Turbolinux Cluster LoadBalancer 10: User Guide
<<< PreviousChapter 6. Configuration FileNext >>>

6.4. Servers and ServerPool

In the configuration file, Servers and ServerPool define the server nodes that the cluster will manage, including IP addresses, the forwarding method, and port numbers.

6.4.1. Servers

Each server node in the cluster is listed in the Servers section. Each line begins with the Server keyword and the name of the server. Then comes the IP address or fully qualified domain name of the server. Following this is the forwarding mechanism to use: `direct', `tunnel', or `nat'. You may also specify a `noping' flag, which will tell the daemon not to ping the server periodically to see if it is up.

Here is an example of what the Servers section may look like.

Servers

  Server node1          node1.turbolinux.usa                       direct

  Server node2          192.168.0.102                              tunnel noping

  Server node3          systemX.turbolinux.usa                     nat

EndServers

6.4.2. ServerPool Section

The ServerPool section corresponds to the `Servers Configuration' | `Server Groups Configuration' section of the configuration tool. It lets you combine several cluster nodes into a more manageable unit. The name of the server pool is given in the ServerPool line.

ServerPool servergroup1

  AddServer node1          ftp/1

  AddServer node2          ftp/3 nntp/1

  CheckServerFrequency     120

  CheckServerTimeout       10

  CheckPortFrequency       240

  CheckPortTimeout         120

EndServerPool

The AddServer lines add cluster nodes to the pool, along with the services they will support, as well as their weighting. The settings are given in the following format:

servicename/weight

where:

servicename is the name of the entry within the Services section

weight is the weight that you want to give to that server in comparison to other servers. A server with a higher weight will have more packets delivered to it than lower-weighted servers.

In addition to the list of server nodes and provided services, this section is also used to specify the frequency of stability checks. The CheckServer settings specify how often to ping each server. The CheckPort settings tell how often to perform the ASA checks. The Frequency settings specify how far apart to perform the checks. The Timeout values indicate how long to wait for a reply from the nodes before assuming that the check failed. All settings are given in seconds. Be aware that the Frequency values must be greater than the Timeout values.

<<< PreviousHomeNext >>>
ServicesUpClusters