Solved

Karbon deployment Failed: failed to deploy the ntnx dvp

  • 5 August 2019
  • 4 replies
  • 305 views

Deploying a karbon cluster we got this error on 8% of process:



Deployment Failed



invalid argument: internal error: internal error: failed to deploy the ntnx dvp: Failed to configure with SSH: Failed to run command: on host: "xx,xx,xx,xx:22" error: "Process exited with status 1"



Has anyone the same problem? Any idea on how to advance?



Regards.
Javier
icon

Best answer by JoseNutanix 5 August 2019, 12:07

Hi Javier,

You will need connectivity from the Karbon VLAN to the AOS, AHV and PC VLAN. Also, from the Karbon VLAN to the internet.

A quick test is to connect to the only provisioned VM via SSH and try to pull an image from Docker Hub.

You can try the following to make sure the VM has connectivity to the Internet.
code:
sudo docker pull alpine

When you get the prompt with the credentials for sudo, use the common Nutanix one.

If after this it doesn't work, I'd suggest you open a ticket with our support.

Regards,
Jose Gomez
View original

4 replies

Userlevel 3
Badge +1
Hi Javier,

Can you share the details for your deployment?

Usually when it fails at 8% is because network problems with IP address allocations or internet connectivity.

Regards,
Jose
Hi Jose.

This is our environment (5 nodes)

AOS 5.10.4.1 | Calm 2.6.0.6 | Epsilon 2.6.0.6 | Karbon 1.0.1 | NCC 3.7.1.2 | PC 5.10.5 | AHV 20170830.270

We have defined a new VLAN for karbon, one different than AOS, AHV and PC. Firewall access from Nutanix VLAN to karbon VLAN is all open. I see in the karbon cluster creation process that one VM is created, started and have an IP assigned from external DHCP (relayed). I'm able to connect on SSH in that machine.

Example of networking definition for karbon VLAN:

Network addresses on Karbon VLAN: 10.x.x.x/16
Service CIDR Ranges: 10.y.x.x/16
Pod CIDR Range: 10.z.x.x/16

Regards
Javier
Userlevel 3
Badge +1
Hi Javier,

You will need connectivity from the Karbon VLAN to the AOS, AHV and PC VLAN. Also, from the Karbon VLAN to the internet.

A quick test is to connect to the only provisioned VM via SSH and try to pull an image from Docker Hub.

You can try the following to make sure the VM has connectivity to the Internet.
code:
sudo docker pull alpine

When you get the prompt with the credentials for sudo, use the common Nutanix one.

If after this it doesn't work, I'd suggest you open a ticket with our support.

Regards,
Jose Gomez
Hi Jose.

Problem solved!!. Loging into VM helped to understand the problem. The problem was a misconfiguration on the DHCP options.

The scope had option 5 (Name Servers) defined instead the option 6 (DNS Servers) for the dns server list. For that reason the VM can't download from internet.

Thanks for your help!

Best Regards
Javier.

Reply