Fastdfs文件存储服务搭建-部署流程（二）

**

2.1 安装libfastcommon

**
shell> cd /usr/local/src
shell> unzip libfastcommon-master.zip
shell> cd /opt/local/fast

编译、安装

shell> ./make.sh
shell> ./make.sh install
执行结果如下：
mkdir -p /usr/lib64
mkdir -p /usr/lib
mkdir -p /usr/include/fastcommon
install -m 755 libfastcommon.so /usr/lib64
安装目录是/usr/bin,如果libfastcommon安装为/usr/local/bin,需要修改文件

**

2.2 安装fastdfs

**
解压安装fastdfs
shell>tar -zxvf /tmp/fastdfs-5.11.tar.gz -C /opt/local/fast
编译 安装
shell>cd /opt/local/fast/fastdfs-5.11
shell>./make.sh
shell>./make.sh install
执行结果如下
mkdir -p /usr/bin
mkdir -p /etc/fdfs
mkdir -p /usr/lib64
mkdir -p /usr/lib
shell>vim /etc/init.d/fdfs_trackerd
shell>vim /etc/init.d/fdfs_storeaged
将/usr/local/bin 替换为/usr/bin
shell>%s+/usr/local/bin+/usr/bin(5.11版本不需要替换)
在tracker和storage节点都安装一遍

**

2.3 配置tracker

**
shell>vim /etc/fdfs/tracker.conf
Port=22122
tracker.conf中 开始使用的默认的bind_addr=0.0.0.0修改为
bind_addr=10.110.60.72（本机ip地址）
base_path=/data/fastdfs/tracker

# the method of selecting group to upload files
# 0: round robin
# 1: specify group
# 2: load balance, select the max free space group to upload file

store_lookup=0
tracker1和tracker2两个节点配置一致，注意bind_addr和本机ip一致

**

2.4 启动和停止tracker

**
启动tracker
shell> /etc/init.d/fdfs_trackerd start
停止tracker
shell> /etc/init.d/fdfs_trackerd stop

**

2.5 配置storage

**
第一组storage配置
shell> mkdir -p /data/fastdfs/storage
shell> vim /etc/fdfs/storage.conf
group_name=group1
base_path=/data/fastdfs/storage
store_path_count=1
store_path0=/data/fastdfs/storage
tracker_server=10.110.60.72:22122
tracker_server=10.110.60.73:22122
http.server_port=8888

第二组storage配置
shell> mkdir -p /data/fastdfs/storage
shell> vim /etc/fdfs/storage.conf
group_name=group2
base_path=/data/fastdfs/storage
store_path_count=1
store_path0=/data/fastdfs/storage
tracker_server=10.110.60.72:22122
tracker_server=10.110.60.73:22122
http.server_port=8888
只是分组group_name不同,同一group为一组，数据相同，互为备份，数据大小以最小节点存储为准；

**

2.6 启动和停止storage

**

启动storage
shell> /etc/init.d/fdfs_storaged start
停止storage
shell> /etc/init.d/fdfs_storaged stop

2.6.1 异常处理
Tracker启动后正常，当storage启动后tracker和storage都报错误；
Tracker错误日志如下：
10.110.60.72错误日志如下：
[2019-07-17 11:40:18] ERROR - file: tracker_service.c, line: 883, client ip: 10.110.60.73, leader 10.110.60.73:22122 not exist
10.110.60.73错误日志如下：
[2019-07-17 11:40:46] INFO - file: tracker_relationship.c, line: 383, selecting leader…
[2019-07-17 11:40:46] ERROR - file: tracker_proto.c, line: 48, server: 10.110.60.72:22122, response status 2 != 0
解决办法
将leader 10.110.60.73的data目录下的文件除fdfs_trackerd.pid复制到10.110.60.72上，重新启动10.110.60.72服务即可；

启动完成后tracker如下：
10.110.60.72日志
[2019-07-17 11:50:16] INFO - file: tracker_relationship.c, line: 383, selecting leader…
[2019-07-17 11:50:16] INFO - file: tracker_relationship.c, line: 422, the tracker leader 10.110.60.73:22122
10.110.60.73日志
[2019-07-17 11:40:56] INFO - file: tracker_relationship.c, line: 401, I am the new tracker leader 10.110.60.73:22122

Storage启动日志
10.110.60.171日志
同一组storage加入后会自动创建连接
[2019-07-17 11:48:09] INFO - file: storage_func.c, line: 254, tracker_client_ip: 10.110.60.171, my_server_id_str: 10.110.60.171, g_server_id_in_filename: -1422103030
[2019-07-17 11:48:09] INFO - file: tracker_client_thread.c, line: 310, successfully connect to tracker server 10.110.60.73:22122, as a tracker client, my ip is 10.110.60.171
[2019-07-17 11:48:09] INFO - file: tracker_client_thread.c, line: 1235, tracker server 10.110.60.73:22122, set tracker leader: 10.110.60.73:22122
[2019-07-17 11:48:10] INFO - file: tracker_client_thread.c, line: 310, successfully connect to tracker server 10.110.60.72:22122, as a tracker client, my ip is 10.110.60.171
[2019-07-17 12:03:09] INFO - file: storage_sync.c, line: 2698, successfully connect to storage server 10.110.60.172:23000
**

2.7 storage集群状态查看

**
所有存储节点都启动之后，可以在任一存储节点上使用如下命令查看集群的状态信息
shell> /usr/bin/fdfs_monitor /etc/fdfs/storage.conf
**

2.8 storage开机启动设置

**
设置storage服务开机启动
shell> chkconfig fdfs_storaged on
shell> chkconfig --list fdfs_storaged