A site for which I provide secondary MX is down for about 2 weeks.
We're about 5 days into this two week period and the queue shows
about 2000 messages in 11MB, or about 2MB per day coming in. Before
the site went down (it is a planned outage for a move) I lengthened
exim's retry for that site to cover the period, and appropriate
changes were made to the other site's DNS expiration..
I've symlinked enough spool space for the rest of the period and a
little more, but:
1) Is there a way to run the queue for that destination separately
from the main queue? (currently I run exim with '-bd -q2h') I'd like
to reduce the CPU load that I see go a little high when exim sifts
through those thousands of messages, by dropping it back to once or
twice a day or so instead of the current 2 hour interval. And the
queue is growing by some 400 or so messages a day... so ideally I'd
like two queue runners, one very infrequent for the known-down site
and another more frequent for all others.
2) I'll be on vacation when the other site comes back up next week.
How will exim handle the delivery of 5000 messages in 30MB to one
host, down my 128k connection? I know some people handle queues this
big routinely, but I don't, and my connection isn't very fast. I'm
wondering if I should adjust deliver_load_max; queue_run_max is at
default of 5.
thanks.
--Rob
--
Rob Lingelbach | "I would give nothing for that man's religion
rob@??? | whose very dog and cat are not the better for it."
www.alegria.com --Rowland Hill, "Village Dialogues", via A. Lincoln
--
*** Exim information can be found at
http://www.exim.org/ ***
