Good morning,

Occasionally a Datapower appliance will crash where it doesn't automatically get restarted by watchdog in our vmware environment, when this happens we get monitoring alerts and have to put in an incident to the server team to reboot it in vsphere (they don't allow us to do that ourselves). This takes quite a bit of time and I feel like vsphere should have some kind of mechanism to either reboot the appliance automatically when it hard crashes like that or it should alert a vsphere admin to do that manually. Does IBM provide any guidance on how to set that up for Datapower appliances? I need to spell out exactly what I'm requesting for the vmware team but I'm not sure exactly what to tell them.

We are on 2018.4.1.10 (upgrade to 2018.4.1.13 in progress).

Hi Jennifer,

• I'm not aware of any such mechanism from a DP perspective that would integrate that tightly with VMware. Sounds like Health/Monitor Operator in K8/OpenShift?
• How about raising a case with IBM Support and investigate for root cause of issue.
• One thought, based on what you describe, shall I suspect VMotion is enabled? Please disable and see if DP stops going frozen.
• So if you have monitoring that can detect this bad DP state and fire a trigger to e.g. invoke some script, how about leveraging vSphere REST API to trigger rebooting the VM in a scripted fashion?