Prefer a currently available signing subkey?

[Daniel Kahn Gillmor]

On Thu 2017-04-20 02:36:16 +0200, Juan Miguel Navarro Martínez wrote:
> On 2017-04-19 at 17:46, Daniel Kahn Gillmor wrote:
>> The open report is https://dev.gnupg.org/T1983
>
> Is it possible that is a duplicate of this report too?
> https://dev.gnupg.org/T1967
>
> Both are about a capable subkey not being used on GnuPG Modern branch
> because it prefers a subkey with its missing secret part. Plus there was
> a patch which seemed to work for 2.1.18.

I agree that these seem related, though T1983 has smartcard-specific
concerns.  I've tested the patch for T1967 in a non-smartcard situation,
but haven't tested it with a smartcard yet.  I'd be happy to hear the
results of such a test, if anyone has a smartcard handy for testing.

    --dkg