请教FastDHT若干个问题！

lifeblood

1.FastDHT安装配置好以后，是直接可以去重了，还是要在客户端写php程序
2.一台FastDHT能支持2个storage吗？
3.fdht_servers.conf这个配置文件里
group_count = 1
group0 = 192.168.1.221:11411
#group0 = 192.168.1.221:11411
这个ip地址是指什么
如果有两台storage要怎么配置？

happy_fish100

1. 内置支持去重，配置文件中要把一个选项打开
2. FDHT server IP，一个group有多台，配置多行即可，例如：
group_count = 1
group0 = 192.168.1.221:11411
group0 = 192.168.1.222:11411

lifeblood

追问，上传图片，出现以下问题

[2014-05-14 17:01:21] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:01:21] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:01:35] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:01:35] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:01:39] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:01:39] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:01:40] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:01:40] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:01:50] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:01:50] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:03:22] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0
[2014-05-14 17:03:22] ERROR - file: tracker_nio.c, line: 212, client ip: 192.168.1.9, recv timeout, recv offset: 0, expect length: 0

lifeblood

[root@localhost fastdfs]# tail -f storage/logs/storaged.log
[2014-05-14 16:53:58] INFO - file: tracker_client_thread.c, line: 1124, tracker server 192.168.1.221:22122, set tracker leader: 192.168.1.221:22122
[2014-05-14 16:56:28] ERROR - file: tracker_client_thread.c, line: 1009, tracker server 192.168.1.221:22122, recv data fail, errno: 110, error info: Connection timed out.
[2014-05-14 16:56:29] INFO - file: tracker_client_thread.c, line: 308, successfully connect to tracker server 192.168.1.221:22122, as a tracker client, my ip is 192.168.1.221
[2014-05-14 16:57:29] ERROR - file: tracker_proto.c, line: 37, server: 192.168.1.221:22122, recv data fail, errno: 110, error info: Connection timed out
[2014-05-14 16:57:59] INFO - file: tracker_client_thread.c, line: 308, successfully connect to tracker server 192.168.1.221:22122, as a tracker client, my ip is 192.168.1.221
[2014-05-14 16:58:59] ERROR - file: tracker_proto.c, line: 37, server: 192.168.1.221:22122, recv data fail, errno: 110, error info: Connection timed out
[2014-05-14 16:59:29] INFO - file: tracker_client_thread.c, line: 308, successfully connect to tracker server 192.168.1.221:22122, as a tracker client, my ip is 192.168.1.221
[2014-05-14 16:59:44] ERROR - file: tracker_proto.c, line: 37, server: 192.168.1.221:22122, recv data fail, errno: 107, error info: Transport endpoint is not connected
[2014-05-14 17:00:14] INFO - file: tracker_client_thread.c, line: 308, successfully connect to tracker server 192.168.1.221:22122, as a tracker client, my ip is 192.168.1.221
[2014-05-14 17:00:44] INFO - file: tracker_client_thread.c, line: 1124, tracker server 192.168.1.221:22122, set tracker leader: 192.168.1.221:22122
[2014-05-14 17:25:44] ERROR - file: tracker_client_thread.c, line: 1009, tracker server 192.168.1.221:22122, recv data fail, errno: 110, error info: Connection timed out.
[2014-05-14 17:25:45] INFO - file: tracker_client_thread.c, line: 308, successfully connect to tracker server 192.168.1.221:22122, as a tracker client, my ip is 192.168.1.221

lifeblood

内置去重，为什么还要编译到php扩展？
还有个问题，这个是不是每台storaged服务器都要装一个FastDHT?还是可以共用

还有：

group_count = 1
group0 = 192.168.1.221:11411

这个gourp0是什么属性，跟哪个属性是关联的

cache_size =32M,这个能存多少要图片记录

cjk87927

回复 5# lifeblood

我也很想知道这个问题,每个storage都要装fastDHT吗?那个group0到底是什么配置?
   

wuyongyu84

lifeblood：
       你的fastdht文件去重解决了吗?我现在也被这个问题困扰住了？（详情请见我的帖子）。

perofu

个人感觉，其实只要在每组的某个Storage上安装DHT，再把Tracker的配置文件中的store_server指定到那个安装DHT的Storage，即可，但是比较危险，万一指定的Storage down了，那么这个组的DHT就没有意义了。。