Re: FIPS seed_pid issue

2012-10-16 Thread Kumar Ghanta
Can somebody please respond to my question? OpenSSH is aborting if 1.1 fips module is used.Here my question is whether we need to comment this logic(by compiling with GETPID_IS_MEANINGLESS) or we need to generate seed again for child process. I want to know under what circumstances we can build

Re: FIPS seed_pid issue

2012-10-16 Thread Dr. Stephen Henson
On Sun, Oct 14, 2012, Kumar Ghanta wrote: Thank you very much for the quick response Stephen. Is it fine if we allow parent and child processes to share the same seed? I just want to know if there are any NIST restrictions. If possible, can you please elaborate on how does openssl takes care

Re: FIPS seed_pid issue

2012-10-14 Thread Kumar Ghanta
Thank you very much for the quick response Stephen. Is it fine if we allow parent and child processes to share the same seed? I just want to know if there are any NIST restrictions. If possible, can you please elaborate on how does openssl takes care automatically after 1.2? On Fri, Oct 12, 2012

Re: FIPS seed_pid issue

2012-10-12 Thread Dr. Stephen Henson
On Fri, Oct 12, 2012, Kumar Ghanta wrote: Hi, Earlier versions of openssl-fips (versions 1.1.2 etc) have the following checks in the fips_rand.c. It looks this check is being removed in the later versions. I just want to know whether we need this check in earlier versions as per the NIST