摘要:Linux 4.0的switchdev一切坏消息在Linux 4.0内核中终结。 Linux 4.0引入了一个switchdev框架,它代表一类拥有“交换”能力芯片的多网口设备的抽象。其中每一个网口就是一个port,在switchdev框架中被注册成一个net_device。除此之外,内核中自带了一
阅读全文
摘要:[root@evpn2 ~]# tc qdisc add dev p0a ingress -bash: tc: command not found [root@evpn2 ~]# yum install iproute-tc [root@localhost ~]# tc filter show de
阅读全文
摘要:[root@localhost vpp]# ethtool -k enp125s0f0 | grep hw-tc-offload hw-tc-offload: off [fixed] [root@localhost vpp]# [root@localhost vpp]# ethtool -k enp
阅读全文
摘要:make -C /lib/modules/4.14.0-115.el7a.0.1.aarch64/build M=/root/vrouter/contrail-vrouter modules make: *** /lib/modules/4.14.0-115.el7a.0.1.aarch64/bui
阅读全文
摘要:Linux开启coredump mkdir -p /home/data/corefile chmod 777 /home/data/corefile echo "ulimit -S -c unlimited > /dev/null 2>&1" >> /etc/profile source /etc/
阅读全文
摘要:In the first part of our new series, we will be building our first VPP platform plug-in, using basic examples. We will start with a first-dive into pl
阅读全文
摘要:查看路由表 DBGvpp# show ip fib ipv4-VRF:0, fib_index:0, flow hash:[src dst sport dport proto ] epoch:0 flags:none locks:[default-route:1, nat-hi:2, ] 0.0.0
阅读全文
摘要:[root@localhost vpp]# find ./ -name router-plugin_plugin.so ./lib/vpp_plugins/router-plugin_plugin.so [root@localhost vpp]#
阅读全文
摘要:/* *INDENT-OFF* */ typedef CLIB_PACKED (struct { mac_address_t mac; ip4_address_t ip4; }) ethernet_arp_ip4_over_ethernet_address_t; /* *INDENT-ON* */
阅读全文
摘要:Could not find a package configuration file provided by "Vpp" with any of the following names: VppConfig.cmake vpp-config.cmake Add the installation p
阅读全文
摘要:查询VRF邻接表和路由 ip neigh show vrf vrf-blue ip addr show vrf vrf-blue ip -br addr show vrf vrf-blue ip route show vrf vrf-blue [root@evpn1 ~]# ip neigh sho
阅读全文
摘要:一开始只有一个vrf [root@evpn2 ~]# vtysh Hello, this is FRRouting (version 7.3-MyOwnFRRVersion). Copyright 1996-2005 Kunihiro Ishiguro, et al. evpn2.novalocal
阅读全文
摘要:export RTE_TARGET=arm64-armv8a-linuxapp-gcc export RTE_SDK=`pwd` insmod ${RTE_SDK}/${RTE_TARGET}/kmod/igb_uio.ko make -j 96 [root@localhost dpdk-19.11
阅读全文
摘要:A bridge-domain can be associated with only one BVI. However a vrf can contain routes to multiple BVIs. To configure a BVI interface, create it using
阅读全文
摘要:DBGvpp# show int addr UnknownEthernet5/0/0 (up): L2 bridge bd-id 1 idx 1 shg 0 host-ns1_tap (up): L2 bridge bd-id 1 idx 1 shg 0 host-ns7_tap (up): L2
阅读全文
摘要:DBGvpp# show inter addr UnknownEthernet5/0/0 (up): L2 bridge bd-id 1 idx 1 shg 0 host-ns1_tap (up): L2 bridge bd-id 1 idx 1 shg 0 host-ns7_tap (up): l
阅读全文
摘要:默认table 0 set interface ip table loop2 vrf1 loop1 in vrf1 and loop2 in vrf2 nat44-out2in no translation with multiple tenants Example Usage Example of
阅读全文
摘要:DBGvpp# show int Name Idx State MTU (L3/IP4/IP6/MPLS) Counter Count UnknownEthernet5/0/0 1 down 9000/0/0/0 rx-miss 649 local0 0 down 0/0/0/0 DBGvpp# l
阅读全文
摘要:#!/bin/bash # # set up a network env with 2 namespaces: # # [ vpp1 ] [vpp] [ vpp2 ] # # vpp1: 192.168.0.1 # vpp2: 192.168.0.2 ip netns delete vpp1 ip
阅读全文
摘要:DBGvpp# show ip arp show ip: unknown input `arp' DBGvpp# sh run Time 63301.6, 10 sec internal node vector rate 1.00 vector rates in 5.3712e1, out 2.27
阅读全文
摘要:#!/bin/bash PATH=$PATH:./build-root/build-vpp-native/vpp/bin/ if [ $USER != "root" ] ; then echo "Restarting script with sudo..." sudo $0 ${*} exit fi
阅读全文
摘要:DBGvpp# sh inter addr UnknownEthernet5/0/0 (up): L3 192.168.1.229/24 host-ns1_tap (up): local0 (dn): DBGvpp# nat44 add interface address UnknownEthern
阅读全文
摘要:This is the last article in the series, we will finish this lab with the VXLAN configuration and testing connectivity between servers. VIRTUAL EXTENSI
阅读全文
摘要:ln -s /root/vppsb/vppsb/router/router.mk build-data/packages/router.mk ln -s /root/vppsb/vppsb/router/netlink.mk build-data/packages/netlink.mk find .
