ipsec keyblobtoid, splitkeytoid - generate key IDs from RSA keys
size_t keyblobtoid(const unsigned char * blob, size_t bloblen,
char * dst, size_t dstlen);
size_t splitkeytoid(const unsigned char * e, size_t elen,
const unsigned char * m, size_t mlen, char * dst,
Keyblobtoid and splitkeytoid generate key IDs from RSA keys, for use in
messages and reporting, writing the result to dst. A key ID is a short
ASCII string identifying a key; currently it is just the first nine
characters of the base64 encoding of the RFC 2537/3110 “byte blob”
representation of the key. (Beware that no finite key ID can be
collision-proof: there is always some small chance of two random keys
having the same ID.)
Keyblobtoid generates a key ID from a key which is already in the form
of an RFC 2537/3110 binary key blob (encoded exponent length, exponent,
Splitkeytoid generates a key ID from a key given in the form of a
separate (binary) exponent e and modulus m.
The dstlen parameter of either specifies the size of the dst parameter;
under no circumstances are more than dstlen bytes written to dst. A
result which will not fit is truncated. Dstlen can be zero, in which
case dst need not be valid and no result is written, but the return
value is unaffected; in all other cases, the (possibly truncated)
result is NUL-terminated. The freeswan.h header file defines a constant
KEYID_BUF which is the size of a buffer large enough for worst-case
Both functions return 0 for a failure, and otherwise always return the
size of buffer which would be needed to accommodate the full conversion
result, including terminating NUL; it is the caller´s responsibility to
check this against the size of the provided buffer to determine whether
truncation has occurred.
With keys generated by ipsec_rsasigkey(3), the first two base64 digits
are always the same, and the third carries only about one bit of
information. It´s worse with keys using longer fixed exponents, e.g.
the 24-bit exponent that´s common in X.509 certificates. However, being
able to relate key IDs to the full base64 text form of keys by eye is
sufficiently useful that this waste of space seems justifiable. The
choice of nine digits is a compromise between bulk and probability of
RFC 3110, RSA/SHA-1 SIGs and RSA KEYs in the Domain Name System (DNS),
Eastlake, 2001 (superseding the older but better-known RFC 2537).
Fatal errors are: key too short to supply enough bits to construct a
complete key ID (almost certainly indicating a garbage key); exponent
too long for its length to be representable.
Written for the FreeS/WAN project by Henry Spencer.