OpenShift installer kills all VMs

[sgtfoo]

I've had trouble tracking down the cause, but at the point in the 3.11 OpenShift installer, the task called `TASK [Initialize openshift.node.sdn_mtu]` is the last one before suddently all my Proxmox VMs shutdown.
I can't think of what this could possibly be aside from an excess of network activity, or perhaps a collision due to the SDN starting up.
Can anyone advise where I should look aside from /var/log/messages ?

Has anyone else had a successful OpenShift cluster build in Proxmox? If so, what did you inventory look like?

 

[LnxBil]

To clearify: Where are you going to install OpenShift: Hypervisor (Proxmox VE), LX(C) container or KVM VM?

 

[sgtfoo]

To clearify: Where are you going to install OpenShift: Hypervisor (Proxmox VE), LX(C) container or KVM VM?

KVM VMs

 

[Gilang Ifandanu]

I've had trouble tracking down the cause, but at the point in the 3.11 OpenShift installer, the task called `TASK [Initialize openshift.node.sdn_mtu]` is the last one before suddently all my Proxmox VMs shutdown.
I can't think of what this could possibly be aside from an excess of network activity, or perhaps a collision due to the SDN starting up.
Can anyone advise where I should look aside from /var/log/messages ?

Has anyone else had a successful OpenShift cluster build in Proxmox? If so, what did you inventory look like?

I successfully installed openshift cluster (3.11) on Proxmox 5.1-41. the instalation is complete without any error. I upload my inventory file.

The problem I'm facing is when build/deploy a new apps, the pulling and pushing images is realy slow. It took almost an hour.
But my friend just tried it out on proxmox version 4.xx, the new apps deployment run smoothly using the same inventory.

 

[Alwin]

`TASK [Initialize openshift.node.sdn_mtu]`

Sounds like the MTU might get changed and if not all network hops have the same MTU, the packets get fragmented.