Re: [exim-dev] Per-Message Retry Semantics?

Páxina inicial
Borrar esta mensaxe
Responder a esta mensaxe
Autor: Philip Hazel
Data:  
Para: Marc Sherman
CC: exim-dev, Tim Wilde
Asunto: Re: [exim-dev] Per-Message Retry Semantics?
On Thu, 16 Jun 2005, Marc Sherman wrote:

> exim_tidydb -t 9d /var/spool/exim retry
> exipick -i 'message_age >= 10d' > xargs exim -Mrm
>
> The first command gets rid of any retry info older than 9d, so exim will
> effectively keep retrying infinitely


Nope. 'Fraid not. If the host has been tried today (for instance), the
retry information won't be older than 9d, even though the host has been
down for months. And if you did delete the record, Exim would revert to
the initial retry timings (every 15 mins, etc.).

You could just set a humungously large final cut-off time and then
delete messages after 10d as suggested. Or perhaps -Mg them rather than
-Mrm them.

But it will be very inefficient if the queue is of any size.

-- 
Philip Hazel            University of Cambridge Computing Service,
ph10@???      Cambridge, England. Phone: +44 1223 334714.
Get the Exim 4 book:    http://www.uit.co.uk/exim-book