Docker Swarm

 

♠  Docker swarm 에 대한 내용 참조

https://docs.docker.com/engine/swarm/

https://docs.docker.com/reference/cli/docker/swarm/init/

Swarm mode overview

 

 

 

▶ Docker swarm 생성 ( manager 노드가 되었다 )

>> 두번째 장비에는 아래 내용중 docker swarm join --token ~ 명령어를 입력한다

>> join 을 위해서는 token 이 필요하다 >>> 그대로 복사 붙여넣기 ( 보안상 좋지는 않다 )

>> 명령어 그대로를 파일로 만들어서 넘겨주는게 낫다

[vagrant@docker1 ~]$ docker swarm init --advertise-addr 192.168.25.10
Swarm initialized: current node (ln6o25x1z658c6gwga1axt5wl) is now a manager.

To add a worker to this swarm, run the following command:

    docker swarm join --token SWMTKN-1-1ks4brrs0tgvukyutgu7rsmv3tsaraz0um2afcrb71zym2x2p3-1ao8wy3buqmjqmxwfs0pu2pwa 192.168.25.10:2377

To add a manager to this swarm, run 'docker swarm join-token manager' and follow the instructions.

 

 

 

 

▶ 파일로 만들어서 넘겨준다

[vagrant@docker1 ~]$ echo 'docker swarm join --token SWMTKN-1-1ks4brrs0tgvukyutgu7rsmv3tsaraz0um2afcrb71zym2x2p3-1ao8wy3buqmjqmxwfs0pu2pwa 192.168.25.10:2377' > swarm-token

 

>> scp 로 복사해준다

[vagrant@docker1 ~]$ scp swarm-token 192.168.25.20:/home/vagrant
vagrant@192.168.25.20's password:
swarm-token                                                       100%  131   219.6KB/s   00:00

 

 

 

 

▶ ubuntu 에서 권한 줘서 실행한다

vagrant@docker2:~$ ll
-rw-rw-r-- 1 vagrant vagrant       131 Mar  7 10:05 swarm-token

vagrant@docker2:~$ chmod u+x swarm-token
-rwxrw-r-- 1 vagrant vagrant       131 Mar  7 10:05 swarm-token*

vagrant@docker2:~$ ./swarm-token
This node joined a swarm as a worker.

 

 

 

 

▶centos 에서의 포트 번호를 확인해보자

>> 2377 포트번호가 도커데몬으로 사용된다는 것을 확인 가능하다

[vagrant@docker1 ~]$ sudo yum install -y net-tools

[vagrant@docker1 ~]$ sudo netstat -nltp
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name
tcp        0      0 0.0.0.0:22              0.0.0.0:*               LISTEN      690/sshd
tcp        0      0 127.0.0.1:25            0.0.0.0:*               LISTEN      790/master
tcp        0      0 0.0.0.0:111             0.0.0.0:*               LISTEN      380/rpcbind
tcp6       0      0 :::22                   :::*                    LISTEN      690/sshd
tcp6       0      0 ::1:25                  :::*                    LISTEN      790/master
tcp6       0      0 :::2377                 :::*                    LISTEN      750/dockerd
tcp6       0      0 :::7946                 :::*                    LISTEN      750/dockerd
tcp6       0      0 :::111                  :::*                    LISTEN      380/rpcbind

 

 

▷ docker node 들이 묶여 있는 걸 확인할 수 있다

>> * 표시가 있으면 manager

[vagrant@docker1 ~]$ docker node ls
ID                            HOSTNAME              STATUS    AVAILABILITY   MANAGER STATUS   ENGINE VERSION
ln6o25x1z658c6gwga1axt5wl *   docker1.example.com   Ready     Active         Leader           25.0.3
257jeo37jv4q0ewrbvi9e84l0     docker2               Ready     Active                          25.0.3

 

○ ubuntu 에서 같은 명령어 실행시, worker node 라고 한다

vagrant@docker2:~$ docker node ls
Error response from daemon: This node is not a swarm manager. Worker nodes can't be used to view or modify cluster state. Please run this command on a manager node or promote the current node to a manager.

