一、ETCD 概述
ETCD 是一個分布式一致性k-v存儲系統，可用於服務注冊發現與共享配置。具有一下優點:
- 簡單： 相比於晦澀難懂的paxos算法，etcd基於相對簡單且易實現的raft算法實現一致性，並通過gRPC提供接口調用
- 安全：支持TLS通信，並可以針對不同的用戶進行對key的讀寫控制
- 高性能：10，000/秒的寫性能
二、overlay網絡模式
容器在兩個跨主機通信的時候，是使用overlay network這個網絡模式進行通信，如果使用host也可以實現跨主機進行通信，直接使用這個物理的ip地址就可以進行通信。overlay它會虛擬出一個網絡比如10.0.9.3 這個ip地址，在這個overlay網絡模式里面，有一個類似於服務網關的地址，然后這個包轉發到物理服務器這個地址，最終通過路由和交換，到達另一個服務器的ip地址。
在docker容器里面overlay是怎么實現的呢？
我們會有一個服務發現，比如說consul，會定義一個ip地址池，比如10.0.9.0/24 之類的，上面會有容器，容器ip地址會從上面去獲取，獲取完了后，會通過eth1進行通信，賊這實現跨主機的東西。
三、部署ETCD集群
node1部署
[root@node1 ~]# wget https://github.com/coreos/etcd/releases/download/v3.0.12/etcd-v3.0.12-linux-amd64.tar.gz
[root@node1 ~]# tar zxf etcd-v3.0.12-linux-amd64.tar.gz
[root@node1 ~]# cd etcd-v3.0.12-linux-amd64/
[root@node1 etcd-v3.0.12-linux-amd64]# nohup ./etcd --name docker-node1 --initial-advertise-peer-urls http://10.211.55.12:2380 \
> --listen-peer-urls http://10.211.55.12:2380 \
> --listen-client-urls http://10.211.55.12:2379,http://127.0.0.1:2379 \
> --advertise-client-urls http://10.211.55.12:2379 \
> --initial-cluster-token etcd-cluster \
> --initial-cluster docker-node1=http://10.211.55.12:2380,docker-node2=http://10.211.55.13:2380 \
> --initial-cluster-state new&
[1] 32505
node2部署
[root@node2 ~]# wget https://github.com/coreos/etcd/releases/download/v3.0.12/etcd-v3.0.12-linux-amd64.tar.gz
[root@node2 ~]# tar zxf etcd-v3.0.12-linux-amd64.tar.gz
[root@node2 ~]# cd etcd-v3.0.12-linux-amd64/
[root@node2 etcd-v3.0.12-linux-amd64]# nohup ./etcd --name docker-node2 --initial-advertise-peer-urls http://10.211.55.13:2380 \
> --listen-peer-urls http://10.211.55.13:2380 \
> --listen-client-urls http://10.211.55.13:2379,http://127.0.0.1:2379 \
> --advertise-client-urls http://10.211.55.13:2379 \
> --initial-cluster-token etcd-cluster \
> --initial-cluster docker-node1=http://10.211.55.12:2380,docker-node2=http://10.211.55.13:2380 \
> --initial-cluster-state new&
[1] 19240
檢查cluster狀態
[root@node2 etcd-v3.0.12-linux-amd64]# ./etcdctl cluster-health
member 98da03f1eca9d9d is healthy: got healthy result from http://10.211.55.12:2379
member 63a987e985acb514 is healthy: got healthy result from http://10.211.55.13:2379
cluster is healthy
參數說明
參數說明：
● –data-dir 指定節點的數據存儲目錄，若不指定，則默認是當前目錄。這些數據包括節點ID，集群ID，集群初始化配置，Snapshot文件，若未指 定–wal-dir，還會存儲WAL文件
● –wal-dir 指定節點的was文件存儲目錄，若指定了該參數，wal文件會和其他數據文件分開存儲
● –name 節點名稱
● –initial-advertise-peer-urls 告知集群其他節點的URL，tcp2380端口用於集群通信
● –listen-peer-urls 監聽URL，用於與其他節點通訊
● –advertise-client-urls 告知客戶端的URL, 也就是服務的URL，tcp2379端口用於監聽客戶端請求
● –initial-cluster-token 集群的ID
● –initial-cluster 集群中所有節點
● –initial-cluster-state 集群狀態，new為新創建集群，existing為已存在的集群
四、管理etcd集群
1.查看集群版本
# etcdctl --version
# etcdctl --help
1
2
2.查看集群健康狀態
# etcdctl cluster-health
1
3.查看集群成員
# etcdctl member list
1
在任一節點上執行，可以看到集群的節點情況，並能看出哪個是leader節點
4.更新一個節點
如果你想更新一個節點的IP(peerURLS)，首先你需要知道那個節點的ID
# etcdctl member list
# etcdctl member update memberID http://ip:2380
1
2
5.刪除一個節點（Etcd集群成員的縮）
# etcdctl member list
# etcdctl member remove memberID
# etcdctl member list
# ps -ef|grep etcd //在相關節點上kill掉etcd進程
1
2
3
4
6.增加一個新節點（Etcd集群成員的伸）
注意:步驟很重要，不然會報集群ID不匹配
# etcdctl member add --help
1
a. 將目標節點添加到集群
# etcdctl member add etcd3 http://10.1.2.174:2380
Addedmember named etcd3 with ID 28e0d98e7ec15cd4 to cluster
ETCD_NAME="etcd3"
ETCD_INITIAL_CLUSTER="etcd0=http://10.1.2.61:2380,etcd1=http://10.1.2.172:2380,etcd2=http://10.1.2.173:2380,etcd3=http://10.1.2.174:2380"
ETCD_INITIAL_CLUSTER_STATE="existing"
