Mesos集群安装

Author： F嘉阳
发布时间：October 3, 2018
904views
No comments
8114 words
Categories：服务器开发

基础环境

4台服务器，一台外部客户机

安装指南

官方支持C++源码编译安装和Docker安装

停止所有容器命令

docker stop $(docker ps -a -q)
docker rm $(docker ps -a -q)

安装

mesos主节点

node1

docker pull mesosphere/mesos-master:1.7.0

node2(host)

docker run --name mesos-zookeeper -p 2181:2181 --restart always -d zookeeper:3.5

node1执行脚本(需替换zk IP地址)

MESOS_QUORUM=1即1为单机运行，3个节点为2，计算方式为 n÷2+1 \

若不手动指定hostname映射的IP，则所有节点默认通过主机名进行通信

使用host通信脚本

docker run -d --net=host \
  -e MESOS_PORT=5050 \
  -e MESOS_ZK=zk://192.168.0.32:2181/mesos \
  -e MESOS_QUORUM=1 \
  -e MESOS_REGISTRY=in_memory \
  -e MESOS_LOG_DIR=/var/log/mesos \
  -e MESOS_WORK_DIR=/var/tmp/mesos \
  -v "$(pwd)/log/mesos:/var/log/mesos" \
  -v "$(pwd)/work/mesos:/var/tmp/mesos" \
  -p 5050:5050 \
  --name=mesos-master \
  mesosphere/mesos-master:1.7.0 \

使用IP通信脚本

docker run -d --net=host \
  --hostname=192.168.0.32 \
  -e MESOS_PORT=5050 \
  -e MESOS_ZK=zk://192.168.0.32:2181/mesos \
  -e MESOS_QUORUM=1 \
  -e MESOS_REGISTRY=in_memory \
  -e MESOS_LOG_DIR=/var/log/mesos \
  -e MESOS_WORK_DIR=/var/tmp/mesos \
  -v "$(pwd)/log/mesos:/var/log/mesos" \
  -v "$(pwd)/work/mesos:/var/tmp/mesos" \
  -p 5050:5050 \
  --name=mesos-master \
  mesosphere/mesos-master:1.7.0 \
  --no-hostname_lookup \
  --ip=192.168.0.33

查看日志

docker logs mesos-master

成功输出日志

I1003 08:41:29.868827    14 group.cpp:341] Group process (zookeeper-group(2)@192.168.0.33:5050) connected to ZooKeeper
I1003 08:41:29.868902    14 group.cpp:831] Syncing group operations: queue size (joins, cancels, datas) = (0, 0, 0)
I1003 08:41:29.868921    14 group.cpp:419] Trying to create path '/mesos' in ZooKeeper
I1003 08:41:29.902194    14 contender.cpp:268] New candidate (id='1') has entered the contest for leadership
I1003 08:41:29.904409     8 detector.cpp:152] Detected a new leader: (id='1')
I1003 08:41:29.904513     8 group.cpp:700] Trying to get '/mesos/json.info_0000000001' in ZooKeeper
I1003 08:41:29.912968     8 zookeeper.cpp:262] A new leading master (UPID=master@192.168.0.33:5050) is detected
I1003 08:41:29.913091     8 master.cpp:2113] Elected as the leading master!
I1003 08:41:29.913115     8 master.cpp:1668] Recovering from registrar
I1003 08:41:29.916913    11 registrar.cpp:383] Successfully fetched the registry (0B) in 3.734016ms
I1003 08:41:29.917054    11 registrar.cpp:487] Applied 1 operations in 26us; attempting to update the registry
I1003 08:41:29.918582    11 registrar.cpp:544] Successfully updated the registry in 1.449984ms
I1003 08:41:29.918630    11 registrar.cpp:416] Successfully recovered registrar
I1003 08:41:29.918772     8 master.cpp:1782] Recovered 0 agents from the registry (121B); allowing 10mins for agents to reregister

访问IP:5050，对于PWD，可以访问http://SSHIP-5050.direct.labs.play-with-docker.com/#/，将SSH ip替换成实际提供的IP即可

mesos从节点

node3 (slave-1)、node4(slave-2)

启动脚本

使用host通信脚本

mkdir /cgroup
docker run -d --net=host --privileged \
  -e MESOS_PORT=5051 \
  -e MESOS_MASTER=zk://192.168.0.32:2181/mesos \
  -e MESOS_SWITCH_USER=0 \
  -e MESOS_CONTAINERIZERS=docker,mesos \
  -e MESOS_LOG_DIR=/var/log/mesos \
  -e MESOS_WORK_DIR=/var/tmp/mesos \
  -v "$(pwd)/log/mesos:/var/log/mesos" \
  -v "$(pwd)/tmp/mesos:/var/tmp/mesos" \
  -v /var/run/docker.sock:/var/run/docker.sock \
  -v /cgroup:/cgroup \
  -v /sys:/sys \
  -v "$(which docker):/usr/local/bin/docker" \
  --name=mesos-slave \
  -p 5051:5051 \
  mesosphere/mesos-slave:1.7.0 \
  --no-systemd_enable_support

