Emails in active recipients increases after executing delivernow

[Sergio Eric Oshiro]

Hello,

I have a problem only with 1 domain in the active recipients. This domain is increasing in the queue, sometimes when i execute the delivernow command while the active recipients # is increasing, nothing happens and sometimes it does.

When the emails start decreasing and reaches 0, after 1-2 minutes later the # of active recipients start to increase again until it reaches aprox. 250 emails.

This only happens with 1 domain and has been like that for the last 2 weeks and it continues.

What could be the reason of that?

Thanks for your help.

[Andreas Mueller]

Hello Eric,

an active recipient is a message that has been processed by the normal mailqueue (HAT, RAT, filters, policies, etc) and is now queued to be delivered. This also counts for messages where the appliance already attempted to delliver them, but was not able to yet, because of a connection error, soft bounce, or the like. When you issue the delivernow command from the CLI for a specific domain, the messages for this domain will be scheduled for instant delivery, and the system will try to send them to the next host. Such am message currently being delivered is not counted as an active recipient any longer, however, if delivery fails, will be put back in the queue, and will count as active recipient again. Thats why you sometimes see the number of them decreasing after a 'delivernow', and  building up after a while. My guess is that there is a bandwith issue with the specific domain where messages send via concurrent connection won't get trough all together, so they are timing out eventually. My suggestion when that happens again:

1. Use 'cli: hoststatus' for the domain and check if it is up or down

2. At the end of hoststatus, the latest error message is displayed, check that for any possible reason

3. Also at the end of hoststatus, the used IPs for the domain are displayed. Use  "CLI: grep IP mail_logs" to find any errors related to those destination IPs , i.e. timeouts, network errors, etc.

Hope that helps,

Andreas

[Sergio Eric Oshiro]

Hello Andreas,

Thanks for the info. I executed the 'hoststatus' and it show that the domain is 'up', but at the end i see this error:

Last 5XX Error:

    ----------

    501 Syntax error in parameters or arguments

    (at Tue Nov 29 18:23:50 2011 PET) IP: 74.6.140.31

    ----------

Then i searched that IP, and there are a lot of errors that are repeating, which are:

Fri Dec 16 09:43:08 2011 Info: Connection Error: DCID: 9841409 domain: 'domainwithproblems.com' IP: 74.6.140.31 port: 25 details: 421-'4.7.0 [TS01] Messages from 190.81.189.194 temporarily deferred due to user complaints - 4.16.55.1; see http://postmaster.yahoo.com/421-ts01.html' interface: 192.168.70.2 reason: unexpected SMTP response

Fri Dec 16 09:59:09 2011 Info: Connection Error: DCID: 9841685 domain: 'domainwithproblems.com' IP: 74.6.140.31 port: 25 details: [Errno 54] Connection reset by peer interface: 192.168.70.2 reason: network error

What is the result of this errors? How can i explain to the client if the error is from the client and not from the IronPort.

Thanks for your help.

[Andreas Mueller · Accepted solution]

Hello Eric,

I'd show the customer the error message with the deferrred error, as it clearly states what the problem is here:

Temporarily deferred due to user complaints - 4.16.55.1; see  http://postmaster.yahoo.com/421-ts01.html'

So this is obviously a connection to yahoo that got throttled. May be caused by spam the customer sends, but most likely it's a problem of sending too many message too fast, which most of the major mail service do not appreciate.So maybe a destination control for yahoo.com would at least ease this. We had some success with these settings:

Destination: yahoo.com

- 20 concurrent connections

- 5 messages per connection

- 20 recipients per 1 minute

Regards,

Andreas

[Sergio Eric Oshiro]

Hello Andreas

Thank you very much for your help.

It is clear now.

Thanks!