1
2
3
4
5
6
b. 查看新增成員列表，etcd3狀態為unstarted
# etcdctl member list
d4f257d2b5f99b64[unstarted]:peerURLs=http://10.1.2.174:2380
1
2
c. 清空目標節點etcd3的data-dir
節點刪除后，集群中的成員信息會更新，新節點是作為一個全新的節點加入集群，如果data-dir有數據，etcd啟動時會讀取己經存在的數據，仍然用老的memberID會造成無法加入集群，所以一定要清空新節點的data-dir。
# rm -rf /path/to/etcd/data
1
d. 在目標節點上啟動新增加的成員
這里的initial標記一定要指定為existing，如果為new，則會自動生成一個新的memberID，這和前面添加節點時生成的ID不一致，故日志中會報節點ID不匹配的錯。
# vim etcd3.sh
1
修改etcd3.sh，腳本中–advertise-client-urls 和 –initial-advertis-peer-urls 參數修改為etcd3的，–initial-cluster-state改為existing
# nohup ./etcd3.sh &
# etcdctl member list
五、Docker使用ETCD分布式存儲
node1
[root@node1 ~]#service docker stop
[root@node1 ~]# /usr/bin/dockerd -H tcp://0.0.0.0:2375 -H unix:///var/run/docker.sock --cluster-store=etcd://10.211.55.12:2379 --cluster-advertise=10.211.55.12:2375&
node2
[root@node2 ~]#service docker stop
[root@node2 ~]# /usr/bin/dockerd -H tcp://0.0.0.0:2375 -H unix:///var/run/docker.sock --cluster-store=etcd://10.211.55.13:2379 --cluster-advertise=10.211.55.13:2375&
六、創建overlay network
在node1上創建一個demo的overlay network
[root@node1 ~]# docker network ls
NETWORK ID NAME DRIVER SCOPE
550d5b450fe3 bridge bridge local
cca92be73cc6 host host local
d21360748bfc none null local
[root@node1 ~]# docker network create -d overlay demo
97e959031044ec634d61d2e721cb0348d7ff852af3f575d75d2988c07e0f9846
[root@node1 ~]# docker network ls
NETWORK ID NAME DRIVER SCOPE
550d5b450fe3 bridge bridge local
97e959031044 demo overlay global
cca92be73cc6 host host local
d21360748bfc none null local
[root@node1 ~]# docker network inspect demo
[
{
"Name": "demo",
"Id": "97e959031044ec634d61d2e721cb0348d7ff852af3f575d75d2988c07e0f9846",
"Created": "2018-08-01T22:22:01.958142468+08:00",
"Scope": "global",
"Driver": "overlay",
"EnableIPv6": false,
"IPAM": {
"Driver": "default",
"Options": {},
"Config": [
{
"Subnet": "10.0.0.0/24",
"Gateway": "10.0.0.1"
}
]
},
"Internal": false,
"Attachable": false,
"Ingress": false,
"ConfigFrom": {
"Network": ""
},
"ConfigOnly": false,
"Containers": {},
"Options": {},
"Labels": {}
}
]
我們會看到在node2上，這個demo的overlay network會被同步創建
[root@node2 etcd-v3.0.12-linux-amd64]# cd
[root@node2 ~]# docker network ls
NETWORK ID NAME DRIVER SCOPE
c6af37ef6765 bridge bridge local
97e959031044 demo overlay global
de15cdab46b0 host host local
cc3ec612fd29 none null local
# 說明etcd分布式存儲已經起作用，兩個節點數據已同步
通過查看etcd的key-value, 我們獲取到，這個demo的network是通過etcd從node1同步到node2的
[root@node2 etcd-v3.0.12-linux-amd64]# ./etcdctl ls /docker
/docker/nodes
/docker/network
[root@node2 etcd-v3.0.12-linux-amd64]# ./etcdctl ls /docker/nodes
/docker/nodes/10.211.55.12:2375
/docker/nodes/10.211.55.13:2375
[root@node2 etcd-v3.0.12-linux-amd64]# ./etcdctl ls /docker/network/v1.0/network
/docker/network/v1.0/network/97e959031044ec634d61d2e721cb0348d7ff852af3f575d75d2988c07e0f9846
[root@node2 etcd-v3.0.12-linux-amd64]# ./etcdctl get /docker/network/v1.0/network/97e959031044ec634d61d2e721cb0348d7ff852af3f575d75d2988c07e0f9846
{"addrSpace":"GlobalDefault","attachable":false,"configFrom":"","configOnly":false,"created":"2018-08-01T22:22:01.958142468+08:00","enableIPv6":false,"generic":{"com.docker.network.enable_ipv6":false,"com.docker.network.generic":{}},"id":"97e959031044ec634d61d2e721cb0348d7ff852af3f575d75d2988c07e0f9846","inDelete":false,"ingress":false,"internal":false,"ipamOptions":{},"ipamType":"default","ipamV4Config":"[{\"PreferredPool\":\"\",\"SubPool\":\"\",\"Gateway\":\"\",\"AuxAddresses\":null}]","ipamV4Info":"[{\"IPAMData\":\"{\\\"AddressSpace\\\":\\\"GlobalDefault\\\",\\\"Gateway\\\":\\\"10.0.0.1/24\\\",\\\"Pool\\\":\\\"10.0.0.0/24\\\"}\",\"PoolID\":\"GlobalDefault/10.0.0.0/24\"}]","labels":{},"loadBalancerIP":"","name":"demo","networkType":"overlay","persist":true,"postIPv6":false,"scope":"global"}
