Getting access token error when connecting from azure databricks to GCS bucket

kiko_roy
Contributor

I am creating a data frame by reading a table's data residing in Azure backed unity catalog. I need to write the dataframe or file to GCS bucket. I have configured the spark cluster config using the GCP service account json values. Also tried uploading it to dbfs and accessed the same through .config ()

 

on running : 

df1.write.format("parquet").save("gs://dev-XXXX-analyt-XXXXXXXX") getting error :
Error getting access token from metadata server at: http://169.254.169.254/computeMetadata/v1/instance/service-accounts/default/token
what could be the reason or resolution. Need help

KristiLogos
Contributor

DId you find out why this was happening? I'm having the same issue

@KristiLogos - Use the cluster access mode as : Single User and not shared and try 

KristiLogos
Contributor

@kiko_roy  unfortunately that didnt' work. the error is stating its trying to get the access token from metadata server, I wonder why from the metadata server?