bug-anubis
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [bug-anubis] Questions about gpg-encrypt


From: Dominique Bordereaux
Subject: Re: [bug-anubis] Questions about gpg-encrypt
Date: Wed, 20 Aug 2003 12:18:17 -0700
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624

Good day Sergey,

> Yes, the fact is that signing and encrypting a message in one
> operation (gpg -se) produces different results compared
> to sequentially signing it (gpg -s) and then encrypting it (gpg -e).
> Unfortunately, we have not yet provided a special command in Anubis
> that would allow you to encrypt and sign a message at once. However,
> this can be done using external-body-processor directive. The rule for
> that would look like
>
> rule "se:(.[^;]*);"
>   external-body-processor /usr/local/bin/gpg --armor --batch -se -r \1
> done
>
> Then, any message whose "Subject:" header ends with "@@se:RECIPIENT;" will
> be signed and encrypted using gpg (of course, you will have to
> adjust the actual path to the utility).
>
> I would also suggest you to try to use alpha version of Anubis. It
> contains a lot of improvements. If you wish to give it a try,
> please download it from anonymous
>
>    ftp://mirddin.farlep.net/pub/alpha/anubis-3.9.91.tar.gz
>
> (the official location alpha.gnu.org is not fully operational
> yet). Please read the file NEWS and accompanying docs for the
> information about the user-visible changes. We would appreciate
> your comments on this release.

I am currently testing the rule above on anubis 3.9.91 and I can't get
it work.
Probably not because of gpg related stuff.
I am under the impression that it comes from the way the anubis and
identd interact:

Before yesterday, with anubis-3.6.2, I had the following sequence in my
/var/log/maillog file
Aug 18 16:43:16 localhost anubis[6457]: Connection from 127.0.0.1:35445
Aug 18 16:43:16 localhost anubis[11997]: Welcome user dbordereaux !
Aug 18 16:43:16 localhost anubis[11997]: Connected to smtp.xxxx.com:25
Aug 18 16:43:16 localhost anubis[11997]: Connection closed successfully.

Today, with anubis-3.9.91, I have the following sequence:
Aug 20 11:39:00 localhost anubis[8954]: Connection from 127.0.0.1:32963
Aug 20 11:39:00 localhost anubis[9410]: IDENT: connected to 127.0.0.1:113
Aug 20 11:39:00 localhost anubis[9410]: IDENT: data probably encrypted
with DES...
Aug 20 11:39:00 localhost anubis[9410]: idecrypt -- No such file or
directory.
Aug 20 11:39:00 localhost anubis[9410]: UID:99, GID:99, EUID:99, EGID:99
Aug 20 11:39:00 localhost anubis[9410]: Getting remote host information...
Aug 20 11:39:00 localhost anubis[9410]: Initializing socket...
Aug 20 11:39:00 localhost anubis[9410]: Connecting to smtp.xxxx.com:25
Aug 20 11:39:00 localhost anubis[9410]: Connected to smtp.xxxx.com:25
Aug 20 11:39:00 localhost anubis[9410]: Transferring message(s)...
Aug 20 11:39:00 localhost anubis[9410]: Connection closed successfully.
Aug 20 11:39:00 localhost anubis[8954]: Child [9410] finished. Exit
status: OK.
0 clients left.

I am using pidentd 3.0.14 packaged for Redhat 8.0 (package name
pidentd-3.0.14-8)
Might it come from that?

> I can understand -and I respect- your reluctance to answer to a guy if
> you have the impression that he is a slacker.
>
> I'm afraid you have misinterpreted Wojciech's reply. There was surely
> no reluctance to help you.

Wojciech, please accept my apologies if I hurt you with my words.
I regret it all the more that the last thing I want is to inoculate
negative "waves" in this project:
you gave a us a nice and clean piece of software and I feel more
grateful than anything else.

> If I don't get the answer, I'll tell you and if I really don't get the
> answer, I'll drop the ball and somebody else will pick it.
>
> Be sure you will always get the proper answer. Notice, however,
> my remark at the start of this letter: for pure technical reasons
> we may fail to receive your letters if we weren't mentioned in
> the Cc list. We hope these problems will soon be solved.

English is not my mother language.
I was using the verb "get" as a synonym of "understand" not of "receive".
With "get == receive", the sentence becomes indeed rude.
My bad!

> I agree. That is a good point. Again, please take a look at the
> new documentation. Does it look better?

The version I have currently in front of me is the anubis.ps generated
by the build process.
It definitely looks great!

> Your help will be highly appreciated.

:-)

Regards,

-- 
Dominique






reply via email to

[Prev in Thread] Current Thread [Next in Thread]