| |
| | [No title] (Site not responding. Last check: 2007-10-08) |
 | | Weak keys are now explicitly permitted in the rsa-md5-des, rsa-md4-des, and des-mac checksum types, because we aren't currently checking for weak "variant" keys at key generation time, and may have stored keys with weak variants. |
 | | Weak keys are now explicitly corrected in 3des key generation, because Heimdal does it, and while MIT doesn't, it wouldn't actually work with such keys anyways, and users' keys may be stored long term in KDCs. |
 | | The nature of the weak keys is such that it is extremely + Many of the choices concerning when weak-key corrections are + performed relate more to compatibility with existing implementations + than to any risk analysis. |
| www.mit.edu /~raeburn/draft-ietf-krb-wg-crypto-03-04-xdiffs.txt (601 words) |
|