 

 

 

■ docker swarm 관련 명령어

docker swarm init : docker swarm 초기화
docker node ls  : cluster node 확인
docker service create --name myalpine --replicas 2 alpine /bin/sleep 3600s
docker service ls
docker service rm myalpine
docker service create --name mysql --replicas 2 -e MYSQL_ROOT_PASSWORD=mypass  -p 9900:3306 mysql

 

 

 

◎ 도커 클러스트 해제할 때에는 시간이 좀 걸린다

 

○ ubuntu 에서 떠난다

vagrant@docker2:~$ docker swarm leave
Node left the swarm.

 

○ centos 확인

[vagrant@docker1 ~]$ docker node ls
ID                            HOSTNAME              STATUS    AVAILABILITY   MANAGER STATUS   ENGINE VERSION
ln6o25x1z658c6gwga1axt5wl *   docker1.example.com   Ready     Active         Leader           25.0.3
257jeo37jv4q0ewrbvi9e84l0     docker2               Down      Active                          25.0.3

 

 

 

 

▷ 도커 서비스로 컨테이너를 만든다

>> replicas 몇개의 컨테이너를 띄울 것인지 결정

>> 서비스를 올리면 verify 로도 확인을 한다 ( 검증까지 한다 )

>> 자동으로 부하분산도 해준다 

>>> 예를 들면, 4개의 서비스를 올리면, node worker 에 2개 manaver 에 2개를 올려준다

[vagrant@docker1 ~]$ docker service create --name alpine-service --replicas 2 alpine /bin/sleep 3600s
iwwk3qn9pn3zeqx6w5ouf2hbg
overall progress: 2 out of 2 tasks
1/2: running
2/2: running
verify: Service converged

 

 

□ docker service 옵션

[vagrant@docker1 ~]$ docker service

Usage:  docker service COMMAND

Manage Swarm services

Commands:
  create      Create a new service
  inspect     Display detailed information on one or more services
  logs        Fetch the logs of a service or task
  ls          List services
  ps          List the tasks of one or more services
  rm          Remove one or more services
  rollback    Revert changes to a service's configuration
  scale       Scale one or multiple replicated services
  update      Update a service
  
  
[vagrant@docker1 ~]$ docker ps
CONTAINER ID   IMAGE     COMMAND   CREATED   STATUS    PORTS     NAMES

 

▩ 도커 서비스를 지웠는데도 docker ps 에서는 출력되는데, 아마도 꺼지는데 시간이 걸리는듯 싶다

[vagrant@docker1 ~]$ docker service rm alpine-service
alpine-service
[vagrant@docker1 ~]$ docker ps
CONTAINER ID   IMAGE           COMMAND              CREATED         STATUS         PORTS     NAMES
bc82f062282e   alpine:latest   "/bin/sleep 3600s"   2 minutes ago   Up 2 minutes             alpine-service.1.sym7ufyokqfoh110218dvxfes
8d1db1510090   alpine:latest   "/bin/sleep 3600s"   2 minutes ago   Up 2 minutes             alpine-service.2.o4bbfm4b627n5qz3s75c38pd9

 

 

 

□ 부하 분산 확인을 위해서 다시 클러스터에 묶는다

vagrant@docker2:~$ ./swarm-token
This node joined a swarm as a worker.

[vagrant@docker1 ~]$ docker node ls
ID                            HOSTNAME              STATUS    AVAILABILITY   MANAGER STATUS   ENGINE VERSION
ln6o25x1z658c6gwga1axt5wl *   docker1.example.com   Ready     Active         Leader           25.0.3
257jeo37jv4q0ewrbvi9e84l0     docker2               Down      Active                          25.0.3
qn99hviik4t1hlh16bvot4ula     docker2               Ready     Active                          25.0.3

 

 

 

 

