| 1 |
Common: |
| 2 |
- don't "die" on PRINT or CLOSE, find better ways to report the error |
| 3 |
|
| 4 |
DKIM-Signature: |
| 5 |
- allow version tag (DONE) |
| 6 |
- accept q=dns/txt (DONE) |
| 7 |
- method to set/get "z" tag |
| 8 |
|
| 9 |
DKIM Public Key Records: |
| 10 |
- enforce t=s option (if present) |
| 11 |
- provide method for caller to get (to check the "testing" flag) |
| 12 |
|
| 13 |
Verifier: |
| 14 |
- verify multiple signatures (ietf05 6.1) (DONE) |
| 15 |
- check that From header is signed (ietf05 6.1.1) |
| 16 |
- check public key "granularity" (DONE) |
| 17 |
- handle no response from first DNS server listed in resolv.conf |
| 18 |
(currently it goes to the second server after 5 seconds, |
| 19 |
but it does this for EVERY signature, so this will badly affect |
| 20 |
overall throughput) |
| 21 |
- **minor bug**- when Debug_Canonicalization=1 on a message with |
| 22 |
multiple signatures, the canonicalized output is recorded multiple times. |
| 23 |
Probably only the first valid signature should receive the |
| 24 |
Debug_Canonicalization option |
| 25 |
- provide semi-standard mechanism to report results of verification |
| 26 |
(including what, if any, of header.from and header.sender can be trusted) |
| 27 |
- provide mechanism in the API to run the DNS lookups in parallel with |
| 28 |
other processing (e.g. the SpamAssassin plugin would want to start the |
| 29 |
DNS queries as early as possible, but continue processing other aspects |
| 30 |
of the message while waiting for the DNS queries to complete). |
| 31 |
Net::DNS::Async may be useful here. |
| 32 |
|
| 33 |
Policy: |
| 34 |
- make it possible to determine explicit vs implicit default policy (DONE) |
| 35 |
- lookup BOTH draft-allman-dkim-ssp policy AND rfc4870(historical) policy |
| 36 |
(DONE) |
| 37 |
- this will probably be: lookup allman policy, and if not found, then try |
| 38 |
rfc4870(historical) policy (REJECTED) |
| 39 |
|
| 40 |
Signer: |
| 41 |
- allow DomainKeys signatures without using a policy object |
| 42 |
- allow adding chained signatures in one pass |
| 43 |
(e.g. allow adding a DomainKeys signature, and a DKIM signature, |
| 44 |
with the new DKIM signature signing the new DomainKeys signature) (REJECTED) |
| 45 |
- allow creation of i= and x= tags (DONE) |
| 46 |
- allow creation of l=, t=, and z= tags |
| 47 |
- do header-wrapping to signature before signing (DONE) |
| 48 |
- allow signer policy to change which private key is used |
| 49 |
|
| 50 |
Testing (some of this may already be implemented): |
| 51 |
- test public key errors: |
| 52 |
- DNS timeout |
| 53 |
- SERVFAIL |
| 54 |
- syntax error in public key record |
| 55 |
- test DNS timeout for signing policy |
| 56 |
- test key records composed of fragmented TXT records |
| 57 |
- test signature options: |
| 58 |
- unspecified query type |
| 59 |
- query type of "dns/" |
| 60 |
- bad query type (DONE) |
| 61 |
- bad algorithm (DONE) |
| 62 |
- unspecified algorithm |
| 63 |
- bad canonicalization |
| 64 |
- unspecified canonicalization |
| 65 |
- test presence of version tag in signature |
| 66 |
- IMPORTANT- allow `make test' to work when DNS is not available |
| 67 |
- test various components of verifying, so better diagnostics can be |
| 68 |
made when the verify.t script reports a bunch of unexplained failures |
| 69 |
- test absense of h= tag in DKIM signature |
| 70 |
- test use of non-ASCII characters in header names and h= tag |
| 71 |
|
| 72 |
Possible issues in base-10 draft: |
| 73 |
- 6.1.2 - check g= tag of public key against i= tag of signature (DONE) |
| 74 |
- 6.1.2 - check h= tag of public key against a= tag of signature (DONE) |
| 75 |
- 3.5 - t= tag, create it when signing messages, check it when verifying |
| 76 |
- 3.5 - x= tag, create it when signing messages |
| 77 |
- check it when verifying (DONE) |
| 78 |
- 5.4 - allow better control of which headers to sign |
| 79 |
- 5.5 - recommended headers to sign and NOT to sign (DONE) |
| 80 |
- 3.3.1 - what's an RSA exponent? |
| 81 |
- 6.1.1 - configurable list of unacceptable signing domains, |
| 82 |
e.g. "com" and "co.uk" |
| 83 |
|
| 84 |
Possible issues in RFC 4871: |
| 85 |
- 3.6.1 - g= should be case-sensitive (see 3.2 "tag values must be |
| 86 |
processed as case sensitive unless...", and 3.6.1, "g=", which does |
| 87 |
NOT mention case-sensitivity) |
| 88 |
- 3.6.1 - g= tag using irregular characters |
| 89 |
- 3.5 - i= tag, should allow quoted-printable encoding |
| 90 |
- 3.5 - i= tag, internationalized domains? |
| 91 |
- 3.5 - l= tag, what happens if the number is REALLY big, |
| 92 |
or doesn't contain a number? |
| 93 |
- 3.5 - q= tag, should skip signature if subtype is not "txt" |
| 94 |
(I think I do this, but do other verifiers?) |
| 95 |
- rationale- if a dns/foo type comes out, then it will be WRONG |
| 96 |
to lookup the txt record |
| 97 |
- 3.2 - "if a tag name does occur more than once, the entire tag-list |
| 98 |
is invalid" |