Understand that at the moment I am sitting down to read rfc0821.txt and a rather thick Oreilly Sendmail book. I am only hoping that one of you can direct me to a shorter path of reasoning...
I am using Sendmail on my networks proxy/gateway to relay mail to the world from my internal network. For some odd reason it takes forever to get a message from a client to the smtp gateway.
Before I got hired our network configuration was absolutely retarded... not that I am the all-knowing authority on network confiurations and all but...
We had 24 port hubs cascaded together before they went into our backbone switch. To be more specific, we had 2 24 port hubs going into one 24 port hub that then fed into our backbone switch. I have since ordered/installed a new 24 port switch to replace our old on-it's-last-leg 12 port switch and placed each hub into a port on the switch. So basically, we no longer have any cascaded hubs.
Here is the kicker...
Our end-users are complaining about how it is taking much longer for them so send email then prior to when I made the network changes.
To test this, I sent an internal host an email from my internal machine. I then tailed the mail log on our smtp gateway. It took about 10-20 seconds for this mail to be received by the smtp gateway.
Any ideas why this is happening? I am of course doing my homework and reading what I can but I have an enormous amount of literature to sift through. Any help you have to steer me in the right direction would be GREATLY appreciated...
Thanks in advance!
What's the time difference between the packets arriving at the smtp gateway and the log appearing from the gateway?
Usually these delays are tardy dns requests, sendmail can't resolve for some reason.
Add a firewall LOG rule in the nat PREROUTING chain and watch for mail and dns packets.
|All times are GMT -5. The time now is 11:53 PM.|