[lamps] draft-ietf-lamps-samples: PKCS12 expertise needed (including objects for comparison)

[Daniel Kahn Gillmor <[e-mail]>]

Hi folks--

We discussed draft-ietf-lamps-samples at the IETF 111 second LAMPS
session.

One concern is that the PKCS#12 objects in the current draft were
unimportable by MacOS's "Keychain Access" application.

Here's a screenshot of that failure, trying to ingest the DER-encoded
form of `bob.p12` from draft -06:

As the draft says, this PKCS#12 object is encrypted with a simple
3-letter password "bob".

However, Thunderbird's certificate manager *does* ingest the object
correctly, and can then export the associated objects as another PKCS#12
object, also encrypted with the password "bob".

This Thunderbird-laundered PKCS#12 object *is* ingestable by Keychain
Access for some reason.

I'm attaching here the original object as `bob.p12` and the
thunderbird-laundered version as `bob.laundered.p12`.  Both objects are
encoded in DER form.

Both objects *should* contain:

 - Bob's decryption secret key
 - Bob's corresponding encryption certificate
 - Bob's signing secret key
 - Bob's corresponding verification certificate
 - The intermediate CA certificate that permits verification from the
   ed25519 root CA

The secret keys in both PKCS#12 objects should be locked with the
password "bob".

Are we seeing a bug in Keychain Access?  Or is "bob.p12" somehow
malformed?  If "bob.p12" is malformed, is there a bug in Thunderbird
that it accepts it?

If anyone has any tooling that can make a clear and concise picture of
the differences between the objects (or can condense the screenfuls of
text that most detailed debuggers produce to highlight the specific
changes), that would be super useful.

          --dkg