Adam Fabian

It seems like we would generally want ntp running and enabled from the start. You would want the time right on compute nodes for the VMs, and between nodes and...

- [`kube-ovn-controller` pods](https://kube-ovn.readthedocs.io/zh-cn/stable/en/reference/architecture/#kube-ovn-controller) play a critical central role - However, _Kube-OVN_ seems to allow them to schedule on any kind of node - You can see this by contrasting the...

The OVN pod doesn't have Python or a lot of tools. Perl looked like the best thing available for parsing the JSON to get the object-store URL out of the...

Please review and merge or request changes. The OVN backup stuff started looking a little cluttered, so I put it in a subdirectory. That also helped me use: ``` commonLabels:...

@cloudnull I mostly did it this way to avoid installing dependencies. From container images generally available, I could get one with the OpenStack command, or one with kubectl-ko to actually...

aedan, maybe the2hill, showed me this in the standup: https://github.com/rackerlabs/genestack/tree/main/Containerfiles I think the entire process of getting an image there seems a bit involved for this. Another option seems like...

closing in favor of https://github.com/rackerlabs/genestack/pull/304

I managed to deploy keystone on my test cluster as below. I copied the command from the genestack wiki, and de-indented the line I added: ``` helm upgrade --install keystone...

I found install-kube-ovn.sh installing chart version v1.12.31 with image v1.12.32 from /opt/genestack/base-helm-configs/kube-ovn/kube-ovn-helm-overrides.yaml

I tried enabling this on a fresh hyperconverged lab with _Kube-OVN_ _1.14.11_ . With ping going between two instances, this went smoothly and caused about 12 seconds of data plane...