Difference between revisions of "VPP/HostStack"

From fd.io
< VPP
Jump to: navigation, search
(→‎Tutorials/Test Apps)
(→‎Getting Started)
Line 39: Line 39:
  
 
For more details on how to further configure the client/server apps to do throughput and CPS testing see [[VPP/HostStack/EchoClientServer | here]]
 
For more details on how to further configure the client/server apps to do throughput and CPS testing see [[VPP/HostStack/EchoClientServer | here]]
−
 
−
=== External Echo Server/Client (vpp_echo) ===
 
−
 
−
The VPP external echo server/client test application (vpp_echo), is a bespoke performance test application which supports all of the VPP HostStack transport layer protocols:
 
−
* tcp
 
−
* udp
 
−
* quic 
 
−
Vpp_echo utilizes the 'native HostStack APIs' to verify performance and correct handling of connection/stream events with uni-directional and bi-directional streams of data.
 
−
It can be found in the build tree at
 
−
 
−
<code>$ ./build-root/build-vpp[_debug]-native/vpp/bin/vpp_echo</code>
 
−
 
−
and as of VPP 20.01 is included in the main vpp debian package (e.g. vpp_20.01-release-amd64.deb)
 
−
 
−
Start vpp1 and attach the server application:
 
−
 
−
<code>$ ./build-root/build-vpp_debug-native/vpp/bin/vpp_echo uri tcp://vpp1_ip/port</code>
 
−
 
−
Then start vpp2 and attach the client:
 
−
 
−
<code>$ ./build-root/build-vpp_debug-native/vpp/bin/vpp_echo client uri tcp://vpp1_ip/port</code>
 
−
 
−
The current command line options for vpp_echo are:
 
−
 
−
<code>
 
−
Usage: vpp_echo [socket-name SOCKET] [client|server] [uri URI] [OPTIONS]
 
−
Generates traffic and assert correct teardown of the hoststack
 
−
 
−
  socket-name PATH    Specify the binary socket path to connect to VPP
 
−
  use-svm-api        Use SVM API to connect to VPP
 
−
  test-bytes[:assert] Check data correctness when receiving (assert fails on first error)
 
−
  fifo-size N[K|M|G]  Use N[K|M|G] fifos
 
−
  mq-size N          Use mq with N slots for [vpp_echo->vpp] communication
 
−
  max-sim-connects N  Do not allow more than N mq events inflight
 
−
  rx-buf N[K|M|G]    Use N[Kb|Mb|GB] RX buffer
 
−
  tx-buf N[K|M|G]    Use N[Kb|Mb|GB] TX test buffer
 
−
  appns NAMESPACE    Use the namespace NAMESPACE
 
−
  all-scope          all-scope option
 
−
  local-scope        local-scope option
 
−
  global-scope        global-scope option
 
−
  secret SECRET      set namespace secret
 
−
  chroot prefix PATH  Use PATH as memory root path
 
−
  sclose=[Y|N|W]      When stream is done, send[Y]|nop[N]|wait[W] for close
 
−
  nuris N            Cycle through N consecutive (src&dst) ips when creating connections
 
−
  lcl IP              Set the local ip to use as a client (use with nuris to set first src ip) 
 
−
 
−
  time START:END      Time between evts START & END, events being :
 
−
                      start - Start of the app
 
−
                      qconnect    - first Connection connect sent
 
−
                      qconnected  - last Connection connected
 
−
                      sconnect    - first Stream connect sent
 
−
                      sconnected  - last Stream got connected
 
−
                      lastbyte    - Last expected byte received
 
−
                      exit        - Exiting of the app
 
−
  rx-results-diff    Rx results different to pass test
 
−
  tx-results-diff    Tx results different to pass test
 
−
  json                Output global stats in json
 
−
  stats N            Output stats evry N secs
 
−
  log=N              Set the log level to [0: no output, 1:errors, 2:log]
 
−
  crypto [engine]    Set the crypto engine [openssl, vpp, picotls, mbedtls] 
 
−
 
−
  nclients N          Open N clients sending data
 
−
  nthreads N          Use N busy loop threads for data [in addition to main & msg queue]
 
−
  TX=1337[K|M|G]|RX  Send 1337 [K|M|G]bytes, use TX=RX to reflect the data
 
−
  RX=1337[K|M|G]      Expect 1337 [K|M|G]bytes
 
−
 
−
-- QUIC specific options --
 
−
  quic-setup OPT      OPT=serverstream : Client open N connections.
 
−
                        On each one server opens M streams
 
−
                      OPT=default : Client open N connections.
 
−
                        On each one client opens M streams
 
−
  qclose=[Y|N|W]      When connection is done send[Y]|nop[N]|wait[W] for close
 
−
  uni                Use unidirectional streams
 
−
 
−
  quic-streams N      Open N QUIC streams (defaults to 1)
 
−
 
−
Default configuration is :
 
−
  server nclients 1 [quic-streams 1] RX=64Kb TX=RX
 
−
  client nclients 1 [quic-streams 1] RX=64Kb TX=64Kb
 
−
</code>
 
  
 
=== VCL socket client/server ===
 
=== VCL socket client/server ===

Revision as of 15:36, 4 November 2020

Description

VPP's host stack is a user space implementation of a number of transport, session and application layer protocols that leverages VPP's existing protocol stack. It roughly consists of four major components:

  • Session Layer that accepts pluggable transport protocols
  • Shared memory mechanisms for pushing data between VPP and applications
  • Transport protocol implementations (e.g. TCP, QUIC, TLS, UDP)
  • Comms Library (VCL) and LD_PRELOAD Library
VPP HostStack Arch

Documentation

Set Up Dev Environment - Explains how to set up a VPP development environment and the requirements for using the build tools

Session Layer Architecture - Goes over the main features of the session layer

TLS Application - Describes the TLS Application Layer protocol implementation

Getting Started

Applications can link against the following APIs for host-stack service:

  • Builtin C API. It can only be used by applications hosted within VPP
  • "Raw" session layer API. It does not offer any support for async communication
  • VCL API that offers a POSIX-like interface. It comes with its own epoll implementation.
  • POSIX API through LD_PRELOAD

A number of test applications can be used to exercise these APIs. For all the examples below, it is assumed that two VPP instances have been brought up and properly configured to ensure networking connectivity between them. To test that network connectivity is available, the builtin ping tool can be used. As a convention, we consider the first vpp instance (vpp1) to be the one the server is attached to and the second instance (vpp2) to be the one where the client application is attached. For illustrative purposes all examples use TCP as a transport protocol but other available protocols could be used.

Builtin Echo Server/Client

On vpp1, from the cli do: 

# test echo server uri tcp://vpp1_ip/port

and on vpp2: 

# test echo client uri tcp://vpp1_ip/port

For more details on how to further configure the client/server apps to do throughput and CPS testing see here

VCL socket client/server

For more details see the tutorial here

Tutorials/Test Apps

Test Builtin Echo Client/Server Apps

Test External Echo Client/Server Apps

Test HTTP Server App

Test Proxy App

iperf3 with LD_PRELOAD

sshd with LD_PRELOAD

nginx with LD_PRELOAD

Running List of Presentations

Retrieved from "https://wiki.fd.io/index.php?title=VPP/HostStack&oldid=10903"