阅读全文
摘要:DBGvpp# show vxlan tunnel [0] instance 0 src 192.168.1.229 dst 192.168.1.82 vni 10 fib-idx 0 sw-if-idx 2 encap-dpo-idx 1 DBGvpp# DBGvpp# show vxlan tu
阅读全文
摘要:在frr中查看vrf 给host-ns1_tap配个ip DBGvpp# set interface ip address host-ns1_tap 192.168.1.230/24 set interface ip address: failed to add 192.168.1.230/24 o
阅读全文
摘要:DBGvpp# show int Name Idx State MTU (L3/IP4/IP6/MPLS) Counter Count UnknownEthernet5/0/0 1 down 9000/0/0/0 local0 0 down 0/0/0/0 DBGvpp# set int state
阅读全文
摘要:[root@localhost ~]# lshw -class network -businfo Bus info Device Class Description pci@0000:05:00.0 enp5s0 network Hi1822 Family (2*25GE) pci@0000:06:
阅读全文
摘要:https://docs.mellanox.com/pages/releaseview.action?pageId=15053848 Prerequisites 1. Install CentOS 7.72. Set up an account to use the Linux Foundation
阅读全文
摘要:gument (errno 22) /root/vpp/vpp/build-root/install-vpp_debug-native/vpp/bin/vpp[73384]: register_node:485: process stack: Invalid argument (errno 22)
阅读全文
摘要:./build/external/packages/dpdk.mk 先要将网卡down才能去绑定vpp DPDK_VERSION ?= 19.08 参考一 https://fd.io/docs/vpp/master/usecases/vppinazure.html $ git clone https
阅读全文
摘要:[root@localhost frr-stable-7.3]# make true make all-am make[1]: Entering directory `/root/frr/frr-stable-7.3' CLIPPY bgpd/bgp_debug_clippy.c /bin/sh:
阅读全文
摘要:[root@localhost vpp-stable-2001]# make install-dep Loaded plugins: fastestmirror Loading mirror speeds from cached hostfile * centos-sclo-rh: mirror-h
阅读全文
摘要:Lets see how to deploy a server using ironic. To do that we are going to define those variables: # Ironic client variables export IRONIC_URL=http://IR
阅读全文
摘要:现在有两个ironic node, node82上成功运行裸金属,node187上没有,现在查看他们的boot device [root@localhost workspace]# ironic node-list The "ironic" CLI is deprecated and will be
阅读全文
摘要:iSCSI (Internet Small Computer Systems Interface) is an Internet Protocol based storage networking for connecting remote storage devices over the netw
阅读全文
摘要:dib制作deploy镜像会构造一个VIRTUAL_ENV环境,会通过chroot进入临时目录 export TMP_MOUNT_PATH="$TMP_BUILD_DIR/mnt" sudo -E chroot $TMP_MOUNT_PATH env -u TMPDIR -u VIRTUAL_ENV
阅读全文
摘要:This document describes how to build and install Open vSwitch using a DPDK datapath. Open vSwitch can use the DPDK library to operate entirely in user
阅读全文
摘要:绑定前网卡驱动 [root@localhost ~]# ethtool -i enp5s0 driver: hinic version: firmware-version: expansion-rom-version: bus-info: 0000:05:00.0 supports-statisti
阅读全文
摘要:网卡配置如下 注意: 使用VFIO需要BIOS开启SMMU [root@localhost dpdk-19.11]# ethtool -i enp6s0 driver: hinic version: firmware-version: expansion-rom-version: bus-info:
阅读全文
摘要:无法爬楼楼的可以从这里下: http://static.dpdk.org/rel/ [root@kunpeng82 data1]# cat /etc/default/grub |grep GRUB_CMDLINE_LINUX GRUB_CMDLINE_LINUX="crashkernel=auto"
阅读全文
摘要:yum install docker [root@kunpeng82 devuser]# uname -aLinux kunpeng82.bogon 4.14.0-115.10.1.el7a.aarch64 #1 SMP Tue Jul 30 14:50:37 UTC 2019 aarch64 aa
阅读全文
摘要:新增一个节点10.10.18.227 10.10.16.82通过vxlan vni=48和vxlan vni=32分别连接10.10.18.227和10.10.18.216 10.10.28.227网络配置如下 #/bin/bash set -e #添加ovs网桥: br-tun/br-int #o
