Hi Martin, Did this OCSP enhancement request ever gain any traction? I am adding X.509 support to an embedded system. Due to limited memory, my product is forced to use OCSP for revocation checking rather than CRLs. This enhancement should benefit anyone building an embedded system or who would like to use OCSP for revocation checking.
I have successfully applied your diff to openssl-0.9.8r, but with a few minor changes. 1) Consider replacing the new check_ocsp_times() with a call to the OCSP_check_validity(). This existing function does much of the same work, with the added benefit of clock skew which is provided by the "nsec" parameter. 2) get_ocsp()'s call to d2i_AUTHORITY_INFO_ACCESS() corrupts certificate memory. See http://www.openssl.org/support/faq.html#PROG3. Consider replacing it with: unsigned char* p; ... p = (unsigned char*)data->data; if (!(aia = d2i_AUTHORITY_INFO_ACCESS(NULL, &p, data->length))) ... 3) check_ocsp() should also check for the id-pkix-ocsp-nocheck extension in the OCSP responder's signing certificate. If present, the CA is stating you can trust a responder for the lifetime of the responder's certificate. Therefore if id-pkix-ocsp-nocheck is present, then we can skip revocation checking of the responder's signing certificate. Finally, thank you for sharing your enhancement request with the openssl-dev community. Sincerely, - Mike Siedzik
|
Hi Martin, Did this OCSP enhancement request ever gain any traction? I am adding X.509 support to an embedded system. Due to limited memory, my product is forced to use OCSP for revocation
checking rather than CRLs. This enhancement should benefit anyone building an embedded system or who would like to use OCSP for revocation checking. I have successfully applied your diff to openssl-0.9.8r, but with a few minor changes. 1) Consider replacing the new check_ocsp_times() with a call to the OCSP_check_validity(). This existing function does much of the same work, with the added benefit of clock
skew which is provided by the “nsec” parameter. 2) get_ocsp()’s call to d2i_AUTHORITY_INFO_ACCESS() corrupts certificate memory. See
http://www.openssl.org/support/faq.html#PROG3. Consider replacing it with: unsigned char* p; ... p = (unsigned char*)data->data; if (!(aia = d2i_AUTHORITY_INFO_ACCESS(NULL, &p, data->length))) ... 3) check_ocsp() should also check for the
id-pkix-ocsp-nocheck extension in the OCSP responder’s signing certificate. If present, the CA is stating you can trust a responder for the lifetime
of the responder’s certificate. Therefore if id-pkix-ocsp-nocheck is present, then we can skip revocation checking of the responder’s signing certificate. Finally, thank you for sharing your enhancement request with the openssl-dev community. Sincerely, - Mike Siedzik |
