Which option do you choose to start a VM business process?

Asked by xingyijin on 2020-12-19

Well, when I'm using Tacker to instantiated VNF, although the underlying image of the different VM is the same, like Ubuntu, different VM have different business processes.
My question is: how do these VMs automatically load different business processes? My personal idea is:
(1) Based on the shared Ubuntu image, different VMs make different images and load different business processes.
(2) Implement the development of the management port driver and plug-in by myself, and meet the requirements by re-implementing mgmt_call or mgmt_create_post interface.
Which plan? Or is there any other plan? Thank you!
By the way, in the second solution, I only saw examples of VNF config, which was not seen using the automatic startup business process.

Question information

Language:
English Edit question
Status:
Answered
For:
tacker Edit question
Assignee:
No assignee Edit question
Last query:
2021-01-15
Last reply:
2021-01-15
Launchpad Janitor (janitor) said : #1

This question was expired because it remained in the 'Open' state without activity for the last 15 days.

xingyijin (xingyijin) said : #2

Can someone help me?

Yoshito Ito (ito.yoshito) said : #3

Thank you for your question! And sorry for my late reply.

It's highly depends on the requirements of your image management. You can handle the image in Tacker with both of them; creating dedicated images for each business process, or implementing MgmtDriver for each case on the same image.

Also you can put all your scripts in HOT as user-data boot scripts with cloud-init as third choice. However, this is a little tricky because you need to use UserData operation, which is just released in Tacker [1].

In terms of image management, if you would like to start your business quickly, it seems better to create different images for each process because it's simple. On the other hand, if you assume to have frequent upgrade of business process, it seems better to keep the same image because replacing the image of VNF is more complex than respawn the process.

The sample of MgmtDriver is very limited in Tacker repository so far, and we'll release a new MgmtDriver implementation for SOL-based implementation soon! This provides you a good sample for more complex operation.

[1] https://docs.openstack.org/tacker/latest/user/etsi_vnf_deployment_as_vm_with_user_data.html

Can you help with this problem?

Provide an answer of your own, or ask xingyijin for more information if necessary.

To post a message you must log in.