Hi Jinlei,
Could you please send me a pcap capture of the grat-arp that VPP receives. Then
I can duplicate the case in the UT.
Thanks,
neale
From: on behalf of Jinlei Li
Date: Saturday 19 September 2020 at 11:56
To: "vpp-dev@lists.fd.io"
Subject: [vpp-dev] The gratuitous ARP issue
Hi guys,
Hello,
People may be interested in knowing that Calico/VPP 0.8.1 is available here:
https://github.com/projectcalico/vpp-dataplane
It now supports native VPP AF_XDP integration for smoother linux integration.
It requires a recent kernel though.
Other than that, it also improves support for ICMP
Folks,
I am in the process of deploying new docker images produced via
automation that I have built [0]. In the README.md file, you will find
documentation of the new docker hub repos and tags as well as the
corresponding Jenkins labels.
While making changes to deploy the new labels and
Thank you! I will do further tests.
At 2020-09-21 12:37:50, "Florin Coras" wrote:
Hi,
TCP coalesces both when sending and when receiving. If the sender really sends
only every second a few bytes, those should be the only ones delivered,
assuming the receiver reads the data as soon as it
Hi,
As stated in the title, I am wondering if VPP stamps every packet coming in and
right before it is TX'd out? If so, where can I find this in the code base?
Thanks,
Mauricio
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#17465):
On Fri, Sep 18, 2020 at 08:06 AM, Benoit Ganne (bganne) wrote:
>
> I am trying to report the hardware interface queue depth.
As in, TX interface FIFO/Queue occupancy.
>
> You can use vnet_get_sup_hw_interface() for this, however HW queues are
> managed differently by each driver.
There is
What is the OpenSSL version you use?
> On 21 Sep 2020, at 14:55, Ray Kinsella wrote:
>
> Anybody else seeing this build failure.
>
> [17/496] Building C object
> plugins/crypto_openssl/CMakeFiles/crypto_openssl_plugin.dir/main.c.o
> FAILED:
> On 21 Sep 2020, at 08:31, Solis JR, M. (Mauricio) via lists.fd.io
> wrote:
>
> Hi,
> As stated in the title, I am wondering if VPP stamps every packet coming in
> and right before it is TX'd out? If so, where can I find this in the code
> base?
no, it doesn’t.
-=-=-=-=-=-=-=-=-=-=-=-
Using 1:1.1.1d-2.fc31
Upgrade to 1:1.1.1g-1.fc31 is available.
Ray K
On 21/09/2020 14:37, Damjan Marion via lists.fd.io wrote:
>
> What is the OpenSSL version you use?
>
>> On 21 Sep 2020, at 14:55, Ray Kinsella wrote:
>>
>> Anybody else seeing this build failure.
>>
>> [17/496] Building C
If not, would adding a patch to time stamp the packets at ethernet-input as a
starting point make sense? What I am trying to do is collect the time it takes
for the packet to traverse VPP. I'm using this and placing it in the iOAM
telemetry data. The ending time stamp would then be somewhere
Dave,
Good for you.
Let me take this opportunity to say Thank You!
for your wise stewardship and wry sense of humour over the past few years of
the FD.io project,
for your bootstrapping of multiple development efforts, like the FD.io VPP TCP
Stack with Florin and Team, static analysis, too
Anybody else seeing this build failure.
[17/496] Building C object
plugins/crypto_openssl/CMakeFiles/crypto_openssl_plugin.dir/main.c.o
FAILED: plugins/crypto_openssl/CMakeFiles/crypto_openssl_plugin.dir/main.c.o
ccache /usr/bin/clang-9 --target=x86_64-linux-gnu -DHAVE_FCNTL64 -DHAVE_GETCPU
> On 21 Sep 2020, at 14:39, Solis JR, M. (Mauricio) via lists.fd.io
> wrote:
>
> If not, would adding a patch to time stamp the packets at ethernet-input as a
> starting point make sense?
no, as it cost cycles, and majority of users doesn’t need that data.
You can always develop smal plugin
13 matches
Mail list logo