k8s1.4.3安装实践记录（1）-etcd、docker、flannel安装配置

来源：互联网发布：java 射击游戏编辑：程序博客网时间：2024/06/08 06:57

由于刚开始学习k8s，本次软件的安装，我们都采用最简单的方式，能用yum 安装的尽量采用yum安装

1、ETCD安装

ETCD官方文档：https://github.com/coreos/etcd/blob/master/Documentation/docs.md

1.1 检查ETCD版本

[root@localhost ~]# yum list|grep etcdetcd.x86_64                                2.3.7-4.el7                 @extras  [root@localhost ~]#

1.2 安装ETCD

yum install etcd

1.3 修改ETCD配置

安装好后，系统会自动生成etcd.service文件（路径为/usr/lib/systemd/system/），修改对应的配置

[Unit]Description=Etcd ServerAfter=network.targetAfter=network-online.targetWants=network-online.target[Service]Type=notifyWorkingDirectory=/var/lib/etcd/EnvironmentFile=-/etc/etcd/etcd.confUser=etcd# set GOMAXPROCS to number of processorsExecStart=/bin/bash -c "GOMAXPROCS=$(nproc) /usr/bin/etcd \--name=\"${ETCD_NAME}\" \--data-dir=\"${ETCD_DATA_DIR}\" \--listen-peer-urls=\"${ETCD_LISTEN_PEER_URLS}\" \--advertise-client-urls=\"${ETCD_ADVERTISE_CLIENT_URLS}\" \--initial-cluster-token=\"${ETCD_INITIAL_CLUSTER_TOKEN}\" \--initial-cluster=\"${ETCD_INITIAL_CLUSTER}\"  \--initial-cluster-state=\"${ETCD_INITIAL_CLUSTER_STATE}\" \--listen-client-urls=\"${ETCD_LISTEN_CLIENT_URLS}\""Restart=on-failureLimitNOFILE=65536[Install]WantedBy=multi-user.target

并配置其配置文件

ETCD_NAME=zwetcd_2ETCD_DATA_DIR="/var/lib/etcd/default.etcd"ETCD_LISTEN_PEER_URLS="http://192.168.37.131:2380"ETCD_LISTEN_CLIENT_URLS="http://192.168.37.131:2379,http://127.0.0.1:2379"#[cluster]ETCD_INITIAL_ADVERTISE_PEER_URLS="http://192.168.37.131:2380"# if you use different ETCD_NAME (e.g. test), set ETCD_INITIAL_CLUSTER value for this name, i.e. "test=http://..."ETCD_INITIAL_CLUSTER="zwetcd_2=http://192.168.37.131:2380,zwetcd_1=http://192.168.37.130:2380"ETCD_INITIAL_CLUSTER_STATE="new"ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"ETCD_ADVERTISE_CLIENT_URLS="http://192.168.37.131:2379"

如果使用firewalld作为防火墙，则需要开放端口：

1
2
3
4
firewall-cmd --zone=public --add-port=2379/tcp --permanent
firewall-cmd --zone=public --add-port=2380/tcp --permanent
firewall-cmd --reload
firewall-cmd --list-all

　　问题：

1、本地连接报错
[root@localhost system]# etcdctl ls /
Error: client: etcd cluster is unavailable or misconfigured
error #0: dial tcp 127.0.0.1:2379: getsockopt: connection refused
error #1: dial tcp 127.0.0.1:4001: getsockopt: connection refused

如果出现如上的错误，是因为ETCD_LISTEN_CLIENT_URLS参数没有配置http://127.0.0.1:2379而导致的，不过已经配置了具体的IP，还需要配置本地链路，这个就有点奇怪了。

2、Docker安装

2.1、检查docker版本

yum list |grep docker

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
[root@localhost ~]# yum list|grep docker
docker.x86_64                              1.10.3-46.el7.centos.14     @extras 
docker-common.x86_64                       1.10.3-46.el7.centos.14     @extras 
docker-selinux.x86_64                      1.10.3-46.el7.centos.14     @extras 
cockpit-docker.x86_64                      0.114-2.el7.centos          extras  
docker-devel.x86_64                        1.3.2-4.el7.centos          extras  
docker-distribution.x86_64                 2.4.1-2.el7                 extras  
docker-forward-journald.x86_64             1.10.3-44.el7.centos        extras  
docker-latest.x86_64                       1.12.1-2.el7.centos         extras  
docker-latest-logrotate.x86_64             1.12.1-2.el7.centos         extras  
docker-latest-v1.10-migrator.x86_64        1.12.1-2.el7.centos         extras  
docker-logrotate.x86_64                    1.10.3-46.el7.centos.14     extras  
docker-lvm-plugin.x86_64                   1.10.3-46.el7.centos.14     extras  
docker-novolume-plugin.x86_64              1.10.3-46.el7.centos.14     extras  
docker-python.x86_64                       1.4.0-115.el7               extras  
docker-registry.noarch                     0.6.8-8.el7                 extras  
docker-registry.x86_64                     0.9.1-7.el7                 extras  
docker-unit-test.x86_64                    1.10.3-46.el7.centos.14     extras  
docker-v1.10-migrator.x86_64               1.10.3-46.el7.centos.14     extras  
python-docker-py.noarch                    1.7.2-1.el7                 extras  
[root@localhost ~]#

2.2 安装docker

1 yum install docker -y

2.3 检查docker安装信息

[root@localhost ~]# docker versionClient: Version:         1.10.3 API version:     1.22 Package version: docker-common-1.10.3-46.el7.centos.14.x86_64 Go version:      go1.6.3 Git commit:      cb079f6-unsupported Built:           Fri Sep 16 13:24:25 2016 OS/Arch:         linux/amd64Cannot connect to the Docker daemon. Is the docker daemon running on this host?

