samshers's questions

From - man capabilities

UNIX implementations distinguish two categories of processes: privileged processes (whose effective user ID is 0, referred to as superuser or root), and unprivileged processes (whose effective UID is nonzero).

and

Starting with kernel 2.2, Linux divides the privileges traditionally associated with superuser into distinct units, known as capabilities, which can be independently enabled and disabled. Capabilities are a per-thread attribute.

Does any Linux command or technique exist to determine the capabilities associated with a running process (or thread).

From this article: Virtualization: What is KVM? and other resources, I learned that it is possible to migrate a Live VM from one host to another?

Live migration
KVM supports live migration, which is the ability to move a running VM between physical hosts with no service interruption. The VM remains powered on, network connections remain active, and applications continue to run while the VM is relocated.

Can some one eloborate on how this is made possible?
In general, A running VM is tied to some hardware (real or virtual) like memory, cpus, disks, graphics, network etc. And it's tied to the Host OS (in case of type-2 hypervisors) or Hypervisor/KVM(in case of type-1 hypervisors). On the new host, the VM has to rebind itself with the above said resources/drivers etc - which could indicate a downtime (might be short).
So how is a live migration with zero service disruption made possible.

Thanks for any responses in advance.

In Virtual Networking, I have seen two techniques to connect the guest machine with the Host machines network.

In VMWare/VirtualBox - Bridged Networking is used to connect the guest machine with the host machines network.
Ex - if the host is on 172.16.0.1/12 subnet with IP 172.16.0.2 as host IP,
then using Bridge networking any guest running on the above host can be connected to the host network i.e 172.16.0.1/12 subnet and the guest will receive an IP on this subnet say 172.16.0.6 (just picked a random valid IP in this subnet).

In Docker, the same is achieved using IPVLAN or MACVLAN.
Ex: - if host is on 172.16.0.1/12 subnet with IP 172.16.0.2 as host IP,
then using MACVLAN or IPVLAN any container running on this host can be connected to host network i.e 172.16.0.1/12 subnet and the container will receive an IP on this subnet say 172.16.0.6 (again, just picked a random valid IP in this subnet).

Though the end result is same, the techniques used seem to be different. Am I right? So just exploring to understand the difference between these two approches and how bridging differs from IPVLAN/MACVLAN?

I am thinking if it's possible to create a bridge while allowing each bridged network to exist in it's own subnet.
Network 1 - Say, I have a physical ethernet based network eth01 - with subnet 192.168.x.x/16.
Network 2 - I want to create another virtual network veth01(based on virtual interface) with subnet 172.16.x.x/12.

Now is it possible to create a linux bridge (virtual not physical) between Network1 and Network2, such that even after bridging the above two networks have there on subnets and there own DHCP.

Any reference or steps/commands on how to create this setup (if possible) are highly welcomed.

I have seen that in virtual networking, guest VMs/containers can be connected with host and internet either by using kernel bridges(linux) or Virtual Switches.

Example - Bridges

docker uses linux kernel bridges to connect the guest containers to host and internet. Even, VMWare/VirtualBox allows to use bridge networking to connect the guest VM's with the host and internet. Example - Virtual Switches

VMWare ESX uses virtual switches to connect VM's with host and internet. Microsoft Hyper-V uses virtual switches to connect VM's with host and internet. So, I started to wonder what's the difference between these two (bridges and virtual switches) and which approach is better than the other.

(please feel free to edit the question to expand or make the question more meaningful.)