Hi Zhuo,
Yes, that is the scenario memif was build for.
Thanks,
Damjan
On 28 Mar 2018, at 01:59, Zhuo Chen (zchen)
mailto:zc...@cisco.com>> wrote:
Hi experts,
Has anyone had one application container work with vpp in a separate
container over memif? For example
https://github.com/matfa
Did you checkout the latest vppsb? There should be a patch merged to fix
this issue.
2018-03-30 3:25 GMT-04:00 Hamid Rasool :
> Hi,
>
> I want to enable dynamic routing protocols (using Quagga or FRR as VPP
> does not have BGP and IS-IS built-in) to perform routing on my VPP VM
> (built using sta
Hi,
"tx frame not ready"? means VPP didn't find empty queue at linux side. Linux is
processing too slow which results in that behavior and VPP drops those packets.
While when you use dpdk, you bypass linux kernel (bottleneck), so you don't
find such behavior.
If you still want to use the vi
hi, i used VPP+afsocket to run ip4 forward case between pcA and pcB. After i
used iperf to send 400Mb/s 64B packets，the iperf server did not show the
received bandwidth. So i inputed show error，the result showed many packets were
not handled normally because of tx frame not ready. when i used dp
hi, i used VPP+afsocket to run ip4 forward case between pcA and pcB. After i
used iperf to send 400Mb/s 64B packets，the iperf server did not show the
received bandwidth. So i inputed show error，the result showed many packets were
not handled normally because of tx frame not ready. when i used dp
On Thu, Mar 29, 2018 at 2:58 PM, Jon Loeliger wrote:
> Matus, et al,
>
> Is there no way to remove either the B4 or AFTR tunnel endpoint once it is
> set?
>
So, I see one can use "::". Is that the expectation here?
jdl
Ok. Thank you.
Seems 1kbps steps looks fine for me too.
Regards,
--Dmitry
On Thu, Mar 29, 2018 at 10:46 PM, Damjan Marion
wrote:
>
> Yes, but too late for 18.04. We can do that change later after rc1 is out.
> I personally prefer that we simply keep speed as numeric value in 1kbps
> steps.
>