● alpine-service 를 2개 올려보았다

>> 분산되었다

[vagrant@docker1 ~]$ docker service create --name alpine-service --replicas 2 alpine /bin/sleep 3600s
edsz3usp7n5zow5kee8t1tvd6
overall progress: 2 out of 2 tasks
1/2: running
2/2: running
verify: Service converged

[vagrant@docker1 ~]$ docker service ls
ID             NAME             MODE         REPLICAS   IMAGE           PORTS
edsz3usp7n5z   alpine-service   replicated   2/2        alpine:latest

vagrant@docker2:~$ docker ps
CONTAINER ID   IMAGE           COMMAND              CREATED          STATUS          PORTS     NAMES
886075ccd787   alpine:latest   "/bin/sleep 3600s"   23 seconds ago   Up 23 seconds             alpine-service.2.wevodxi53c4umjpyjp7rrfrwn

 

 

 

● 서비스 도중에 추가 업데이트도 가능하다

[vagrant@docker1 ~]$ docker service update  alpine-service --replicas 4
alpine-service
overall progress: 4 out of 4 tasks
1/4: running
2/4: running
3/4: running
4/4: running
verify: Service converged

 

>> 확인이 된다

[vagrant@docker1 ~]$ docker service ls
ID             NAME             MODE         REPLICAS   IMAGE           PORTS
edsz3usp7n5z   alpine-service   replicated   4/4        alpine:latest

 

 

▷ 추가로 늘어난 2개 service 도 부하분산 되었다

[vagrant@docker1 ~]$ docker ps
CONTAINER ID   IMAGE           COMMAND              CREATED              STATUS              PORTS     NAMES
8269c211cc4c   alpine:latest   "/bin/sleep 3600s"   About a minute ago   Up About a minute             alpine-service.3.zrzu6u8po4xq5kyciycpo98yi
21b930b09fa5   alpine:latest   "/bin/sleep 3600s"   3 minutes ago        Up 3 minutes                  alpine-service.1.4chr33kfwhorqrcsh9z0wijhy

vagrant@docker2:~$ docker ps
CONTAINER ID   IMAGE           COMMAND              CREATED              STATUS              PORTS     NAMES
7af2f1916d2d   alpine:latest   "/bin/sleep 3600s"   About a minute ago   Up About a minute             alpine-service.4.h915m1ift4bpmpoxzpbhh1t5y
886075ccd787   alpine:latest   "/bin/sleep 3600s"   2 minutes ago        Up 2 minutes                  alpine-service.2.wevodxi53c4umjpyjp7rrfrwn

 

 

 

▶ 아파치 서비스를 1개 더 올려본다

[vagrant@docker1 ~]$ docker service create --name apache-service --replicas 1 httpd
sujvqf79xra3pan562250wf9b
overall progress: 1 out of 1 tasks
1/1: running
verify: Service converged

[vagrant@docker1 ~]$ docker service ls
ID             NAME             MODE         REPLICAS   IMAGE           PORTS
edsz3usp7n5z   alpine-service   replicated   4/4        alpine:latest
sujvqf79xra3   apache-service   replicated   1/1        httpd:latest

 

▶ 서비스 확인

[vagrant@docker1 ~]$ docker service ps apache-service
ID             NAME               IMAGE          NODE                  DESIRED STATE   CURRENT STATE           ERROR     PORTS
7ds4fe3q8yeo   apache-service.1   httpd:latest   docker1.example.com   Running         Running 2 minutes ago

[vagrant@docker1 ~]$ docker ps
CONTAINER ID   IMAGE           COMMAND              CREATED          STATUS          PORTS     NAMES
e144adc96623   httpd:latest    "httpd-foreground"   2 minutes ago    Up 2 minutes    80/tcp    apache-service.1.7ds4fe3q8yeodrjzbz73crp76
8269c211cc4c   alpine:latest   "/bin/sleep 3600s"   18 minutes ago   Up 18 minutes             alpine-service.3.zrzu6u8po4xq5kyciycpo98yi
21b930b09fa5   alpine:latest   "/bin/sleep 3600s"   20 minutes ago   Up 20 minutes             alpine-service.1.4chr33kfwhorqrcsh9z0wijhy

