hanzoai/s3 was an AGPL-3.0 MinIO fork whose network-copyleft blocks embedding
and reselling as a managed service. Replace it wholesale with a fork of
SeaweedFS (github.com/seaweedfs/seaweedfs, Apache-2.0, Go) which ships a
built-in S3 gateway + Filer.
- License: AGPL-3.0 -> Apache-2.0 (upstream LICENSE kept verbatim, incl.
"Copyright 2025 Chris Lu"); NOTICE added crediting SeaweedFS. No AGPL/BUSL/
SSPL/GPL or enterprise ("ee") files remain.
- Module: go.mod path set to github.com/hanzoai/s3 with a self-replace
(replace github.com/seaweedfs/seaweedfs => ./) so the ~1800 internal Go files
keep their upstream import paths with zero churn. Internal identifiers and
import paths are intentionally NOT renamed.
- CLI rebrand: the binary is built as `s3` (was `weed`); root usage, per-command
help/usage/flag strings, shell autocomplete program name, and the product name
in CLI output now read "Hanzo S3" / `s3`. Users never see `weed`.
(e.g. `s3 server`, `s3 filer`, `s3 master`, `s3 volume`, `s3 s3`, `s3 mount`.)
- Build: `make` (or `go build -o s3 ./weed`) builds the `s3` binary green;
`cmd/weed-db` and `cmd/weed-sql` also build; `go vet ./weed/command/` clean.
- README/LLM.md rewritten for Hanzo S3; AGENTS.md/CLAUDE.md symlinks preserved.
- Dropped a 33MB prebuilt Mach-O `stress` binary upstream had committed; `.gitignore`
updated to ignore the `s3` build output.
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
-
Start SeaweedFS with default ports:
# Terminal 1: Start master weed master # Terminal 2: Start volume server weed volume -master=localhost:9333 # Terminal 3: Start filer weed filer -master=localhost:9333 -
Verify services are running:
- Master: http://localhost:9333
- Filer HTTP: http://localhost:8888
- Filer gRPC: localhost:18888
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 totrueto 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
-
Copy the built JAR to your Hadoop classpath:
cp target/seaweedfs-hadoop3-client-*.jar $HADOOP_HOME/share/hadoop/common/lib/ -
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> -
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:
-
Configuration Tests Only (default, no SeaweedFS required):
mvn test -Dtest=SeaweedFileSystemConfigTest -
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:
- Configuration tests (no SeaweedFS required)
- Integration tests (with SEAWEEDFS_TEST_ENABLED guard)
- Documentation updates