Chris Green | 18 Oct 12:37 2014

Why is mutt choosing the HTML and what are those strange 'A A ' sequences?

This only happens occasionally, from specific people.  I have
forwarded an example with this, it's below.  There are two problems:-

1 - Mutt shows me the HTML version in preference to the plain text one.

2 - What is causing those odd 'A  A  ' sequences?  It looks as if some
systems version of TAB is getting mis-translated between character sets.

How do I get mutt to default to the text version and how do I get
those spaces displayed correctly?

----- Forwarded message from Mark Horn <mhorn71 <at>> -----

Return-Path: <mhorn71 <at>>
X-Original-To: chris <at>
Delivered-To: chris <at>
Received: from (
 []) by (Postfix) with ESMTPS id 04FA1380082 for
 <chris <at>>; Tue, 14 Oct 2014 10:41:06 +0100 (BST)
Received: from exim by with spam-scanned (Exim 4.63)
 (envelope-from <mhorn71 <at>>) id 1XdybH-0001Hp-JV for
 chris <at>; Tue, 14 Oct 2014 10:41:04 +0100
X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on
X-Spam-Status: No, score=-4.2 required=4.0 tests=BAYES_00,DKIM_SIGNED,
 HTML_MESSAGE,RCVD_IN_DNSWL_LOW,RDNS_NONE autolearn=ham version=3.3.1
Received: from [] ( by with esmtps (TLSv1:RC4-SHA:128) (Exim 4.63)
(Continue reading)

Brandon Amos | 18 Oct 11:18 2014

mutt on OSX 10.10 with gssapi smtp_authenticator error.

Hi, after upgrading to OSX 10.10, I started getting the following
error when sending emails with smtp using smtp_authenticators='gssapi'.

    TLSv1 connection using TLSv1/SSLv3 (RC4-SHA)
    No authenticators available

I was first using mutt 1.5.23 provided with Homebrew,
and I've tried reinstalling with no success.
I've also built from source with the following
configure options and still get the same error.

Configure options: '--with-sasl' '--with-gss' '--enable-imap'
'--enable-smtp' '--with-ssl' '--enable-hcache'

I have some version of gss headers and the dynamic library:

    ~ » ls /usr/include/gssapi*

    gssapi.h         gssapi_generic.h gssapi_krb5.h

    ~ » ls /usr/lib/libgssapi_krb5.dylib

The mutt configuration files I'm using (without users/pws) are
uploaded at if these
provide further info.

Does anybody know how I can fix this?
(Continue reading)

Elias Diem | 13 Oct 19:20 2014

Group replying: set To: instead of Cc:

Hi all

I often group reply to messages. I'd like to set the To: 
field with the reply addresses instead of the Cc: field. Is 
this possible? I didn't find anything on the list archive.



Chris Green | 12 Oct 11:14 2014

send-hook usage - how to say "did you really mean that?"

I want to try and catch an error I often make, I assume that someone
(or a mailing list) has an alias when it doesn't so I send a message
to 'tin' for example when it should have been to 'tin-users'.

It's easy enough to catch the error because the To: address will be to
my local domain rather than to the right place.  So I can implement a
send-hook matching ''.  I virtually *never* send mail to anyone
or anything at so that's not a problem.

However, having caught the error, I'm not quite sure how I can make
the send-hook interactive.  I want it to say "Did you really mean
that?" and, if possible offer alternatives.  If I just write an
interactive shell script that gets called by the send-hook will it
work as expected?  How does mutt handle this, i.e. where will it put
the interactions?

The sequence will be something like:-

    Sending to 'tin <at>', did you really mean that?
    [answer 'n']
    Did you mean 'tin-users <at> tin-org'?
    [answer 'y']

The script then tells mutt to change the To: address and all is well.
If you answer 'n' to the second question it prompts for a new To:

While I'm about it, how does one change the To: address from the
send-hook script?

(Continue reading)

Chris Green | 10 Oct 17:03 2014

Can one run mutt, via ssh, on a system without ncurses (or similar)?

