[ovs-dev] [RFC] Extremely crude conntrack resubmit test case

Russell Bryant russell at ovn.org
Mon Nov 23 18:36:25 UTC 2015


On 11/23/2015 01:27 PM, Ben Pfaff wrote:
> On Thu, Nov 05, 2015 at 09:06:32PM -0500, Russell Bryant wrote:
>> This patch includes a really crude test case demonstrating the problem
>> I'm seeing with resubmitting to a table that does ct() multiple times.
>> It only seems to work on the first resubmit.
>>
>> The test case is REALLY crude, but I think it shows the issue.  The test
>> case will always fail as written so I can look at the flows.  I run this
>> with:
>>
>>   $ sudo make check-kmod TESTSUITEFLAGS="-k resubmit -v"
> 
> Trying to catch up on email.  I think this was fixed by:
> 
> commit e37b8437e915a02a88116e5ea7af1a7e716bd597
> Author: Joe Stringer <joestringer at nicira.com>
> Date:   Fri Nov 6 16:16:47 2015 -0800
> 
>     ofproto-dpif-xlate: Don't stop processing after ct.
>     
>     If conntrack recirculates, it should not stop processing the current
>     pipeline. The cloned packet will begin processing in the table specified
>     with the current metadata and action set; The current copy of the packet
>     will continue processing, including to return back to prior resubmit()
>     calls.
>     
>     Reported-by: Russell Bryant <rbryant at redhat.com>
>     Signed-off-by: Joe Stringer <joestringer at nicira.com>
>     Acked-by: Jarno Rajahalme <jrajahalme at nicira.com>
> 
> Let me know if I'm wrong and I'll investigate.

Correct, and a cleaned up version of the test case went in after it:

https://github.com/openvswitch/ovs/commit/c4e34c6114bcf4cf9248fe910ae8f202b6293f40

-- 
Russell Bryant



More information about the dev mailing list