使用IP通信脚本，--ip需要改成实际IP

mkdir /cgroup
docker run -d --net=host --privileged \
  -e MESOS_PORT=5051 \
  -e MESOS_MASTER=zk://192.168.0.32:2181/mesos \
  -e MESOS_SWITCH_USER=0 \
  -e MESOS_CONTAINERIZERS=docker,mesos \
  -e MESOS_LOG_DIR=/var/log/mesos \
  -e MESOS_WORK_DIR=/var/tmp/mesos \
  -v "$(pwd)/log/mesos:/var/log/mesos" \
  -v "$(pwd)/tmp/mesos:/var/tmp/mesos" \
  -v /var/run/docker.sock:/var/run/docker.sock \
  -v /cgroup:/cgroup \
  -v /sys:/sys \
  -v "$(which docker):/usr/local/bin/docker" \
  --name=mesos-slave \
  -p 5051:5051 \
  mesosphere/mesos-slave:1.7.0 \
  --no-systemd_enable_support \
  --no-hostname_lookup \
  --ip=192.168.0.31

查看日志

docker logs mesos-slave

启动成功日志

I1003 09:05:08.615039    12 group.cpp:341] Group process (zookeeper-group(1)@192.168.0.31:5051) connected to ZooKeeper
I1003 09:05:08.615104    12 group.cpp:831] Syncing group operations: queue size (joins, cancels, datas) = (0, 0, 0)
I1003 09:05:08.615121    12 group.cpp:419] Trying to create path '/mesos' in ZooKeeper
I1003 09:05:08.618291    13 state.cpp:66] Recovering state from '/var/tmp/mesos/meta'
I1003 09:05:08.618405    15 slave.cpp:6930] Finished recovering checkpointed state from '/var/tmp/mesos/meta', beginning agent recovery
I1003 09:05:08.618475     8 task_status_update_manager.cpp:207] Recovering task status update manager
I1003 09:05:08.618628     8 docker.cpp:908] Recovering Docker containers
I1003 09:05:08.626359    13 detector.cpp:152] Detected a new leader: (id='3')
I1003 09:05:08.626487    15 group.cpp:700] Trying to get '/mesos/json.info_0000000003' in ZooKeeper
I1003 09:05:08.628664    15 zookeeper.cpp:262] A new leading master (UPID=master@192.168.0.33:5050) is detected
I1003 09:05:08.997036    15 docker.cpp:921] Got the list of Docker containers
I1003 09:05:08.997324    15 docker.cpp:1157] Finished processing orphaned Docker containers
I1003 09:05:08.997436     8 slave.cpp:7159] Recovering executors
I1003 09:05:08.997553     8 slave.cpp:7312] Finished recovery
I1003 09:05:08.998044    11 task_status_update_manager.cpp:181] Pausing sending task status updates
I1003 09:05:08.998045     8 slave.cpp:1275] New master detected at master@192.168.0.33:5050
I1003 09:05:08.998159     8 slave.cpp:1329] No credentials provided. Attempting to register without authentication
I1003 09:05:08.998181     8 slave.cpp:1340] Detecting new master
I1003 09:05:09.446756    11 slave.cpp:1500] Registered with master master@192.168.0.33:5050; given agent ID b597695d-6e37-418a-9879-4ffe46117e17-S0
I1003 09:05:09.449091    10 task_status_update_manager.cpp:188] Resuming sending task status updates

访问主节点web界面可看见资源邀约有效

节点情况

marathon安装

node5

启动脚本

docker run -d --net=host \
  -p 8080:8080 \
  mesosphere/marathon:latest \
  --master zk://192.168.0.32:2181/mesos \
  --zk zk://192.168.0.32:2181/marathon

通过IP:8080查看运行结果

marathon-lb安装

node6

启动脚本

docker run -d -e PORTS=9090 --net=host \
  -p 9090:9090 \
  mesosphere/marathon-lb sse \
  --group external \
  --marathon http://192.168.0.29:8080

访问IP:9090/haproxy?stats可查看启动状态

Last modification：October 3rd, 2018 at 10:10 pm

如果觉得我的文章对你有用，请随意赞赏

Mesos集群安装

F嘉阳 • 2018 年 10 月 03 日

基础环境

4台服务器，一台外部客户机

安装指南

官方支持C++源码编译安装和Docker安装

C++源码编译安装地址

Docker镜像仓库

停止所有容器命令

docker stop $(docker ps -a -q)
docker rm $(docker ps -a -q)

