Thayne Harbaugh <[EMAIL PROTECTED]> writes:
> More specificly, I notice that lshd exits with the following error:
Thanks for the bug reports. I'll look into it in the week.
> I'm also wondering what's up with unix_random.c. Why is there so much
> effort exerted trying to stumble through vmstat, ps, netstat and the
> like
> on machines that have a good /dev/random? Will there be a compile
> time option for using /dev/random instead of the current functionality?
Some reasons for not making the new code optional are
1. It doesn't hurt to use more input for the generator.
2. If the code is used everywhere, it is more likely that bugs are
noticed and fixed.
3. It's not entirely trivial to figure out if a /dev/urandom that
happens to exist in the file system really is good. In particular
not for the configure script.
If you really need a compile time option, I could add one. What should
it be called?
/Niels
