配置ganglia出了点问题,高手帮帮忙

Question

这几天装了ganglia 3.0.2, 装上后可以正常远程用网页打开, 可以正常显示收集到到计算机的一些资源,但有一个问题一直没法解决,我的服务器有两个IP段，eth0: 192.168.0.100, eth1: 10.10.1.106 这两个IP段分别接的两个交换机，我装上gangia后，它只能显示出10.10.1.X IP段的所以机子,而192.168.0.X段的机子，没法显示，而两个IP段的机子我都已经正常安装了gmond客户端，且都可以正常起动，运行．哪位高手指点一下，是不是我的gmetad.conf的gmond.conf哪些地方设置的不对，非常感受谢！

Gmetad.conf:
data_source "rcmm1" localhost
gridname "RCMM1"
all_trusted on
setuid_username "ren"
其他配置都是默认的

Gmond.conf:
globals {
  setuid = yes
  user =  root
  cleanup_threshold = 300 /*secs */
}

cluster {
  name = "rcmm1"  
}   
   
host {  
  location = "unspecified"
}  
     
udp_send_channel {
  mcast_join = 239.2.11.71  
  port = 8649  
}  
   
udp_recv_channel {
  mcast_join = 239.2.11.71
  port = 8649
  bind = 239.2.11.71
}   
   
tcp_accept_channel {
  port = 8649
}

collection_group {
  collect_once = yes
  time_threshold = 20
  metric {
    name = "heartbeat"
  }
}

collection_group {
  collect_once = yes
  time_threshold = 1200
  metric {
    name = "cpu_num"
  }
  metric {
    name = "cpu_speed"
  }
  metric {
    name = "mem_total"
}  
  /* Should this be here? Swap can be added/removed between reboots. */
  metric {
    name = "swap_total"
  }
  metric {  
    name = "boottime"  
  }  
  metric {  
    name = "machine_type"
  }  
  metric {  
    name = "os_name"  
  }  
  metric {  
    name = "os_release"
  }  
  metric {  
    name = "location"
  }
}

collection_group {  
  collect_once = yes  
  time_threshold = 300
  metric {  
    name = "gexec"
  }  
}   
collection_group {
  collect_every = 20
  time_threshold = 90
  /* CPU status */
  metric {
    name = "cpu_user"
    value_threshold = "1.0"
  }
  metric {
    name = "cpu_system"
    value_threshold = "1.0"
  }
  metric {
    name = "cpu_idle"
    value_threshold = "5.0"
  }
  metric {
    name = "cpu_nice"
    value_threshold = "1.0"
  }
  metric {
    name = "cpu_aidle"
    value_threshold = "5.0"
  }
  metric {
    name = "cpu_wio"
    value_threshold = "1.0"
  }
   metric {
    name = "cpu_intr"
    value_threshold = "1.0"
}  
  metric {  
    name = "cpu_sintr"
    value_threshold = "1.0"
  }   
  */  
}   
     
collection_group {
  collect_every = 20
  time_threshold = 90  
  /* Load Averages */  
  metric {
    name = "load_one"
    value_threshold = "1.0"
  }  
  metric {
    name = "load_five"
    value_threshold = "1.0"
  }  
  metric {
    name = "load_fifteen"
    value_threshold = "1.0"
  }  
}  
   
collection_group {   
  collect_every = 80
  time_threshold = 950  
  metric {  
    name = "proc_run"
    value_threshold = "1.0"
  }  
metric {
    name = "proc_total"
    value_threshold = "1.0"
  }
}

collection_group {
  collect_every = 40
  time_threshold = 180
  metric {
    name = "mem_free"
    value_threshold = "1024.0"
  }
  metric {
    name = "mem_shared"
    value_threshold = "1024.0"
  }
  metric {
    name = "mem_buffers"
    value_threshold = "1024.0"
  }
  metric {
    name = "mem_cached"
    value_threshold = "1024.0"
  }
  metric {
    name = "swap_free"
    value_threshold = "1024.0"
  }
}

Answer 1

好像ganglia发现IP有一定的协议,不是你想显示哪个IP就可以显示的.(也许我理解的不正确)
可以在安装gmetad的服务器上
telnet localhost 8651 | grep "ip"
ip为你安装gmond机器的地址,如果有多个可以都试验一下,
看看gmetad能够识别哪一个IP
你的问题

data_source "rcmm1" localhost
localhost是本机,你要根据你安装了gmond的机器来设置它
比如:data_source "rcmm1" 15 10.10.1.2  10.10.1.3……10.10.1.X
       data_source "rcmm"   15 192.168.0.1……192.168.0.X
它们默认端口都是8649，你也可以更改它们。在gmond.conf
udp_send_channel {
  mcast_join = 239.2.11.71  
  port = 8648
}  
   
udp_recv_channel {
  mcast_join = 239.2.11.71
  port = 8648
  bind = 239.2.11.71
}   
   
tcp_accept_channel {
  port = 8648

在Gmetad.conf:
data_source "rcmm"   15 192.168.0.1：8648
这里只是简单的例子，供你参考。