Bon Matin Judy,

Hmmm... The Kangaroo Syndrom ... When I saw this it was almost always a
local configuration problem at the user site.

When the primary mail server is down, the mail is "served" by the
backup mail server. So, if Host A is primary and has B as backup and B
has A listed as backup. If The Primary is down it tries to find the
Backup Host on B... If it is misconfigured and A is listed, it loops
from B to A ... The scenario I give is propably not correct but it has
to do with something of that nature ... (I believe the component in
trouble are "MX" records content)

In short, bad configuration may create a loop  that ends when the
maximum "hops" (ie the max number of Received: lines is busted ...)

The problem may be different but the only thing you can do is forward
them and postmaster@fqdn a copy of the bounces ...  The problem is local
to them ... (at least most of the times)

You can check it by looking at the bounce for recurring appearances of
the same domain names ... You may also forward a copy of the bounce to
each "postmaster@" host listed in the repeating Received: sequence ...

If they can't ask to their mail manager to look into that then you may
advise them that you will have to ignore their complaints... If it
annoys you set them NOMAIL/REVIEW or DELETE them according to your
preferences.
And let them solve *their* problems.

HTH,

--
Eric Paul-Hus , BLUES-L