I find it interesting that you'd like pop3 access to a bunch of listed *webmail* providers. Who provide access via the web - NOT pop3. I also agree with the below statement - your mail queues themselves will provide far more accurate information. The issue of 'successful mail delivery' would be a support issue, not an operations issue, as long as you as an operations staffer can verify that your mailserver is not holding the mail for undue periods. (Mail doesnt have a delivery SLA in general, after all.) And if your support staff need offsite mail accounts to verify delivery delay times, they should be able to purchase them or otherwise obtain them, and keep a database of mail account access details that they can use *on demand*. This would of course entail them becoming customers and footing the regular monthly bill, or whatever. In every ISP i've ever worked for, all we've needed to do is verify that we're delivering the mail to the advertised MX in a timely fashion, and our responsibility ends at that point. Outside of the usual responsibilities of a good ISP - like providing responsive abuse/security staff, like providing a valid abuse@ contact :-) This would be one of the primary reasons your mail doesn't get accepted, and prevention is better than cure... So in all honesty - if your company wants to extend its monitoring-ability outside of its own network to those of other providers, it should do so in the 'usual fashion' - sign up. How is this something that yourself as an operations employee got involved in anyway, as it strikes me as a support issue? - Mark. On Tue, 14 Sep 2004, Hannigan, Martin wrote:

Ross,

There are a lot of knowledgeable people on this list [tier 3].

Why can't you already tell if you aren't getting through to major providers? Wouldn't your queues backup, or are you being blocked and the messages are being rejected and you are trying to track that?

-M<

-----Original Message----- From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu]On Behalf Of Hosman, Ross Sent: Tuesday, September 14, 2004 2:41 PM To: 'paul@adelphiacom.net'; Roy Cc: nanog@merit.edu Subject: RE: Email Complexes

Your right this isn't my department and it's not my place to tell them how to do their job. If Roy would like to send me a valid abuse complaint I'll make sure to forward it on or even walk it over to the abuse department supervisor.

I would also like to say I'm suprised at how many people have been attacking me on/off the list for asking a simple question. I'm just trying to ensure people are able to recieve/send mail from/to charter's mail complex. I didn't think asking for some help would be a big deal (as I thought there might be some people on the list from the companies that were knowledgeable instead of having to deal with typical tier-1 support).

-----Original Message----- From: Paul Bradford [mailto:paul@adelphiacom.net] Sent: Tuesday, September 14, 2004 1:31 PM To: Roy Cc: Hosman, Ross; nanog@merit.edu Subject: RE: Email Complexes

Come on... what is this.... Ross doesn't have the ability to put more "clueful" people in abuse, he's prolly an engineer like you and me.... we just want to fix the network.... why take this e-mail as a chance to bash Charter?

Ease up, Paul

On Tue, 2004-09-14 at 14:01, Roy wrote:

...

I suggest you concentrate some resources in your abuse department. One charter IP address hit my firewall 1617 times so far today. Repeated complaints to abuse@charter.net just get ignored.

According to the local newspaper, my fellow citizens consider Charter the worst company in town.

Roy

Why can't you already tell if you aren't getting through to major providers? Wouldn't your queues backup, or are you being blocked and the messages are being rejected and you are trying to track that?

It is all in the mail logs. Here is a quick hack to take a peak at your mail queue (for sendmail) & it should not be hard to get provider names that are backed up in your queue: #!/bin/sh # Do the Mqueues ls -d /var/spool/mqueue* | while read queue_name do total=`(cd $queue_name ; find . -name 'q*' -print | wc -l )` deferred=`(cd $queue_name ; find . -name 'q*' -exec grep MDeferred "{}" ";" | wc -l )` deferred=`echo "$deferred 2 / p " | dc` deferred_400=`(cd $queue_name ; find . -name 'q*' -exec egrep -e 'MDeferred: 4[0-9][0-9]' "{}" ";" | wc -l )` deferred_400=`echo "$deferred_400 2 / p " | dc` deferred_refused=`(cd $queue_name ; find . -name 'q*' -exec egrep -e 'MDeferred: Connection refused' "{}" ";" | wc -l )` deferred_refused=`echo "$deferred_refused 2 / p " | dc` deferred_reset=`(cd $queue_name ; find . -name 'q*' -exec egrep -e 'MDeferred: Connection reset' "{}" ";" | wc -l )` deferred_reset=`echo "$deferred_reset 2 / p " | dc` deferred_timeout=`(cd $queue_name ; find . -name 'q*' -exec egrep -e 'MDeferred: Connection timed out' "{}" ";" | wc -l )` deferred_timeout=`echo "$deferred_timeout 2 / p " | dc` echo "=========================================================================== " echo "Stats for: $queue_name" echo "Total messages in queue : $total" echo "Total messages in deferred: $deferred" echo "Total messages 400 error: $deferred_400" echo "Total messages conn. refused: $deferred_refused" echo "Total messages conn. reset: $deferred_reset" echo "Total messages conn. timeout: $deferred_timeout" echo "Total messages deferred other: `echo "$deferred_400 $deferred_refused $deferred_reset $deferred_timeout $total - - - - p" | dc`" echo "=========================================================================== " done