Problem gcc'ing to nnimap

Problem gcc'ing to nnimap

[Ulf Stegemann]

Gnus/5.090012 (Oort Gnus v0.12) XEmacs/21.4

When sending an e-mail (or posting an article) I gcc the message to a local
nnml group _as well_ as to a nnimap group. This is working fine except when
the message or parts of it have to be signed (using pgp/mime or s/mime).

What's happening then is that I have to enter the passphrase for signing and
sending the message and enter it again for signing and gcc'ing it to the imap
server (which is slightly annoying).

Strange thing is that this does not happen always (but most of the time) and
sometimes it's sufficient to enter the passphrase once.

Furthermore, when gcc'ing to a nnimap group alone or to multiple nnml
groups and no nnimap group the problem does not occur, as well.


Ulf

-- 
Bubble bubble, toil and trouble; cast that float into a double.

Re: Problem gcc'ing to nnimap

[Lars Magne Ingebrigtsen]

Ulf Stegemann <ulf@zeitform.de> writes:

> When sending an e-mail (or posting an article) I gcc the message to a local
> nnml group _as well_ as to a nnimap group. This is working fine except when
> the message or parts of it have to be signed (using pgp/mime or s/mime).
>
> What's happening then is that I have to enter the passphrase for signing and
> sending the message and enter it again for signing and gcc'ing it to the imap
> server (which is slightly annoying).
>
> Strange thing is that this does not happen always (but most of the time) and
> sometimes it's sufficient to enter the passphrase once.

Are you able to determine what differs between the messages that you
have to sign twice and the ones you have to sign once?

(I'm not at all familiar with the signing machinery, so somebody else
is probably better as the person to track down this problem.  If it
isn't a feature, that is.  :-)

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Problem gcc'ing to nnimap

[Ulf Stegemann]

Lars Magne Ingebrigtsen <larsi@gnus.org> wrote:

> Ulf Stegemann <ulf@zeitform.de> writes:
>
>> Strange thing is that this does not happen always (but most of the time) and
>> sometimes it's sufficient to enter the passphrase once.
>
> Are you able to determine what differs between the messages that you
> have to sign twice and the ones you have to sign once?

Sorry, I did not notice any difference between messages I had to sign once
and those that required a passphrase twice. Nevertheless, I'll try to monitor
this behaviour and maybe there's really a difference between the messages.

One more hint: I discovered that after entering the passphrase for the first
time (but before the second time) the buffer got updated, i.e. the headers
where set up. I suppose that this means that message-send-hook has been
already executed at this point. Maybe this information is of any help?

Ulf

-- 
And so farewell, till we shall meet again.