[ovs-dev] [PATCH v2 0/6] Add OVS DPDK keep-alive functionality

Bodireddy, Bhanuprakash bhanuprakash.bodireddy at intel.com
Mon May 1 23:03:24 UTC 2017


>There two ways to go with the design.
>
>1) Make it generic, so that it is not so PMD specific, as it is now.
>2) If it stays PMD specific, make it stronger; right now, the health check is
>limited – it detects that a PMD thread is proceeding or not.
>For something like DPDK, I don’t think that will be enough in the long run.
>This can result in some false negatives, as well.
>              Maybe, we want to know that the ports and queues are getting
>processed, PMD/port/queue mappings as expected, time spent processing
>packets per PMD, port state changes, packet stats, queue depths, etc This
>information could be correlated by the final receiver of the data.

Darrell,

Thanks for the feedback. I am implementing this for OvS-DPDK and so it would be PMD specific at this point.
However I completely agree with your suggestions here and would try extending the monitoring to factor in ports & packet stats and other parameters to make sure there aren't any false negatives.

>
>I also agree that socket communication is preferred over shm, although I don’t
>think any shm usage will necessarily lead to a meltdown.

As the consensus is to have Unix domain sockets over SHM,  I will make necessary changes.

Regards,
Bhanuprakash. 



More information about the dev mailing list