Re: OpenSSL Hacks
- From: Marcus Streets <marcus@xxxxxxxxxxxxxxx>
- Date: Tue, 06 Jun 2006 09:40:48 +0100
Douglas A. Gwyn wrote:
~David~ wrote:
... OpenSSL is widely considered very secure software and recently won a
government approval; breaks on it at any level are pretty uncommon if not rare.
I haven't looked at OpenSSL for a couple of years, but back then
(when we were considering using it in a "product"), after I found
several vulnerabilities I decided against using it. Maybe the
vulnerabilities have since been repaired, or maybe that
"approval" was not the result of a thorough, competent cecurity
review, or maybe something else.
The approval is a FIPS 140-2 level 1 approval - which means that it
meets the security requirements set out in the standard, no more and no
less.
FIPS 140-x does not include a full security review.
For a level 1 software validation, such as OpenSSL has achieved, this
essentially means that the cryptographic algorithms have been correctly
implemented and perform sufficient self tests to determine that they are
operating correctly.
In particular the current standard does not include any testing of
mechanisms to protect against timing, power or emissions analysis -
though these will probably be included in the FIPS 140-3 standard.
Note the FIPS 140-2 validation is for a specific code base - see the
following caveat on the NIST web site.
"When built, installed, protected and initialized as specified in the
provided Security Policy. Appendix B of the provided Security Policy
specifies the complete set of source files of this module. There shall
be no additions, deletions or alterations of this set as used during
module build. All source files, including the specified OpenSSL
distribution tar file, shall be verified as specified in Appendix B of
the provided Security Policy. Installation, protection, and
initialization shall be completed as specified in Appendix C of the
provided Security Policy. Any deviation from specified verification,
protection, installation and initialization procedures will result in a
non FIPS 140-2 compliant module."
Having said all that congratulation must go to the OpenSSL team and to
the staff at Domus, for persuading NIST and CSE to accept an open source
project as suitable for FIPS 140-2 validation.
Marcus
.
- References:
- OpenSSL Hacks
- From: John E. Hadstate
- Re: OpenSSL Hacks
- From: ~David~
- Re: OpenSSL Hacks
- From: Douglas A. Gwyn
- OpenSSL Hacks
- Prev by Date: Re: RSA: more than one secret exponent d exists ???
- Next by Date: Re: RSA: more than one secret exponent d exists ???
- Previous by thread: Re: OpenSSL Hacks
- Next by thread: Re: OpenSSL Hacks
- Index(es):
Relevant Pages
|