Re: reject_unverified_recipient case sensitive?

11 msg	SMTP authentication with saslauthd against PAM
7 msg	Error in SMTP sequence
2 msg	dkim-milter on relay server unknown-jobid exter...
11 msg	Stubborn Cert issue
3 msg	Problems with false positives

reject_unverified_recipient case sensitive?
\ Nick Smith (12 Apr 2007)
. \ Jorey Bump (12 Apr 2007)
. . \ Jorey Bump (12 Apr 2007)
. . . \ Nick Smith (12 Apr 2007)

3 msg	dkim=fail (verification error: invalid key gran...
5 msg	? Using header checks on (Postini) X-pstn-levels
2 msg	Can't find error in my config file. Can you?
1 msg	chroot postgresql ssl problem
15 msg	What cert to buy
3 msg	Address re-writing
9 msg	Mail Rejected when we relay for a client.
8 msg	Queueing up archive messages
2 msg	Cannot send e-mails from any e-mail clients
1 msg	Re: outbound mail failure - need to fix asap -S...
4 msg	outbound mail failure - need to fix asap
3 msg	Postfix alias problem
2 msg	problem to send a mail to a command in aliases
3 msg	smtpd_auth

Subject:	Re: reject_unverified_recipient case sensitive?
Group:	Postfix-users
From:	Nick Smith
Date:	12 Apr 2007

On 4/12/07, Jorey Bump <list> wrote:
> Jorey Bump wrote:
> > Nick Smith wrote:
> >> It appears as though the reject_unverified_recipient code requires an
> >> exact case sensitive match for an active cached entry before it will
> >> return the result from cache rather than initiate a fresh probe
> >>
> >> Since <Invalid.Recipient> and
> >> <INVALID.RECIPIENT> are equivalent, it seems that this
> >> behaviour causes additional probes to be generated unnecessarily
> >
> > Unfortunately, they are not equivalent.
>
> From RFC 2821:
>
> "The local-part of a mailbox
> MUST BE treated as case sensitive. Therefore, SMTP implementations
> MUST take care to preserve the case of mailbox local-parts. Mailbox
> domains are not case sensitive. In particular, for some hosts the
> user "smith" is different from the user "Smith". However, exploiting
> the case sensitivity of mailbox local-parts impedes interoperability
> and is discouraged."
>
>
Oh thanks - that is something I did not know

...although <invalid.user> is still equivalent to
<invalid.user> and each of these requires an individual
probe

Nick