On Sun, 3 Aug 2014, Paul Heinlein wrote: > Further down this thread you mentioned that you may have fixed this problem, > in which please ignore me...
Paul, Not ignoring; I think they're two separate issues. > My experience is that this sort of error is usually the result of a > SMTP-level filter (e.g, a milterized SpamAssassin) that is rejecting the > inbound mail. That message can't be delivered anywhere else since there's > no appropriate local user, and the remote SMTP server (even if "remote" > means 127.0.0.1 or ::1) is rejecting it in transit. > > So Sendmail panics; rather than discard the message, it renames them > within /var/spool/mqueue (or clientmqueue) so it won't try to process them > in the future. The more I tested and searched man pages and Web sites the more I realized the severe errors were not related to the logwatch issue. So, your explanation is very useful in teaching me from where they come. I kept going back and forth between logwatch and sendmail as the source of the problem because I've not changed the .conf files since I first installed logwatch many years ago. Yet, if changing the Output from stdout to mail fixes the problem (which it appears to do), then I'll stop looking for the reason why and return to more important things. :-) Much appreciated, Rich _______________________________________________ PLUG mailing list [email protected] http://lists.pdxlinux.org/mailman/listinfo/plug
