3. Installing containers

This chapter describes container installation tasks. Container installation tasks include:

  1. Launching a container manually
    Launches a single SR Linux container using a manual procedure.
  2. Launching a container topology
    Launches a container topology using an automated script.

3.1. Container installation prerequisites

Ensure that prerequisites are met before installing an SR Linux container or container topology.

Minimum system requirements:

  1. CentOS 7.4 system, kernel 4.17
  2. 8 GB RAM
  3. 8 core CPU
  4. The host machine user must have sudo privileges (preferably passwordless)

Minimum software requirements:

  1. Docker CE installed, minimum version 18.09:
    https://docs.docker.com/get-docker/
  2. Python version 3.6 or later
  3. docker-topo-master.tgz file (if building an SR Linux container topology)
  4. srlinux-vX.Y.Z-N.tar.xz file (for the SR Linux Docker image)
    where X = major, Y = minor, Z = patch, and N = build number

3.2. Launching a container manually

This procedure manually launches a single container.

  1. Copy the srlinux-vX.Y.Z-N.tar.xz file into ~/ of the CentOS 7 host machine.
  2. Load the Docker image. To load the image, the user must have root privilege, or be part of the Docker group.
    $ docker image load -i ~/srlinux-vX.Y.Z-N.tar.xz
  3. Check that the Docker image was imported correctly:
    $ docker images
    Example:
    REPOSITORY TAG          IMAGE ID              CREATED         SIZE
    srlinux 19.11.1-        3e77d45745f2          7 hours ago     1.3GB
  4. Launch an instance of the SR Linux container on the host using the options in the following command line. This command must be entered in a single line. See the Note that follows about copying text from a PDF file without broken lines.
    $ docker run -t -d --rm --privileged \
    --sysctl net.ipv6.conf.all.disable_ipv6=0 \
    --sysctl net.ipv4.ip_forward=0 \
    --sysctl net.ipv6.conf.all.accept_dad=0 \
    --sysctl net.ipv6.conf.default.accept_dad=0 \
    --sysctl net.ipv6.conf.all.autoconf=0 \
    --sysctl net.ipv6.conf.default.autoconf=0 \
    -u $(id -u):$(id -g) \
    --name srlinux_dut1 srlinux:vX.Y.Z-N 
    sudo bash -c ‘/opt/srlinux/bin/sr_linux'
    where X = major, Y = minor, Z = patch, and N = build Number for srlinux:vX.Y.Z-N
    Note:

    Copying a long command string from a PDF file will introduce line breaks. As a workaround, copy the text string and place into Notepad++. Highlight the text and select CTRL+J. The result is a single line with no returns.

