Network in Google Cloud


Projects are the outermost container and are used to group resources that share the same trust boundary. Many developers map Projects to teams since each Project has its own access policy (IAM) and member list. Projects also serve as a collector of billing and quota details reflecting resource consumption. Projects contain Networks which contain Subnetworks, Firewall rules, and Routes (see below architecture diagrams for illustration).


Networks directly connect your resources to each other and to the outside world. Networks, using Firewalls, also house the access policies for incoming and outgoing connections. Networks can be Global (offering horizontal scalability across multiple Regions) or Regional (offering low-latency within a single Region).


Subnetworks allow you to group related resources (Compute Engine instances) into RFC1918 private address spaces. Subnetworks can only be Regional. A subnetwork can be in auto mode or custom mode. An auto mode network has one subnet per region, each with a predetermined IP range and gateway. These subnets are created automatically when you create the auto mode network, and each subnet has the same name as the overall network. A custom mode network has no subnets at creation. In order to create an instance in a custom mode network, you must first create a subnetwork in that region and specify its IP range. A custom mode network can have zero, one, or many subnets per region.

The Default Network

When any Project is created, a single Network named default is created for you. The default Network has the following 2 firewall rules defined for network traffic:
: Deny all incoming traffic
: Allow all outbound traffic
The following firewall rules are created for specific protocols and the roles they play in the default Network.
: Allows network connections of any protocol and port between instances on the network.
: Allows SSH connections from any source to any instance on the network over TCP port 22.
: Allows RDP connections from any source to any instance on the network over TCP port 3389.
: Allows ICM


Each firewall rule has a Priority value from 0-65535 inclusive. Relative priority values are used to determine precedence of conflicting rules. Lower priority value implies higher precedence. When unspecified, a priority value of 1000 is used. If a packet matches conflicting rules with the same priority, the deny policy takes precedence. Important: Since the default Network allows relatively open access, it is a recommended best practice that you delete it.You cannot delete the default Network unless another Network is present. A Project requires at least one Network. Because this lab has the user-created Network mynetwork, we can delete the default Network.

Stateful Firewalls

GCP Firewalls are stateful: for each initiated connection tracked by allow rules in one direction, the return traffic is automatically allowed, regardless of any rules.

Firewall Rules and IAM

The privilege of creating, modifying, and deleting firewall rules is reserved for the compute.securityAdmin role by IAM. Users assigned to the compute.networkAdmin role are able to safely view and list firewall rules that might apply to their projects. See the Network-specific IAM roles section later in this lab for more on this topic. Remember: All instances are configured with a "hidden" firewall rule that drops TCP connections after 10 minutes of inactivity. TCP keep-alives can be used to alter this behavior. See documentation for details on how to do that.

Allow Ingress Rules

First mynetwork is created by
gcloud compute networks create mynetwork --mode=auto
Because by default all ingress are denied, so logical next step towards opening network one at a step is allowing ingress. Example of adding some ingress firewall rules to allow us to SSH and ping the instances using the gcloud command line in Cloud Shell:
Network=mynetwork, icmp firewall rule name=mynetwork-allow-icmp, ssh rule name=mynetwork-allow-ssh, internal allow rule name=mynetwork-allow-internal
gcloud beta compute firewall-rules create mynetwork-allow-icmp --network mynetwork --action ALLOW --direction INGRESS --rules icmp
gcloud beta compute firewall-rules create mynetwork-allow-ssh --network mynetwork --action ALLOW --direction INGRESS --rules tcp:22
gcloud beta compute firewall-rules create mynetwork-allow-internal --network mynetwork --action ALLOW --direction INGRESS --rules all --source-ranges
gcloud beta compute firewall-rules list filter="network:mynetwork" 
When you create a VM on this netwrok, you create by
gcloud compute instances create mynet-us-vm --zone=us-central1-a --network=mynetwork
gcloud compute instances create mynet-eu-vm --zone=europe-west1-b --network=mynetwork
Because it is custom network subnetwork is not automatically created per region. To create it,
gcloud compute networks subnets create privatesubnet --network=privatenet \ --region=us-central1 --range= --enable-private-ip-google-access
TO create a VM in that network,gcloud compute instances create privatenet-us-vm --zone=us-central1-f \ --subnet=privatesubnetTo create firewall rule in a custom subnet networks First create a privatenet with custom subnet:gcloud compute networks create privatenet --mode=custom
gcloud beta compute firewall-rules create privatenet-allow-icmp \ --network privatenet --action ALLOW --direction INGRESS --rules icmp
gcloud beta compute firewall-rules create privatenet-allow-ssh \ --network privatenet --action ALLOW --direction INGRESS --rules tcp:22
gcloud beta compute firewall-rules create privatenet-allow-internal \ --network privatenet --action ALLOW --direction INGRESS --rules all \ --source-ranges
To create VMs in that network,
gcloud compute instances create privatenet-us-vm --zone=us-central1-f \ --subnet=privatesubnet