安装

mesos主节点

node1

docker pull mesosphere/mesos-master:1.7.0

node2(host)

docker run --name mesos-zookeeper -p 2181:2181 --restart always -d zookeeper:3.5

node1执行脚本(需替换zk IP地址)

MESOS_QUORUM=1即1为单机运行，3个节点为2，计算方式为 n÷2+1 \

若不手动指定hostname映射的IP，则所有节点默认通过主机名进行通信

使用host通信脚本

docker run -d --net=host \
  -e MESOS_PORT=5050 \
  -e MESOS_ZK=zk://192.168.0.32:2181/mesos \
  -e MESOS_QUORUM=1 \
  -e MESOS_REGISTRY=in_memory \
  -e MESOS_LOG_DIR=/var/log/mesos \
  -e MESOS_WORK_DIR=/var/tmp/mesos \
  -v "$(pwd)/log/mesos:/var/log/mesos" \
  -v "$(pwd)/work/mesos:/var/tmp/mesos" \
  -p 5050:5050 \
  --name=mesos-master \
  mesosphere/mesos-master:1.7.0 \

使用IP通信脚本

docker run -d --net=host \
  --hostname=192.168.0.32 \
  -e MESOS_PORT=5050 \
  -e MESOS_ZK=zk://192.168.0.32:2181/mesos \
  -e MESOS_QUORUM=1 \
  -e MESOS_REGISTRY=in_memory \
  -e MESOS_LOG_DIR=/var/log/mesos \
  -e MESOS_WORK_DIR=/var/tmp/mesos \
  -v "$(pwd)/log/mesos:/var/log/mesos" \
  -v "$(pwd)/work/mesos:/var/tmp/mesos" \
  -p 5050:5050 \
  --name=mesos-master \
  mesosphere/mesos-master:1.7.0 \
  --no-hostname_lookup \
  --ip=192.168.0.33

查看日志

docker logs mesos-master

成功输出日志

I1003 08:41:29.868827    14 group.cpp:341] Group process (zookeeper-group(2)@192.168.0.33:5050) connected to ZooKeeper
I1003 08:41:29.868902    14 group.cpp:831] Syncing group operations: queue size (joins, cancels, datas) = (0, 0, 0)
I1003 08:41:29.868921    14 group.cpp:419] Trying to create path '/mesos' in ZooKeeper
I1003 08:41:29.902194    14 contender.cpp:268] New candidate (id='1') has entered the contest for leadership
I1003 08:41:29.904409     8 detector.cpp:152] Detected a new leader: (id='1')
I1003 08:41:29.904513     8 group.cpp:700] Trying to get '/mesos/json.info_0000000001' in ZooKeeper
I1003 08:41:29.912968     8 zookeeper.cpp:262] A new leading master (UPID=master@192.168.0.33:5050) is detected
I1003 08:41:29.913091     8 master.cpp:2113] Elected as the leading master!
I1003 08:41:29.913115     8 master.cpp:1668] Recovering from registrar
I1003 08:41:29.916913    11 registrar.cpp:383] Successfully fetched the registry (0B) in 3.734016ms
I1003 08:41:29.917054    11 registrar.cpp:487] Applied 1 operations in 26us; attempting to update the registry
I1003 08:41:29.918582    11 registrar.cpp:544] Successfully updated the registry in 1.449984ms
I1003 08:41:29.918630    11 registrar.cpp:416] Successfully recovered registrar
I1003 08:41:29.918772     8 master.cpp:1782] Recovered 0 agents from the registry (121B); allowing 10mins for agents to reregister

访问IP:5050，对于PWD，可以访问http://SSHIP-5050.direct.labs.play-with-docker.com/#/，将SSH ip替换成实际提供的IP即可

mesos从节点

node3 (slave-1)、node4(slave-2)

启动脚本

使用host通信脚本

mkdir /cgroup
docker run -d --net=host --privileged \
  -e MESOS_PORT=5051 \
  -e MESOS_MASTER=zk://192.168.0.32:2181/mesos \
  -e MESOS_SWITCH_USER=0 \
  -e MESOS_CONTAINERIZERS=docker,mesos \
  -e MESOS_LOG_DIR=/var/log/mesos \
  -e MESOS_WORK_DIR=/var/tmp/mesos \
  -v "$(pwd)/log/mesos:/var/log/mesos" \
  -v "$(pwd)/tmp/mesos:/var/tmp/mesos" \
  -v /var/run/docker.sock:/var/run/docker.sock \
  -v /cgroup:/cgroup \
  -v /sys:/sys \
  -v "$(which docker):/usr/local/bin/docker" \
  --name=mesos-slave \
  -p 5051:5051 \
  mesosphere/mesos-slave:1.7.0 \
  --no-systemd_enable_support

