Reputation: 482
I want to access cross gcp project's cloud storage using hadoop file system APIs to read parquet, avro and sequence files. If i setup a dataproc cluster in a gcp project named "proj1", how can i read cloud storage files in other gcp project named "proj2" using the dataproc cluster in "proj1"?
Upvotes: 1
Views: 671
Reputation: 75940
You can use the service accounts to achieve this. Create one, set the permission that you want, on the bucket and the project that you want (use its email address as any email address, as you could do with a user email address), and that's all!
Upvotes: 3