Michael Ströder wrote, On 2008-08-06 04:07:
> Nelson B Bolyard wrote:
>>> cmsutil -D -d ~/.mozilla/xxxxxxx/ -c name.tar.gz -i name.tar.gz.p7m -o test
>> I remember running into this long ago.  As I recall, the pass/fail result
>> is very subtle.  It may be nothing more than the program's result code.
>>
>> What did you get in the "test" file? 
> 
> It's the same file (here name.tar.gz) like given with -c.

identical?  same length and sum?  Nothing extra on the beginning or end?

>> Is the pass/fail indication there?
> 
> Nope. The file given with -o seems to be the "decoded" file.
> 
> If I invoke cmsutil with a wrong input file I get the following message:
> ------------------------------ snip ------------------------------
> signer 0 status = DigestMismatch
> cmsutil: problem decoding: Signature verification failed: no signer 
> found, too many signers found, or improper or corrupted data.
> ------------------------------ snip ------------------------------

OK, so the failure result is verbose and explicit, and the success result
is rather terse (:-).

Did the -v option improve that any?

> Strange enough this works as expected giving correct results:
> 
> signver -V -v -d ~/.mozilla/xxxxxxx/ -i name.tar.gz < name.tar.gz.p7m

It doesn't surprise me that that works.  I am surprised that the other
command fails in the fashion you've documented.  Looking at the NSS source
code I see no way for it to open the file named with the -s option for
output (writing), yet your strace results show that it did.  This makes
me wonder if the program you have was built from official NSS sources,
or if someone has modified the sources from which the distribution you
used was built.  :-(

I wish I could point you to a distribution that I know was built from
Mozilla's sources, unmodified, but alas, I don't know of any such for 3.12.

The binaries for the NSS 3.11.4 release may be obtained from
ftp://ftp.mozilla.org/pub/security/nss/releases/NSS_3_11_4_RTM/
If the -s option also behaves as you found with those binaries, I'd like
to know that.
_______________________________________________
dev-tech-crypto mailing list
dev-tech-crypto@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-tech-crypto

Reply via email to