FIPS mode confusion, and resolution

2006-02-10 Thread Kyle Hamilton
I have a slightly non-standard (and non-security-policy-compliant) FIPS mode setup, generated with: cvs -d/home/kyle/openssl/repository co -rOpenSSL_FIPS_1_0 \ -d /home/kyle/openssl/work openssl ./config fips --prefix=/home/kyle make make test make install (the 'non-security-policy-compliant'

Re: FIPS mode confusion, and resolution

2006-02-10 Thread Dr. Stephen Henson
On Fri, Feb 10, 2006, Kyle Hamilton wrote: (the 'non-security-policy-compliant' part comes from the extra parameter to ./config -- I can't touch the OpenSSL build that's already in the system directories, and the security policy states that no other parameters can be passed to ./config. [if

[openssl.org #1281] Solaris9: 'make test' failure in evp_test

2006-02-10 Thread [EMAIL PROTECTED] via RT
attached you can find the 'make report' output from one of my Solaris9 maschines. I used the SunStudio11 compiler on Solaris9 (SUN Fire V890) Is there any fix available for this problem ? thanks in advance ! regards, Steffen -- Steffen Unger [EMAIL PROTECTED]

Re: [openssl.org #1282] error setting AuthorityKeyIdentifier

2006-02-10 Thread Dr. Stephen Henson
On Fri, Feb 10, 2006, Tim Bond via RT wrote: I am doing some interop testing with a toolkit that performs PKIX certificate verification and it is having a problem validating a chain I built with OpenSSL. What appears to be happening is that when 'ca' copies in the authority key information

Re: [openssl.org #1282] error setting AuthorityKeyIdentifier

2006-02-10 Thread Kyle Hamilton
I haven't checked the archives, but if I'm not mistaken, it's because it's (presumably) the rootCA that is the original trusted authority (the 'trust anchor'), and thus the authorityKeyIdentifier is the anchor rather than the CA that derives its trust from the anchor? (Also: if the question has

Re: [openssl.org #1282] error setting AuthorityKeyIdentifier

2006-02-10 Thread Dr. Stephen Henson
On Fri, Feb 10, 2006, Kyle Hamilton wrote: I haven't checked the archives, but if I'm not mistaken, it's because it's (presumably) the rootCA that is the original trusted authority (the 'trust anchor'), and thus the authorityKeyIdentifier is the anchor rather than the CA that derives its