I have a blade server (currently 2 blades powered on) each server in this blade has around 8 VMs , they are connected to a NetApp filer 270 with iScsi , I just want to implement HA to this blade , it should be a simple plane , I am planning to do it like this :
1-Create new Cluster
2-Give the new cluster a name
3-Check the VMware HA box
4-Secify the number of host failures allowed (I am planning to choose 1)
5-Admission Control (I am confused which one I will choose , don't start virutial machine or Allow virtual machines ) and which one is better.
6-Drag and Drop the 2 Hosts into the Cluster.
Do I have to do something else ?
Thanks
Your instructions look good. There is not much to setting up HA. You may want to tinker with the Admissions Control for VM startup after failure. It is pretty cut and dry if you look at it like giving your VMs priority to start up or not startup at all on failure to control resources on the host(s) that these VMs are moved to.
In other words you I personally would hate to bring the "standby" host(s) to their knees by starting up all VMs with a high priority if their "home" host were to fail so I give them priority or do not start them at all depending on criticallity.
Also you may consider implementing DRS if nothing but to allow for recommendations. I find this is the safest choice while allowing you to be mindful of the recommendation fro VC itself.
Help me help you by scoring points...
Your instructions look good. There is not much to setting up HA. You may want to tinker with the Admissions Control for VM startup after failure. It is pretty cut and dry if you look at it like giving your VMs priority to start up or not startup at all on failure to control resources on the host(s) that these VMs are moved to.
In other words you I personally would hate to bring the "standby" host(s) to their knees by starting up all VMs with a high priority if their "home" host were to fail so I give them priority or do not start them at all depending on criticallity.
Also you may consider implementing DRS if nothing but to allow for recommendations. I find this is the safest choice while allowing you to be mindful of the recommendation fro VC itself.
Help me help you by scoring points...
Thanks , I appreciate your help
I make sure that the settings in HA are as follows;
Number of hosts failure allowed = 1
Admission Controll = Do not power on if violation occurs.
Then go to the Virtual machine Options and set your priority. This determines the order in which virtual machines are restarted upon host failure. Restart priority is always considered, but is especially important in the following cases:
If you've set host failure to a certain number of hosts (for example, three) and more hosts fail (for example, four)
If you've turned off strict admission control and have started more virtual machines than HA has been set up to support
Do this in the settings window. Double click on the default settings "Medium" you'll see a menu to chose from, make your choice. Also double click on the "Power Off" and make your choice from the menu there.
Also remember, everytime you add a host to a cluster, all virtual machines in the cluster default to a restart priority of MEDIUM and an isolation response of SHUTDOWN.
Hope that helped.
Do I have to schedule down time for to implement HA , I don't have testing lab to do this that's why I am so careful about everything !
No , This is done without affecting your hosts.
Hope that helped.
Nope, creating a cluster and enabling DRS and HA can be done on the fly.
One gotcha is if you add your servers to the newly created cluster and you want to remove them, that is a little more tricky and requires extra effort on your part, so if you accidentally add the wrong one, it is not as easy as dragging it out for removal. I did ithat once and learned to double check which server I have selected. This is for people with more than 2 esx hosts..