Hi Suresh,
Thanks for the quick reply! Does exim_tinydb simply resort/compress the db
file? Is this safe to do whilst the server is running and active exim
processes?
We retry queue data every 5 hours. This seems reasonable doesn't it? Our
queue contains over 1 milllion files. Any mail older than 10 days is auto
purged from the queue.
Below are our exim stats (24 hour period) in case this helps gauge our mail
system...
TOTAL Volume Messages Hosts Delayed Failed
Received 22GB 579283 42797 61713 10.7% 112149
19.4%
Delivered 19GB 485471 5103
Deliveries by transport
-----------------------
Volume Messages
local_delivery 8347KB 691
remote_smtp 989MB 44362
virtual_delivery 18GB 440418
----- Original Message -----
From: "Suresh Ramasubramanian" <linux@???>
To: "Paul" <paul@???>
Cc: <exim-users@???>
Sent: Friday, December 12, 2003 12:14 PM
Subject: Re: [Exim] large retry and wait-remote_smtp files
> >>>>> "Paul" == Paul <paul@???> writes:
>
> Paul> Hi All, We are using 3.36 on a large mail server and are
> Paul> trying to reduce some bottlenecks on our system. I've
> Paul> noticed our retry and wait-remote_smtp files are quiet
> Paul> large. Both around above the 60mb + mark. Can these be
>
> # exim_tidydb /var/spool/exim retry
> # exim_tidydb /var/spool/exim wait-remote_smtp
>
>
> Paul> /usr/local/exim/bin/exim -bd (for smtp daemon) and
> Paul> /usr/local/exim/bin/exim -q5h (for mail queue runner)
>
> Why 5 hours? -bd -q15m might be a better idea.
>
> And run a local caching nameserver.
>
> And shunt deferred mail to a secondary server for delivery.
>
> srs
>
> --
> linux@??? (Suresh Ramasubramanian)
> jaharkes@ravel:/usr/src$ mv linux Gnu/Linux
> mv: cannot move `linux' to `Gnu/Linux': No such file or directory
> jaharkes @ cs.cmu.edu in reply to RMS on linux.kernel
>
>