  5. Check that the docker container has been created with the name ‘srlinux’:
    $ docker ps
    Example:
     CONTAINER ID  IMAGE               COMMAND             CREATED     STATUS   PORTS   NAMES
     9d5dbd03f7f8  srlinux:19.11.1     "/tini--fixuid-q.." 3 mins ago  Up3 mins.     srlinux_dut1
  6. Turn off the Docker0 Tx checksum offload:
    # sudo ethtool --offload docker0 tx off
  7. Open an SSH session to the DUT using the following credentials:
    1. username: admin
    2. password: admin
    Example:
    # ssh admin@$(docker inspect srlinux_dut1 --format {{.NetworkSettings.IPAddress}})
    admin@172.17.0.4's password: 
          ___         ___                               ___         ___         ___
         /  /\       /  /\                  ___        /__/\       /__/\       /__/|
        /  /:/_     /  /::\  ___           /  /\       \  \:\      \  \:\     |  |:|
       /  /:/ /\   /  /:/\:\/__/\         /  /:/        \  \:\      \  \:\    |  |:|
      /  /:/ /::\ /  /:/~/:/\  \:\     __/__/::\    _____\__\:\ ___  \  \:\ __|__|:|
     /__/:/ /:/\:/__/:/ /:/__\  \:\   /  \__\/\:\  /__/::::::::/__/\  \__\:/__/::::\____
     \  \:\/:/~/:\  \:\/:::::/\  \:\ /  /:/ \  \:\/\  \:\~~\~~\\  \:\ /  /:/  ~\~~\::::/
      \  \::/ /:/ \  \::/~~~~  \  \:\  /:/   \  \:\/\  \:\  ~~~ \  \:\  /:/    |~~|:|~~
       \__\/ /:/   \  \:\       \  \:\/:/     \__\::/\  \:\      \  \:\/:/     |  |:|
         /__/:/     \  \:\       \  \::/      /__/:/  \  \:\      \  \::/      |  |:|
         \__\/       \__\/        \__\/       \__\/    \__\/       \__\/       |__|/
    Hello admin,
    Welcome to the srlinux CLI.
    Type 'help' (and press <ENTER>) if you need any help using this.
    --{ running }--[  ]--
  8. Verify the application versions running on the system:
    # info from state system app-management application * | as table | filter fields pid author version
    Example:
    A:3-node-srlinux-A# info from state system app-management application * | 
    as table | filter fields pid author version
    +---------------------+---------+------------+------------------------+
    |        Name         |    Pid  |  Author    |       Version          |
    +=====================+=========+============+========================+
    | aaa_mgr             |    2189 | Nokia      | v20.6.0-13-gc6a313b84c |
    | acl_mgr             |    2198 | Nokia      | v20.6.0-13-gc6a313b84c |
    | app_mgr             |    2131 | Nokia      | v20.6.0-13-gc6a313b84c |
    | arp_nd_mgr          |    2207 | Nokia      | v20.6.0-13-gc6a313b84c |
    | bfd_mgr             |         |            |                        |
    | bgp_mgr             |    2650 | Nokia      | v20.6.0-13-gc6a313b84c |
    | chassis_mgr         |    2216 | Nokia      | v20.6.0-13-gc6a313b84c |
    | dev_mgr             |    2155 | Nokia      | v20.6.0-13-gc6a313b84c |
    | dhcp_client_mgr     |    2228 | Nokia      | v20.6.0-13-gc6a313b84c |
    | fib_mgr             |    2237 | Nokia      | v20.6.0-13-gc6a313b84c |
    | gnmi_server         |    2473 | Nokia      | v20.6.0-13-gc6a313b84c |
    | idb_server          |    2180 | Nokia      | v20.6.0-13-gc6a313b84c |
    | isis_mgr            |         |            |                        |
    | json_rpc            |    2476 | Nokia      | v20.6.0-13-gc6a313b84c |
    | l2_mac_learn_mgr    |    2248 | Nokia      | v20.6.0-13-gc6a313b84c |
    | l2_mac_mgr          |    2258 | Nokia      | v20.6.0-13-gc6a313b84c |
    | l2_static_mac_mgr   |         |            |                        |
    | lag_mgr             |    2267 | Nokia      | v20.6.0-13-gc6a313b84c |
    | linux_mgr           |    2276 | Nokia      | v20.6.0-13-gc6a313b84c |
    | lldp_mgr            |    2657 | Nokia      | v20.6.0-13-gc6a313b84c |
    | log_mgr             |    2285 | Nokia      | v20.6.0-13-gc6a313b84c |
    | mcid_mgr            |    2294 |            |                        |
    | mgmt_server         |    2303 | Nokia      | v20.6.0-13-gc6a313b84c |
    | mpls_mgr            |         |            |                        |
    | net_inst_mgr        |    2312 | Nokia      | v20.6.0-13-gc6a313b84c |
    | oam_mgr             |    2325 | Nokia      | v20.6.0-13-gc6a313b84c |
    | ospf_mgr            |         | Nokia      |                        |
    | plcy_mgr            |    2667 | Nokia      | v20.6.0-13-gc6a313b84c |
    | qos_mgr             |    2471 | Nokia      | v20.6.0-13-gc6a313b84c |
    | sdk_mgr             |    2335 | Nokia      | v20.6.0-13-gc6a313b84c |
    | sshd-mgmt           |    2923 |            |                        |
    | static_route_mgr    |    2674 | Nokia      | v20.6.0-13-gc6a313b84c |
    | supportd            |    2140 | Nokia      | v20.6.0-13-gc6a313b84c |
    | vrrp_mgr            |         |            |                        |
    | xdp_cpm             |    2349 | Nokia      | v20.6.0-13-gc6a313b84c |
    | xdp_lc_1            |    2365 | Nokia      | v20.6.0-13-gc6a313b84c |
    +---------------------+---------+------------+------------------------+

3.3. Launching a container topology

This procedure launches a container topology using an automated script. See Figure 1 for a topology example.

