Debugging object storage issues
Configuration and connectivity problems can be addressed by using various debugging guidelines.
Technical background
The flow of object storage information within Maximo® Visual Inspection includes the following details:
- User input
- The user provides the bucket name and enabled flag information in the application custom resource.
- Maximo Visual Inspection operator
- The Maximo Visual Inspection operator reads the user input information when the workspace is reconciled. The information is validated, including the bucket name that was added into the application custom resource.
- vision-service API pod
-
The
vision-service
API pod reads the bucket name from the workspace custom resource.
Debugging object storage issues
For issues with the flow of object storage information, check the following parameters:
URL region
Check that the storage bucket is in the same URL region that was created when you configured the IBM® Maximo Application Suite object storage entity.
A specific URL region is created for the IBM Cloud® Object Storage and Amazon Simple Storage Service (Amazon S3) storage providers only. The storage bucket must be created in the same URL region.
If the storage bucket was not created in the correct region, move the bucket to the correct region or delete and re-create the bucket in the correct region.
For more information, see the Installing Object Storage procedure in the Maximo Application Suite documentation.
Bucket name
Check that the bucket name exists and was entered correctly into the Maximo Visual Inspection application custom resource.
If these conditions are not correct, the status in Maximo Visual Inspection reports as Not Ready
.
If these conditions are correct, but image uploads continue to fail, check that the bucket is created in the correct region. For more information, see URL region.
Workplace custom resource
Check that the bucket name is in the workspace custom resource.
When the workspace is activated, the operator reconciles the workspace. The reconciliation includes adding the bucket name in the workspace custom resource so that the Maximo Visual Inspection vision-service
pod can retrieve it.
Open the Maximo Visual Inspection project namespace and search for resources that match the pattern via
. In the Status section, the workspace name is listed in the COSBucketID field.
To recreate the workspace custom resource, open the Red Hat® OpenShift® web console and delete the workspace ConfigMap.
vision-service pod
The vision-service
pod looks for the bucket name when it starts only.
Check that the vision-service
API pod has the correct bucket name by running the following curl command:
curl -k -H “x-auth-token: ${USER_API_KEY}” ${MVI_API_URL}/profiles
The command output provides the following JSON information:
{
"workspace_id": "cos",
"base_uri": "https://workspace.visualinspection.myinstance.somewhere.com",
"bucket_name": "n/a",
"object_store_enabled": false,
"last_commit": "53a6f17475c090e9325ad47adbed0cdf163f45ae",
"docker_manager": "k8s",
"encrypt_model_export": false,
"supported_cic_scenarios": [
"Flower",
"Landscape",
"Food",
"Action",
"Scene",
"Vehicle",
"Others"
]
}
Check the following parameters in the JSON information:
- The
object_store_enabled
field is set totrue
. - The
bucket_name
field matches the expected bucket name.
When these fields are not correct and all other checks are successful, restart the vision-service-pod
by deleting the pod in the Red Hat OpenShift web console.