svn commit: r273872 - in head: etc/defaults etc/rc.d libexec/save-entropy share/examples/kld/random_adaptor sys/conf sys/dev/glxsb sys/dev/random sys/kern sys/modules sys/modules/padlock_rng sys/mo...

[Adrian Chadd]

Hi,

Right - but I'm not even getting to anywhere near this point.

It's hanging in hostapd when hostapd tries reading random data.

again, if i try ctrl-C'ing whilst the random device is inside
random_adaptor_read(), it's just .. broken. The read doesn't finish,
the process doesn't accept any new input, signals or anything - it
just doesn't work.

So, is there some way to go back to whatever the previous revision
before this did? It at least comes up and doesn't hang things.

Thanks,


-adrian

On 1 November 2014 16:57, Ian Lepore <ian at freebsd.org> wrote:
> On Sun, 2014-11-02 at 00:33 +0100, Dag-Erling Smørgrav wrote:
>> Dag-Erling Smørgrav <des at des.no> writes:
>> > From the code.  This is a portion of the rc script that only runs at
>> > shutdown.  If random_stop() runs during boot, something is seriously
>> > wrong *somewhere else* and you are welcome to help find out what.
>>
>> I found it - postrandom triggers it.  For now, just remove
>> /etc/rc.d/postrandom (I'm not sure it's really needed, there are better
>> ways to do this).
>>
>> DES
>
> It appears another quick easy workaround is to set entropy_file=NO in
> rc.conf, which may be the right thing to do for people using a varmfs
> that isn't going to survive a reboot anyway, even after this glitch is
> fixed.
>
> -- Ian
>
>