network_namespaces

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision		 Previous revision
network_namespaces [2018/08/19 08:51] – [Overview] rpjdaynetwork_namespaces [2018/08/19 16:24] (current) – [Roman Mashak] rpjday
Line 3: Line 3:
 How to use network namespaces and ''ip netns'' to tie two ports together in order to ping out one port and to the other. How to use network namespaces and ''ip netns'' to tie two ports together in order to ping out one port and to the other.
  
-Updated links:+Links:
  
   * [[http://man7.org/linux/man-pages/man8/ip-netns.8.html|ip-netns (8) man page]]   * [[http://man7.org/linux/man-pages/man8/ip-netns.8.html|ip-netns (8) man page]]
   * [[https://blog.scottlowe.org/2013/09/04/introducing-linux-network-namespaces/|Introducing Linux Network Namespaces (Scott Lowe's Blog, Sep 2013)]]   * [[https://blog.scottlowe.org/2013/09/04/introducing-linux-network-namespaces/|Introducing Linux Network Namespaces (Scott Lowe's Blog, Sep 2013)]]
 +  * [[http://www.opencloudblog.com/?p=42|Linux Network Namespaces (Open Cloud Blog, Sep 2013)]]
 +  * [[https://lwn.net/Articles/580893/|Namespaces in operation, part 7: Network namespaces (LWN.net, Jan 2014)]]
 +    * [[https://lwn.net/Articles/531114/#series_index|Full namespace series]]
   * [[https://serverfault.com/questions/568839/linux-network-namespaces-ping-fails-on-specific-veth|Linux network namespaces - ping fails on specific veth (Server Fault, 2014)]]   * [[https://serverfault.com/questions/568839/linux-network-namespaces-ping-fails-on-specific-veth|Linux network namespaces - ping fails on specific veth (Server Fault, 2014)]]
 +  * [[https://www.slideshare.net/MikeWilson35/linux-network-namespaces|Linux network namespaces (Mike Wilson Slideshare (Aug 2015)]]
 +  * [[https://www.dasblinkenlichten.com/an-introduction-to-network-namespaces/|An introduction to network namespaces (Das Blinken Lichten, Oct 2015)]]
   * [[https://blogs.igalia.com/dpino/2016/04/10/network-namespaces/|Network namespaces (Unweaving the web, Apr 2016)]]   * [[https://blogs.igalia.com/dpino/2016/04/10/network-namespaces/|Network namespaces (Unweaving the web, Apr 2016)]]
 +  * [[http://abregman.com/2016/09/29/linux-network-namespace/|Linux: Network Namespace (Arie Bregman, Sep 2016}]]
 +  * [[https://unix.stackexchange.com/questions/391193/how-to-forward-traffic-between-linux-network-namespaces|How to forward traffic between Linux network namespaces? (Stack Exchange, 2017)]]
 +  * [[https://docs.openstack.org/neutron/pike/admin/intro-network-namespaces.html|Network namespaces (OpenStack, Aug 2018)]]
 +  * [[https://docker-k8s-lab.readthedocs.io/en/latest/docker/netns.html|Linux Network Namespace Introduction (Docker Kubernetes Lab)]]
  
-Links: 
- 
-  * [[http://abregman.com/2016/09/29/linux-network-namespace/|Linux: Network Namespace (Arie Bregman, 2016)]] 
-  * [[http://www.opencloudblog.com/?p=42|Linux Networking Namespaces (Open Cloud Blog, 2013)]] 
-  * [[https://lwn.net/Articles/580893/|Namespaces in operation, part 7: Network namespaces (LWN.net, 2014)]] 
-    * [[https://lwn.net/Articles/531114/#series_index|Full namespace series]] 
  
 ===== man ip-netns ===== ===== man ip-netns =====
Line 89: Line 92:
 Namespaces is a good solution. Something like this should work: Namespaces is a good solution. Something like this should work:
  
-ip netns add namespace1 +ip netns add ns1 
-ip netns add namespace2+ip netns add ns2
  
-ip link set eth1 netns namespace1 +ip link set eth1 netns ns1 
-ip link set eth2 netns namespace2+ip link set eth2 netns ns2
  
-ip netns exec namespace1 \+ip netns exec ns1 \
         ip addr add 10.42.42.42/24 dev eth1         ip addr add 10.42.42.42/24 dev eth1
  
-ip netns exec namespace1 \+ip netns exec ns1 \
         ip link set eth1 up         ip link set eth1 up
  
