About Nodes

Nodes

Overview

A node is a logical object on F5 BIG-IP Local Traffic Manager (LTM) that identifies the IP address of a physical resource on your network. Nodes are the foundation of load balancing pools. Before you can add a server to a pool as a pool member, you must first create a node for that server.

You can create a node in two ways:

• Explicitly — Create the node yourself through the BIG-IP Configuration utility.
• Automatically — Let the system create the node when you add a pool member to a pool.

The Nodes section of the BIG-IP Configuration utility has three screens. Access them from the left navigation menu under Local Traffic > Nodes:

Nodes compared to pool members

A node uses only an IP address to identify a device (for example, 10.10.10.10). A pool member combines an IP address with a service port (for example, 10.10.10.10:80).

How health monitors work with nodes

You can associate health monitors with both nodes and pool members. The difference is what they check:

• A health monitor on a node checks the status of the device itself.
• A health monitor on a pool member checks the status of a specific service running on the device.

Health monitors can be assigned to nodes in two ways:

• Default monitors — Applied automatically to every node on the system. Configure these on the Default Monitor screen.
• Node-specific monitors — Applied to an individual node. Configure these when you create or update a node.

Note: Any monitor that you associate with a node must reside either in the /Common partition or in the same partition as the node.

Node status

You can check the status of a node at any time in the BIG-IP Configuration utility. Go to Local Traffic > Nodes > Node List and view the Status column, or select a node and view the Availability property.

The system displays status icons that use shape and color:

• The shape of the icon indicates the status reported by the health monitor.
• The color of the icon indicates the current operational status of the node.

Tip: You can manually set the availability of a node with the Manual Resume attribute of the associated health monitor.

Before you begin

• You have access to the BIG-IP Configuration utility with the Administrator role or equivalent permissions.
• You know the IP address or fully qualified domain name (FQDN) of the server you want to add as a node.

Configure default monitors for nodes

Default monitors apply automatically to every node on the system. When BIG-IP LTM creates a node automatically (for example, when you add a pool member), the system assigns the default monitors to the new node. This saves you from assigning monitors to each node individually.

The Default Monitor screen has two configuration modes: Basic and Advanced. Advanced mode adds an Availability Requirement setting.

Keep the following in mind when you work with default monitors:

• You must have the Administrator user role to configure default monitors.
• Default monitors must reside in the /Common partition.
• If a user with permission to manage objects in the /Common partition disables a default monitor (such as the icmp monitor), this affects all nodes on the system.
• If all nodes reside in the same partition, the default monitor must reside in that partition or in /Common. If nodes reside in different partitions, the default monitor must reside in /Common.

Configure default monitors (Basic)

1. On the Main tab, go to Local Traffic > Nodes > Default Monitor. The Default Monitor screen opens.

2. From the Configuration dropdown, select Basic.

3. In the Health Monitors section, assign monitors using the Active and Available lists:

   List	Description
   Active	Monitors currently assigned as defaults. These monitors automatically apply to every node on the system
   Available	Monitors available for assignment. The list shows all monitors in the /Common partition (for example, arp, gateway_icmp, https_443, icmp)

   Use the arrow buttons to move monitors between lists:

   • Select a monitor in the Available list and select « to move it to Active.
   • Select a monitor in the Active list and select » to move it to Available.

4. Select Update.

Configure default monitors (Advanced)

1. On the Main tab, go to Local Traffic > Nodes > Default Monitor. The Default Monitor screen opens.

2. From the Configuration dropdown, select Advanced. The screen displays the same Health Monitors section as Basic mode, plus an additional field.

3. Assign monitors using the Active and Available lists as described in the Basic configuration steps.

4. Configure the additional setting:

   Setting	Default	Description
   Availability Requirement	All	Set the minimum number of health monitors that must report a node as available before BIG-IP LTM marks the node as up. Select All to require every active monitor to pass. Select At Least and enter a number to set a specific minimum

5. Select Update.

Create a node

1. On the Main tab, go to Local Traffic > Nodes > Node List. The Node List screen opens.

2. Select Create. The New Node screen opens.

3. In the General Properties section, complete the following fields:

   Field	Description
   Name	Type a unique name for the node
   Description	(Optional) Type a description to help identify the node’s purpose
   Address	Select one of the following options, then type the value in the field below:

   • Address — Type the IP address of the server (for example, 10.10.10.10). If you use a route domain other than route domain 0, append the route domain ID (for example, 10.10.10.10%1).
   • FQDN — Type the fully qualified domain name of the server. If you select this option, additional FQDN fields appear (see step 5).

4. In the Configuration section, adjust the following settings or keep the defaults:

   Setting	Default	Description
   Health Monitors	Node Default	Select the health monitor assignment method. Node Default applies the system-wide default monitors to this node. Node Specific lets you choose individual monitors for this node
   Ratio	1	Set the ratio weight used by the Ratio load balancing method. At least one node in the pool must have a value greater than 1 for this setting to take effect. Otherwise, the system uses Round Robin behavior
   Connection Limit	0	Set the maximum number of concurrent connections allowed for this node. A value of 0 means no limit
   Connection Rate Limit	0	Set the maximum number of new connections per second allowed for this node. A value of 0 means no limit
   Delay Timeout	0 seconds	Enter the number of seconds the system waits before it terminates active connections on a node that has been forced offline. This setting only takes effect when the node’s pool member belongs to a pool where Action On Service Down is set to Reject or Drop. If Action On Service Down is set to None (the default), the Delay Timeout has no effect and existing connections persist until they close naturally regardless of the value. When Reject or Drop is active, a value of 0 means zero grace period—the system terminates existing connections immediately. See Node state behavior for details