使用IP通信脚本，--ip需要改成实际IP

mkdir /cgroup
docker run -d --net=host --privileged \
  -e MESOS_PORT=5051 \
  -e MESOS_MASTER=zk://192.168.0.32:2181/mesos \
  -e MESOS_SWITCH_USER=0 \
  -e MESOS_CONTAINERIZERS=docker,mesos \
  -e MESOS_LOG_DIR=/var/log/mesos \
  -e MESOS_WORK_DIR=/var/tmp/mesos \
  -v "$(pwd)/log/mesos:/var/log/mesos" \
  -v "$(pwd)/tmp/mesos:/var/tmp/mesos" \
  -v /var/run/docker.sock:/var/run/docker.sock \
  -v /cgroup:/cgroup \
  -v /sys:/sys \
  -v "$(which docker):/usr/local/bin/docker" \
  --name=mesos-slave \
  -p 5051:5051 \
  mesosphere/mesos-slave:1.7.0 \
  --no-systemd_enable_support \
  --no-hostname_lookup \
  --ip=192.168.0.31

查看日志

docker logs mesos-slave

启动成功日志

I1003 09:05:08.615039    12 group.cpp:341] Group process (zookeeper-group(1)@192.168.0.31:5051) connected to ZooKeeper
I1003 09:05:08.615104    12 group.cpp:831] Syncing group operations: queue size (joins, cancels, datas) = (0, 0, 0)
I1003 09:05:08.615121    12 group.cpp:419] Trying to create path '/mesos' in ZooKeeper
I1003 09:05:08.618291    13 state.cpp:66] Recovering state from '/var/tmp/mesos/meta'
I1003 09:05:08.618405    15 slave.cpp:6930] Finished recovering checkpointed state from '/var/tmp/mesos/meta', beginning agent recovery
I1003 09:05:08.618475     8 task_status_update_manager.cpp:207] Recovering task status update manager
I1003 09:05:08.618628     8 docker.cpp:908] Recovering Docker containers
I1003 09:05:08.626359    13 detector.cpp:152] Detected a new leader: (id='3')
I1003 09:05:08.626487    15 group.cpp:700] Trying to get '/mesos/json.info_0000000003' in ZooKeeper
I1003 09:05:08.628664    15 zookeeper.cpp:262] A new leading master (UPID=master@192.168.0.33:5050) is detected
I1003 09:05:08.997036    15 docker.cpp:921] Got the list of Docker containers
I1003 09:05:08.997324    15 docker.cpp:1157] Finished processing orphaned Docker containers
I1003 09:05:08.997436     8 slave.cpp:7159] Recovering executors
I1003 09:05:08.997553     8 slave.cpp:7312] Finished recovery
I1003 09:05:08.998044    11 task_status_update_manager.cpp:181] Pausing sending task status updates
I1003 09:05:08.998045     8 slave.cpp:1275] New master detected at master@192.168.0.33:5050
I1003 09:05:08.998159     8 slave.cpp:1329] No credentials provided. Attempting to register without authentication
I1003 09:05:08.998181     8 slave.cpp:1340] Detecting new master
I1003 09:05:09.446756    11 slave.cpp:1500] Registered with master master@192.168.0.33:5050; given agent ID b597695d-6e37-418a-9879-4ffe46117e17-S0
I1003 09:05:09.449091    10 task_status_update_manager.cpp:188] Resuming sending task status updates

访问主节点web界面可看见资源邀约有效

节点情况

marathon安装

node5

启动脚本

docker run -d --net=host \
  -p 8080:8080 \
  mesosphere/marathon:latest \
  --master zk://192.168.0.32:2181/mesos \
  --zk zk://192.168.0.32:2181/marathon

通过IP:8080查看运行结果

marathon-lb安装

node6

启动脚本

docker run -d -e PORTS=9090 --net=host \
  -p 9090:9090 \
  mesosphere/marathon-lb sse \
  --group external \
  --marathon http://192.168.0.29:8080

访问IP:9090/haproxy?stats可查看启动状态

Mesos集群安装

基础环境

安装指南

停止所有容器命令

安装

mesos主节点

mesos从节点

marathon安装

marathon-lb安装

Leave a Comment Cancel reply

GraphQL实践10——Netflix Dgs Graphql异步订阅

hadoop分布式搭建的一些坑

LeetCode刷题记录

Typecho Pinghsu 主题加入备案信息

GitLab跨版本升级

Jenkins 安装配置

DevOps GitLab CICD 实践3——CI文件编写

GraphQL实践6——Netflix Dgs Graphql N+1问题

mybatis中遇到的坑

gitlab-ce安装配置

Mesos集群安装

基础环境

安装指南

停止所有容器命令

安装

mesos主节点

mesos从节点

marathon安装

marathon-lb安装