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

weird OpenPGP expiration dates in AM reports [was: Re: AM report for Daniel Kahn Gillmor]



Woo hoo!  thanks, Patrick!

On 04/22/2009 08:36 AM, Patrick Schoenfeld wrote:

>    Let's test if its a version 4 or greater key
>    Key is ok
>    Check for key expire stuff
>    Key has an expiration date of 2009-06-19
>    2012-05-31.

Just a note on the weird-looking key expiration here: my primary key
expires in 2012.  The only bit that expires in june is my smaller
authentication-capable subkey:

0 dkg@pip:~$ gpg --with-colons --list-key 0xd21739e9 | cut -d: -f
1,3,7,12 | grep '^[ps]ub'
pub:4096:2012-05-31:scESCA
sub:4096:2012-05-31:e
sub:2048:2009-06-19:a
0 dkg@pip:~$

I suspect that the scripts used to generate these reports aren't used to
dealing with keys with different expiration dates.  Since the debian
project probably currently only cares about OpenPGP usage for signing
(signing .dsc and .changes files, bts messages, etc), the correct logic
for this script is something like:

 * if the primary key has the "Signing" usage flag set (the lower-case
"s" flag in field 12 above), use the primary key's expiration date.

 * if the primary key does *not* have the "Signing" flag, but one or
more non-revoked subkeys do have the signing flag set, use:

 min(exp. of pubkey, max(exp. of non-revoked, sign-capable subkeys))

 * if neither the primary key nor any non-revoked subkeys have the
"Signing" flag set, the key should probably not be considered to be OK.

If someone could to point me to the scripts, i can try to take a look at
writing this logic in something more effective than pseudocode.

Regards,

	--dkg

Attachment: signature.asc
Description: OpenPGP digital signature


Reply to: