svn commit: r315656 - head/tests/sys/netinet

[Alan Somers]

On Mon, Mar 20, 2017 at 6:55 PM, Ngie Cooper <yaneurabeya at gmail.com> wrote:
> On Mon, Mar 20, 2017 at 4:07 PM, Alan Somers <asomers at freebsd.org> wrote:
>> Author: asomers
>> Date: Mon Mar 20 23:07:34 2017
>> New Revision: 315656
>> URL: https://svnweb.freebsd.org/changeset/base/315656
>>
>> Log:
>>   Fix back-to-back runs of sys/netinet/fibs_test;slaac_on_nondefault_fib6
>>
>>   This test was failing if run twice because rtadvd takes too long to die.
>>   The rtadvd process from the first run was still running when the
>>   second run created its interfaces.  The solution is to use SIGKILL during
>>   the cleanup instead of SIGTERM so rtadvd will die faster.
>>
>>   While I'm here, randomize the addresses used for the test, which makes bugs
>>   like this easier to spot, and fix the cleanup order to be the opposite of
>>   the setup order
>>
>>   PR:           217871
>>   MFC after:    18 days
>>   X-MFC-With:   315458
>>   Sponsored by: Spectra Logic Corp
>
> Hi Alan,
>     I wonder if another seatbelt needs to be added for this scenario
> because running a separate copy of rtadvd could adversely affect the
> running system behavior...
> Thanks,
> -Ngie

I thought about that, but I decided against it, because rtadvd(8)
requires you to specify which interfaces it should operate on, either
on the command line or with rtadvctl(8).  This test always creates a
new interface and only runs rtadvd on the new interface.  Also,
rtadvd(8) allows you to override the default pidfile, configfile, and
socket file, which this test does.  So I don't think that the test
rtadvd and the system rtadvd can ever affect each other.

-Alan