-ip netns exec namespace2 \+ip netns exec ns2 \
         ip addr add 10.42.42.24/24 dev eth2         ip addr add 10.42.42.24/24 dev eth2
  
-ip netns exec namespace2 \+ip netns exec ns2 \
         ip link set eth2 up         ip link set eth2 up
  
-ip netns exec namespace1 \+ip netns exec ns1 \
         ping 10.42.42.24         ping 10.42.42.24
  
Line 120: Line 123:
 the same without involving namespaces. It's a bit hackish but sometimes the same without involving namespaces. It's a bit hackish but sometimes
 convenient. I can dig if someone is interested. convenient. I can dig if someone is interested.
 +</code>
 +
 +==== Roman Mashak ====
 +
 +<code>
 +I used this in the past to test dual-port NIC over loopback cable, you
 +will need to ajust the script:
 +
 +#!/bin/bash -x
 +
 +ip="sudo $HOME/bin/ip"
 +eth1=192.168.2.100
 +eth2=192.168.2.101
 +
 +dev1=eth1
 +dev2=eth2
 +dev1mac=00:1b:21:9b:24:b4
 +dev2mac=00:1b:21:9b:24:b5
 +
 +# fake client interfaces and addresses
 +dev=dummy0
 +dev_mac=00:00:00:00:00:11
 +
 +# max fake clients supported for simulation
 +maxusers=3
 +
 +## Create dummy device
 +## Accepted parameters:
 +##    $1 - devname
 +##    $2 - devmac
 +##    $3 - subnet (e.g. 10.10.10)
 +##    $4 - max number of IP addresses to create on interface
 +setup_dummy()
 +{
 +#   sudo sh -c "echo 1 > /proc/sys/net/ipv4/ip_forward"
 +   # Enable tc hardware offload
 +#   ethtool -K $SGW_DEV hw-tc-offload on
 +
 +   $ip link add $1 address $2 type dummy
 +   $ip link set $1 up
 +   for i in `seq 1 $4`;
 +   do
 +      $ip addr add $3.$i/32 dev $1
 +   done
 +}
 +
 +## Delete dummy device
 +## Accepted parameters:
 +##    $1 - devname
 +delete_dummy()
 +{
 +  $ip link del $1 type dummy
 +}
 +
 +setup_network()
 +{
 +  # Send traffic eth3 <-> eth4 over loopback cable, where both interfaces
 +  # eth3 and eth4 are in the same subnet.
 +  #
 +  # We assume that NetworkManager is not running and eth3/eth4 are configured
 +  # via /etc/network/interfaces:
 +  #
 +  # 192.168.1.100/32 dev eth3
 +  # 192.168.1.101/32 dev eth4
 +  #
 +  # Specify source IP address when sending the traffic:
 +  # ping -I 192.168.1.100 192.168.1.101
 +  #
 +  #
 +  $ip neigh add $eth2 lladdr $dev2mac nud permanent dev $dev1
 +  $ip neigh add $eth1 lladdr $dev1mac nud permanent dev $dev2
 +  $ip route add table main $eth1 dev $dev2
 +  $ip route add table main $eth2 dev $dev1
 +  $ip rule add from all lookup local pref 100
 +  $ip rule del pref 0
 +  $ip rule add from $eth2 to $eth1 iif $dev1 lookup local pref 1
 +  $ip rule add from $eth1 to $eth2 iif $dev2 lookup local pref 2
 +  $ip rule add from $eth2 to $eth1 lookup main pref 3
 +  $ip rule add from $eth1 to $eth2 lookup main pref 4
 +
 +#  $ip rule add from 10.10.10.0/24 to $eth1 iif $dev1 lookup local pref 5
 +#  $ip rule add from 10.10.10.0/24 to $eth2 iif $dev2 lookup local pref 6
 +#  $ip rule add from $eth1 to 10.10.10.0/24 iif $dev2 lookup local pref 7
 +#  $ip rule add from $eth2 to 10.10.10.0/24 iif $dev1 lookup local pref 8
 +}
 +
 +restore_network()
 +{
 +  # FIX: hangs connections
 +  $ip rule flush
 +  $ip rule add priority 32767 lookup default
 +}
 +
 +#delete_dummy dummy0
 +#delete_dummy dummy1
 +
 +#setup_dummy dummy0 00:00:00:00:00:11 10.10.10 3
 +#setup_dummy dummy1 00:00:00:00:00:22 20.20.20 3
 +setup_network
 </code> </code>
  • network_namespaces.1534668680.txt.gz
  • Last modified: 2018/08/19 08:51
  • by rpjday