VM Self Healing 6.7

[randophbar]

Hey all... Happy Holidays first and foremost. Question. I have an environment on 6.7 with 100 VM's running both windows and Centos7 OS. I get a call at-least 2x a week to reset VMs due to inaccessible by the users for both OS types. I have to go in and issue a "reset OS" and on some occasions a full reboot for them to have access returned. Is there some sort of self-healing software or setting I can apply in my environment that will see that a VM is hung and will reset its self? Thanks in advance!!

[ashilkrishnan]

Hi @randophbar ,

If the issue is with VMs becoming unresponsive/hung, consider enabling VM monitoring in vSphere HA --> VM and application monitoring 

Enabling VM monitoring 

Hope that helps

[randophbar]

Awesome!!! this is exactly what I am looking for. The link you shared doest show how to actually enable it. Do you know where I can find the Doc?

[randophbar]

To enable Application Monitoring, you must first obtain the appropriate SDK. Do you know where I can find the SDK?

[scott28tt]

@randophbar 

Moderator: Note the following:

1. Please do not create multiple threads on the same topic, your duplicate created today has been archived

2. This thread has been moved to the Availability: HA & FT area

[randophbar]

I was told To enable Application Monitoring, you must first obtain the appropriate SDK (or be using an application that supports VMware Application Monitoring) and use it to set up customized heartbeats for the applications you want to monitor;

How is this exactly done? the above reply only talks about what needs to be done and not the actual documentation on the process. The exact steps in setting up the Monitoring.

I am not a Developer so I do not know how to use SDK. Thanks in advance.

[scott28tt]

@randophbar 

Moderator: Marking every single reply as the “solution” is confusing for others who might have the same issue, plus it shows the thread as “solved” which will put off others who might be able to help you from viewing the thread. I have unmarked all the replies as “solutions”.

[randophbar]

understood. Sorry about that. Thanks for letting me know for future use.

[Ajay1988]

https://docs.vmware.com/en/VMware-vSphere/6.5/com.vmware.vsphere.avail.doc/GUID-4F623D19-814B-4799-B...

http://www.yellow-bricks.com/2010/09/10/application-monitoring-ha/    >>>>>Check comments section ..might be helpful

[SureshKumarMuth]

Check this link for steps 

https://docs.vmware.com/en/VMware-vSphere/6.5/com.vmware.vsphere.avail.doc/GUID-4F623D19-814B-4799-B...

There are two options "VM monitoring only" and "VM Monitoring and Application Monitoring". 

VM monitoring uses vmware tools to check the heartbeats, in your use case when a VM is hung or not responding , VMware tools also will not be in responsive state which can trigger VM reboot.

For application monitoring we need SDK as said in article, you may refer this link for more information on that 

https://vdc-repo.vmware.com/vmwb-repository/dcr-public/0c24f040-d644-4cc6-ac10-f3624e8ff98e/1c3d34e7...

https://code.vmware.com/docs/4078/guest-and-ha-application-monitoring-sdk-programming-guide