On Mon, Nov 17, 2008 at 11:06:32AM -0600, Noel Jones wrote:
> Anvil sounds like the wrong tool for this job - it will defer
> all connections from the client until the rate falls below the
> level you specify. If your goal is to force connections to go
> to another server in your farm, anvil might work.
I think anvil is a poor fit for the problem description. Anvil monitors
plant-wide rates, and may trigger client backoff, while the OP wants
a single connection re-use limit, which does not require an external
counting process, each SMTP server alrwady has all the data it needs
locally.
The OP largely wants the upstream to not cache at all (yielding fair
load). Caching is for situations when the target domain has multiple
MX hosts, some of which may be down (blackhole for SYN), or may perform
expensive connection setup processing. To avoid high delivery latency,
one caches existing connections.
In the OPs case, there may be only 1 target IP (behind the LB) or or
perhaps a few IPs, but all *highly* available. In this case it is far
better to just ask the upstream to disable their connection cache
(though with multiple LB sites, the upstream will be slow if a site
fails completely).
--
Viktor.
Disclaimer: off-list followups get on-list replies or get ignored.
Please do not ignore the "Reply-To" header.
To unsubscribe from the postfix-users list, visit
http://www.postfix.org/lists.html or click the link below:
<mailto:[EMAIL PROTECTED]>
If my response solves your problem, the best way to thank me is to not
send an "it worked, thanks" follow-up. If you must respond, please put
"It worked, thanks" in the "Subject" so I can delete these quickly.
