250234 - CERT_EncodeAuthKeyID produces malformated DER without KeyID set to something

Status: RESOLVED DUPLICATE of bug 244922

(NSS :: Libraries, defect)

Description

[Emil Assarsson]

User-Agent:       Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040413 Debian/1.6-5
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040413 Debian/1.6-5

Sorry about this long and boring text but I don't know it I can explain it better.

I'm using CERT_EncodeAuthKeyID to add the Authority Key Extension to generated
certificates. I use the issuers Subject Public Key ID if it exist, but when it's
not I want to use the issuers subject and serial only.

I guess that the KeyID (the first item in the AKI Sequence) should be NULL. But
I get a malformated item here. It is 2 bytes with HEAD, witch leaving nothing to
the DATA. But the Length is set to 1. So this will overlap the next item.




Reproducible: Always
Steps to Reproduce:
1.
2.
3.

Actual Results:  
Output from a generated certificate:
$openssl asn1parse -inform der -in outbox/mupp.der -i

[...]
 374:d=4  hl=2 l=  18 cons:     SEQUENCE
  376:d=5  hl=2 l=   3 prim:      OBJECT            :X509v3 Basic Constraints
  381:d=5  hl=2 l=   1 prim:      BOOLEAN           :255
  384:d=5  hl=2 l=   8 prim:      OCTET STRING
  394:d=4  hl=2 l=  76 cons:     SEQUENCE
  396:d=5  hl=2 l=   3 prim:      OBJECT            :X509v3 Authority Key Identifier
  401:d=5  hl=2 l=  69 prim:      OCTET STRING
  472:d=1  hl=2 l=  13 cons:  SEQUENCE
  474:d=2  hl=2 l=   9 prim:   OBJECT            :sha1WithRSAEncryption
  485:d=2  hl=2 l=   0 prim:   NULL
  487:d=1  hl=3 l= 129 prim:  BIT STRING
[...]

Output from the generated AuthKeyID octet string found at offset 403:
$openssl asn1parse -inform der -in outbox/mupp.der -i  -offset 403
    0:d=0  hl=2 l=  65 cons: SEQUENCE
    2:d=1  hl=2 l=   1 prim:  cont [ 0 ]
Error in encoding

* This is the problem *
If I set KeyID to an empty SECItem it will end up as a 2 byte long der item with
the length set to 1. This will overlap the next item!


If I go on to offset 407 witch is the subject of the issuer:
$openssl asn1parse -inform der -in outbox/mupp.der -i \
>  -offset 407 -length 60 
    0:d=0  hl=2 l=  60 cons: cont [ 1 ]
    2:d=1  hl=2 l=  58 cons:  cont [ 4 ]
    4:d=2  hl=2 l=  56 cons:   SEQUENCE
    6:d=3  hl=2 l=  11 cons:    SET
    8:d=4  hl=2 l=   9 cons:     SEQUENCE
   10:d=5  hl=2 l=   3 prim:      OBJECT            :countryName
   15:d=5  hl=2 l=   2 prim:      PRINTABLESTRING   :SE
   19:d=3  hl=2 l=  16 cons:    SET
   21:d=4  hl=2 l=  14 cons:     SEQUENCE
   23:d=5  hl=2 l=   3 prim:      OBJECT            :organizationName
   28:d=5  hl=2 l=   7 prim:      PRINTABLESTRING   :Test AB
   37:d=3  hl=2 l=  23 cons:    SET
   39:d=4  hl=2 l=  21 cons:     SEQUENCE
   41:d=5  hl=2 l=   3 prim:      OBJECT            :commonName
   46:d=5  hl=2 l=  14 prim:      PRINTABLESTRING   :Emil Assarsson
 



Expected Results:  
A valid DER encoding with a nice length :-)

Comment 1

[Emil Assarsson]

Attachment: Test program that uses CERT_EncodeAuthKeyID without KeyID

Compile this and run. The program writes a file (authKeyID.der) with an DER
formated AKI. After that you could run DER test programs on this file and you
will find something fishy.

Comment 2

[Wan-Teh Chang]

*** Bug 250241 has been marked as a duplicate of this bug. ***

Comment 3

[Nelson Bolyard (seldom reads bugmail)]

Emil,  This seems like a dup of bug 244922, which is awaiting patch review.
Please review that bug, and tell me (here) if you think this bug is a 
duplicate of that, or not.  The crucial question is whether or not your
template has any "MAY_STREAM" flags.  
Also, you can test out the complete set of 5 patches attached to that bug,
and see if that solves your problem.  

Comment 4

[Emil Assarsson]

The description in bug 244922 matches mine.
CERTAuthKeyIDTemplate uses SEC_OctetStringTemplate on the KeyID witch has the
flag SEC_ASN1_MAY_STREAM.

After applying the 5 patches it works perfectly!

Here is the output after running the test program again:
$ ./akitest
$ openssl asn1parse -inform der -in authkeyid.der -i -dump -length 64
    0:d=0  hl=2 l=  62 cons: SEQUENCE
    2:d=1  hl=2 l=  57 cons:  cont [ 1 ]
    4:d=2  hl=2 l=  55 cons:   cont [ 4 ]
    6:d=3  hl=2 l=  53 cons:    SEQUENCE
    8:d=4  hl=2 l=  11 cons:     SET
   10:d=5  hl=2 l=   9 cons:      SEQUENCE
   12:d=6  hl=2 l=   3 prim:       OBJECT            :countryName
   17:d=6  hl=2 l=   2 prim:       PRINTABLESTRING   :SE
   21:d=4  hl=2 l=  16 cons:     SET
   23:d=5  hl=2 l=  14 cons:      SEQUENCE
   25:d=6  hl=2 l=   3 prim:       OBJECT            :organizationName
   30:d=6  hl=2 l=   7 prim:       PRINTABLESTRING   :test AB
   39:d=4  hl=2 l=  20 cons:     SET
   41:d=5  hl=2 l=  18 cons:      SEQUENCE
   43:d=6  hl=2 l=   3 prim:       OBJECT            :commonName
   48:d=6  hl=2 l=  11 prim:       PRINTABLESTRING   :test person
   61:d=1  hl=2 l=   1 prim:  cont [ 2 ]

This bug should be marked as an duplicate of bug 244922.

Thank you!

Comment 5

[Wan-Teh Chang]

*** This bug has been marked as a duplicate of 244922 ***