On shutting down UDP socket [Re: svn commit: r285910 - in head: lib/libc/sys sys/kern sys/sys]

[Maxim Sobolev]

JFYI. I've opened a follow-up differential for this potential regression:
https://reviews.freebsd.org/D10351

Thanks!

-Max

On Mon, Apr 10, 2017 at 7:43 AM, Maxim Sobolev <sobomax at freebsd.org> wrote:

> Hi Guys, I am sorry to bring this old thread up, but I think Ed's
> comparison with other OSes here and in the relevant differential was not
> entirely correct. What linux does (tested with 4.4.0) when UDP socket is
> shut down is actually shutting down receiving end, so any threads that are
> blocked in recv() on that socket return. Still shutdown() system call
> itself returns ENOTCONN. FreeBSD on the other hand does not do anything for
> the socket, so that the threads just hang. I am pretty sure there are at
> least some software out there that relies on that behavior, at least in our
> case we do. Bumped into this after upgrading to the 11.0.
>
> Therefore, I am curious about possibility to make our behavior match that
> of Linux's, so we are not the odd one with regards to this, that is return
> an error but still shutdown the socket?
>
> Small test case is attached. Both FreeBSD 10.3 and Linux 4.4.0 pass
> (albeit Linux's shutdown() returns with an error), FreeBSD 11.0 fails.
>
> -Max
>
> On Sun, Aug 9, 2015 at 6:08 AM, Ed Schouten <ed at nuxi.nl> wrote:
>
>> Hi Alexander,
>>
>> 2015-08-09 14:55 GMT+02:00 Alexander Kabaev <kabaev at gmail.com>:
>> > On Sun, 9 Aug 2015 09:37:13 +0200
>> > It most definitely does work, this is what I have done to get my
>> > network scripts work again. I wonder if there are other means of
>> > restricting raw sockets that can be used to achieve the result
>> > authors of rtsold had hoped or?
>>
>> Yes, there sure are. We could for example call cap_rights_limit() on
>> the socket and whitelist the exacty set of actions that the program
>> needs.
>>
>> That said, it wouldn't make a difference in the end. It looks like
>> rtsol/rtsold don't seem to drop any privileges or switch credentials
>> after startup, assuming I haven't overlooked anything. Even if we were
>> to restrict the raw socket, the process could always open a new one
>> later on.
>>
>> I think it would make sense for now to just commit the patch that I
>> proposed. Will push it into the tree tomorrow.
>>
>> Thanks,
>> --
>> Ed Schouten <ed at nuxi.nl>
>> Nuxi, 's-Hertogenbosch, the Netherlands
>> KvK/VAT number: 62051717
>>
>>
>