Figure 1:  Three-node container topology example 
  1. Copy the file srlinux-vX.Y.Z-N.tar.xz into ~/ of the CentOS 7 host machine.
  2. Load the Docker image:
    $ docker image load -i ~/srlinux-vX.Y.Z-N.tar.xz
    Example:
    ~ $ docker image load -i srlinux-vX.Y.Z.tar.xz
    ba6c2307b523: Loading layer  329.1MB/329.1MB
    53a1d27f1cbd: Loading layer   5.12kB/5.12kB
    b1cfd1c17e46: Loading layer  311.3kB/311.3kB
    559e14cb1ef4: Loading layer  303.1kB/303.1kB
    99b5d14edc41: Loading layer  4.015MB/4.015MB
    b89e0ee26076: Loading layer  3.072kB/3.072kB
    ba4b6c40087a: Loading layer   2.56kB/2.56kB
    4f529c55a3c9: Loading layer   2.56kB/2.56kB
    9637194042f7: Loading layer   7.68kB/7.68kB
    34eed33aab83: Loading layer   25.6kB/25.6kB
    964b072de0ec: Loading layer   25.6kB/25.6kB
    ac434023ca76: Loading layer  3.072kB/3.072kB
    ae35e17f1206: Loading layer  3.072kB/3.072kB
    7f230658f8b8: Loading layer  3.072kB/3.072kB
    78f9091adeb6: Loading layer   2.56kB/2.56kB
    b8f40a106975: Loading layer  3.072kB/3.072kB
    238c0f5a7a42: Loading layer  3.072kB/3.072kB
    c607a3d0a237: Loading layer  3.072kB/3.072kB
    00f1c823da02: Loading layer  3.072kB/3.072kB
    068f6071eab8: Loading layer  403.9MB/403.9MB
    1963843423a6: Loading layer  3.072kB/3.072kB
    Loaded image: srlinux:v<release>-190
  3. Check that the Docker image was imported correctly:
    $ docker images
    Example:
    REPOSITORY TAG          IMAGE ID              CREATED         SIZE
    srlinux 19.11.1      3e77d45745f2          7 hours ago     1.3GB
  4. Create a docker-topo-master folder with the required files, using the following command:
    # cd ~; tar -zxvf docker-topo-master.tar.gz
  5. Change directories to the docker-topo-master directory:
    cd ~/docker-topo-master
  6. In the docker-topo-master directory, install docker-topo using:
    # sudo python3.6 setup.py install
    Note: If you encounter an error creating a directory or file, check the directory permissions.
  7. The image name SRLINUX_IMAGE: srlinux:vX.Y.Z-N is defined by default. If a different name is required, edit the image name in file ~/docker-topo-master/topo-extra-files/examples/v2/3-node.yml.
    Example (image filename location):
    VERSION: 2
    driver: veth
    PREFIX: 3-node
    SRLINUX_IMAGE: srlinux:19.11.1
    PUBLISH_BASE: 9000
    links:
      - endpoints: ["srlinux-A:e1-1", "srlinux-B:e1-1"]
      - endpoints: ["srlinux-A:e1-2", "srlinux-C:e1-2"]
      - endpoints: ["srlinux-B:e1-3", "srlinux-C:e1-3"]
  8. Important: If there are already existing containers with 3-node.yml, they need to be destroyed before creating the new three-node topology. See Destroying an existing topology for this procedure.
    If there is no existing topology (or an existing topology is destroyed), create a three-node topology using:
    $ docker-topo --create topo-extra-files/examples/v2/3-node.yml
    Example:
    $ docker-topo --create topo-extra-files/examples/v2/3-node.yml
    INFO:__main__:Version 2 requires sudo. Restarting script with sudo
    INFO:__main__:
    alias srlinux-A='ssh -o UserKnownHostsFile=/dev/null -
    o StrictHostKeyChecking=no admin@$(docker inspect 3-node-srlinux-A --
    format "{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}")'
     
    $ alias srlinux-B='ssh -o UserKnownHostsFile=/dev/null -
    o StrictHostKeyChecking=no admin@$(docker inspect 3-node-srlinux-B --
    format "{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}")'
     
    $ alias srlinux-C='ssh -o UserKnownHostsFile=/dev/null -
    o StrictHostKeyChecking=no admin@$(docker inspect 3-node-srlinux-C --
    format "{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}")'
    INFO:__main__:All devices started successfully
    Note:

    Topologies for one node and two nodes can also be built using the following commands:

