I'm going to deploy VIO with Kubernetes (VIOK) features, after some tests:
- if i deploy VIOK in SDDC provider mode i have no problems and it works perfectly
- if i deploy VIOK in OPENSTACK provider mode the cluster will be created inside my VIO deployment but the cluster remains in CREATION state for some minutes (120 min) and after will go in ERROR
I tried a lot of time to deploy VIOK in OPENSTACK provider mode but with no results.
I understand also:
- if i deploy VIOK in SDDC provider mode the ubuntu image for k8s is auto selected
- if i deploy VIOK in OPENSTACK provider mode i must select ubuntu-cloudimage from my VIO deploy
- the two images are completely different!!! one is about2GB, the other 900MB
- i also regenerate VIO.crt as documented
- i have one internal vxlan routed (exclusive) to external network, dns, pings and so on works
So what is the issue???
Help me please, i dont have any other idea
At least I can think that the problem of SSH connectivity between K8S-manager and the nodes is the MTU
After some search i found the log of the cluster deploy (vmware column), its very strange beacuse i dont think there are network issues:
PLAY [all:!localhost] **********************************************************
TASK [Verify SSH connectivity with every nodes] ********************************
skipping: [k8s-master-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0]
skipping: [k8s-node-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0]
skipping: [bastion]
TASK [Verify SSH connectivity with bastion] ************************************
skipping: [k8s-master-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0]
skipping: [k8s-node-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0]
172.18.0.2 - - [26/Oct/2017 09:54:24] "GET /runs/db68310d-03cd-4431-9582-c5e697d408a0 HTTP/1.1" 200 -
ok: [bastion -> localhost]
TASK [Fetch the python path on bastion host] ***********************************
changed: [k8s-master-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0]
changed: [bastion]
changed: [k8s-node-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0]
TASK [set the ansible_python_interpreter for the bastion host] *****************
ok: [k8s-master-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0]
ok: [k8s-node-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0]
ok: [bastion]
TASK [Verify SSH connectivity with every other nodes] **************************
skipping: [k8s-master-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0]
skipping: [bastion]
172.18.0.2 - - [26/Oct/2017 09:54:29] "GET /runs/db68310d-03cd-4431-9582-c5e697d408a0 HTTP/1.1" 200 -
...
...
...
fatal: [k8s-node-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0]: UNREACHABLE! => {"changed": false, "msg": "SSH Error: data could not be sent to the remote host. Make sure this host can be reached over ssh", "unreachable": true}
NO MORE HOSTS LEFT *************************************************************
to retry, use: --limit @/app/ansible/cluster.retry
PLAY RECAP *********************************************************************
bastion : ok=3 changed=1 unreachable=0 failed=0
k8s-master-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0 : ok=2 changed=1 unreachable=0 failed=0
k8s-node-0-34c41420-9b0a-4041-8632-8c52dbf7a3d0 : ok=2 changed=1 unreachable=1 failed=0
localhost : ok=5 changed=1 unreachable=0 failed=0
