Well it looks like a call to PSS has identified and shot down my
problem.
It seems there is a big difference in the way DNS queries are handled by
the SMTP VIRTUAL SERVER over DNS queries made through the Network
Properties function. If a mail admin plugs DNS server IP addresses into
the SMTP VIRTUAL SERVER Delivery -> Advanced ->External DNS page,
Exchange will query these DNS servers rather than use the DNS servers
supplied via the Network Properties page.
That, by itself is not so bad, however, DNS queries made through the
SMTP VIRTUAL SERVER are SINGLE THREAD only. One message queue is
processed at a time and the DNS queries have to resolve or fail before
the next queue is processed. When it can take 20 seconds or more for a
DNS query to fail, that can be a significant amount of time to process a
long line of queues before starting over again.
Now this condition only comes into play when you have DNS entries in the
EXTERNAL DNS page, and use DNS to do the delivery of mail messages (no
smart host entries in either an SMTP connector or the virtual SMTP
Server).
The workaround is to A. remove the External DNS server entries and let
DNS servers be designated via Network Properties, OR B. supply a
SMARTHOST for mail forwarding to the Internet or other dynamic DNS
environment.
PSS will be forwarding me other information today.
I hope you can use this information.
John Matteson
Geac Corporate ISS
(404) 239 - 2981
Atlanta, Georgia, USA.
-----Original Message-----
From: John Matteson
Posted At: Wednesday, October 15, 2003 1:13 PM
Posted To: Exchange 2000 Discussion list
Conversation: Problems with E2K SP3 bridgehead
Subject: [Exchange2000] Problems with E2K SP3 bridgehead
Hey Guys and Gals:
Got a persistent problem that I'm trying to troubleshoot.
I have a bridgehead server with two routing group connectors and
three Internet mail connectors loaded on it, one SMTP virtual server.
The hardware is a DELL 1650, single 1.4 Gbyte processor, 512
Mbytes of RAM, 100Mbyte NIC, OS is Win2K sp 4, E2k SP3 and September
Roll up package.
What happens is that the queue builds up on the bridgehead
server, then delivers mail to local routing group servers, across the
RGC connectors to other machines and out to the Internet (we deliver
directly via DNS rather than use a single outbound gateway machine).
This happens about every 8 to 10 minutes. If I bounce the Routing
Engine, then some of the items get delivered then, but not all.
Winroute tells me that Link State is UP on all machines in all
RG's (All machines are in one admin group, but split into three routing
groups) even as the queue is building.
Perfmon shows little if any network traffic outbound from the
bridgehead server except when the Routing Engine is bounced, or when the
queue dumps on it's own every ten minutes.
Where do I go to help this machine deliver mail a bit more
smoothly? Currently there are no limits on connections or the number of
messages per connection. But I rarely see more than 5 connections on the
server 4 inbound and one outbound.
Anyone have suggestions?
TIA
John Matteson
Geac Corporate ISS
(404) 239 - 2981
Atlanta, Georgia, USA.
[Non-text portions of this message have been removed]
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US &
Canada. http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/eJp0lB/TM
---------------------------------------------------------------------~->
Post message: [EMAIL PROTECTED]
Unsubscribe: [EMAIL PROTECTED]
Exchange 2000 FAQ:
http://www.exchange-mail.org/faq.html
Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
_________________________________________________________________
List posting FAQ: http://www.swinc.com/resource/exch_faq.htm
Web Interface:
http://intm-dl.sparklist.com/cgi-bin/lyris.pl?enter=exchange&text_mode=&lang=english
To unsubscribe: mailto:[EMAIL PROTECTED]
Exchange List admin: [EMAIL PROTECTED]