阅读全文
摘要:10.10.16.82节点上 [root@kunpeng82 devuser]# ip netns exec ns1 ping 192.168.10.22 PING 192.168.10.22 (192.168.10.22) 56(84) bytes of data. ^C 192.168.10.2
阅读全文
摘要:登陆devuser显示-bash-4.2$ 原因是在用useradd添加普通用户时,有时会丢失home目录下的环境变量文件,丢失的文件如下:1、.bash_profile2、.bashrc以上这些文件是每个用户都必备的文件。此时可以使用以下命令从主默认文件/etc/skel/下重新拷贝一份配置信息到
阅读全文
摘要:制作了一个NeoKylin-Server-7.0-aarch64-Release-Build06.euler.的ironic 镜像但是跑不起来 insmod 、lsmod命令都用不了,用losetup 和kpartx 将qcow2镜像mount后,chroot进去lsinitrd试了下 [root@
阅读全文
摘要:Sometimes you just can’t use multicast. Some cloud providers just do not provide it. In that scenario, you need to configure your vxlan layer using un
阅读全文
摘要:1、给table 2 添加一条规则 [root@kunpeng82 devuser]# ovs-ofctl add-flow vswtich1 "table=2 actions=learn(table=10,NXM_OF_VLAN_TCI[0..11], \ NXM_OF_ETH_DST[]=NXM
阅读全文
摘要:1、ping [root@kunpeng82 devuser]# ip netns exec ns7 ping 192.168.1.200 PING 192.168.1.200 (192.168.1.200) 56(84) bytes of data. 64 bytes from 192.168.1
阅读全文
摘要:表 1. 流表常用字段 字段名称说明 in_port=port 传递数据包的端口的 OpenFlow 端口编号 dl_vlan=vlan 数据包的 VLAN Tag 值,范围是 0-4095,0xffff 代表不包含 VLAN Tag 的数据包 dl_src=<MAC>dl_dst=<MAC> 匹配
阅读全文
摘要:[root@kunpeng82 devuser]# ovs-vsctl list port tap6_br _uuid : 70e22269-7254-4300-88fc-b3e8cfcc1df6 bond_active_slave : [] bond_downdelay : 0 bond_fake
阅读全文
摘要:eth0连接internet ns1 [root@kunpeng82 devuser]# ip netns exec ns1 route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface
阅读全文
摘要:1、ovs-vsctl set port tap0_br tag=100 [root@kunpeng82 devuser]# ovs-vsctl list port tap0_br_uuid : f8b41259-a1b3-4226-bd25-afb7c0dfe9e1bond_active_slav
阅读全文
摘要:[root@kunpeng82 devuser]# ovs-ofctl dump-flows vswitch0 cookie=0x0, duration=3.829s, table=0, n_packets=0, n_bytes=0, priority=0 actions=NORMAL [root@
阅读全文
摘要:节点ip改为192.168.2.210/24 [root@ovs216 ~]# ip netns exec ns1 ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen
阅读全文
摘要:[root@kunpeng82 devuser]# ovs-vsctl show 25581fe4-f414-4c0a-a9eb-5d04cc658c0c Bridge "vswitch0" Port "tap3_br" Interface "tap3_br" Port "vswitch0" Int
阅读全文
摘要:[root@kunpeng82 devuser]# ovs-ofctl -O OpenFlow13 dump-group-stats vswitch0 OFPST_GROUP reply (OF1.3) (xid=0x6): [root@kunpeng82 devuser]# ovs-ofctl -
阅读全文
摘要:1、把所有的flow删掉 [root@kunpeng82 devuser]# ovs-ofctl dump-flows vswitch0 table=0 cookie=0x0, duration=132.837s, table=0, n_packets=0, n_bytes=0, priority=
阅读全文
摘要:ip netns add ns1 ip netns add ns2 ip link add tap0 type veth peer name tap0_br ip link add tap3 type veth peer name tap3_br ip link set tap0 netns ns1
阅读全文
摘要:中标麒麟基于euler的裸金属镜像启动失败报dracut initqueue timeout 找个openeuler能启动的裸金属镜像,查看grub.cfg reserve_kbox_mem=16M grub.cfg加上这些参数后 找个正常的centos7镜像发现: 而镜像上没有/dev/disk/
阅读全文
摘要:流表分为两大类: 1、 内核中flow table 也称为fast path 2、找用户态中flow table被称为slow path 一个数据报文接收后,会经过多个流表,每个流表负责特定的功能,ovs中的多级流表存放在用户空间,内核态存放的是流表的缓存。 网卡收到报文时,Openvswitch.
阅读全文
摘要:BOOTAA64.EFI走的是bios模式,grubaa64.efi走的是uefi模式 [root@localhost centos]# md5sum grubaa64.efi dff06670847929df2d469961bca5a275 grubaa64.efi [root@localhost
阅读全文