I'm wondering if there is an option on modern nova (Beyond newton) allowing me to automatically failover an instances that is hosted on a failed node?

I know that the purpose of the cloud is to create a swarm and so don't bother about this kind of issue, but, you know, clients don't always behave like you intend it will.

Additionally, I don't want to force my clients to spawn multiple instances and so paid twice the price just to be sure that nothing bad happen.

Is this achievable without using distributed/shared storage?


Ok, so after digging a little bit, I find out that Openstack labeled this issue as a critical feature lack and is working on it as stated here:

Openstack VM_HA

This purposal and discussion has been opened in 2016 and not yet implemented as for July 2017

Status of this issue can be followed here:

VM_HA status

As explain on both links, some workaround are available through vendor specific solution but not considered as natively implemented.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.