Deploy vRO v8.0.1 Instance

Description:

In this post, I will go over how I deployed a vRO appliance v8.0.1 in on-prem vCenter.

To Resolve:

1. Login to VMWare’s support site and download the .iso image.

2. Deploy it in vCenter with the appropriate hostname, IP Address info, etc.

3. Follow this guide to setup vCenter SSO

   • Essentially, login to https://your_orchestrator_FQDN/vco-controlcenter with root/ password you setup in vCenter when you deployed appliance
   • Join to vCenter using admin account
   • Add the AD group you want to have access

4. In my case, I couldn’t get to vco-controlcenter even on first boot because vco-app failed to even start. So I followed this KB which has you run these 3 commands:

   vracli cluster exec -- bash -c 'echo -e "FROM vco_private:latest\nRUN sed -i s/root:.*/root:x:18135:0:99999:7:::/g /etc/shadow\nRUN sed -i s/vco:.*/vco:x:18135:0:99999:7:::/g /etc/shadow" | docker build - -t vco_private:latest'
   
   vracli cluster exec -- bash -c 'echo -e "FROM db-image_private:latest\nRUN sed -i s/root:.*/root:x:18135:0:99999:7:::/g /etc/shadow\nRUN sed -i s/postgres:.*/postgres:x:18135:0:99999:7:::/g /etc/shadow" | docker build - -t db-image_private:latest'
   
   #Persist the new changes through reboots:
   
   /opt/scripts/backup_docker_images.sh
   

5. Once that was setup, SSH’d into the box and created a script called /root/rs.sh for restart-service:

   /opt/scripts/svc-stop.sh
   sleep 120
   /opt/scripts/deploy.sh --onlyClean
   sleep 60
   /opt/scripts/deploy.sh
   

   • I also ran passwd -x 99999 root because it has something like under the release notes

   • This whole issue started because I would issue reboot commands and apparently that stopped Kubernetes from working properly. I would run kubectl describe nodes and they would all be set to starting. Running kubectl --namespace kube-system describe pod tiller-deploy-5c996fbc66-zxhmt would give me 0/1 nodes are available: 1 node(s) had taints that the pod didn't tolerate.

6. Troubleshooting VRO:

   • Run watch "kubectl -n prelude get all" to see all containers matching prelude and their status
   • Run kubectl describe nodes to get a general idea of containers
   • Run kubectl --namespace kube-system describe pod tiller-deploy-5c996fbc66-zxhmt to see details on a specific pod, just replace tiller-deploy with whatever from previous command
   • See here for generic commands
   • See here for more like:

   kubectl get pods --all-namespaces # List all pods in all namespaces
   kubectl get pods # List all pods in the namespace
   kubectl logs my-pod # dump pod logs (stdout)