Files
upstream-seaweedfs/weed/filer/foundationdb/CONFIGURATION.md
upstream-mirror 89f5be005d
Some checks failed
go: build dev binaries / cleanup (push) Successful in 4s
Codespell / Check for spelling errors (push) Failing after 3s
docker: build dev containers / build-rust-binaries (amd64, x86_64-unknown-linux-musl) (push) Failing after 3s
docker: build dev containers / build-dev-containers (push) Has been skipped
docker: build dev containers / build-rust-binaries (arm64, true, aarch64-unknown-linux-musl) (push) Failing after 3s
End to End / FUSE Mount (push) Failing after 2s
EC Integration Tests / EC Integration Tests (push) Failing after 30s
FUSE DLM Integration Tests / FUSE DLM Integration Tests (push) Failing after 3s
FUSE Volume Server Failover Tests / FUSE Volume Server Failover (push) Failing after 3s
FUSE Integration Tests / FUSE Integration Testing (push) Failing after 2s
FUSE P2P Peer Chunk Sharing Integration Tests / FUSE P2P Peer Chunk Sharing (push) Failing after 3s
go: build binary / Go Vet (push) Failing after 2s
go: build binary / Go Vet 32-bit (push) Failing after 3s
go: build binary / Build (push) Failing after 4s
go: build binary / Build cross-platform (freebsd/amd64) (push) Failing after 3s
go: build binary / Build cross-platform (windows/amd64) (push) Failing after 3s
go: build binary / Build cross-platform (darwin/arm64) (push) Failing after 3s
go: build binary / Build cross-platform (windows/arm64) (push) Failing after 3s
go: build binary / Test (push) Failing after 2s
go: build binary / Test 32-bit (push) Failing after 3s
helm: lint and test charts / lint-test (push) Failing after 3s
Java Client Integration Tests / Java Integration Tests (11) (push) Failing after 3s
Java Client Integration Tests / Java Integration Tests (17) (push) Failing after 3s
go: build dev binaries / build_dev_linux_windows (amd64, linux) (push) Failing after 2s
go: build dev binaries / build_dev_linux_windows (amd64, windows) (push) Failing after 3s
go: build dev binaries / build_dev_darwin (amd64, darwin) (push) Has been skipped
go: build dev binaries / build_dev_darwin (arm64, darwin) (push) Has been skipped
Java Client Unit Tests / Java Unit Tests (11) (push) Failing after 3s
Java Client Unit Tests / Java Unit Tests (17) (push) Failing after 2s
Java Client Unit Tests / Java Unit Tests (21) (push) Failing after 3s
Java Client Unit Tests / Java Unit Tests (8) (push) Failing after 3s
Kafka Quick Test (Load Test with Schema Registry) / Kafka Client Load Test (Quick) (push) Failing after 3s
Kafka Gateway Tests / Kafka Unit Tests (unit-tests-1) (push) Failing after 11s
Kafka Gateway Tests / Kafka Integration Tests (Critical) (integration-1) (push) Failing after 3s
Kafka Gateway Tests / Kafka End-to-End Tests (with SMQ) (e2e-1) (push) Failing after 3s
Kafka Gateway Tests / Kafka Consumer Group Tests (Highly Isolated) (consumer-group-1) (push) Failing after 2s
Kafka Gateway Tests / Kafka Client Compatibility (with SMQ) (client-compat-1) (push) Failing after 3s
Kafka Gateway Tests / Kafka SMQ Integration Tests (Full Stack) (smq-integration-1) (push) Failing after 3s
Kafka Gateway Tests / Kafka Protocol Tests (Isolated) (protocol-1) (push) Failing after 3s
KMS Tests / KMS Provider Integration Tests (push) Failing after 2s
KMS Tests / S3 KMS End-to-End Tests (push) Failing after 3s
Master Cold Start Tests / Master Cold Start Tests (push) Failing after 4s
Metadata Subscribe Integration Tests / Metadata Subscribe Integration Tests (push) Failing after 3s
Multi-Master Tests / Multi-Master Failover Tests (push) Failing after 3s
Performance / CPU and Heap Profile (push) Failing after 2s
Performance / S3 Read/Write Benchmark (rust volume) (push) Failing after 3s
pjdfstest POSIX Compliance / pjdfstest (push) Failing after 2s
