Hi, We are planning to use VPP with DPDK and trying to understand the processing of packet which is received for a destination for which no entry is present in L2 table. Does VPP will queue this packet and perform the ARP query to update its L2 table or it will forward the packet to kernel f
Please clarify whether your question concerns bridging or routing.
If you configure interfaces in L2 mode - bridging - packets sent to unknown MAC
addresses will be flooded across interfaces in the bridge.
In L3 mode, vpp sends ARP requests.
In neither case is the linux kernel involved.
Thanks
Hi Dave,
Thanks for the reply.
I want to know about the L3 mode handling when no MAC address is configured. In the VPP source code, the code to send the ARP request/ ARP response is present, however my question is that what will happen to that data packet.
Do VPP will queue it up till
If VPP receive a packet for which it needs to send an ARP request to resolve
the MAC for its DIP, the ARP request will be sent and the current packet will
be dropped. If an ARP response is received for the DIP of the original packet,
then the following packets received with the same DIP will be
For visibility, updated RH 2544 testing results with 16.12.rc0
This is PVP testing and shows up vhost-user perf compared with OVS/DPDK
Once we have vhost-user merged and tested in CSIT I would expect to see
significant improvement.
--TFH
On 11/03/2016 02:51 PM, Karl Rister wrote:
Hi All
Be
Hi, We are planning to use VPP with DPDK and trying to understand the processing of packet which is received for a destination for which no entry is present in L2 table. Does VPP will queue this packet and perform the ARP query to update its L2 table or it will forward the packet to kernel for
On 11/03/2016 01:07 PM, Edward Warnicke wrote:
At the time we laid the current tag for master, we thought we were
doing a 16.12 release, so the
tag was laid as:
v16.12-rc0
The result is that all of our merge by merge patches for master are
being built as 16.12-rc0 packages.
I would like t
Vpp converts transit-switched packets which hit on an ARP adjacencies to ARP
requests, and sends them.
This is the traditional - and only sensible - solution to the problem.
Thanks… Dave
From: Prerit Jain [mailto:prerit.j...@samsung.com]
Sent: Friday, November 4, 2016 8:55 AM
To: Dave Barach (d
Looping in vpp-dev (as they have more context on your question about vpp
arp handling).
Short answer though... vpp does not use the kernel for arp handling, it has
its own graph nodes for that, and its own arp tables.
Ed
On Thu, Nov 3, 2016 at 9:58 PM, Prerit Jain wrote:
> Hi,
>
>
>
> We are p
Hi all,
I am currently trying to build a fresh Vagrant VM within VPP (for testing
purposes) but during the install I encounter this error..
==> default: Processing triggers for libgdk-pixbuf2.0-0:amd64
(2.30.7-0ubuntu1.6) ...
==> default: Errors were encountered while processing:
==> default:
Yi,
Putting the list back on.
I don’t see significant justification from you to do a respin.
i.e. “NSH doesn’t work” (sic ODL SFC) isn’t enough justification. Is ODL SFC
looking to do a respin?
How come this wasn’t caught sooner, as in, we all got feedback that ODL SFC ->
HC -> VPP + NSH_SFC
Sounds like your vagrant box has an altered /etc/sudoers in there prior to you
booting it and the bootstrap tried to upgrade sudo to a newer version causing
dpkg to burp on it. (read: whoever made the box hasn’t heard of /etc/sudoers.d
which was invented for this sort of thing)
There’s two ways
12 matches
Mail list logo
