Forum Discussion
Shiva_Chetan_34
Nimbostratus
Nimbostratus2 Load balancers with/without High Availability
I'm trying to configure High availability on 2 BIG-IP 4000 series devices with software 12.1.2.0.249 with Active/Standby Configuration. My questions are
- Is HA only pertaining to shutting down of one of the F5 devices or will the standby take over when the link to the servers is shut? When I had shut the link to the servers, the Virtual server was still responding to the queries(with Error 404) and blackholing the traffic.
- Without using HA, is there a way to use 2 F5 load balancers to load balance across 2 ESXi hosts? If the link to server or the link to the router is down(shut), the F5 devices should route the traffic to the other F5 device and service the requests.
eey0re_68979Altocumulus
-
Use the HA Group feature to control what triggers failover of your traffic-group, such as pool monitors.
- Review the Managing Failover chapter of the Device Service Clustering manual.
- Also take note of the article K16947: Best practices for the HA group feature
- Not really (well, I can think of a quite perverse way - but don't do it). Your HA Group configuration should include monitoring of the gateways as well as the nodes. If you want to be able to survive the scenario that link failures mean one appliance can reach one node, the other can reach the other node, but neither can reach both - then the answer is to use HA and configure two traffic-groups. Put the two VIPs into the different traffic-groups, create two different HA groups each monitoring the relevant node, and associate one HA group with one traffic-group.
-
Use the HA Group feature to control what triggers failover of your traffic-group, such as pool monitors.