背景：文件服务器已经上线三个月，使用fastdfs做文件服务器的中间件，一台tracker服务器，两台storage互相备份（小公司你懂得，白手起家，资源有限，想怎么搞怎么搞），
从阿里云40G的存储快用的差不多了，是时候挂载一个数据盘进行扩容了。内心OS：扩容还不简单，挂载个硬盘，修改个storage配置文件不就行了，然而我就这样入坑了。
首先修改storage配置文件，vim /etc/fdfs/storage.conf
修改如下两点
# path(disk or mount point) count, default value is 1 store_path_count=2 # store_path#, based 0, if store_path0 not exists, it's value is base_path # the paths must be exist store_path0=/home/ubuntu/fastdfs/storage/storage0 store_path1=/mnt/data1
(1)store_path_count 文件路径数数：存储文件的硬盘数或挂载点数
(2)文件路径分别是什么
OK，重启storage
/usr/bin/fdfs_storaged /etc/fdfs/storage.conf restart
这个时候查看fdfs的运行情况
/usr/bin/fdfs_monitor /etc/fdfs/clint.conf
问题来了，
重启的这台storage，处于offline状态，另外一台正常active。。。
奔溃，分析日志找原因，
vim /home/ubuntu/fastdfs/tracker/logs/trackerd.log
[2017-12-11 15:20:16] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:17] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:18] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:19] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:20] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:21] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:22] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:23] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:24] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:25] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:26] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:27] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:28] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:29] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
[2017-12-11 15:20:30] ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
报了一堆错：ERROR - file: tracker_service.c, line: 3550, cmd=84, client ip: 106.15.137.142, package size 32 is not correct, expect length: 16
tracker报错了。。。
网上各种搜索，看源码，意思大概是数据包大小超了，但是如何解决，没有下文
各种尝试，最终试出来来了
两台storage的store_path_count 必须一致，吐血，想想也应该这样呀，不然存在M01的数据同步不到另外一台storage上
所以两台一起修改配置文件重启后，就OK了，
监控运行结果，两台storage均为active状态了
配合nginx使用需要修改mod_fastdfs.conf,修改和storage.conf一致，即可，OK，大功告成！
关注微信公众号每天学习一点程序员的职业经
