EVE-NG Customized image for CCIE

[Myccie]

Hello Folks,

 

Does anyone know how to save a customized image, namely vManage, in EVE-NG and reuse it whenever needed, similar to how it's done in CLC?

It takes a lot of time to build it from scratch.

[D4m13n]

Hey. I've never done it myself, but here's what GPT Plus has to say on the matter.

-----------------------------------------------------------------------------------------

Yes, you can do it by creating a "golden" / customized QEMU image.

The safest method is:

1. Clone your original vManage image folder first, so you don't overwrite the clean image.

Example:

cp -a /opt/unetlab/addons/qemu/vtmgmt-20.x.x /opt/unetlab/addons/qemu/vtmgmt-20.x.x-custom
/opt/unetlab/wrappers/unl_wrapper -a fixpermissions

2. Add a new vManage node in a lab using the vtmgmt-20.x.x-custom image.

3. Configure vManage as needed.

4. Properly shut down vManage from inside the OS, do not just kill/stop the node.

5. In EVE-NG, check:

- Lab UUID from Lab Details
- Node ID by right-clicking the node
- POD ID, usually 0 for admin

6. SSH to EVE and go to the temporary node folder:

cd /opt/unetlab/tmp/0/<lab-uuid>/<node-id>/

7. Commit the disk changes back to the image:

/opt/qemu/bin/qemu-img commit virtioa.qcow2
/opt/qemu/bin/qemu-img commit virtiob.qcow2

8. Fix permissions:

/opt/unetlab/wrappers/unl_wrapper -a fixpermissions

After that, every new vManage node created from vtmgmt-20.x.x-custom should already contain your customized state.

Important notes:

- If you commit directly against the original vtmgmt-* image, you overwrite the base image.
- For vManage, remember the second disk: virtiob.qcow2. EVE’s own Viptela guide creates virtioa.qcow2 plus a second 100 GB virtiob.qcow2, so customized vManage data may live there too.
- EVE’s official method for saving QEMU node changes is to shut down the node, locate /opt/unetlab/tmp/<pod>/<lab-uuid>/<node-id>/, then run qemu-img commit on the relevant qcow2 disk.