七、創建連接demo網絡的容器
node1
[root@node1 ~]# docker run -d --name test1 --net demo busybox sh -c "while true; do sleep 3600; done"
Unable to find image 'busybox:latest' locally
latest: Pulling from library/busybox
8c5a7da1afbc: Pull complete
Digest: sha256:cb63aa0641a885f54de20f61d152187419e8f6b159ed11a251a09d115fdff9bd
Status: Downloaded newer image for busybox:latest
4bc3ab1cb7d838e8ef314618e6d3d878e744ef7842196a00b3999e6b6fe8402f
ERRO[2018-08-01T22:26:33.105124642+08:00] Failed to deserialize netlink ndmsg: invalid argument
INFO[0379] shim docker-containerd-shim started address="/containerd-shim/moby/4bc3ab1cb7d838e8ef314618e6d3d878e744ef7842196a00b3999e6b6fe8402f/shim.sock" debug=false pid=6630
[root@node1 ~]# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
4bc3ab1cb7d8 busybox "sh -c 'while true; …" 26 seconds ago Up 24 seconds test1
[root@node1 ~]# docker exec test1 ifconfig
eth0 Link encap:Ethernet HWaddr 02:42:0A:00:00:02
inet addr:10.0.0.2 Bcast:10.0.0.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1450 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
eth1 Link encap:Ethernet HWaddr 02:42:AC:12:00:02
inet addr:172.18.0.2 Bcast:172.18.255.255 Mask:255.255.0.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:31 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:4045 (3.9 KiB) TX bytes:0 (0.0 B)
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:65536 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
node2
[root@node2 ~]# docker run -d --name test1 --net demo busybox sh -c "while true; do sleep 3600; done"
Unable to find image 'busybox:latest' locally
latest: Pulling from library/busybox
56bec22e3559: Pull complete
Digest: sha256:29f5d56d12684887bdfa50dcd29fc31eea4aaf4ad3bec43daf19026a7ce69912
Status: Downloaded newer image for busybox:latest
fad6dc6538a85d3dcc958e8ed7b1ec3810feee3e454c1d3f4e53ba25429b290b
docker: Error response from daemon: service endpoint with name test1 already exists. # 容器已存在
[root@node2 ~]# docker run -d --name test2 --net demo busybox sh -c "while true; do sleep 3600; done"
9d494a2f66a69e6b861961d0c6af2446265bec9b1d273d7e70d0e46eb2e98d20
驗證連通性
[root@node2 etcd-v3.0.12-linux-amd64]# docker exec -it test2 ifconfig
eth0 Link encap:Ethernet HWaddr 02:42:0A:00:00:03
inet addr:10.0.0.3 Bcast:10.0.0.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1450 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
eth1 Link encap:Ethernet HWaddr 02:42:AC:12:00:02
inet addr:172.18.0.2 Bcast:172.18.255.255 Mask:255.255.0.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:32 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:4110 (4.0 KiB) TX bytes:0 (0.0 B)
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:65536 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
[root@node1 ~]# docker exec test1 sh -c "ping 10.0.0.3"
PING 10.0.0.3 (10.0.0.3): 56 data bytes
64 bytes from 10.0.0.3: seq=0 ttl=64 time=0.698 ms
64 bytes from 10.0.0.3: seq=1 ttl=64 time=1.034 ms
64 bytes from 10.0.0.3: seq=2 ttl=64 time=1.177 ms
64 bytes from 10.0.0.3: seq=3 ttl=64 time=0.708 ms
64 bytes from 10.0.0.3: seq=4 ttl=64 time=0.651 ms