On Tue, Feb 18, 2014 at 03:36:09AM +0000, Viktor Dukhovni wrote:
> For the pre-filter delay: you seem to be using an RBL that is not
> responsive, and may cause clients to disconnect due to high delays:
>
> Feb 17 14:37:57 mail postfix/smtpd[14259]: generic_checks:
> name=reject_rbl_client
> Feb 17 14:37:57 mail postfix/smtpd[14259]: reject_rbl:
> Client host 68.230.241.216
> Feb 17 14:38:07 mail postfix/smtpd[14259]: dns_query:
> 216.241.230.68.opm.blitzed.org (A): Host not found, try again
> Feb 17 14:38:07 mail postfix/smtpd[14259]: warning:
> 216.241.230.68.opm.blitzed.org: RBL lookup error:
> Host or domain name not found.
> Name service error for name=216.241.230.68.opm.blitzed.org type=A:
> Host not found, try again
>
> Paying attention to such warnings may be wise.
As well as not blindly including random RBLs from some stale HOWTO.
This particular RBL seems to have been dead for 7 years now.
http://www.dnsbl.com/2007/02/status-of-opmblitzedorg-dead.html
The OP would be wise to keep the set of RBLs used to a short list
of just the most effective ones. The marginal value of additional
RBLs is low, and the latency cost of using too many is high.
Periodically audit your RBL list to make sure none are dead.
The mail server performance will be much better without dead RBLs.
--
Viktor.
