prestodb/presto · error · PrestoException

HIVE_FILESYSTEM_ERROR

HIVE_FILESYSTEM_ERROR

Error message

Failed checking path: 

What it means

isS3FileSystem obtains a Hadoop FileSystem for the given path to test whether the underlying (raw) filesystem is PrestoS3FileSystem; an IOException from the Hadoop filesystem layer is wrapped in HIVE_FILESYSTEM_ERROR with the path appended. This check is part of default-location validation and temporary-path creation.

Source

Thrown at presto-hive/src/main/java/com/facebook/presto/hive/HiveWriteUtils.java:452

                throw new PrestoException(HIVE_DATABASE_LOCATION_ERROR, format("Database '%s' location is not a directory: %s", schemaName, databasePath));
            }
        }

        return new Path(databasePath, tableName);
    }

    private static Database getDatabase(ConnectorIdentity identity, MetastoreContext metastoreContext, SemiTransactionalHiveMetastore metastore, String database)
    {
        return metastore.getDatabase(metastoreContext, database).orElseThrow(() -> new SchemaNotFoundException(database));
    }

    public static boolean isS3FileSystem(HdfsContext context, HdfsEnvironment hdfsEnvironment, Path path)
    {
        try {
            return getRawFileSystem(hdfsEnvironment.getFileSystem(context, path)) instanceof PrestoS3FileSystem;
        }
        catch (IOException e) {
            throw new PrestoException(HIVE_FILESYSTEM_ERROR, "Failed checking path: " + path, e);
        }
    }

    public static boolean isViewFileSystem(HdfsContext context, HdfsEnvironment hdfsEnvironment, Path path)
    {
        try {
            return getRawFileSystem(hdfsEnvironment.getFileSystem(context, path)) instanceof ViewFileSystem;
        }
        catch (IOException e) {
            throw new PrestoException(HIVE_FILESYSTEM_ERROR, "Failed checking path: " + path, e);
        }
    }

    private static FileSystem getRawFileSystem(FileSystem fileSystem)
    {
        if (fileSystem instanceof HadoopExtendedFileSystem) {
            return getRawFileSystem(((HadoopExtendedFileSystem) fileSystem).getRawFileSystem());
        }

View on GitHub (pinned to 55bb57d202)

Solutions

1. Verify the storage cluster is reachable and the NameNode/endpoint is healthy (hdfs dfs -ls <path> from a worker)
2. Check filesystem URI scheme is correct and supported (hdfs://, s3a/s3n, viewfs) in the database location
3. Review worker logs for the wrapped IOException cause (auth, DNS, timeout) and fix credentials/network accordingly

Example fix

// before
location: hdfs://wrongnn:9000/warehouse/db.db -- NameNode unreachable
// after
ALTER DATABASE db SET LOCATION 'hdfs://correctnn:9000/warehouse/db.db';

Path p = new Path(dbLocation);
try (FileSystem fs = FileSystem.get(p.toUri(), conf)) { fs.getUri(); }
catch (IOException e) { throw new IllegalStateException("Filesystem unreachable for " + p, e); }

try {
    createTable(schema, ...);
} catch (PrestoException e) {
    if (e.getErrorCode() == HIVE_FILESYSTEM_ERROR.toErrorCode() && e.getCause() instanceof IOException) {
        // check NN health/auth, back off, retry
    } else throw e;
}

When it happens

Trigger: Any call path (e.g. getTableDefaultLocation, CREATE TABLE location validation) where opening the FileSystem for the database path fails — NameNode unreachable, kerberos/auth failure, unknown URI scheme, network timeout.

Common situations: HDFS NameNode down or in safe mode; wrong URI scheme (no handler for scheme); expired/invalid Kerberos token or missing credentials; network partition to storage.

Related errors

• HIVE_FILESYSTEM_ERROR
• HIVE_FILESYSTEM_ERROR
• HIVE_WRITER_CLOSE_ERROR
• HIVE_WRITER_OPEN_ERROR
• HIVE_BAD_DATA

AI-assisted analysis of prestodb/presto@55bb57d202 (2026-09-04). Data as JSON: /api/errors/bbf9ef11918dfd7a. Report an issue: GitHub.