Cool, also would be glad to hear from the community :p
zhang...@yunshan.net.cn
发件人： 张东亚
发送时间： 2017-06-22 13:30
收件人： zhang...@yunshan.net.cn
抄送： 雷彦章; vpp-dev
主题： Re: Re: [vpp-dev]答复: 【vpp-dev】delay is error in ping with multi worker
thread
Hi zhangpan,
Glad to here feedback, as I written in m
Hi zhangpan,
Glad to here feedback, as I written in my previous mail, after we modify
the code to using RPC mechanism, ping works fine but we have random 10ms
issue while our normal ping delay is <1ms, we think the reason it due to
schedule delay in the main thread but we want the community to con
I didn't investigate the code but there's an assumption I would like to share.
I encountered the same problem before, but what interested me is that I found
if you send multiple ping pkts as fast as you can, the problem went away.
I tried different numbers and if it's a multiple of 4, VPP works j
Hi List,
I am colleague of Leiyanzhang, let me describe share the issues and
investigations found when we test the ping command under multiple worker
thread case.
Stage1:
Issue: Ping does not work under multiple worker thread case.
Version: Vanilla vpp 17.04
(vanilla vp)
Stage2:
Issue: Ping go
Hi All,
I want to collect flow information(src IP -> dst IP) at L2 interface.
But it looks like flowprobe does not collect flow, if it doesn't have IP
address.
For example,
flowprobe params record l3 active 2
flowprobe feature add-del TengigabitEthernet4/0/0 ip4
"show flowprobe table" show n
In interesting news, Sander Stefann the author of DHCPKit has integrated his
software with VPP.
http://dhcpkit-vpp.readthedocs.io/en/stable/
http://dhcpkit.readthedocs.io/en/stable/
It depends on https://gerrit.fd.io/r/#/c/6721/ for sending and receiving
packets.
(I will clean that one up and c
Hi Vladimir,
> On 21 Jun 2017, at 15:41, Vladimir Torgovitsky
> wrote:
>
> Hi,
> I'm testing performance of different NICs and have an issue with Mellanox 40G
> ConnectX-4 device, where VPP performance seem to be similar to Linux and
> doesn't improve.
> I am testing throughput of UDP traffic
I just submitted latest revision of new vppctl implementation.
https://gerrit.fd.io/r/7230
I would like to hear feedback. It can be simply tested with:
make build
make run
and in other window (without sudo):
build-root/install-vpp_debug-native/vpp/bin/vppctl
Additional startup conf is:
unix
Hello Anton,
We did some checks and here are results:
1. ping simulated node from the host itself - ping is OK
2. ping simulated node from other host (i.e. node simulated on virl2, executing
ping command on virl3) - discovered packet loss (see e-mail from Peter below)
- even for successful
Hello vpp-dev team,
Because the root cause of the packet loss on virl1 and virl2 is not known up to
now we decided to put virl1 out of production to not block vpp patches because
of ssh timeouts during vpp-csit-verify tests.
It means that only virl3 server is in production so we can expect some
Hello vpp-dev team,
Because the root cause of the packet loss on virl1 and virl2 is not known up to
now we decided to put virl1 out of production to not block vpp patches because
of ssh timeouts during vpp-csit-verify tests.
It means that only virl3 server is in production so we can expect some
Hi,
I'm testing performance of different NICs and have an issue with Mellanox 40G
ConnectX-4 device, where VPP performance seem to be similar to Linux and
doesn't improve.
I am testing throughput of UDP traffic, packets of size 64B, 128B and 1518B, on
single core.
Two questions:
1. Is there
Hi Chris,
Thanks a lot for the help and clarifying the same.
Sure yes I have not enabled the dpdk plugin while the testing .
Thanks & Regards,
Sreejith
On 21 June 2017 at 18:18, Luke, Chris wrote:
> Do you have the DPDK plugin loaded? It looks like “no”.
>
>
>
> If not, it’s likely because the
Do you have the DPDK plugin loaded? It looks like “no”.
If not, it’s likely because the buffer allocator used when DPDK is not loaded
is not yet thread-safe. I’ve done some early work on this but it’s not going to
make it into the open for a while.
For now, don’t enable worker threads if
Thank you,
In VPP, can we limit the packets across different interfaces. If we can
limit the packets which all the API's are responsible please mention to me.
On Monday 05 June 2017 11:26 PM, Dave Barach (dbarach) wrote:
I stand corrected, at least in a general sense... (😊)...
Thanks… Dave
Hi Dharmaray,
The short answer is that ip4-lookup performs the lookup in a FIB table of the
packet’s destination address, the result of which is an adjacency. The
ip4-rewrite node applies the ‘L2’ encap of an adjacency and transmits the
packet on the output port*. So typically packets will go f
Yeah, that's exactly what is going on. There is a magic value which
was due to be exterminated but I wanted to do it for both L2 and L3 -
and like I said the L2 path needs more thinking, so I did not get
around to it yet...
Could you adjust your local code according to this change:
https://gerrit.
BFD UDP code is interesting example.
I was looking at bfd_udp_input rather and one question in relation to this.
Though both ' bfd_udp_input ()' and ' bfd_send_periodic()' form the IP and UDP
headers using ' bfd_add_transport_layer', they pass the packet to different
next nodes.
bfd_udp_input us
Hi,
> I want to write some VPP API and I decide to create separate directory for
> this APIs.
> Which libraries are required to register this API?
> What is the right way to add and register APIs?
> How to add this directory to Makefile?
Take a look at src/examples/sample-plugin
Typically the A
without vpp-844, the ping cmd not work.the latest master has already have the
patch
now is the time delay value is not right.
-- Original --
From: "wang.hui56";
Date: Wed, Jun 21, 2017 05:19 PM
To: "leiyanzhang";
Cc: "vpp-dev";
Subject: 答复: [vpp-dev] 【v
Hi
I want to write some VPP API and I decide to create separate directory for
this APIs.
Which libraries are required to register this API?
What is the right way to add and register APIs?
How to add this directory to Makefile?
Thanks.
___
vpp-dev mailing
Hi All,
when we try to use ping cmd with multi woker thread, we found sometimes the
ping delay is nearly 1000ms eventhough the two site is directly linked to each
other.
we found in the function signal_ip46_icmp_replay_event, it used
vlib_process_signal_event to notify the ping_response msg.
Hi All,
when we try to use ping cmd with multi woker thread, we found sometimes the
ping delay is nearly 1000ms eventhough the two site is directly linked to each
other.
we found in the function signal_ip46_icmp_replay_event, it used
vlib_process_signal_event to notify the ping_response msg.
but
Denis,
> PPTP connection working well via Hairpin NAT 1:1.
Good to hear!
Now if I could ask you for one favour to settle if we need an ALG or not.
Could you analyse all your PPTP sessions (or all sessions that aren't TCP/UDP
for that matter) and determine the total number of concurrent session
24 matches
Mail list logo
