Multiple vms gets the same ip within the same network

[Eslam-Nawara]

Description

• I am trying to deploy a nomad cluster but deployment results in multiple vms with the same ip.

• I've attempt to set the ip of "server1" to some value "10.1.2.2"

Expected Behavior

• every vm should have different ip with only server1 with ip "10.1.2.2"

Actual Behavior

• this resulted in more than one vm with ip "10.1.2.2"

all of the 3 vms are in the same node and I've tried multible nodes with the same result

[AbdelrahmanElawady]

can you share the tf file please

[Eslam-Nawara]

https://github.com/threefoldtech/terraform-provider-grid/blob/development_create_nomad_module/examples/resources/nomad_with_modules/main.tf

[AbdelrahmanElawady]

It seems a problem with the Go grid client that it assigns the same IP for different deployments when on the same network
threefoldtech/tfgrid-sdk-go#457

[rawdaGastan]

Verification

• Verified using version 1.9.3-rc2. Will be available in the next release 1.9.3

• Tried the single node example (https://github.com/threefoldtech/terraform-provider-grid/tree/development/examples/resources/singlenode)

• It creates 2 VMs in a deployment and both have different IPs ✅

node1_zmachine1_ip = "172.20.3.2"
node2_zmachine1_ip = "172.20.3.3"

Problem

Tried to deploy 2 deployments of 2 VMs on different nodes, but the IPs were the same ❌

node1_zmachine1_ip = "172.20.3.2"
node2_zmachine1_ip = "172.20.3.2"

This is because the parallelism that terraform supports for the resource, so if we deployed 2 vm resources it won't wait for the state to be updated

We should mention this issue as a known one in the README

[rawdaGastan]

Known issue

Tried to deploy 2 deployments of 2 VMs on different nodes, but the IPs were the same ❌

node1_zmachine1_ip = "172.20.3.2"
node2_zmachine1_ip = "172.20.3.2"

This is because the parallelism that terraform supports for the resource, so if we deployed 2 or more vm resources it won't wait for the state to be updated, hence the same IP will be assigned for them