Plugin Worker Integration Tests / Plugin Worker: erasure_coding (push) Failing after 3s
Plugin Worker Integration Tests / Plugin Worker: vacuum (push) Failing after 3s
Rust Volume Server Tests / Rust Integration Tests (push) Failing after 3s
Rust Volume Server Tests / Go Tests with Rust Volume (grpc - Shard 1) (push) Failing after 2s
Rust Volume Server Tests / Go Tests with Rust Volume (http - Shard 1) (push) Failing after 3s
Rust Volume Server Tests / Go Tests with Rust Volume (grpc - Shard 2) (push) Failing after 3s
Rust Volume Server Tests / Go Tests with Rust Volume (http - Shard 2) (push) Failing after 3s
Rust Volume Server Tests / Go Tests with Rust Volume (grpc - Shard 3) (push) Failing after 3s
S3 IAM Integration Tests / S3 IAM Distributed Tests (push) Failing after 2s
S3 IAM Integration Tests / S3 IAM Performance Tests (push) Failing after 3s
Performance / Throughput Benchmark (go volume) (push) Failing after 3s
Performance / Throughput Benchmark (rust volume) (push) Failing after 2s
Performance / S3 Read/Write Benchmark (go volume) (push) Failing after 3s
Plugin Worker Integration Tests / Plugin Worker: volume_balance (push) Failing after 3s
PostgreSQL Gateway Tests / PostgreSQL Basic Tests (push) Failing after 3s
Rust Volume Server Tests / Rust Unit Tests (push) Failing after 2s
Rust Volume Server Tests / Go Tests with Rust Volume (http - Shard 3) (push) Failing after 3s
Rust Plugin Worker Tests / Rust Plugin Worker Build and Unit Tests (push) Failing after 3s
rust: build dev volume server binaries / cleanup (push) Successful in 3s
S3 ETag and ACL Tests / S3 ETag + ACL Integration Tests (push) Failing after 3s
S3 IAM Integration Tests / IAM Unit Tests (push) Failing after 3s
S3 IAM Integration Tests / S3 IAM Integration Tests (advanced) (push) Failing after 3s
S3 IAM Integration Tests / S3 IAM Integration Tests (basic) (push) Failing after 2s
S3 IAM Integration Tests / S3 IAM Integration Tests (group) (push) Failing after 3s
S3 IAM Integration Tests / S3 IAM Integration Tests (policy-enforcement) (push) Failing after 2s
S3 IAM Integration Tests / S3 IAM Integration Tests (sts) (push) Failing after 4s
S3 Keycloak Integration Tests / S3 Keycloak Integration Tests (push) Failing after 3s
S3 PyArrow Parquet Tests / PyArrow Parquet Tests (Python 3.11) (push) Failing after 3s
S3 PyArrow Parquet Tests / Go Unit Tests (Implicit Directory) (push) Has been cancelled
S3 PyArrow Parquet Tests / PyArrow Parquet Tests (Python 3.12) (push) Has been cancelled
rust: build dev volume server binaries / build-rust-volume-dev-darwin (darwin-amd64, x86_64-apple-darwin) (push) Has been cancelled
rust: build dev volume server binaries / build-rust-volume-dev-darwin (darwin-arm64, aarch64-apple-darwin) (push) Has been cancelled
rust: build dev volume server binaries / build-rust-volume-dev-linux (linux-amd64, x86_64-unknown-linux-gnu) (push) Has been cancelled
S3 PyArrow Parquet Tests / PyArrow Parquet Tests (Python 3.9) (push) Has been cancelled
S3 Proxy Signature Tests / S3 Proxy Signature Verification Tests (push) Has been cancelled
S3 SDK V2 Route Disambiguation Tests / S3 SDK V2 Routing Tests (push) Has been cancelled
S3 SSE Tests / S3 SSE Integration Tests (quick) (push) Has been cancelled
S3 SSE Tests / S3 SSE Compatibility Test (push) Has been cancelled
S3 SSE Tests / S3 SSE Metadata Persistence Test (push) Has been cancelled
S3 SSE Tests / S3 SSE Copy Operations Test (push) Has been cancelled
S3 SSE Tests / S3 SSE Multipart Upload Test (push) Has been cancelled
S3 SSE Tests / S3 SSE Performance Test (push) Has been cancelled
S3 SSE Tests / S3 Volume Encryption Test (push) Has been cancelled
S3 SSE Tests / S3 SSE Integration Tests (comprehensive) (push) Has been cancelled
