Re: Dealing with unreliable milters - revisited
- Wietse wrote:
> Mark Martinec:If an administrator explicitly requests an override (despite
> > I would like to request a Postfix configuration setting
> > which could override any attempts of some milter
> > to reject or tempfail a message.
> I also asked this question about ignoring tempfail replies. What
> if the Milter has already modified the message? The result will be
> inconsistent (from the user point of view; Postfix would not create
> a corrupted queue file).
some documented caveats), he should be willing to accept the
risk that a milter might first request some changes to a mail,
then changed its mind and rejected or tempfailed it.
I think a warning in a log would suffice. For me, it would be
acceptable one way or another (with or without milter edits
to a message being applied).
> BTW it should be possible to configure Postfix so that it limitsIn this particular case the header size is limited to 32 kB,
> headers to within Milter-acceptable limits. Most of the necessary
> support for that is already there.
although I don't think it is worth the effort to anticipate
anything that might go wrong in a milter and have a workaround
available. Today is a header size, yesterday was a header field
length, tomorrow will be something else.