VMware Aria Automation Orchestrator SOAP Plugin failing because of NumberFormatException

Products

• VMware Aria Suite

Issue/Introduction

Symptoms:

• The SOAP plugin is not working, in VMware Aria Automation logs /services-logs/prelude/vco-app/files-logs/vco-server-app.log the following error is seen:

java.lang.NumberFormatException: For input string: ""
    at java.lang.NumberFormatException.forInputString(NumberFormatException.java:65) ~[?:?]
    at java.lang.Integer.parseInt(Integer.java:662) ~[?:?]
    at java.lang.Integer.parseInt(Integer.java:770) ~[?:?]
    at org.apache.axis2.transport.http.ProxyConfiguration.configure(ProxyConfiguration.java:191) ~[axis2-transport-http-1.5.6.jar:?]

• During the startup of the vco-app (Aria Automation Orchestrator) pod, the logs show an empty proxy value.

Sysprop: http.proxyPort =

Note: Running the command vracli proxy show will indicate the proxy is disabled.

Environment

VMware Aria Automation Orchestrator 8.12.x

Cause

The user has manually disabled the proxy service. This should only be done under consultation with VMware. The plugin requires the http.proxyPort property to be non-empty.

The current versions of the SOAP plug-in require the proxy service to run.

VMware will include in the future release, SOAP plug-in version 2.0.9 or higher, the ability of the plug-in to work having the proxy stopped. 

Resolution

Prerequisites

• You have taken simultaneous snapshots of each appliance in the cluster with memory disabled.
  • Use the quiesce option for versions 8.9.1 and above.
• You have access to root username and password for SSH connections.

Procedure

1. SSH to one appliance in the cluster.
2. Run the following commands.

vracli proxy enable
vracli proxy apply

4. If the external proxy is not being used, restore the default proxy configuration.

vracli proxy set-default
vracli proxy apply

5. Restart the orchestrator service.

kubectl scale deployment vco-app -n prelude --replicas=0
kubectl scale deployment vco-app -n prelude --replicas=$(kubectl get nodes --no-headers | wc -l)

6. Monitor the restart of the pods until they are running 3/3

kubectl get pods -n prelude -w | grep vco-app

Additional Information

Impact/Risks:

After applying the instructions in this article, the proxy service will be running.