# Deploy User VMs

## Deploy UVMs

Since this is a new environment, we need user VMs to check our routes and generate traffic.  This step typically wouldn't be done in a customers environment as it's assumed they already have VMs.  Run the shell script ``uvm.sh`` in the ``uservm directory`` to create (2) users VMs.  One VM will run in SubnetA and the other in SubnetB.  These subnets were created as part of ``base.sh``.  This script does NOT need to be modified.  The script will output the pubic IP of each VM which you can SSH to using the key stored in ``./keys/vectra``. This is the same key that has been used by cPacket cVu and the Vectra Sensor.

```bash
#!/bin/bash

# Do not modify
key=$(cat ../keys/vectra.pub)
rg=($(jq -r '.parameters.virtualNetwork.value.resourceGroup' ../shared/params.json))
vnet=($(jq -r '.parameters.virtualNetwork.value.name' ../shared/params.json))

# Create User VMs.  Subnets and names hard coded for lab
az deployment group create --resource-group $rg --template-file ../uservm/simplevm.json --parameters projectName=User-A adminUsername=vectra adminPublicKey="$key" vnet="$vnet" subnet=SubnetA
az deployment group create --resource-group $rg --template-file ../uservm/simplevm.json --parameters projectName=User-B adminUsername=vectra adminPublicKey="$key" vnet="$vnet" subnet=SubnetB

# Show Public IPs
echo "Public IP of UVM A"
az vm show --resource-group $rg --name "User-A-vm" --show-details --query publicIps --output tsv
echo "Public IP of UVM B"
az vm show --resource-group $rg --name "User-B-vm" --show-details --query publicIps --output tsv
```

## Modifying User Defined Routes/UDRs 

To get the traffic to cPacket we need to route UVM traffic to the cPacket load balancer.  Vectra has a script called ``cpacket-redirect.sh`` located in the ``udr directory`` which can be used to modify UDRs.  However, some customers may wish to do with their own scripting or manually.  Additionally, cPacket has a function as part of their ``ccloud`` tool that will modify UDRs.  Since routes are being modified a change ticket should be opened by the customer.  **DO NOT DO THIS WITHOUT APPROVAL**.  

- SSH into UVM A with the usersname ``vectra`` and install system tools.  **You will need to download your ssh key from ``./keys/vectra`` to do this.**
```bash
ssh -i <name of local key file> vectra@<public-ip-of-uvm>
```
```bash
sudo apt-get update && sudo apt-get -y install inetutils-traceroute
```
- From UVM A perform a tracroute to UVM B.  You should see UVM B as the next hop.
```bash
traceroute <private-ip-uvm-b>
```
- Add your values for `<cpacket-lb-ip>`, `<vnet-name>`, `<rgname>` and run ``cpacket-redirect.sh`` to change the routes.  
```bash
./cpacket-redirect.sh --action deploy --cpacket-gateway <cpacket-lb-ip> --vnet-name <vnet-name> --resource-grp <rgname>  --src-subnets subnetA,subnetB
```
- Recheck the routes from UVM A to UVM B and they should be modifed.  You will see the IP address of one of the cVu appliances not the ILB similar to below.
```bash
vectra@User-A-vm:~$ traceroute 10.70.4.4
traceroute to 10.70.4.4 (10.70.4.4), 64 hops max
  1   10.70.2.6  1.347ms  0.557ms  0.437ms 
  2   10.70.4.4  3.988ms  1.317ms  1.387ms 
```
- Start a ping to UVM B and leave it running for several minutes

## Verifying Traffic

After some time logon to your Brain and navigate to "Network Stats" >> "Observed IPs" and you should see your subnets.
![Brain Subnets](./images/brainsubnets.png)

## Lab Cleanup once verified

You have successfully deployed Vectra in Azure.  As there are significant costs please delete your resource group(s) **once your screenshot has been sent to group leader and verified** .  To cleanup your resource group(s) type``Resource Groups`` in the search bar and select your ``Resource Group`` and then click ``Delete Resource Group``.  You will delete both Brain and Sensor resource groups.
![Clean](./images/cleanup.png)

- Brain $30 per day
- Sensor $7 per day
- cPacket $8 per day