UnsupportedFileSystemException: No FileSystem for scheme "abfs" when creating external table in Trino with ADLS Gen2

[rajatnt]

I'm attempting to create an external table in Trino (v475) backed by Azure Data Lake Storage Gen2 using the abfs or abfss URI schemes. However, I'm encountering the following errors:

trino:default> SHOW SCHEMAS FROM azure;
       Schema
--------------------
 default
 information_schema
(2 rows)

Query 20250603_164407_00003_g6hjr, FINISHED, 1 node
Splits: 5 total, 5 done (100.00%)
0.73 [2 rows, 35B] [2 rows/s, 48B/s]

trino:default> CREATE TABLE azure.default.existing_customer_data (
            ->     id bigint,
            ->     name varchar(100),
            ->     email varchar(255),
            ->     created_date date
            -> ) WITH (
            ->     format = 'PARQUET',
            ->     external_location = 'abfs://[email protected]/tables/customer_data'
            -> );
Query 20250603_164415_00004_g6hjr failed: Got exception: org.apache.hadoop.fs.UnsupportedFileSystemException No FileSystem for scheme "abfs"

trino:default> CREATE TABLE azure.default.existing_customer_data (
            ->     id bigint,
            ->     name varchar(100),
            ->     email varchar(255),
            ->     created_date date
            -> ) WITH (
            ->     format = 'PARQUET',
            ->     external_location = 'abfss://[email protected]/tables/customer_data'
            -> );
Query 20250603_164427_00005_g6hjr failed: Got exception: org.apache.hadoop.fs.UnsupportedFileSystemException No FileSystem for scheme "abfss"

trino:default> CREATE TABLE azure.default.existing_customer_data (
            ->     id bigint,
            ->     name varchar(100),
            ->     email varchar(255),
            ->     created_date date
            -> ) WITH (
            ->     format = 'PARQUET',
            ->     external_location = 'wasbs://[email protected]/tables/customer_data'
            -> );
Query 20250603_164634_00006_g6hjr failed: External location is not a valid file system URI: wasbs://[email protected]/tables/customer_data

trino:default>

Configuration Details:
catalog/azure.properties

connector.name=hive
hive.metastore=thrift
hive.metastore.uri=thrift://hive-metastore:9083
hive.non-managed-table-writes-enabled=true
azure.auth-type=OAUTH
azure.oauth.tenant-id=<tenant-id>
azure.oauth.endpoint=https://login.microsoftonline.com/<tenant-id>/oauth2/token
azure.oauth.client-id=<client-id>
azure.oauth.secret=<secret-id>
fs.native-azure.enabled=true

Docker Compose (Partial)

services:
  trino:
    image: trinodb/trino:latest
    ports:
      - "8080:8080"
    volumes:
      - ./trino-config:/etc/trino
    environment:
      - JAVA_TOOL_OPTIONS=-Xmx4G
    depends_on:
      - hive-metastore
    
  # Standalone Hive Metastore for Azure
  hive-metastore:
    image: apache/hive:3.1.3
    ...

Azure Access Configuration
Using Service Principal credentials
Storage account has Storage Blob Data Contributor role assigned to the Service Principal

Observations
The SHOW SCHEMAS FROM azure command works, indicating the catalog is recognized.
The error appears to stem from Trino or the underlying Hive connector not recognizing or supporting the abfs / abfss file system schemes.
A similar issue was reported here #25863 and #25919

Is this a known limitation or bug in Trino’s Hive connector for Azure?
Is there additional configuration or plugin/jar required to support abfs[s] in this setup?

[rajatnt]

If we are using an older version of Trino (~ 458), we are still facing issues with the legacy method of connection.

Since 458 is an older version, we have not opened a bug on it, but we have reported this in the discussion forum here - #25918

[rajatnt]

Tried a few more scenarios.

Tried using the iceberg connector.

Below is the properties.

connector.name=iceberg
iceberg.catalog.type=hive_metastore
iceberg.register-table-procedure.enabled=true
iceberg.unique-table-location=true
iceberg.register-table-procedure.enabled=true
hive.metastore.uri=thrift://hive-metastore:9083

# Azure OAuth Configuration (instead of access key)
fs.native-azure.enabled=true
azure.auth-type=OAUTH
azure.oauth.client-id=client-id
azure.oauth.secret=secret-id
azure.oauth.tenant-id=tenant-id3
azure.oauth.endpoint=https://login.microsoftonline.com/client-id/oauth2/v2.0/token

With above, used both latest trino docker image and the 466 docker image.

I was trying different ways to connect to azure storage and thought to register an already-existing Iceberg table. So I pre-created the table using Spark in azure using a spark docker container. Post this tried below which is also resulting in failed: Got exception: org.apache.hadoop.fs.UnsupportedFileSystemException No FileSystem for scheme "abfss"

From the trino CLI -

ruser@rubuntu:~/iceberg$ docker ps -a
CONTAINER ID   IMAGE               COMMAND                  CREATED              STATUS                        PORTS                                                               NAMES
6d50f7f360a6   trinodb/trino:466   "/usr/lib/trino/bin/…"   About a minute ago   Up About a minute (healthy)   0.0.0.0:8080->8080/tcp, [::]:8080->8080/tcp                         trino
c9527a5e07ee   apache/hive:3.1.3   "sh -c /entrypoint.sh"   About a minute ago   Up 6 seconds                  10000/tcp, 0.0.0.0:9083->9083/tcp, [::]:9083->9083/tcp, 10002/tcp   hive-metastore
26ad9f3453f3   postgres:13         "docker-entrypoint.s…"   About a minute ago   Up About a minute             0.0.0.0:5432->5432/tcp, [::]:5432->5432/tcp                         postgres
azureuser@rntubuntu:~/iceberg$ ./trino --server http://localhost:8080 --catalog hive --schema default
trino:default> CALL azure.system.register_table(
            ->     schema_name => 'default',
            ->     table_name => 'customers',
            ->     table_location => 'abfss://[email protected]/tables/default/customers'
            -> );
Query 20250604_125938_00001_295cs failed: Got exception: org.apache.hadoop.fs.UnsupportedFileSystemException No FileSystem for scheme "abfss"

trino:default>