Spark Integration Tests / Spark Integration Tests (push) Failing after 2s
Telemetry Integration Tests / telemetry-integration-test (push) Failing after 3s
Volume Server Integration Tests / Volume Server Integration Tests (grpc - Shard 1) (push) Failing after 3s
Volume Server Integration Tests / Volume Server Integration Tests (http - Shard 1) (push) Failing after 3s
Ceph S3 tests / Basic S3 tests (KV store) (push) Has been cancelled
Ceph S3 tests / S3 Versioning & Object Lock tests (push) Has been cancelled
S3 Policy Integration Tests / S3 Policy Unit Tests (push) Failing after 3s
Ceph S3 tests / S3 CORS tests (push) Has been cancelled
Ceph S3 tests / SeaweedFS Custom S3 Copy tests (push) Has been cancelled
Ceph S3 tests / Basic S3 tests (SQL store) (push) Has been cancelled
test s3 over https using aws-cli / awscli-tests (push) Has been cancelled
TLS Rotation Integration Tests / TLS Rotation Integration Tests (push) Has been cancelled
Vacuum Integration Tests / Vacuum Integration Tests (push) Has been cancelled
S3 Policy Integration Tests / S3 Policy Variables Integration Tests (push) Failing after 3s
S3 Policy Integration Tests / S3 Policy Enforcement Integration Tests (advanced-policy) (push) Failing after 3s
S3 Policy Integration Tests / S3 Policy Enforcement Integration Tests (basic-policy) (push) Failing after 2s
S3 Policy Integration Tests / S3 Policy Enforcement Integration Tests (contextual-policy) (push) Failing after 4s
S3 Policy Integration Tests / Trusted Proxy Detection Tests (push) Failing after 2s
S3 Policy Integration Tests / S3 Policy Shell Integration Tests (push) Failing after 3s
Samba on FUSE Integration / samba-integration (push) Failing after 2s
SFTP Integration Tests / SFTP Integration Testing (push) Failing after 3s
terraform: validate and test modules / fmt, validate, plan-level tests (push) Failing after 3s
terraform: validate and test modules / local cluster smoke test (real weed) (push) Failing after 2s
Volume Server Integration Tests / Volume Server Integration Tests (grpc - Shard 3) (push) Failing after 3s
Volume Server Integration Tests / Volume Server Integration Tests (grpc - Shard 2) (push) Failing after 3s
Volume Server Integration Tests / Volume Server Integration Tests (http - Shard 2) (push) Failing after 2s
Volume Server Integration Tests / Volume Server Integration Tests (http - Shard 3) (push) Failing after 5s
mount: windows conformance / WinFsp conformance (push) Has been cancelled
mount: windows / Mount on Windows (push) Has been cancelled
snapshot of https://github.com/seaweedfs/seaweedfs.git (master, 2026-09-16; blobs > 131072B omitted)
2026-09-16 03:23:42 +05:00

424 lines
10 KiB
Markdown

# FoundationDB Filer Store Configuration Reference
This document provides comprehensive configuration options for the FoundationDB filer store.
## Configuration Methods
### 1. Configuration File (filer.toml)
```toml
[foundationdb]
enabled = true
cluster_file = "/etc/foundationdb/fdb.cluster"
api_version = 740
timeout = "5s"
max_retry_delay = "1s"
directory_prefix = "seaweedfs"
```
### 2. Environment Variables
All configuration options can be set via environment variables with the `WEED_FOUNDATIONDB_` prefix:
```bash
export WEED_FOUNDATIONDB_ENABLED=true
export WEED_FOUNDATIONDB_CLUSTER_FILE=/etc/foundationdb/fdb.cluster
export WEED_FOUNDATIONDB_API_VERSION=740
export WEED_FOUNDATIONDB_TIMEOUT=5s
export WEED_FOUNDATIONDB_MAX_RETRY_DELAY=1s
export WEED_FOUNDATIONDB_DIRECTORY_PREFIX=seaweedfs
# Write batching (disabled by default)
export WEED_FOUNDATIONDB_BATCH_ENABLED=false
export WEED_FOUNDATIONDB_BATCH_SIZE=100
export WEED_FOUNDATIONDB_BATCH_INTERVAL=1ms
```
### 3. Command Line Arguments
While not directly supported, configuration can be specified via config files passed to the `weed` command.