 

 

▷아파치의 IP 확인

>> replica 이기 때문에 apache-service.1 뒤에 긴 내용이 적힌다

[vagrant@docker1 ~]$ docker container inspect apache-service.1.7ds4fe3q8yeodrjzbz73crp76

 

>> apache update

[vagrant@docker1 ~]$ docker service update apache-service --replicas 3
apache-service
overall progress: 3 out of 3 tasks
1/3: running
2/3: running
3/3: running
verify: Service converged

 

 

 

▶ ubuntu 에 있는 service 가 늘어난 상황에서 ubuntu 장비를 강제로 종료해보자

vagrant@docker2:~$ docker ps
CONTAINER ID   IMAGE           COMMAND              CREATED          STATUS          PORTS     NAMES
8b75b9bd5453   httpd:latest    "httpd-foreground"   52 seconds ago   Up 51 seconds   80/tcp    apache-service.3.kbgprr1n5gxjsfxq4rddpgzy4
f452524f1024   httpd:latest    "httpd-foreground"   52 seconds ago   Up 51 seconds   80/tcp    apache-service.2.aa5k0s1na2i7om1gps1n8y8uh
7af2f1916d2d   alpine:latest   "/bin/sleep 3600s"   27 minutes ago   Up 27 minutes             alpine-service.4.h915m1ift4bpmpoxzpbhh1t5y
886075ccd787   alpine:latest   "/bin/sleep 3600s"   29 minutes ago   Up 29 minutes             alpine-service.2.wevodxi53c4umjpyjp7rrfrwn

 

>> centos 장비 service

[vagrant@docker1 ~]$ docker ps
CONTAINER ID   IMAGE           COMMAND              CREATED          STATUS          PORTS     NAMES
e144adc96623   httpd:latest    "httpd-foreground"   10 minutes ago   Up 10 minutes   80/tcp    apache-service.1.7ds4fe3q8yeodrjzbz73crp76
8269c211cc4c   alpine:latest   "/bin/sleep 3600s"   27 minutes ago   Up 27 minutes             alpine-service.3.zrzu6u8po4xq5kyciycpo98yi
21b930b09fa5   alpine:latest   "/bin/sleep 3600s"   29 minutes ago   Up 29 minutes             alpine-service.1.4chr33kfwhorqrcsh9z0wijhy

 

▷ centos 로 모든 서비스가 올라왔다

>> 장애 극복이 가능하다

[vagrant@docker1 ~]$ docker ps
CONTAINER ID   IMAGE           COMMAND              CREATED          STATUS          PORTS     NAMES
4d9ad0dff125   httpd:latest    "httpd-foreground"   7 seconds ago    Up 1 second     80/tcp    apache-service.2.sow07j8o5qpey8hfku7ywwuus
9c6a705b62b7   alpine:latest   "/bin/sleep 3600s"   7 seconds ago    Up 1 second               alpine-service.2.kbugcuounh5d1tw0n46j6mlr9
a34c36c3facd   alpine:latest   "/bin/sleep 3600s"   7 seconds ago    Up 1 second               alpine-service.4.d0v0mwta9fe1i4mjacj9fatgx
461e35705bf0   httpd:latest    "httpd-foreground"   7 seconds ago    Up 1 second     80/tcp    apache-service.3.na1p56atg8yrigc2nuf3114v3
e144adc96623   httpd:latest    "httpd-foreground"   13 minutes ago   Up 12 minutes   80/tcp    apache-service.1.7ds4fe3q8yeodrjzbz73crp76
8269c211cc4c   alpine:latest   "/bin/sleep 3600s"   29 minutes ago   Up 29 minutes             alpine-service.3.zrzu6u8po4xq5kyciycpo98yi
21b930b09fa5   alpine:latest   "/bin/sleep 3600s"   31 minutes ago   Up 31 minutes             alpine-service.1.4chr33kfwhorqrcsh9z0wijhy