etcd安装和所遇到的坑

首先参照 https://www.cnblogs.com/lyzw/p/6016789.html来安装

　　虚拟机：VMware® Workstation 12 Pro

　　系统：CentOS Linux release 7.2.1511 (Core) 3.10.0-327.el7.x86_64

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 etcd
etcd.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 Server
After=network.target
After=network-online.target
Wants=network-online.target

[Service]
Type=notify
WorkingDirectory=/var/lib/etcd/
EnvironmentFile=-/etc/etcd/etcd.conf
User=etcd
# set GOMAXPROCS to number of processors
ExecStart=/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-failure
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target

 并配置其配置文件

ETCD_NAME=zwetcd_2
ETCD_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 version
Client:
 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/amd64
Cannot 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 flannel
flannel.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 agent
After=network.target
After=network-online.target
Wants=network-online.target
After=etcd.service
Before=docker.service

[Service]
Type=notify
EnvironmentFile=/etc/sysconfig/flanneld
EnvironmentFile=-/etc/sysconfig/docker-network
ExecStart=/usr/bin/flanneld -etcd-endpoints=${FLANNEL_ETCD} -etcd-prefix=${FLANNEL_ETCD_KEY} $FLANNEL_OPTIONS
ExecStartPost=/usr/libexec/flannel/mk-docker-opts.sh -k DOCKER_NETWORK_OPTIONS -d /run/flannel/docker
Restart=on-failure

[Install]
WantedBy=multi-user.target
RequiredBy=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 runs
FLANNEL_ETCD="http://127.0.0.1:2379"

# etcd config key.  This is the configuration key that flannel queries
# For address range assignment
FLANNEL_ETCD_KEY="/atomic.io/network"

# Any additional options that you want to pass
FLANNEL_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 runs
FLANNEL_ETCD="http://192.168.37.130:2379"

# etcd config key.  This is the configuration key that flannel queries
# For address range assignment
FLANNEL_ETCD_KEY="/flannel/network"

# Any additional options that you want to pass
FLANNEL_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 Engine
Documentation=http://docs.docker.com
After=network.target rhel-push-plugin.socket
Wants=docker-storage-setup.service

[Service]
Type=notify
NotifyAccess=all
#import flannel configuration 
EnvironmentFile=-/etc/sysconfig/flanneld
EnvironmentFile=-/run/flannel/subnet.env
EnvironmentFile=-/etc/sysconfig/docker
EnvironmentFile=-/etc/sysconfig/docker-storage
EnvironmentFile=-/etc/sysconfig/docker-network
Environment=GOTRACEBACK=crash
ExecStart=/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=1048576
LimitNPROC=1048576
LimitCORE=infinity
TimeoutStartSec=0
MountFlags=slave
Restart=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-reload
systemctl 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 a
1: 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 forever
2: 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 forever
9: 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 forever
10: 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

如果看到到flannel0余docker0的网段相同，则网络配置成功。

在第一次安装完3台机器集群后没有问题，后面过段时间后要加入一个新节点，就碰到很多坑了

如http://www.mamicode.com/info-detail-2194387.html 此文说到的

        systemd启动etcd服务的时候出现错误：Failed at step CHDIR spawning /usr/bin/etcd: No such file or directory

　　　　　　解决办法：etcd.service服务配置文件中设置的工作目录WorkingDirectory=/var/lib/etcd/必须存在，否则会报以上错误

        systemd启动etcd服务的时候出现错误：cannot assign requested address

　　　　　　解决办法：绑定阿里云的私网IP

也有https://blog.csdn.net/u010087956/article/details/53670468

通过systemd托管的etcd数据备份还原无法启动服务并且报错

error listing data dir: /var/lib/etcd/default.etcd

    1

但是单独执行启动命令可以

/usr/bin/etcd --debug  --name=default --data-dir=/var/lib/etcd/default.etcd --listen-client-urls http://0.0.0.0:2379 --advertise-client-urls http://0.0.0.0:2380

    1

主要是还原目录时没有注意权限问题,systemd默认是以etcd用户执行的,这里需要修改default.etcd文件夹权限

chown etcd:etcd -R /var/lib/etcd/default.etcd

    1

参考文档

etcd can’t start due to status=1/FAILURE or status=200/CHDIR · Issue #3331 · coreos/etcd · GitHub