5. (FQDN only) If you selected FQDN in step 3, an FQDN section appears below the Configuration section. Configure the following settings:

   Setting	Default	Description
   Address Type	IPv4	Select whether the FQDN resolves to an IPv4 or IPv6 address
   Auto Populate	Disabled	Controls how the system handles DNS responses for this FQDN. When set to Enabled, the system creates a temporary node for each IP address returned by the DNS query. When an IP address is no longer returned, the system deletes the temporary node. When set to Disabled, the system resolves the FQDN to a single IP address
   Interval	3600 seconds	Set how often, in seconds, the system queries DNS when the server is up. The monitor probes three times within this interval and marks the server as down if there is no response. Select the Use TTL checkbox to use the DNS time-to-live value instead of a fixed interval
   Down Interval	5 seconds	Set how often, in seconds, the system queries DNS when the server is down. The system continues to poll at this interval until the server responds

6. Select one of the following buttons:

   Button	Description
   Finished	Save the node and return to the Node List screen. The new node appears in the list
   Repeat	Save the node and keep the New Node screen open so you can create another node
   Cancel	Discard your changes and return to the Node List screen without creating a node

Tip: You can also create a node by adding a pool member to a pool. When you create a pool member, BIG-IP LTM automatically creates the corresponding node.

Update a node

After you create a node, you can change its description, state, health monitor assignments, and configuration settings. You can also delete a node you no longer need.

1. On the Main tab, go to Local Traffic > Nodes > Node List. The Node List screen opens.

2. Select the name of the node you want to update. The node properties screen opens with three tabs:

   Tab	Description
   Properties	View and edit the node’s general properties, state, and configuration settings
   Pool Membership	View the pools that include this node as a member
   Statistics	View traffic and performance statistics for this node

3. On the Properties tab, review the read-only fields in the General Properties section:

   Field	Description
   Name	The name assigned to the node when it was created
   Address	The IP address of the node
   Partition / Path	The administrative partition where the node resides
   Availability	The current availability status reported by health monitors. Displays the monitor state, a description, and a timestamp of the most recent status check
   Health Monitors	The monitors currently associated with this node. Each monitor displays a colored status icon. To change monitor assignments, use the Configuration section in step 6
   Current Connections	The number of connections currently active on this node

4. Update the editable fields in the General Properties section:

   Field	Description
   Description	Update the text to reflect the current purpose of the node
   Monitor Logging	Select the Enable checkbox to log health monitor events for this node. Clear the checkbox to turn off logging

5. In the State section, select the desired state for the node:

   State	Description
   Enabled (All traffic allowed)	The node accepts all new connections and traffic
   Disabled (Only persistent or active connections allowed)	The node stops accepting new connections. Existing connections and connections that belong to a persistence session continue until they end normally
   Forced Offline (Only active connections allowed)	The node doesn’t accept new connections, including connections that match existing persistence records. Only already-active connections continue. The actual effect on those connections depends on the Action On Service Down setting of the associated pool. See Node state behavior for details

6. In the Configuration section, update the following settings:

   Setting	Default	Description
   Health Monitors	Node Default	Select the monitor assignment method from the dropdown. Node Default applies the system-wide default monitors. Node Specific lets you choose individual monitors for this node

   When you select Node Specific, the Select Monitors panel appears with two lists:

   List	Description
   Active	Monitors currently assigned to this node
   Available	Monitors available for assignment from the /Common partition

   Use the arrow buttons to move monitors between lists:

   • Select a monitor in the Available list and select « to move it to Active.
   • Select a monitor in the Active list and select » to move it to Available.

   Configure the remaining settings:

   Setting	Default	Description
   Availability Requirement	All	Set the minimum number of health monitors that must report the node as available before BIG-IP LTM marks the node as up. Select All to require every active monitor to pass. Select At Least and enter a number to set a specific minimum
   Ratio	1	Adjust the ratio weight used by the Ratio load balancing method
   Connection Limit	0	Adjust the maximum number of concurrent connections. A value of 0 means no limit
   Connection Rate Limit	0	Adjust the maximum number of new connections per second. A value of 0 means no limit
   Delay Timeout	0 seconds	Enter the number of seconds the system waits before it terminates active connections on a node that has been forced offline. This setting only takes effect when the node’s pool member belongs to a pool where Action On Service Down is set to Reject or Drop. If Action On Service Down is set to None (the default), the Delay Timeout has no effect and existing connections persist until they close naturally regardless of the value. When Reject or Drop is active, a value of 0 means zero grace period—the system terminates existing connections immediately. See Node state behavior for details

7. Select one of the following buttons:

   Button	Description
   Update	Save your changes to the node
   Delete	Delete this node from the system

Node state behavior

The Delay Timeout setting interacts with the Action On Service Down setting on the associated pool. The behavior described below for Forced Offline only applies when Action On Service Down is set to Reject or Drop. When Action On Service Down is set to None (the default), existing connections persist until they close naturally, and the Delay Timeout value has no effect.

The following table describes how each node state handles connections when Action On Service Down is set to Reject or Drop:

When a node is set to Forced Offline and Action On Service Down is Reject or Drop:

1. The system immediately rejects all new connections, including connections that match existing persistence records.
2. Already-active connections continue during the Delay Timeout period.
3. After the Delay Timeout expires, the system terminates any remaining active connections.
4. If Delay Timeout is set to 0, there is zero grace period—the system terminates existing connections immediately.

Important: When Action On Service Down is set to None (the default pool setting), forcing a node offline prevents new connections but doesn’t terminate existing ones. Connections remain active until they close naturally, regardless of the Delay Timeout value. To use Delay Timeout to control connection draining, set Action On Service Down to Reject or Drop on the associated pool.