I have a dlink NAS which has mutt installed on it.  It works in
command line mode, e.g.:-

    mutt -s Subject john.smith <at> <message.txt

but it would be useful if I could run it interactively.  At present it
fails as follows:-

    ~ $ mutt
    Error opening terminal: xterm.
    ~ $ 

As I understand it this is because there's no termcap/terminfo on the
dlink NAS.  However I'm running via ssh so the terminal *is* actually
an xterm and understands all the appropriate bits and pieces.

Is there any way I can tell the remote mutt that it has a proper
terminal to run in?

...oh, and here is the result of 'mutt -v':-

Mutt 1.5.18 (2008-05-17)
Copyright (C) 1996-2008 Michael R. Elkins and others.
Mutt comes with ABSOLUTELY NO WARRANTY; for details type `mutt -vv'.
Mutt is free software, and you are welcome to redistribute it
under certain conditions; type `mutt -vv' for details.

System: Linux (armv5tel)
ncurses: ncurses 5.7.20081102 (compiled with 5.7)
libiconv: 1.9
(Continue reading)

Mark Filipak | 18 Sep 22:19 2014

How is mutt with multi-mega-byte mboxes?

How is mutt with multi-mega-byte mboxes? Have you found that having tens
of thousands of messages in a single box is dangerous?

Thank You.

Matthias Apitz | 18 Sep 12:24 2014

display text when a main contains HTML and text


From time to time I'm getting mails having the message as text and HTML
at the same time in the body; ofc this is bad posting style of the sender and
maybe one should not care about and just delete such mails;

to display HTML attachments I have in my .mailcap:

$ fgrep lynx .mailcap 
text/html; lynx -dump -display_charset UTF-8 %s ; copiousoutput; nametemplate=%s.html

Why is mutt presenting the HTML part of the mail and not just the text?
Is this somehow a config issue?



Matthias Apitz               |  /"\   ASCII Ribbon Campaign:
E-mail: guru <at>     |  \ /   - No HTML/RTF in E-mail
WWW: |   X    - No proprietary attachments
phone: +49-170-4527211       |  / \   - Respect for open standards

Óscar Pereira | 13 Sep 20:48 2014

Format flowed equals no space in depth 1

Dear all,

When hitting reply, in the editor the lines of the quoted text are shown
like this: ">foo", but after sending the mail, the content is shown like
"> foo". Furthermore, if when composing the reply, I change from ">foo"
(which is done automatically) to "> foo", the end result still shows
like "> foo". This happens with vim and nano (and, I presume, all other

I thought this might be related with the inner workings of flowed text
(which I'm using), but checking the relevant RFC [1], I still see no
reason to quote the text (while composing) as ">foo" (instead of 
"> foo"). Can anyone shed some light?

---Óscar Pereira

Russell Urquhart | 13 Sep 16:08 2014

SMTP Authentication fails with 1.5.23 on Verizon

I can confirm that i am still having the SMTP Authentication error with version 1.5.23 of Mutt when trying to
send mail on a Verizon network.

I get around it, like I described earlier by explicitly naming all Authications to use.

I compiled my version of Mutt 1.5.23 with the debug option in hopes that i would check various debug logs,
after the error occurred.

Can someone tell me where these log files are/what their names are? I can't seem to find them readily.



John Niendorf | 9 Sep 15:05 2014

Mutt and Exchange

Hello Mutters,

I'm using Mutt to access my personal IMAP account, but was wondering if 
there is a way to use Mutt to connect to my work Exchange 2010 account?

I'm using Thunderbird that has an add-on that allows T-bird to connect to 
Exchange using the Outlook Web Access settings.

Has anyone ever connected to Exchange with Mutt?


Thank you,


Andre Klärner | 9 Sep 11:50 2014

when does re-scoring occour

Hi all,

is there a specific event / command to re-evaluate the message scoring? I
have a set of scores that rely on "~d >2d", but as long as my mutt instance
lives it is never re-evaluated. So as my mutt runs mostly for weeks the
score gets stays at the value that correct days ago, no matter if I change
the mailbox, or resource the whole muttrc.

Also a small side question: why is the minimal date range 1 day, and no
hours / minutes available? And does anybody know a real use for superseded
messages, I think I never saw one in the wild?

Thanks and regards,


Andre Klärner
Attachment (smime.p7s): application/x-pkcs7-signature, 5585 bytes