    1. For one node: $ docker-topo --create topo-extra-files/examples/v2/1-node.yml
    2. For two nodes: $ docker-topo --create topo-extra-files/examples/v2/2-node.yml
  9. Using the output from the previous step, create login shortcut commands.
    Example:
    $ alias srlinux-A='ssh -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no admin@$(docker inspect 3-node-srlinux-A --format "{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}")'
    $ alias srlinux-B='ssh -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no admin@$(docker inspect 3-node-srlinux-B --format "{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}")'
    $ alias srlinux-C='ssh -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no admin@$(docker inspect 3-node-srlinux-C --format "{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}")'
  10. After the shortcut is created, use the alias command to open the CLI with the alias name. Enter admin as the username and password. For example, to open srlinux-A CLI, enter:
    $ srlinux-A
    Example:
    $ srlinux-A
    Warning: Permanently added '172.18.18.2' (RSA) to the list of known hosts.
    admin@172.18.18.2's password:
    Hello admin,
    Welcome to the srlinux CLI.
    Type 'help' (and press <ENTER>) if you need any help using this.
    --{ running }--[  ]-- 
  11. Check that new nodes are up and running:
    $ docker ps -a
    Example:
    $ docker ps -a
    CONTAINER ID    IMAGE           COMMAND                 CREATED         STATUS       PORT       NAMES
    2505ea85a702    srlinux:v0.1.0-198  " /tini -- fixuid -q …"2 mins ago   Up 2 minutes            3-
    node-srlinux-A
    f8a0b21f126d    srlinux:v0.1.0-198  " /tini -- fixuid -q …"2 mins ago   Up 2 minutes            3-
    node-srlinux-B
    b30f6326c87b    srlinux:v0.1.0-198  " /tini -- fixuid -q …"2 mins ago   Up 2 minutes            3-
    node-srlinux-C
  12. Access the CLI for the nodes.
    1. Using the shortcut command created in Step 9:
      srlinux-A
      1. If the shortcut was not created, use the docker exec command:
        $ docker exec -it 3-node-srlinux-A sr_cli
    2. Using the full SSH command, as shown in the output after the container creation:
      $ ssh -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no admin@$(docker inspect 3-node-srlinux-A --format "{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}")
    Note:

    If the CLI prompt does not come up immediately, adjust the terminal window size. This is a known Docker issue.

  13. Verify that the interface and IP addresses exist on all nodes:
    1. enter running
    2. info interface <interface name>
    3. info network-instance <instance name>
    Example:
    # enter running
    --{ running }--[  ]--
    # info interface *
        interface ethernet-1/1 {
            admin-state enable
            subinterface 1 {
                admin-state enable
                ipv4 {
                    dhcp-client false
                    address 192.168.11.1/30 {
                    }
                }
            }
        }
        interface ethernet-1/2 {
            admin-state enable
            subinterface 1 {
                admin-state enable
                ipv4 {
                    dhcp-client false
                    address 192.168.12.1/30 {
                    }
                }
            }
        }
        interface mgmt0 {
            admin-state enable
            subinterface 0 {
                admin-state enable
                ipv4 {
                    dhcp-client true
                }
                ipv6 {
                    dhcp-client true
                }
            }
        }
    --{ running }--[  ]-
    # info network-instance *
        network-instance mgmt {
            type ip-vrf
            admin-state enable
            description "Management network instance"
            ip-forwarding {
                receive-ipv4-check true
                receive-ipv6-check true
            }
            interface mgmt0.0 {
            }
            protocols {
                linux {
                    export-routes true
                    export-neighbors true
                }
            }
        }
        network-instance red {
            type ip-vrf
            admin-state enable
            ip-forwarding {
                receive-ipv4-check true
                receive-ipv6-check true
            }
            interface ethernet-1/1.1 {
            }
            interface ethernet-1/2.1 {
            }
            protocols {
                linux {
                    export-routes true
                    export-neighbors true
                }
            }
            static-routes {
                route 192.168.13.0/24 {
                    admin-state enable
                    metric 1
                    preference 5
                    next-hop-group static-ipv4-grp
                }
            }
            next-hop-groups {
                ecmp {
                    max-paths-level-2 1
                }
                group static-ipv4-grp {
                    admin-state enable
                    collect-stats false
                    nexthop 1 {
                        ip-address 192.168.12.2
                        admin-state enable
                        resolve true
                    }
                    nexthop 2 {
                        ip-address 192.168.11.2
                        admin-state enable
                        resolve true
                    }
                }
            }
        }
    --{ running }--[  ]--
  14. Ping different nodes to verify connectivity:
    ping <IP Address> network-instance <instance name>
    Example:
    --{ running }--[  ]--
    # ping 192.168.11.1 network-instance red
    Pinging 192.168.11.1 in srbase-red
    PING 10.0.0.0 (192.168.11.1) 56(84) bytes of data.
    64 bytes from 192.168.11.1: icmp_seq=1 ttl=64 time=1.74 ms
    64 bytes from 192.168.11.1: icmp_seq=2 ttl=64 time=0.596 ms
    64 bytes from 192.168.11.1: icmp_seq=3 ttl=64 time=0.664 ms
    64 bytes from 192.168.11.1: icmp_seq=4 ttl=64 time=0.710 ms
    ^C
    --- 192.168.11.1 ping statistics ---
    4 packets transmitted, 4 received, 0% packet loss, time 3001ms
    rtt min/avg/max/mdev = 0.596/0.927/1.740/0.471 ms

3.4. Destroying an existing topology

Destroy an existing topology using the docker-topo --destroy command. Destroy the created topology using the following command:

$ docker-topo --destroy topo-extra-files/examples/v2/3-node.yml