TTL question & sendmail

[Bill Moseley]

Hi,

I'm looking for only a guess:

I have a secondary mail server and every once in a while I get mail
delivered to it instead of the primary MX machine.  Both are running the
exact same setup for bind (and sendmail), and both machines are very, very
low load:

 7:09am  up 18 days, 21:52,  2 users,  load average: 0.00, 0.00, 0.00

And that's at a busy time ;)

In my primary's mail log I see the reason the mail was delivered to the
secondary:  

  Domain of sender address assist at infopeople.org does not resolve

But in that same second of time it gets delivered to the secondary MX
machine (so the domain does resolve on the secondary machine, which
sendmail does check).

Can anyone venture a guess why two machines on the same network segment,
both running exact same copies of bind (one's primary master, and the other
a slave) and exact same copies of sendmail, and both very bored and lonely
machines, yet one is able to resolve the name and the other not within a
second of each other?

When sendmail says "domain does not resolve" I wonder if that means it
failed to find the A RR or the MX RR.

Thanks,







Bill Moseley
mailto:moseley at hank.org