 15:49:21 +0100
Received: from [192.168.101.40] (helo=pih-community01.plus.net)
	 by fhw-relay07.plus.net with esmtp (Exim) id 1IaYCV-00061w-IZ
	for ********@********; Wed, 26 Sep 2007 15:49:19 +0100
Received: from www-data by pih-community01.plus.net with local (Exim 4.63)
	(envelope-from <www-data@pih-community01.plus.net>)
	id 1IaYCV-0006t3-FB
	for ********@********; Wed, 26 Sep 2007 15:49:19 +0100

X-Daemon-Classification: INNOCENT
Envelope-to: ********@********
Delivery-date: Thu, 27 Sep 2007 00:36:36 +0100       4 hours
Received: from ptb-relay02.plus.net ([212.159.14.213])
	  by pih-sunmxcore12.plus.net with esmtp (PlusNet MXCore v2.00) id 1IadL3-0007bP-Sg 
	  for ********@********; Wed, 26 Sep 2007 21:18:30 +0100
Received: from [192.168.101.42] (helo=fhw-community03.plus.net)
	 by ptb-relay02.plus.net with esmtp (Exim) id 1IadL3-0008NY-Ah
	for ********@********; Wed, 26 Sep 2007 21:18:29 +0100
Received: from www-data by fhw-community03.plus.net with local (Exim 4.63)
	(envelope-from <www-data@fhw-community03.plus.net>)
	id 1IadL3-0001vn-7i
	for ********@********; Wed, 26 Sep 2007 21:18:29 +0100

I received a Usergroup message and an external ones during the same period without dealys.

[#5]

Hi guys,
Thanks for the examples. It's clear that this mornings workaround hasn't stopped the occurrences of these delays (although our reporting suggests it has helped reduce the frequency)
We've looked into the examples we've received and reckon we might have found the root cause. One of the Clam virus filters performs an update every 10-20 minutes or so. When this happens it seems that something is locking up.
There's an Exim script that reads files generated by the virus filter. For a minute or so each time Clam performs a database update the Exim script errors saying it can't find the cirus filter file it's looking for. This file exists though as Clam itself doesn't return any errors to suggest it's had a problem writing it.
We're going to look at scheduling a period of maintenance next week to update the version of Clam in use. If this does not resolve the issue then we may reduce the frequency of the database updates and work from there. I'll update Service Status shortly.
It might be an idea to keep posting examples of delays to ensure that we're on the right track throughout.

[#6]

This remains a problem for me and I suspect it is more serious than people think.
I know I am missing at least 3 emails already today.  Some from 2 days ago took 10 hours to arrive.  Not much cop when you're trying to run a business and are paying £31.71 a month to PN for connectivity and email.
I just ran a test - sent myself 6 emails, 2 of them delayed.  That's quite a percentage.
Because of queries from clients as to why I hadn't responded to their emails (these are on top of the ones from the last few days when people rang up asking why their emails had been bounced back to them) I am starting to look like a complete idiot and my £31.71 isn't looking like money well spent!

[#7]

Quote from: wuggywugrat

I just ran a test - sent myself 6 emails, 2 of them delayed.  That's quite a percentage.

Can you PM me the headers of the delayed emails please?

[#8]

Hi - I'm not sure if this is still an ongoing issue with everyone but it is for me. Can you let me know if this is part of plusnet or qmail please.
Thanks,
K
X-Daemon-Classification: INNOCENT
Envelope-To: *******@*******.plus.com
Delivery-Date: Thu, 04 Oct 2007 18:47:50 +0100
Received: from smtp-pri.interdns.co.uk ([212.241.249.218]) by pih-sunmxcore10.plus.net with esmtp (PlusNet MXCore v2.00) id 1IdUne-0007iO-GC  for *******@*******.plus.com; Thu, 04 Oct 2007 18:47:50 +0100
Received: (qmail 70163 invoked by uid 68); 4 Oct 2007 17:35:38 -0000
Received: (qmail 53993 invoked from network); 4 Oct 2007 17:32:25 -0000
Received: from unknown (HELO smtp-sec.interdns.co.uk) (62.105.76.12) by smtp-pri.interdns.co.uk with SMTP; 4 Oct 2007 17:32:25 -0000
Received: (qmail 975 invoked from network); 4 Oct 2007 17:27:48 -0000
Received: from unknown (HELO mq3.ifl.net) (213.18.248.110) by smtp-sec.interdns.co.uk with SMTP; 4 Oct 2007 17:27:48 -0000
Received: from mx1.segfl.ifl.net (mx1.segfl.ifl.net [62.171.198.108]) by mq3.ifl.net (8.13.8/8.13.8) with ESMTP id l94HRHXa005214 for <*******@*******.co.uk>; Thu, 4 Oct 2007 17:27:17 GMT
Received: from sme-server.sme-admin.local ([10.137.15.1]) by mx1.segfl.ifl.net (8.13.7/8.13.7) with ESMTP id l94HREXv016477 for <*******@*******.co.uk>; Thu, 4 Oct 2007 17:27:15 GMT

[#9]

Hi Katw and welcome to the community forums.
That email was received by our servers at 1847 and delivered almost immediately.

[#10]

Thanks Mark, and I thought that might be the case - any ideas on how I can get this resolved at all.......what's qmail?
Thanks,
K

[#11]

I think the delay was with the servers between the source and PlusNet, so there's not much really you could do AFAIK (Mark will know better though).

[#12]

Quote from: Katw

Received: from unknown (HELO smtp-sec.interdns.co.uk) (62.105.76.12) by smtp-pri.interdns.co.uk with SMTP; 4 Oct 2007 17:32:25 -0000

Would appear to me that this email was delayed about 20 minutes on the server smtp-pri.interdns.co.uk.
That seems to be owned by this outfit.
Edit: Just seen this:
"As you may already be aware there are severe mail delays on our mail clusters. This is primary due to the level of messages being almost double the normal amount primary due to spam deluges.
We have been witnessing similar behaviour with our spam clusters over the past week.
We sincerely apologise for the delays you have had to experience and be rest assured everything that can be done is being done to clear the backlogs.
Looking forward we are bring forward our spending plans to increase the scale of our mail clusters to better deal with issues such as that today. These will be heavily over engineered to ensure that even the heaviest of unexpected mail delivery can be catered for."

[#13]

Thanks Bob, appreciate you finding that 
K

[#14]

[OT]
<rant>

Quote from: Bob

this outfit. Edit: Just seen this: "As you may already be aware there are severe mail delays on our mail clusters. This is primary due to the level of messages being almost double the normal amount primary due to spam deluges. We have been witnessing similar behaviour with our spam clusters over the past week. We sincerely apologise for the delays you have had to experience and be rest assured everything that can be done is being done to clear the backlogs. Looking forward we are bring forward our spending plans to increase the scale of our mail clusters to better deal with issues such as that today. These will be heavily over engineered to ensure that even the heaviest of unexpected mail delivery can be catered for."

That wasn't written in an English speaking country - sounds more like the instructions that came with my ancient video.
</rant>

[#15]

I appear to be experience extreme email delays at the moment (down from 100 per day to about 10 today and nothing since 1pm, anyone else experienceing problems?