Forum Discussion
Hello.
Take into account that each blade will behave as an individual Linux VM that will work coordinate with the rest of the blades instances.
If one blade fails, you will have half of resources.
Regarding the network interfaces, you need to allow traffic in both blades in order to let them work together. If you only allow traffic in one of them, only this blade will process traffic.
Regards,
Dario.
Hello,
thank you for the reply. And what about that disk. If I am running two Linux VMs, as you say, what about the disk allocation. Its placed on both blades or just one of them?
If I loose half of resources, then I loose half of RAM as well, which is needed for modules. If I am using few modules and I loose one blade, it cannot work anymore.
Ok understand, one guest over two blades is like two Linuxes and I must connect intreface from both blades to network, in order to make it work.
Best regards
- Jul 15, 2020
Hello.
Imagine for one moment that you have a vcmp guest that uses 2 blades and one of them crash. If this happends you will have the vcmp guest running with only half of resources (CPU and memory).
When you replace that blade, VM of the working unit will be replicated into the disk of the second unit and a new linux instance will boot up. When this unit will be ready for working, the vcmp guest will also work with the complete set of resources.
If you have many modules running in that guest, you should take into account that you should have enough resources from each blade to be working without troubles in case of crashing.
Regards,
Dario.