High availability between 2 virtualserver

[DmitriyGlu95]

CIS Version : 2.10.1
AS3 Version: 3.42.0-5
Agent Mode: AS3
Orchestration: K8S

We have two virtual servers deployed via CIS on one F5, each located in its own partition (each virtual server is used for one Kubernetes cluster). They use the same anycast IP (routeAdvertisement: "enable"). How can we ensure high availability between these two virtual servers? The problem is that when there are no members behind one virtual server (pods behind server in k8s are scaled to 0), the virtual server still announces its route. In a standard situation, if all members failed the health check, the virtual server would stop announcing the route. However, in our situation, the members are absent, and the virtual server does not stop announcing the route. But second virtual server with the same IP has working members. We cannot specify an additional health check for the service (it's check only endpoints for service). The annotation "cis.f5.com/f5.pool.action: reject" does not work. We cannot add a service from another cluster to the second pool VirtualServer.

What should we do in this situation?

[DmitriyGlu95]

And how set "Default Pool" for virtual server via cis? By default it's create pool but not set it in default for virtual server. Default Pool - None.

[DmitriyGlu95]

https://github.com/F5Networks/k8s-bigip-ctlr/issues/2377 "Recommend using a TransportServer as an option.", working, but I need use profiles and monitoring at http level.

[trinaths]

@DmitriyGlu95 Do you still have this issue ?

[trinaths]

No reply for user. Closing this issue.

[DmitriyGlu95]

@trinaths I stilll have problem.

[trinaths]

@DmitriyGlu95 Please mail automation_toolchain_pm@f5.com this issue with required CIS configuration and resource examples.

[trinaths]

@DmitriyGlu95 Please mail automation_toolchain_pm@f5.com this issue with required CIS configuration and resource examples.

[trinaths]

Closing this issue. No further communication.