On Mon, 10 Aug 1998, Nigel Metheringham wrote:
> When exim attempts to deliver a bounce to a user, and that user is unable
> to receive mail for whatever reason (ie the user doesn't exist). A
> message is logged "error ignored" meaning that no bounce message is
> generated.
Only if you have configured it that way by setting ignore_errmsg_errors.
By default, the message gets frozen, for postmaster attention.
> However I think the "error ignored" message is misleading, and the text
> here should be changed.
Suggest an alternative, please!
> Here's an example of one of these messages (with some information changed
> to protect the guilty)
>
> 1998-08-10 16:06:37 0z5tWv-0001Ia-00 <= <> H=(relay21.mx.aol.com)
> [198.81.19.139] P=esmtp S=6406
> id=199808101505.LAA10571@???
> T="Returned mail: User unknown" for xxx@???
> 1998-08-10 16:21:33 0z5tWv-0001Ia-00 SMTP error from remote mailer
> after RCPT TO: <xxx@???>: host mailhost.yyy.zzz [1.2.3.4]:
> 550 sorry, recipient does not exist.
> 1998-08-10 16:21:35 0z5tWv-0001Ia-00 ** xxx@??? R=lookuphost
> T=remote_smtp: SMTP error from remote mailer after RCPT TO:
> <xxx@???>: host mailhost.yyy.zzz [1.2.3.4]:
> 550 sorry, recipient does not exist.
> 1998-08-10 16:21:35 0z5tWv-0001Ia-00 xxx@???: error ignored
> 1998-08-10 16:21:36 0z5tWv-0001Ia-00 Completed
>
> [actually is there any good reason for logging the 2nd entry when it will
> be repeated as a ** entry immediately after. Its nice for debug, but
> pointless otherwise and my logs are *huge* anyhow]
Yes, there is a good reason. The first entry is the error from the host.
If it were a 4xx error, and there were other available MX hosts, Exim
would go on to try them. So you might get a sequence of host errors. The
2nd entry is really the final entry for the whole address. I suppose
Exim could be clever, and notice if there is only one host, or if
there's a 5xx error, and refrain from putting out the host error on the
grounds that it will be repeated as an address error. I'll take a look
at that sometime. Thanks for the suggestion.
--
Philip Hazel University of Cambridge Computing Service,
ph10@??? Cambridge, England. Phone: +44 1223 334714.
--
*** Exim information can be found at
http://www.exim.org/ ***
