On 2010-09-20 15:24 +0200, Stefan Monnier wrote: > Package: emacs23 > Version: 23.2+1-4 > Severity: critical > Tags: patch > Justification: causes serious data loss > > I have lost a few days's worth of outgoing email because Emacs decided that it > did not need to pass "-f" to sendmail any more
For the record, Debian's emacs2[123] packages always behaved this way, so "any more" just refers to you suddenly using them instead of vanilla Emacs. > (resulting in the envelope-from > address in this outgoing mail to have a form that my relayhost rejects as > potential spam). > > The (simple) patch is attached, > > > Stefan > > > --- mine/message.el.gz > +++ other/message.el.gz > @@ -10,14 +10,9 @@ > (const never) > (const ask))) > > -;; message-sendmail-f-is-evil is nil here in the upstream source, but > -;; sendmail works right under Debian Linux, so we want t. [was > -;; orignally reported for emacs 19 as debian#7051] > -(defcustom message-sendmail-f-is-evil t > +(defcustom message-sendmail-f-is-evil nil > "*Non-nil means don't add \"-f username\" to the sendmail command > -line, because adding it would be more evil than leaving it out. Under > -Debian/GNU/Linux, sendmail works right, so it should be safe for this > -to be set to true." > +line, because adding it would be more evil than leaving it out." > :group 'message-sending > :link '(custom-manual "(message)Mail Variables") > :type 'boolean) This patch just reverts debian-adjust-mail-from-addresses.diff, so the reasonable action would be to drop that patch from the series. See #397757 why this would be correct. Sven -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
