1、环境准备

3台centos7虚拟机

参照单机方式安装

2、修改配置文件

tracker.conf

和单机版本一致

storage.conf

192.168.244.128（group1）, 192.168.244.129（group1）两个配置文件

# 组名称 group_name=group1 # 基础路径，和单机保持一致 base_path=/home/fastdfs # 磁盘或挂载点数量 store_path_count=1 # 磁盘路径 store_path0=/home/fastdfs # tracker server集群地址，有几台服务器就使用几行配置 tracker_server=192.168.244.128:22122 tracker_server=192.168.244.129:22122 tracker_server=192.168.244.130:22122 # http访问端口 http.server_port=8888

192.168.244.130（group2）

# 组名称 这里是另外一个group group_name=group2 # 基础路径，和单地址机保持一致 base_path=/home/fastdfs # 磁盘或挂载点数量 store_path_count=1 # 磁盘路径 store_path0=/home/fastdfs # tracker server集群地是否址，有几台服务器就使用几行配置 tracker_server=192.168.244.128:22122 tracker_server=192.168.244.129:22122 tracker_server=192.168.244.130:22122 # http访问端口 http.server_port=8888

mod_fastdfs.conf

192.168.244.128（group1）, 192.168.244.129（group1）两个配置文件

# 跟踪服务器集群地址 tracker_server=192.168.244.128:22122 tracker_server=192.168.244.129:22122 tracker_server=192.168.244.130:22122 # storage服务器端口 storage_server_port=23000 # 组名称，自己本地的storage属于哪个组，就配置那个组的名称 group_name=group1 # 磁盘或挂载点数量 store_path_count=1 # 磁盘路径 store_pat方式h0=/home/fastdfs

192.168.244.130（group2）

#是否 跟踪服务器集群方式地址 tracker_server=192.168.244.128:22122 tracker_server=192.168.244.129:22122 tracker_server=192.168.244.130:22122 # storage服务器端口 storage_server_port=23000 # 组名称，自己本地的storage属于哪个组，就配置那个组的名称 group_name=group2 # 磁盘或挂载点数量 store_path_count=1 # 磁盘路径 store_path0=/home/fastdfs

client.conf

三台机器一样

# 基础路径 base_path = /home/fastdfs # 跟踪服务器集群地址 tracker_server = 192.168.244.128:22122 tracker_server = 192.168.244.129:22122 tracker_server = 192.168.244.130:22122

3、重启tracker， storage, nginx

分别在三台机器执行

# 重启tracker fdfs_trackerd /etc/fdfs/tracker.conf restart # 重启storage fdfs_storaged /etc/fdfs/storage.conf restart # 重启nginx /usr/local/nginx/sbin/nginx -s reload

4、测试上传下载

使用client上传

[root@centos01 fdfs]# fdfs_upload_file /etc/fdfs/client.conf /opt/test/text.jpg group1/M00/00/00/wKj0gGIm_qSAGrEKAABDhrJAkTU406.jpg [root@centos01 fdfs]# fdfs_upload_file /etc/fdfs/client.conf /opt/test/text.jpg group2/M00/00/00/wKj0gmIm_qiAcjR_AABDhrJAkTU7使用52.jpg

使用client查看文件信息

# 执行命令查看group1 fdfs_file_info /etc/fdfs/client.conf group1/M00/00/00/wKj0gGIm_qSAGrEKAABDhrJAkTU406.jpg # 结果显示 source storage id: 0 source ip address: 192.168.244.128 file create timestamp: 2022-03-08 14:58:44 file size: 17286 file crc32: 2990575925 (0xB24意思09135) # 执行命令查看group2 fdfs_file_info /etc/fdfs/client.conf group2/M00/00/00/wK是什么j0gmIm_qiAcjR_AABDhrJAkTU752.jpg # 结果显示 source storage id: 0 source ip address: 192.168.244.130 file create timestamp: 2022-03-08 14:58:48 file size: 17286 file crc32: 2990575925 (0xB2409135)

使用nginx访问图片

5、FastDFS集群优化

1、最大使用连接数优化

在tracker.conf和storage.conf中均存在该配置项

# max concurrent connections this server support # you should set this parameter larger, eg. 10240 # default value is 256 # 默认值是256，表示一次性预分配这么集群多。到那时不能超过/etc/secerity/limits.conf中软硬件的限制 max_connections = 1024

可以配置limits.conf

* soft nofile 65535 * hard nofile 65535

2、工作线程数量配置

在tracker.conf和storage.conf中均存在该配置项

# work thread count # work threads to deal network io # default value is 4 # since V2.00 一般和CPU的数量相等即可 work_threads = 4

3、storage目录数设置

在storage.conf中存在该配置项

# subdir_count * subdir_count directories will be auto created under each # store_path (disk), value c是什么an be 1 to 256, default value is 256。适当减小效率会更高 subdir_count_per_p意思ath=256

4、storage磁盘读写线程

在storage.conf中存在该配置项

# if disk read / write separated ## false for mixed read and write ## true for separated read and write # default value is 地址true # since V2.00 # 读写分离是否开启，默认开启 disk_rw_separated = true # disk reader thread count per store base path # for mixed read / write, this parameter can be 0 # default value is 1 # since V2.00 # 单个磁盘读线程数量 disk_reader_threads = 1 # disk writer thread count per store base path # for 集群mixed read / write, this parameter can be 0 # default value is 1 # since V2.00 # 单个磁盘写线程数量 disk_writer_threads = 1

如果机器性能好，可以适当增加读写线程的数量。