Deny Egress Rules

Now the firewall is created for basic ingress, we can establish specific deny rule. We were able to ping from mynet-eu-vm to mynet-us-vm, If we want to deny that, Egress firewall rules require a --destination-ranges flag.
gcloud beta compute firewall-rules create mynetwork-deny-icmp \ --network mynetwork --action DENY --direction EGRESS --rules icmp \ --destination-ranges --priority 500
gcloud beta compute firewall-rules list \ --filter="network:mynetwork AND name=mynetwork-deny-icmp"
Command Output:
project1-net-deny-icmp project1-net EGRESS 500 icmp
Go back to the mynet-us-vm SSH shell and re-run the ping command. You should no longer see the nice icmp results printing, the command seems to hang. This is the deny firewall rule in action. Ctrl+c to exit.Note 1: This rule was created with the direction EGRESS. Since ping is a bi-directional protocol, this will block ICMP as traffic leaves the VMs virtual NIC. If this rule were to be created with the INGRESS direction, the ICMP packets would be allowed to leave the VM's virtual NIC.
Note 2: If you supply both --source-ranges and --source-tags, you can allow traffic from both external IP ranges and internal instances identified by tags. The firewall matches either-or the source-range or source-tag.

Cloud Routes

If you want traffic from specific instances to specific ranges to be routed in a specific way, you can use Google Cloud Routes to set up the destination for this traffic. You can route traffic based on instance tags and destination range, and you can set the next hop to either:
  • A specific instance (by instancename or IP)
  • A VPN Tunnel
  • The default internet gateway
If multiple routes exist, the more specific route will be used. If there are multiple of those, the lowest priority value is used. Between Subnetworks, routes are automatically created implicitly at lowest priority value. Those routes cannot be changed or deleted.

Convert to a NAT gateway

Let's convert an instance to a NAT gateway so privatenet-us-vm can talk to the Internet without having an IP assigned. first create the VM:
gcloud compute instances create privatenet-ksakib --zone=us-central1-c \ --subnet=privatesubnet --can-ip-forward
Within privatenet, there are 2 instances: privatenet-nat and privatenet-us-vm. Both have a public IP, but you will remove the public IP from privatenet-us-vm. After the public IP is removed, you can SSH into privatenet-us-vm through privatenet-nat. You might want to use a NAT gateway either for additional filtering or if you want to egress from specific static IP addresses. In this case we just set it up without specific functionality.


The first step was already taken when the instance was created: the IP forwarding was set to On, which we did via the--can-ip-forward flag. This flag cannot be changed after the VM has been created. Now you will tag privatenet-us-vm and create a route through the gateway. In Cloud Shell run the following:
gcloud compute instances add-tags privatenet-us-vm --zone us-central1-f --tags nat-me
gcloud compute routes create nat-route --network privatenet --destination-range --next-hop-instance privatenet-nat --next-hop-instance-zone us-central1-c --tags nat-me --priority 800
SSH privatenet-bastion and activate IP forwarding and NAT on the Linux side:
sudo sysctl -w net.ipv4.ip_forward=1
sudo iptables -t nat -A POSTROUTING -o eth0 -j MASQUERADE
ssh privatenet-us-vm

Test if eberything is working properly by running follcowing test commands
gsutil ls gs://gcp-next2017-security-bootcamp/README

gsutil cat gs://gcp-next2017-security-bootcamp/README
Last updated: Wed 2017-02-22
Files used in the Google Cloud Platform Next 2017 Security Bootcamp codelabs.

(Output:) (IP of privatenet-bastion)
NAT is working: privetnet-us-vm appears to the outside from privatenet-bastion IP address.

Post a Comment