We are trying to integrate OpenNebula with Edge Clusters in AWS but we are having a lot of problems. First of all, the default Ansible playbook is not working properly as @ahuertas tells in this post. A lot of customization is required to work properly if for example you need use Ubuntu 20.04 in terms of set provision template or Ansible playbooks.
Besides that, we also found a bug in CLI version 6.2 where documentation and the cluster operations guide site says hosts can be added to a deployed provision but it is not available in CLI.
if for example you need use Ubuntu 20.04 in terms of set provision template or Ansible playbooks.
This has been changed by default in the new release.
Besides that, we also found a bug in CLI version 6.2 where documentation and the cluster operations guide site says hosts can be added to a deployed provision but it is not available in CLI.
This has been also added to the new release, so this is present only in 6.2.1 version.
Another question is, can we use OpenNebula to deploy EC2 instances instead of create an KVM Node to host VMs?
No, this driver was deprecated in favor of OneProvision.
Hi @david.gar.dom! EE and CE follow a separate maintenance cycle until a new major/minor release is produced, when both are syncronized and âOpenNebula Systems contributes to the Community Edition all bug fixes and software patches with minor enhancements that have been developed and incorporated into an Enterprise Edition version during the previous maintenance cycleâ (see this for more details). The beta of the new 6.4 will be released by the end of March.
I am testing 6.4.0. The problem I had previously with not being able to add or remove hosts is solved but now I have other problems:
Providers site is not accesible from Fireedge main page. You have to know the exactly path (/fireedge/provision/dashboard) if not you are redirected to Sunstone site.
It is not possible to create virtual providers from the website, only via CLI.
Related with previous point, it is not possible to create Provisions in virtual Providers even when they have been created via CLI and appears in the website.
A lot of customization is still required in Ansible files.
I will keep trying diferent features to see if I detect some more things!
Following the aws.yml playbook, you need to be consisten in the whole dependent roles: ddc (OS family and version), opennebula-* âŚ
I really donât know if that is the team idea, but for final user it is not an simple setup. Ansible knowledge is required to start using the software. Furthermore, you canât edit this files from the website: the gear icon is confusing because it is not for edit, it is to reload the configâŚ
The playbooks are not supposed to be modified to launch a provision. They come with a predefined set of instance types, OS families and opennebula packages that are tested and work out of the box.
What issue are you having that forces you to change the playbooks?
Hello, please can any one assist in solving this error gotten after setting up AWS as provider and tried creating EC2 cluster, pls how can this be solved? Thank you.
Thank you, the developer has abandon the deployment and integration so I think I would have to delete the installation and maybe find an alternative to achieve the project objectives if I canât find someone to complete the work.
