Can not read data from GCS

[shihs]

I am trying to use Databricks to read data on Google Cloud Storage (GCS) with Databricks on Google Cloud. I followed the steps from https://docs.gcp.databricks.com/storage/gcs.html.

I have tried Access GCS buckets using Google Cloud service accounts on clusters, but I still couldn't read the file on GCS with the code below

from pyspark.sql import SparkSession

spark = SparkSession.builder.appName("test").getOrCreate()

df = spark.read.format("csv").load("gs://mybucket/test.csv")

The error message I got

```
"xxx@xxx.iam.gserviceaccount.com does not have storage.objects.get access to the Google Cloud Storage object. Permission 'storage.objects.get' denied on resource (or it may not exist).",
```

I also tried Access a GCS bucket directly with a Google Cloud service account key. I stucked in Step 4 & 5. Since step 5 uses `{{secrets/scope/gsa_private_key}}` and `{{secrets/scope/gsa_private_key_id}}` to get thegsa_private_key and gsa_private_key_id. I am not quite sure where should I do the step 4? I think it doesn't make to do it on local computer, however, it is also weird to do it on the cluster terminal. 

Please help me solve this problem. Thanks in advance! 

[Kaniz_Fatma]

Hi @shihs , 

- Check your Service Account Permissions: Ensure the service account has "storage.objects.get" permission for the GCS bucket. Add the "Storage Object Viewer" role to your service account via the GCS console.

 - Use External Locations: Databricks recommends using external locations to connect to GCS. Create an external place combining a cloud storage path with a storage credential. Grant permissions like CREATE TABLE, READ FILES, WRITE FILES on it.