## Configuration Options
### Basic Options
| Option | Type | Default | Description |
|--------|------|---------|-------------|
| `enabled` | boolean | `false` | Enable the FoundationDB filer store |
| `cluster_file` | string | `/etc/foundationdb/fdb.cluster` | Path to FoundationDB cluster file |
| `api_version` | integer | `740` | FoundationDB API version to use |
### Connection Options
| Option | Type | Default | Description |
|--------|------|---------|-------------|
| `timeout` | duration | `5s` | Transaction timeout duration |
| `max_retry_delay` | duration | `1s` | Maximum delay between retries |
### Storage Options
| Option | Type | Default | Description |
|--------|------|---------|-------------|
| `directory_prefix` | string | `seaweedfs` | Directory prefix for key organization |
### Write Batching Options
| Option | Type | Default | Description |
|--------|------|---------|-------------|
| `batch_enabled` | boolean | `false` | Enable write batching. Disabled by default for optimal S3 PUT latency. |
| `batch_size` | integer | `100` | Maximum number of operations per batch (when batching is enabled) |
| `batch_interval` | duration | `1ms` | Maximum time to wait before flushing a batch (when batching is enabled) |
**Note:** Write batching is **disabled by default**. Each write commits immediately in its own
transaction, providing optimal latency for S3 PUT operations. Enable batching only for
high-throughput bulk ingestion workloads where you can tolerate slightly higher per-operation latency.
## Configuration Examples
### Development Environment
```toml
[foundationdb]
enabled = true
cluster_file = "/var/fdb/config/fdb.cluster"
api_version = 740
timeout = "10s"
max_retry_delay = "2s"
directory_prefix = "seaweedfs_dev"
```
### Production Environment
```toml
[foundationdb]
enabled = true
cluster_file = "/etc/foundationdb/fdb.cluster"
api_version = 740
timeout = "30s"
max_retry_delay = "5s"
directory_prefix = "seaweedfs_prod"
```
### High-Performance Setup (Low Latency)
For S3 workloads requiring low latency per operation:
```toml
[foundationdb]
enabled = true
cluster_file = "/etc/foundationdb/fdb.cluster"
api_version = 740
timeout = "5s"
max_retry_delay = "1s"
directory_prefix = "sw" # Shorter prefix for efficiency
# Batching disabled (default) for optimal per-operation latency
batch_enabled = false
```
### High-Throughput Bulk Ingestion
For bulk data loading where throughput matters more than per-operation latency:
```toml
[foundationdb]
enabled = true
cluster_file = "/etc/foundationdb/fdb.cluster"
api_version = 740
timeout = "60s"
max_retry_delay = "10s"
directory_prefix = "sw"
# Enable batching for higher throughput
batch_enabled = true
batch_size = 100
batch_interval = "1ms"
```
### Path-Specific Configuration
Configure different FoundationDB settings for different paths:
```toml
# Default configuration
[foundationdb]
enabled = true
cluster_file = "/etc/foundationdb/fdb.cluster"
directory_prefix = "seaweedfs_main"
# Backup path with different prefix
[foundationdb.backup]
enabled = true
cluster_file = "/etc/foundationdb/fdb.cluster"
directory_prefix = "seaweedfs_backup"
location = "/backup"
timeout = "120s"
# Archive path with extended timeouts
[foundationdb.archive]
enabled = true
cluster_file = "/etc/foundationdb/fdb.cluster"
directory_prefix = "seaweedfs_archive"
location = "/archive"
timeout = "300s"
max_retry_delay = "30s"
```
## Configuration Validation
### Required Settings
The following settings are required for FoundationDB to function:
1. `enabled = true`
2. `cluster_file` must point to a valid FoundationDB cluster file
3. `api_version` must match your FoundationDB installation
### Validation Rules
- `api_version` must be between 600 and 740
- `timeout` must be a valid duration string (e.g., "5s", "30s", "2m")
- `max_retry_delay` must be a valid duration string
- `cluster_file` must exist and be readable
- `directory_prefix` must not be empty
### Error Handling
Invalid configurations will result in startup errors:
```
FATAL: Failed to initialize store for foundationdb: invalid timeout duration
FATAL: Failed to initialize store for foundationdb: failed to open FoundationDB database
FATAL: Failed to initialize store for foundationdb: cluster file not found
```
## Performance Tuning
### Timeout Configuration
| Use Case | Timeout | Max Retry Delay | Notes |
|----------|---------|-----------------|-------|
| Interactive workloads | 5s | 1s | Fast response times |
| Batch processing | 60s | 10s | Handle large operations |
| Archive operations | 300s | 30s | Very large data sets |
### Connection Pool Settings
FoundationDB automatically manages connection pooling. No additional configuration needed.
