Autor: Eli Data: A: exim-users Assumpte: RE: [exim] Exim doesn't seem to be following my retry rules properly?
Tony wrote: > When this happens you should also pull the relevant entries out of the
> retry database. Note that delays before retry are per-host rather than
> per-message, so if the host has been down for a while the
> message won't be retried every 5 minutes as you seem to expect.
Oh. I thought having the retry_use_local_part setting (or whatever it is, I
don't have my config in front of me atm) on a router (or transport) would
change the retry entries to be specific per recipient email address, rather
than domain/host based? I'm assuming I didn't understand that setting
correctly then and that the retries aren't happening when I expect because
many other messages are failing to be delivered to the remote hosts as well.
I'll take a good look at my retry db for those failing hosts and see what
their entries are... Does it not retry a delivery until the LAST retry entry
for a host? As in, say I do "exinext 66.165.106.102" (that IP being one of
the failing hosts) and it spits out multiple entries with different "next
try" times for each - will it wait until the last one before retrying a
delivery attempt for *all* messages destined to that host? I was hoping
that my retry settings would kind of guarantee me a physical retry to the
failed host at least every 15 minutes (... Or is it but because I'm looking
at just *one* message, I don't see the retries for others?).
All I can say is, I hate windows based SMTP daemons :P They're nothing but
a load of trouble.
