Previous  |  Next  >  
Product: Cluster Server Guides   
Manual: Cluster Server 4.1 Bundled Agents Reference Guide   

MultiNICB Agent

Works with the IPMultiNICB agent. Allows IP addresses to fail over to multiple NICs on the same system, before VCS attempts to fail over to another system.

When you use the MultiNICB agent, you must plumb the NIC before putting it under the agent's control. You must configure all the NICs on a single IP subnet inside a single MultiNICB resource.

The agent monitors the interfaces it controls by sending packets to other hosts on the network and checking the link status of the interfaces.

If a NIC goes down, the MultiNICB agent notifies the IPMultiNICB agent, which then fails over the virtual IP addresses to a different NIC on the same system. When the original NIC comes up, the agents fail back the virtual IP address.

Each NIC must have its own unique and exclusive base IP address, which the agent uses as the test IP address.

If multiple service groups have IPMultiNICB resources associated with the same MultiNICB resource, only one group should have the MultiNICB resource. The other groups can have a proxy resource pointing to it.

MultiNICB uses the following criteria to determine if an interface is working:

  • Interface status: The interface status as reported by driver of the interface (assuming the driver supports this feature). This test is skipped if the attribute IgnoreLinkStatus = 1.
  • ICMP echo: ICMP echo request packets are sent to one of the network hosts (if specified). Otherwise, the agent uses ICMP broadcast and caches the sender of the first reply as a network host. While sending and receiving ICMP packets, the IP layer is completely bypassed.

The MultiNICB agent writes the status of each interface to an export information file, which other agents (like IPMultiNICB) or commands (like haipswitch) can read.


Failover and Failback

During an interface failure, the MultiNICB agent fails over all logical IP addresses to a working interface under the same resource. The agent remembers the first physical interface from which an IP address was failed over. This physical interface becomes the "original" interface for the particular logical IP address. When the original interface is repaired, the logical IP address fails back to it.

Entry Points

  • Open---Allocates an internal structure to store information about the resource.
  • Close---Frees the internal structure used to store information about the resource.
  • Monitor---Checks the status of each physical interface. Writes the status information to the export information file for IPMultiNICB resources to read it. Performs failover. Performs failback if failback is set to 1.

State Definitions

  • ONLINE---Indicates that one or more of the network interfaces listed in the Device attribute of the resource is in working condition.
  • FAULTED---Indicates that all of the network interfaces listed in the Device attribute failed.
  • UNKNOWN---Indicates that the configuration is incorrect.

Type Definition


type MultiNICB (
        static int MonitorInterval = 10
        static int OfflineMonitorInterval = 60
        static int MonitorTimeout = 60
        static int Operations = None
        static str ArgList[] = { Device, NetworkHosts,
        LinkTestRatio, IgnoreLinkStatus, NetworkTimeout,
        OnlineTestRepeatCount, OfflineTestRepeatCount, NoBroadcast,
         DefaultRouter, Failback}

        str Device{}
        str NetworkHosts[]
        int LinkTestRatio = 1
        int IgnoreLinkStatus = 1
        int NetworkTimeout = 100
        int OnlineTestRepeatCount = 3
        int OfflineTestRepeatCount = 3
        int NoBroadcast = 0
        str DefaultRouter = "0.0.0.0"
        int Failback = 0
)
 ^ Return to Top Previous  |  Next  >  
Product: Cluster Server Guides  
Manual: Cluster Server 4.1 Bundled Agents Reference Guide  
VERITAS Software Corporation
www.veritas.com