> On 7 May 2026, at 10:33, Dumitru Ceara <[email protected]> wrote:
> 
> On 5/6/26 10:09 PM, Martin Kalčok wrote:
>> 
>> 
>>>> On 6 May 2026, at 10:11, Dumitru Ceara <[email protected]> wrote:
>>> 
>>> On 5/5/26 11:28 PM, Frode Nordahl wrote:
>>>> On 5/4/26 14:27, Dumitru Ceara via dev wrote:
>>>>> On 4/23/26 6:34 PM, Martin Kalcok wrote:
>>>>>> Commit 985551d "northd: Don't recognize UDP packets as health check
>>>>>> responses."
>>>>>> [0] fixed an issue that caused UDP packets to be delivered to the
>>>>>> controller in response to the health check probe. It also introduced a
>>>>>> standalone test for this use case.
>>>>>> However, the same behavior can be tested by existing
>>>>>> "Load balancer health checks with LRP IP" tests. These tests generated
>>>>>> "unsupported protocol" warning in logs before the introduction of the
>>>>>> fix [0], but the message was explicitly ignored.
>>>>>> Now that the fix [0], is applied, the warning message can be removed
>>>>>> from "ignored warnings" list, essentially accomplishing the same thing
>>>>>> as the standalone test.
>>>>>> 
>>>>>> [0] https://github.com/ovn-org/ovn/
>>>>>> commit/985551d482694e39d51ed290a4607f71e5b65be8
>>>>>> 
>>>>>> Signed-off-by: Martin Kalcok <[email protected]>
>>>>>> ---
>>>>> 
>>>>> Hi Martin,
>>>>> 
>>>>> Thanks for the follow up!  Applied to main and 26.03.
>> 
>> Thanks for the review and apply Dumitru.
>> 
>>>> 
>>>> Thanks for fix, review and merges all.  This means we no longer need to
>>>> do something about LP: #2147468 [1], other than removing the skip.
>> 
>> Hi Frode o/ I’m glad that this bug got caught in the crossfire :) I
>> originally tried to trigger the "UDP nc cleanup issue” in the “/
>> Unsupported protocol message/” test, but I wasn’t successful [2],
>> perhaps there are some more contributing factors that trigger it.
>> 
>> In any case, do you all think that I’d would be worth to create test
>> macro for cleanly starting/stopping UDP service, or is it unlikely that
>> more tests will use it? Currently the “clean way” lives as a "hardcoded
>> hack” in the only two tests that use it [3] with some comments for
>> explanation.
>> 
> 
> Hi Martin,
> 
> Two occurrences is enough to justify a small helper IMO.  I don't know
> if you have time to work on this; if you don't I can add it to my list
> for at some point later.

Hi Dumitru,
I’m happy to take care of this. I likely won’t have time to look at it this 
week, but next week or week after at latest.

Martin.
 
> 
>> [2] https://mail.openvswitch.org/pipermail/ovs-dev/2026-
>> March/431287.html <https://mail.openvswitch.org/pipermail/ovs-dev/2026-
>> March/431287.html>
>> [3] https://github.com/ovn-org/ovn/blob/
>> e6ffc4919c388534bf10067ad563d71f8fbc2fa3/tests/system-ovn.at#L4568-L4580
>> 
>> Martin.
>> 
> 
> 
> Regards,
> Dumitru
> 
>>>> 
>>> 
>>> Hi Frode,
>>> 
>>> Cool, nice to hear that!
>>> 
>>> Regards,
>>> Dumitru
>>> 
>>>> 1: https://launchpad.net/bugs/2147468
>>>> 
>>> 
>> 
> 
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Reply via email to