Unable to ping and connect cloud stack vm instance ip from the same local network

Question

I am a beginner to cloud computing and I am now learning cloud stack to build private cloud.I have installed management server together with storage on one node and KVM compute server on the other node according to this documentation .I have also set up management server and added zones , pods , clusters and hosts.Everything seems properly working.But when I create a instance in cloud stack ,I can get into that vm with console provided by cloud stack but I am unable to ping and connect that instance with it's IP address via ssh.Could someone now that problem?Thanks for your attention.

Answer 1

If you can access the VM's console that is provided by the Console Proxy VM than CloudStack has no role in you unable to ping the VM on your local machine. Can you provide some more details regarding your network? I think as you have just started you would be using L2 (flat) networking right now (providing IP of the same network CIDR or subnet) There has got to be a problem with your network configuration either on the local machine from where you are trying or Cloudstack. I suggest you see if both your machine and VM can ping the gateway. That would be the first step, if they can, try accessing internet from both (if you can use internet in your network you used to setup Cloudstack).

The quickstart guide you are using is specificly being designed for configuring management server, NFS storage and agent in one machine only. When you use a different machine as a KVM agent, it's configuration changes slightly I have encountered similar issues earlier. I suggest you follow this guide: http://docs.cloudstack.apache.org/projects/cloudstack-installation/en/4.9/hypervisor/kvm.html As you are using separate machine as KVM agent, a frequent issue is out of sync iptables rules. Following should be taken care of in KVM agent's OS environment:

1. Disable firewalld - don't be afraid, CloudStack uses iptables which is indeed used by firewalld service to manage network rules. Leaving firewalld service running is the most common root cause of the issue. You can do it by the following commands ( These disable and stops running of firewalld currently and after reboot)

   systemctl disable firewalld && systemctl stop firewalld

2. Restart the iptables and cloudstack-agent services

   systemctl restart iptables systemctl restart cloudstack-agent

If you still see any issues, I suggest you load the default iptables rules that are given here in ACS documentation and restart the iptables service: http://docs.cloudstack.apache.org/projects/cloudstack-installation/en/4.9/hypervisor/kvm.html

Also, if you are just starting, common places to see and understand the errors are the /var/log/cloudstack/management/management-server.log in management machine and /var/log/cloudstack/agent/agent.log in agent machine. Here you will see the most probable errors or warnings (if they exist) in your cloudstack setup. Tell me if it works. All the best.

PS: Use Cloudstack 4.11 if you are just starting, that's the latest stable LTS release (although there are some minor bugs here and there, but nothing cloud breaking ;)