brad at python.org
Fri Feb 17 20:23:29 CET 2006
At 10:17 AM -0800 2006-02-17, Mark Sapiro wrote:
> These delays were between bag.python.org and servers named
> smtp-*.xs4all.nl (which is I think where python.org is hosted). So
> missing posts, posts out of sequence etc. from approximately noon GMT
> on FEB 16 to about noon GMT on Feb 17 appear to be due to an outgoing
> mail issue at xs4all.nl and not due to anyone's particular method of
> reading them (unless you read the pipermail archive which I think has
> been current throughout).
This is correct. I just found out this morning that they've been
doing some maintenance on their systems, and depending on the precise
circumstances of when a particular message came in and what was
happening at that time, there may have been an attempt to deliver a
given outgoing message which was temporarily rejected, causing that
message to be re-queued and retried at a later time. The issue is
that we do enough volume that this caused a fairly significant
backlog of messages, which I believe is now mostly cleared up.
Of course, I've been away from my computer all day, so I was not
able to help work on resolving the problem at a time when that would
have done some good.
I will tweak our configuration so that we should be more immune
to these kinds of problems in the future, and we try to guarantee
that we retry delivery on each and every message faster than we had
done in the past.
Brad Knowles <brad at python.org> Python.org Postmaster Team
More information about the Mailman-Users