john.gill at computer.org
Mon Apr 9 23:01:21 CEST 2012
I know that gpg chooses common algos between the sender and recipient.
(I've not tested what will happen with recipients who have no preferences
in common with my enabled algos, but that's a problem for a new day.) I'm
not trying to out-think the intelligence codified in the application. I am
analyzing my keyring contents to identify how using "disable-cipher-algo"
and similar options, may impact exchanges with my recipients.
My original question was to clarify my understanding of the output from
"list-packets", so I could finish writing out the report.
On Apr 9, 2012 1:16 PM, "David Shaw" <dshaw at jabberwocky.com> wrote:
> On Apr 9, 2012, at 10:52 AM, John Gill wrote:
> > I'm assuming the the signatures indicate, roughly the set of options
> that my recipients will not receive an error about ignored preferences.
> For instance, symmetric algo 9 has been around for the last 10 years at
> least. but if I force it on someone who doesn't have it as a preference,
> the recipient will get a message about my ignoring preferences. For systems
> that are automated, this message may have repercussions, depending on how
> they were coded.
> > I'm identifying any recipients in my keyring that have preferences that
> conflict with my disabling of specific algorithms and functions.
> You don't need to do that. GnuPG does it for you automatically. When
> encrypting to a particular set of user IDs, GPG ensures that the algorithms
> and features that are chosen are acceptable to all recipients.
> In your example, if algo 9 (AES256) isn't available for a particular
> recipient, GPG will use something else.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Gnupg-users