3 flannel

3.1 检查flannel版本

[root@localhost etcd]# yum list |grep flannelflannel.x86_64                             0.5.3-9.el7                 @extras

3.2 安装flannel

yum install flannel

3.3 修改service配置

查看flannel的配置文件（使用yum安装会自动生成此文件，如果下载的执行文件则需要手动生成，在使用systemctl命令执行service 的时候会用到），可以看到flannel的service配置如下：

[root@localhost etcd]# more /usr/lib/systemd/system/flanneld.service[Unit]Description=Flanneld overlay address etcd agentAfter=network.targetAfter=network-online.targetWants=network-online.targetAfter=etcd.serviceBefore=docker.service[Service]Type=notifyEnvironmentFile=/etc/sysconfig/flanneldEnvironmentFile=-/etc/sysconfig/docker-networkExecStart=/usr/bin/flanneld -etcd-endpoints=${FLANNEL_ETCD} -etcd-prefix=${FLANNEL_ETCD_KEY} $FLANNEL_OPTIONSExecStartPost=/usr/libexec/flannel/mk-docker-opts.sh -k DOCKER_NETWORK_OPTIONS -d /run/flannel/dockerRestart=on-failure[Install]WantedBy=multi-user.targetRequiredBy=docker.service

其中所有的参数都配置在/etc/sysconfig/flanneld文件中，修改此文件，初始文件如下：

# Flanneld configuration options  # etcd url location.  Point this to the server where etcd runsFLANNEL_ETCD="http://127.0.0.1:2379"# etcd config key.  This is the configuration key that flannel queries# For address range assignmentFLANNEL_ETCD_KEY="/atomic.io/network"# Any additional options that you want to passFLANNEL_OPTIONS=""

其中

　　FLANNEL_ETCD：为ETCD的地址，

　　FLANNEL_ETCD_KEY：为在etcd中配置的网络参数的key　

　　FLANNEL_OPTIONS：为flannel的启动参数，我在这里加上了监听的网卡

根据前面步骤中etcd的配置，我们修改配置文件如下：

# Flanneld configuration options  # etcd url location.  Point this to the server where etcd runsFLANNEL_ETCD="http://192.168.37.130:2379"# etcd config key.  This is the configuration key that flannel queries# For address range assignmentFLANNEL_ETCD_KEY="/flannel/network"# Any additional options that you want to passFLANNEL_OPTIONS="--iface=eno16777736"

3.4 启动FLANNEL

可以使用service flanneld start 或者systemctl start flannel启动flannel

3.5 修改docker网络

因为docker需要使用flanneld的网络，因此需要修改docker的service文件：

[Unit]Description=Docker Application Container EngineDocumentation=http://docs.docker.comAfter=network.target rhel-push-plugin.socketWants=docker-storage-setup.service[Service]Type=notifyNotifyAccess=all#import flannel configuration EnvironmentFile=-/etc/sysconfig/flanneldEnvironmentFile=-/run/flannel/subnet.envEnvironmentFile=-/etc/sysconfig/dockerEnvironmentFile=-/etc/sysconfig/docker-storageEnvironmentFile=-/etc/sysconfig/docker-networkEnvironment=GOTRACEBACK=crashExecStart=/usr/bin/docker-current daemon \          --exec-opt native.cgroupdriver=systemd \          $OPTIONS \          $DOCKER_STORAGE_OPTIONS \          $DOCKER_NETWORK_OPTIONS \          $ADD_REGISTRY \          $BLOCK_REGISTRY \          $INSECURE_REGISTRY \          --bip=${FLANNEL_SUBNET}LimitNOFILE=1048576LimitNPROC=1048576LimitCORE=infinityTimeoutStartSec=0MountFlags=slaveRestart=on-abnormal[Install]WantedBy=multi-user.target

在执行前增加配置文件

EnvironmentFile=-/etc/sysconfig/flanneld

EnvironmentFile=-/run/flannel/subnet.env

执行命令增加参数 --bip=${FLANNEL_SUBNET}

重启docker

systemctl daemon-reloadsystemctl restart docker

3.6 问题

1、Failed to retrieve network config: 104: Not a directory (/flannel/network/config)

问题原因：在初次配置的时候，把flannel的配置文件中的etcd-prefix-key配置成了/flannel/network/config，实际上应该是/flannel/network

注意：如上配置需要在集群的所有机器上执行，完成后，上述安装的各个系统的启动顺序应该是：

systemctl start etcd

systemctl start flannel

systemctl start docker

配置完检查：

使用ip a检查当前的网络的准备情况：

[root@localhost system]# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00    inet 127.0.0.1/8 scope host lo       valid_lft forever preferred_lft forever    inet6 ::1/128 scope host        valid_lft forever preferred_lft forever2: eno16777736: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000    link/ether 00:0c:29:79:cf:e3 brd ff:ff:ff:ff:ff:ff    inet 192.168.37.130/24 brd 192.168.37.255 scope global dynamic eno16777736       valid_lft 1554sec preferred_lft 1554sec    inet6 fe80::20c:29ff:fe79:cfe3/64 scope link        valid_lft forever preferred_lft forever9: flannel0: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1472 qdisc pfifo_fast state UNKNOWN qlen 500    link/none     inet 172.17.75.0/16 scope global flannel0       valid_lft forever preferred_lft forever10: docker0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN     link/ether 02:42:e3:f0:0d:05 brd ff:ff:ff:ff:ff:ff    inet 172.17.75.1/24 scope global docker0       valid_lft forever preferred_lft forever

阅读全文

0 0