For reasons unconnected with this question, I've just changed
the WORLDWIDEWORDS list experimentally from Full Auto to Manual.
LISTSERV reported just before the change that the list - which
is used to mail out a newsletter once a week - has about 8,300
subscribers, and that 16 subscribers were being monitored. After
Saturday's mailing, I received about 500 error reports via the
LISTSERV, with 417 of them "user unknown".

Could somebody explain the very large difference between these
two figures? Does the LISTSERV not act on the error reports it
receives? Presumably these dead subscribers would be eliminated
as a result of our three-monthly passive probe, but if the
LISTSERV is not acting on the weekly bounces, will it not also
fail to see those in response to the probes? To end up with more
than 400 dead subscribers on a list of 8000 within a three-month
period seems high.

--
Michael Quinion