Evgeniy Berdnikov <bd4@???> (Di 14 Jun 2016 22:16:57 CEST):
> On Tue, Jun 14, 2016 at 10:45:14AM +0100, Jeremy Harris wrote:
> > > ??? id=<generated id> # logged as always, because it *is* a message id
> > No. Don't change existing behaviour. Log generated only if configured.
> >
> > > fixed=id,sender,date # logged if +fixes is active
> >
> > Perhaps "fh=ISD" - "fh" for "fixed headers" and whatever subset
> > of the letters I,S,D are appropriate to indicate fixup of
> > Message-ID: / Sender: / Date: header(s)?
> >
> > Only to keep the overall line length down.
>
> I propose to generate separate log line, filled with readable fixed_id=<..>,
> fixed_date=... and other fixed_ items. Their sense should be obvious and
> user should not spent time guessing what the abbreviation "fh=..." means.
> This would be similar to logging of address rewrite rules.
Hm. fh=[I][S][D] isn't as easy greppable as fixed_id=…, fixed_date=…
But these fixed_* are quite noisy. OTOH they shouldn't occur that often.
What about two log selectors
submission_fixups log submission mode message fixups on a separate line
fixed_message_id log submission mode fixed message id as id=
But I do not see any harm in logging the fixed message id w/o a new log
selector. What reason do we have not the use id=… even for a generated
(fixed) id?
Best regards from Dresden/Germany
Viele Grüße aus Dresden
Heiko Schlittermann
--
SCHLITTERMANN.de ---------------------------- internet & unix support -
Heiko Schlittermann, Dipl.-Ing. (TU) - {fon,fax}: +49.351.802998{1,3} -
gnupg encrypted messages are welcome --------------- key ID: F69376CE -
! key id 7CBF764A and 972EAC9F are revoked since 2015-01 ------------ -
