Configured Alerts for Load Balancers

Selector implements the following 5 alerts to monitor the health of Load Balancers. While these alerts will work for all F5 load balancers, for this example document, Selector has enabled these alerts only for two example devices : ABCDEFGHIJ01 and ABCDEFGHIJ02.

This document describes the specific device conditions that trigger these alerts:

Alert 1: VIPServerAvailState down alert at device level

Alert 2: VIPAddrAvailState “unavailable” alert at device level

Alert 3: LTMPoolAvailState down alert at device level

ALert 4: LTMPoolMemberAvailState down alert at device level

Alert 5: Load Balancer Cluster down alert at cluster level

Alert 1: VIPServerAvailState down alert at device level

Alert Name	Priority	Queryable	Notes
LoadBalancerDeviceVIPServerAvailState	P2	vip_server_avail_state_5m_controlled as honeycomb where device=~ ABCDEFGHIJ01 \| ABCDEFGHIJ01, vip_server_avail_state_5m_controlled_violation>0	Load Balancer VIPServerAvailState at device level.

Alert 1 Condition

The Selector platform tracks the Virtual Server Availability Status on a continuous basis using SNMP polling of the OID ltmVirtualServStatAvailabilityState variable. If that polling shows that the server is in a down state for 5 minutes, this alert fires at the P2 level.

Selector LB Figure 1

Other parameters related to this alert are shown below:

Selector LB Figure 2

Back to Table of Contents

Alert 2: VIPAddrAvailState “unavailable” alert at device level

Alert Name	Priority	Queryable	Notes
LoadBalancerDeviceVIPAddrAvailState	P2	vip_addr_avail_state_5m_controlled as honeycomb where device=~ABCDEFGHIJ01 \| ABCDEFGHIJ02, vip_addr_avail_state_5m_controlled_violation>0	Load Balancer VIP address availability state at device level

Alert 2 Condition

The Selector platform tracks the Virtual IP Address Availability Status on a continuous basis using SNMP polling of the OID ltmVAddrStatusAvailState variable. If that polling shows that the VIP address is in a “down” state for 5 minutes, this alert fires at P2 level.

Selector LB Alert 2 Figure

Other parameters related to this alert are shown below:

Selector LB Alert 2 Figure 2

Back to Table of Contents

Alert 3: LTMPoolAvailState down alert at device level

Alert Name	Priority	Queryable	Notes
LoadBalancerDeviceLTMPoolAvailState	P2	ltm_pool_avail_state_5m as honeycomb where device=~ABCDEFGHIJ01 \| ABCDEFGHIJ02, ltm_pool_avail_state_5m_violation>0 show-by device , ltm_pool_name	Load Balancer LTM pool availability state at device level. Violation when pool availability below expected threshold or pool down for 5 minutes

Alert 3 Condition

The Selector platform tracks the Load Balancer LTM pool availability state at the device level by continuous SNMP polling of the OID ltmPoolStatAvailabilityState variable. A violation occurs when pool availability falls below the expected threshold or the pool is down for 5 minutes. A P2 Alert fires when the parameters are violated.

Selector LB Alert Figure 3

Other parameters related to this alert are shown below:

Selector LB Alert 3 Figure 2

Back to Table of Contents

Alert 4: LTMPoolMemberAvailState down alert at device level

Alert Name	Priority	Queryable	Notes
LoadBalancerDeviceLTMPoolMemberAvailState	P2	ltm_pool_member_avail_state_5m as honeycomb where device=~ABCDEFGHIJ01 \| ABCDEFGHIJ 02, ltm_pool_member_avail_state_5m_violation>0	Load Balancer LTM pool member availability state at device level (5m).

Alert 4 Condition

The Selector platform tracks the Load Balancer LTM Pool Member Availability State at device level by continuous SNMP polling of the OID ltmPoolMemberStatAvailabilityState variable. A violation occurs when one or more pool members is unavailable or below the threshold for a 5-minute window. A P2 Alert fires when the parameters are violated.

Selector LB Alert 4 Figure

Other parameters related to this alert are shown below:

Selector LB Alert 4 Figure 2

Back to Table of Contents

Alert 5: Load Balancer Cluster down alert at cluster level

Alert Name	Priority	Queryable	Notes
LoadBalancerHAClusterDown	P2	site1_lb_ha_cluster_harddown as threshold-violation-matrix where site2_lb_ha_cluster_harddown_violation>0	Indicates the HA cluster is fully unavailable; all devices within the cluster are down.

Alert 5 Condition

The Selector platform tracks the status of all load balancers in a cluster by continuously monitoring the status of each device in a cluster. When ALL devices in the cluster go down the entire cluster unavailable. A P2 Alert fires when ALL LB devices in the cluster are down.

Selector LB Alert 5 Figure

Back to Table of Contents