On 08/12/2016 14:11, Bastian Blank wrote:
On Thu, Dec 08, 2016 at 07:39:00AM +0000, Dominic Raferd wrote:
Is there a way to get it instead to try again, after a delay, to the
primary onward server(s) (whether the one specified as relayhost or,
if no relayhost is specified, the MXs per the recipient address),
and only if these fail again (or after n times) then fallback?
No. The fallback is there to free the primary server from mail that
can't be delivered immediately. To queue it for some time first and
only then use the fallback would defeat this purpose.
Thanks for your response. I thought the logic was to help me and my
server but I was not considering the wider picture ;-)
If nothing was specified as smtp_fallback_relay would it wait and
try again on a transient error or would it just give up immediately?
If the former, this would give me the behaviour I want but I really
need to have the fallback option too (in case of permanent errors,
or transient errors that don't go away).
Why do you think a transient error should ever produce an immediate
bounce?
Well I agree it would not be logical, but I was hoping that Postfix
could treat the fallback_relay as a '2nd tier', and try for some time to
deliver at the '1st tier' level before falling back, but I understand
now that it doesn't (by design).
The reality is that (with my small volumes, all to our own mailboxes)
the wait time to succeed with delivery direct to gmail is short, so if I
had no fallback_relay it would presumably quickly succeed in sending
directly instead of having to use an intermediate relay.
