docker compose部署cassandra集群

docker compose 配置

假设有两台电脑
A电脑的ip为192.168.1.100
B电脑的ip为192.168.1.103
A电脑的docker compose 配置

version: '3'

services:

  cassandra:
    restart: always
    image: cassandra:3.11.10
    hostname: cassandra1
    container_name: cassandra-node-1
    environment:
      - CASSANDRA_BROADCAST_ADDRESS=cassandra1
      - CASSANDRA_SEEDS=cassandra1,cassandra3
    extra_hosts:
      - "cassandra1:192.168.1.100"
      - "cassandra3:192.168.1.103"
    ports:
      - "9042:9042"
      - "7000:7000"
    volumes:
      - cassandra_data:/var/lib/cassandra

volumes:
  cassandra_data:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

B电脑的docker compose 配置

version: '3'

services:

  cassandra:
    restart: always
    image: cassandra:3.11.10
    hostname: cassandra3
    container_name: cassandra-node-3
    environment:
      - CASSANDRA_BROADCAST_ADDRESS=cassandra3
      - CASSANDRA_SEEDS=cassandra1,cassandra3
    extra_hosts:
      - "cassandra1:192.168.1.100"
      - "cassandra3:192.168.1.103"
    ports:
      - "9042:9042"
      - "7000:7000"
    volumes:
      - cassandra_data:/var/lib/cassandra

volumes:
  cassandra_data:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

设置内存

临时

sysctl -w vm.max_map_count=262144
1

永久

Disable memory paging and swapping performance on the host to improve performance.
禁用主机上的内存分页和交换性能以提高性能。
ps：这个看情况，内存大的话，也可以关掉

sudo swapoff -a
1

Increase the number of memory maps available to OpenSearch.
增加OpenSearch可用的内存映射数量。

# Edit the sysctl config file
sudo vi /etc/sysctl.conf

# Add a line to define the desired value
# or change the value if the key exists,
# and then save your changes.
vm.max_map_count=262144

# Reload the kernel parameters using sysctl
sudo sysctl -p

# Verify that the change was applied by checking the value
cat /proc/sys/vm/max_map_count
1
2
3
4
5
6
7
8
9
10
11
12
13

两边同时启动

docker compose up -d
1

查看集群状态

#在A机器pingB机器
docker exec -ti cassandra-node-1 cqlsh -u cassandra -pcassandra cassandra3 -e "DESCRIBE CLUSTER"
1
2

返回

Cluster: Test Cluster
Partitioner: Murmur3Partitioner
1
2

连接成功，可以使用数据库连接工具插入一条数据，分别连入A和B的数据库，查看数据是否一致