A quick search didn’t come up with any previous conversations on this
topic… but it seems to me to be the sort of thing that should have
been discussed at some point in the past… so if I’ve missed
something a point is more than welcome.
Has anyone ever addressed the potential for auto-reply loops between
instances of RT that get directed at each other? I’m wondering if RT
is smart enough to recognize when an incoming message was generated by
another instance of RT (or even other ticket systems) and suppress an
auto-response.
Here’s the specific case I’m worried about:
We’re moving a bunch of our ‘role’ email addresses over to RT, away
from being just distribution lists. Among these is the contact
address used for several domains. In one case, our registrar will
send a confirmation email from inside their instance of RT to the tech
contact of the domain in question, which (if we move it to RT) is
likely to send an auto-response to their ticketing system, and
potentially start an auto-response loop. Even if a full loop isn’t
formed just a single misplaced auto response would be unwelcome.
Obviously I can fix this specific instance by building in custom rules
to that queue suppressing an autoresponse if the Requester is the
registrar address that I’m concerned with… however I’d love to know
if there’s a more elegant solution that will avoid the same problem
with other cases I can’t predict.
Thoughts, anyone? 
Matt
PGP.sig (194 Bytes)
A quick search didn’t come up with any previous conversations on
this topic… but it seems to me to be the sort of thing that
should have been discussed at some point in the past… so if I’ve
missed something a point is more than welcome.
Has anyone ever addressed the potential for auto-reply loops between
instances of RT that get directed at each other? I’m wondering if
RT is smart enough to recognize when an incoming message was
generated by another instance of RT (or even other ticket systems)
and suppress an auto-response.
Here’s the specific case I’m worried about:
We’re moving a bunch of our ‘role’ email addresses over to RT, away
from being just distribution lists. Among these is the contact
address used for several domains. In one case, our registrar will
send a confirmation email from inside their instance of RT to the
tech contact of the domain in question, which (if we move it to RT)
is likely to send an auto-response to their ticketing system, and
potentially start an auto-response loop. Even if a full loop isn’t
formed just a single misplaced auto response would be unwelcome.
Obviously I can fix this specific instance by building in custom
rules to that queue suppressing an autoresponse if the Requester is
the registrar address that I’m concerned with… however I’d love to
know if there’s a more elegant solution that will avoid the same
problem with other cases I can’t predict.
RT is set up to not autoreply to another RT or another ticketing
system. The feature isn’t perfect, but it does work. And you can
always set up individual addresses to not get mail.
PGP.sig (186 Bytes)
Has anyone ever addressed the potential for auto-reply loops
between instances of RT that get directed at each other? I’m
wondering if RT is smart enough to recognize when an incoming
message was generated by another instance of RT (or even other
ticket systems) and suppress an auto-response.
RT is set up to not autoreply to another RT or another ticketing
system. The feature isn’t perfect, but it does work. And you can
always set up individual addresses to not get mail.
Ah, okay that’s what I was hoping. In the specific case I’m concerned
about, the other ticketing system is RT so I’d expect it to work
pretty well. If any others come up that don’t, exception lists are a
possibility.
Thanks Jesse,
Matt
PGP.sig (194 Bytes)