关于Redhat AS 3.0 cluster的heartbeat的问题

fxp1126

请教：
我最近按照七月飞雪的文档安装设置了在redhat AS 3上的cluster. 文中提到了要在两台机器上分别安装设置一个用于heartbeat的网卡。但是我安装后，发现在设置HA的时候，没有用到这两个网卡，只是选择是用broadcast还是multicast heartbeat。同时我发现在/var/log/messages中出现很多这样的错误信息：
Kernel: sent error broadcast ICMP to eth1 (eth1是心跳网卡,具体错误信息记不得了)。
请问这是为什么？在什么地方可以规定cluster使用eth1作为heartbeat?
(veritas cluster在安装的时候可以设置规定那几个作为heartbeat).

七月飞雪

还要更改HOSTS文件,还有其它的一些设置..

设置心跳地址

其它的都是REDHAT自己就可以认出来的..

q1208c

heartbeat 就是你写 member 时用的那个hostname 对应的网卡呀!

fxp1126

七月飞雪：照你这么说，redhat cluster在设置的时候，不用指定心跳网卡地址？为什么我的/var/log/messages文件里很多这种信息：
Aug 27 18:11:49 tst1 kernel: 192.168.0.68 sent an invalid ICMP type 11, code 0 error to a broadcast: 192.168.0.255 on eth1.
192.168.0.68是lan地址，eth1是心跳网卡。

q1208c：我仔细看了看redhat的文档，输入的应该是lan的地址而不是心跳的地址。

q1208c

嘿嘿，  那是你没看好。再好好看看。

fxp1126

First of all, I thank you for your response. I will try your suggestion.

Second, I have read the doc for ten times. It clearly states: Enter the name or address of a system on the CLUSTER SUBNET. Note that each member must be on the same subnet as the system on which you are running the Cluster Configuration Tool and must be defined either in DNS or in each cluster system\'s /etc/hosts file. ( I think CLUSTER SUBNET means the subnet that connects with both LAN interfaces of two nodes).

Below is the step-by-step of adding a member to a new cluster copied and pasted from redhat web site:

1. Ensure that the Members tab is selected and click New. It prompts for a member name.

2. Enter the name or address of a system on the cluster subnet. Note that each member must be on the same subnet as the system on which you are running the Cluster Configuration Tool and must be defined either in DNS or in each cluster system\'s /etc/hosts file

3. The system on which you are running the Cluster Configuration Tool must be explicitly added as a cluster member; the system is not automatically added to the cluster configuration as a result of running the Cluster Configuration Tool.

4. Leave Enable SW Watchdog checked. (A software watchdog timer enables a member to reboot itself.)

5. Click OK.

6. Choose File =>; Save to save the changes to the cluster configuration.

fxp1126

I think I found why there is error message like this: Aug 27 18:11:49 tst1 kernel: 192.168.0.68 sent an invalid ICMP type 11, code 0 error to a broadcast: 192.168.0.255 on eth1.

Because in VMWare, both eth0 and eth1 are actually in the same subnet.

luyanluyan14575

第几块网卡做heat，有什么关系嘛？

luyanluyan14575

第几块网卡做heat，有什么关系嘛？

q1208c

第几块网卡没关系统. 但也很少会有人把 heartbeat 和 work net 设在一个网段.

一般的做法是把 heartbeat 设成私有地址, 比如 10.0.0.1 之类的.

不知道 楼主是怎样理解 cluster net 的?

我的理解是, 做 heartbeat 那个子网. 而真正对外, 应该是工作网络了吧.