Data collection for vSphere cloud accounts is failing in Aria Automation. Communication to the accounts and password is working as image synchronization and provisioning status is still completing successfully.

search cancel

book

calendar_today

VCF Operations/Automation (formerly VMware Aria Suite)

The data collection can run for a long time and eventually fails.
Provisioning service logs located under /services-logs/prelude/provisioning-service-app/file-logs/provisioning-service-app.log contains the following error messages.
- java.util.concurrent.CompletionException: java.lang.IllegalStateException: Failed to retrieve properties.
- Fault received from vCenter [...] The session is not authenticated.
- Existing connection has been closed or is invalid. Removing it from pool.

Transient communication issues between Aria Automation and vCenter an cause data collection to become stuck requiring a restart to resolve

Where the data collection job is getting stuck due to issues on the service pods, we can resolve it in two ways:

You can mark the data collection for a cloud account as Failed using this API call. It will restart automatically within 10 minutes.
- curl -kv --location --request PATCH 'https://<FQDN>/iaas/api/cloud-accounts/<cloud_account_id>?apiVersion=2021-07-15' \ --header 'Authorization: Bearer <access_token>' \ --header 'Content-Type: application/json' \ --data-raw '{ "customProperties": { "enumerationTaskState": "FAILED" }, "privateKey":"<vSphere_password>", "privateKeyId":"<vSphere_username>" }'
Alternatively, delete the 3 provisioning-service-app pods to restart this service which is responsible for data collection (among other things)
Note that VM provisioning and other functions will be unavailable for a few minutes while the pods restart.
1. Get pod names:
  - kubectl -n prelude get pods | grep provisioning-service-app
2. Restart pods:
  - kubectl -n prelude delete pods <POD1> <POD2> <POD3>

thumb_up Yes

thumb_down No