Re: normalizing part numbering across PGP/MIME processing

Subject: Re: normalizing part numbering across PGP/MIME processing

Date: Fri, 03 Jun 2011 14:34:17 -0700

To: Jameson Graef Rollins, notmuch@notmuchmail.org

Cc:

From: Carl Worth


On Sat, 28 May 2011 14:31:08 -0700, Jameson Graef Rollins <jrollins@finestructure.net> wrote:
> On Fri, 27 May 2011 17:53:44 -0700, Carl Worth <cworth@cworth.org> wrote:
> >   * Should we set the crypto option to verify/decrypt by default?
...
> I don't really have an opinion on this.  I have it set now, so whether
> or not it's set by default doesn't make much difference to me.

I'm inclined to set it by default. But I'll wait until I get it fully
working, (which looks to be problems in my environment). See below.

> >   * I can't actually get decryption to work for me. :-(
> > 
> > 	When I run "notmuch show --decrypt" on a message encrypted with
> > 	my public key I get a segfault within libgmime, specifically in
> > 	the g_mime_session_request_passwd function.

I'm still getting this. I'll start debugging libgmime next.

> > 	$ gpg-agent
> > 	gpg-agent: symbol lookup error: /usr/lib/libassuan.so.0: undefined symbol: gpg_err_set_errno

That part I at least figured out. This behavior can be replicated (for me) with:

	LD_LIBRARY_PATH=/usr/lib gpg-agent

or:

	LD_LIBRARY_PATH=/usr/lib:/lib gpg-agent

And it goes away with:

	LD_LIBRARY_PATH=/lib:/usr/lib gpg-agent

It's still a mystifying bug in gpg-agent to me. But I've at least got my
environment to no longer trigger it, and I've opened a Debian bug report
for it at least.

-Carl
part-000.sig (application/pgp-signature)

Thread: