Wavefront does not start when using Aria Automation 8.16.1

Products

• VMware Aria Suite

Issue/Introduction

• When Wavefront is enabled using "vracli wavefront internal" command the wavefront-proxy pods are unable to start.
• In the output of "kubectl get pods -n wavefront" you see the status is ErrImagePull or ImagePullBackOff.

Environment

• VMware Aria Automation 8.16.1

Cause

This is caused by a misconfiguration in version 8.16.1.

Impact:

This only affects versions 8.16.1 where the internal wavefront proxy has been enabled via the vracli command.

Resolution

This issue is resolved in versions 8.16.2 and above.

Workaround:

Prerequisites:

• You have taken simultaneous snapshots for each virtual machine participating in the Aria Automation cluster.
• You have access to root username and password.
• You have access to SSH tools such as PuTTy.

Procedure:

1. SSH into one node in the cluster.
2. Run the following command once in order to create a profile for wavefront:
   

   export WAVEFRONT_URL='<WAVEFRONT_URL>';export WAVEFRONT_TOKEN='<WAVEFRONT_TOKEN>';base64 -d <<< 'IyEvYmluL2Jhc2gKCm1rZGlyIC1wICIvZXRjL3Ztd2FyZS1wcmVsdWRlL3Byb2ZpbGVzL3JlZ2lzdHJ5L2hlbG0vd2F2ZWZyb250IgoKY2F0IDw8RU9GID4gL2V0Yy92bXdhcmUtcHJlbHVkZS9wcm9maWxlcy9yZWdpc3RyeS9oZWxtL3dhdmVmcm9udC81MC1yZXNvdXJjZXMueWFtbAppbWFnZToKICByZWdpc3RyeTogIiIKRU9GCgpoZWxtIHVwZ3JhZGUgXAogIC0taW5zdGFsbCAtLWZvcmNlIC1uIHdhdmVmcm9udCBcCiAgLS1zZXQtc3RyaW5nICJ3YXZlZnJvbnQudXJsPSRXQVZFRlJPTlRfVVJMIiBcCiAgLS1zZXQtc3RyaW5nICJ3YXZlZnJvbnQudG9rZW49JFdBVkVGUk9OVF9UT0tFTiIgXAogIC1mIC9ldGMvdm13YXJlLXByZWx1ZGUvcHJvZmlsZXMvcmVnaXN0cnkvaGVsbS93YXZlZnJvbnQvNTAtcmVzb3VyY2VzLnlhbWwgXAogIHdhdmVmcm9udC1wcm94eSAvb3B0L2NoYXJ0cy93YXZlZnJvbnQtcHJveHkK' | bash -

   Note: Replace <WAVEFRONT_URL> and <WAVEFRONT_TOKEN> with their respective values used when enabling it with the vracli wavefront command.
   Note: If wavefront is disabled, delete the custom profile directory tree from each node. The below command only needs to be run once on one appliance in the cluster:
   

   vracli cluster exec – bash -c 'rm -rf /etc/vmware-prelude/profiles/registry/helm/wavefront'