Files
seaweedfs/other/java/hdfs3
chrislu a3cf4eb843 debug: track stream lifecycle and total bytes written
Added comprehensive logging to identify why Parquet files fail with
'EOFException: Still have: 78 bytes left'.

Key additions:
1. SeaweedHadoopOutputStream constructor logging with 🔧 marker
   - Shows when output streams are created
   - Logs path, position, bufferSize, replication

2. totalBytesWritten counter in SeaweedOutputStream
   - Tracks cumulative bytes written via write() calls
   - Helps identify if Parquet wrote 762 bytes but only 684 reached chunks

3. Enhanced close() logging with 🔒 and  markers
   - Shows totalBytesWritten vs position vs buffer.position()
   - If totalBytesWritten=762 but position=684, write submission failed
   - If buffer.position()=78 at close, buffer wasn't flushed

Expected scenarios in next run:
A) Stream never created → No 🔧 log for .parquet files
B) Write failed → totalBytesWritten=762 but position=684
C) Buffer not flushed → buffer.position()=78 at close
D) All correct → totalBytesWritten=position=684, but Parquet expects 762

This will pinpoint whether the issue is in:
- Stream creation/lifecycle
- Write submission
- Buffer flushing
- Or Parquet's internal state
2025-11-23 11:54:01 -08:00
..

SeaweedFS Hadoop3 Client

Hadoop FileSystem implementation for SeaweedFS, compatible with Hadoop 3.x.

Building

mvn clean install

Testing

This project includes two types of tests:

1. Configuration Tests (No SeaweedFS Required)

These tests verify configuration handling and initialization logic without requiring a running SeaweedFS instance:

mvn test -Dtest=SeaweedFileSystemConfigTest

2. Integration Tests (Requires SeaweedFS)

These tests verify actual FileSystem operations against a running SeaweedFS instance.

Prerequisites

  1. Start SeaweedFS with default ports:

    # Terminal 1: Start master
    weed master
    
    # Terminal 2: Start volume server
    weed volume -mserver=localhost:9333
    
    # Terminal 3: Start filer
    weed filer -master=localhost:9333
    
  2. Verify services are running:

Running Integration Tests

# Enable integration tests
export SEAWEEDFS_TEST_ENABLED=true

# Run all tests
mvn test

# Run specific test
mvn test -Dtest=SeaweedFileSystemTest

Test Configuration

Integration tests can be configured via environment variables or system properties:

  • SEAWEEDFS_TEST_ENABLED: Set to true to enable integration tests (default: false)
  • Tests use these default connection settings:
    • Filer Host: localhost
    • Filer HTTP Port: 8888
    • Filer gRPC Port: 18888

Running Tests with Custom Configuration

To test against a different SeaweedFS instance, modify the test code or use Hadoop configuration:

conf.set("fs.seaweed.filer.host", "your-host");
conf.setInt("fs.seaweed.filer.port", 8888);
conf.setInt("fs.seaweed.filer.port.grpc", 18888);

Test Coverage

The test suite covers:

  • Configuration & Initialization

    • URI parsing and configuration
    • Default values
    • Configuration overrides
    • Working directory management
  • File Operations

    • Create files
    • Read files
    • Write files
    • Append to files
    • Delete files
  • Directory Operations

    • Create directories
    • List directory contents
    • Delete directories (recursive and non-recursive)
  • Metadata Operations

    • Get file status
    • Set permissions
    • Set owner/group
    • Rename files and directories

Usage in Hadoop

  1. Copy the built JAR to your Hadoop classpath:

    cp target/seaweedfs-hadoop3-client-*.jar $HADOOP_HOME/share/hadoop/common/lib/
    
  2. Configure core-site.xml:

    <configuration>
      <property>
        <name>fs.seaweedfs.impl</name>
        <value>seaweed.hdfs.SeaweedFileSystem</value>
      </property>
      <property>
        <name>fs.seaweed.filer.host</name>
        <value>localhost</value>
      </property>
      <property>
        <name>fs.seaweed.filer.port</name>
        <value>8888</value>
      </property>
      <property>
        <name>fs.seaweed.filer.port.grpc</name>
        <value>18888</value>
      </property>
      <!-- Optional: Replication configuration with three priority levels:
           1) If set to non-empty value (e.g. "001") - uses that value
           2) If set to empty string "" - uses SeaweedFS filer's default replication
           3) If not configured (property not present) - uses HDFS replication parameter
      -->
      <!-- <property>
        <name>fs.seaweed.replication</name>
        <value>001</value>
      </property> -->
    </configuration>
    
  3. Use SeaweedFS with Hadoop commands:

    hadoop fs -ls seaweedfs://localhost:8888/
    hadoop fs -mkdir seaweedfs://localhost:8888/test
    hadoop fs -put local.txt seaweedfs://localhost:8888/test/
    

Continuous Integration

For CI environments, tests can be run in two modes:

  1. Configuration Tests Only (default, no SeaweedFS required):

    mvn test -Dtest=SeaweedFileSystemConfigTest
    
  2. Full Integration Tests (requires SeaweedFS):

    # Start SeaweedFS in CI environment
    # Then run:
    export SEAWEEDFS_TEST_ENABLED=true
    mvn test
    

Troubleshooting

Tests are skipped

If you see "Skipping test - SEAWEEDFS_TEST_ENABLED not set":

export SEAWEEDFS_TEST_ENABLED=true

Connection refused errors

Ensure SeaweedFS is running and accessible:

curl http://localhost:8888/

gRPC errors

Verify the gRPC port is accessible:

# Should show the port is listening
netstat -an | grep 18888

Contributing

When adding new features, please include:

  1. Configuration tests (no SeaweedFS required)
  2. Integration tests (with SEAWEEDFS_TEST_ENABLED guard)
  3. Documentation updates