### Directory Organization
Use meaningful directory prefixes to organize data:
```toml
# Separate environments
directory_prefix = "prod_seaweedfs" # Production
directory_prefix = "staging_seaweedfs" # Staging
directory_prefix = "dev_seaweedfs" # Development
# Separate applications
directory_prefix = "app1_seaweedfs" # Application 1
directory_prefix = "app2_seaweedfs" # Application 2
```
## Security Configuration
### Cluster File Security
Protect the FoundationDB cluster file:
```bash
# Set proper permissions
sudo chown root:seaweedfs /etc/foundationdb/fdb.cluster
sudo chmod 640 /etc/foundationdb/fdb.cluster
```
### Network Security
FoundationDB supports TLS encryption. Configure in the cluster file:
```
description:cluster_id@tls(server1:4500,server2:4500,server3:4500)
```
### Access Control
Use FoundationDB's built-in access control mechanisms when available.
## Monitoring Configuration
### Health Check Settings
Configure health check timeouts appropriately:
```toml
[foundationdb]
enabled = true
timeout = "10s" # Reasonable timeout for health checks
```
### Logging Configuration
Enable verbose logging for troubleshooting:
```bash
# Start SeaweedFS with debug logs
WEED_FOUNDATIONDB_ENABLED=true weed -v=2 server -filer
```
## Migration Configuration
### From Other Filer Stores
When migrating from other filer stores:
1. Configure both stores temporarily
2. Use path-specific configuration for gradual migration
3. Migrate data using SeaweedFS tools
```toml
# During migration - keep old store for reads
[leveldb2]
enabled = true
dir = "/old/filer/data"
# New writes go to FoundationDB
[foundationdb.migration]
enabled = true
location = "/new"
cluster_file = "/etc/foundationdb/fdb.cluster"
```
## Backup Configuration
### Metadata Backup Strategy
```toml
# Main storage
[foundationdb]
enabled = true
directory_prefix = "seaweedfs_main"
# Backup storage (different cluster recommended)
[foundationdb.backup]
enabled = true
cluster_file = "/etc/foundationdb/backup_fdb.cluster"
directory_prefix = "seaweedfs_backup"
location = "/backup"
```
## Container Configuration
### Docker Environment Variables
```bash
# Docker environment
WEED_FOUNDATIONDB_ENABLED=true
WEED_FOUNDATIONDB_CLUSTER_FILE=/var/fdb/config/fdb.cluster
WEED_FOUNDATIONDB_API_VERSION=740
```
### Kubernetes ConfigMap
```yaml
apiVersion: v1
kind: ConfigMap
metadata:
name: seaweedfs-config
data:
filer.toml: |
[foundationdb]
enabled = true
cluster_file = "/var/fdb/config/cluster_file"
api_version = 740
timeout = "30s"
max_retry_delay = "5s"
directory_prefix = "k8s_seaweedfs"
```
## Troubleshooting Configuration
### Debug Configuration
```toml
[foundationdb]
enabled = true
cluster_file = "/etc/foundationdb/fdb.cluster"
timeout = "60s" # Longer timeouts for debugging
max_retry_delay = "10s"
directory_prefix = "debug_seaweedfs"
```
### Test Configuration
```toml
[foundationdb]
enabled = true
cluster_file = "/tmp/fdb.cluster" # Test cluster
timeout = "5s"
directory_prefix = "test_seaweedfs"
```
## Configuration Best Practices
### 1. Environment Separation
Use different directory prefixes for different environments:
- Production: `prod_seaweedfs`
- Staging: `staging_seaweedfs`
- Development: `dev_seaweedfs`
### 2. Timeout Settings
- Interactive: 5-10 seconds
- Batch: 30-60 seconds
- Archive: 120-300 seconds
### 3. Cluster File Management
- Use absolute paths for cluster files
- Ensure proper file permissions
- Keep backup copies of cluster files
### 4. Directory Naming
- Use descriptive prefixes
- Include environment/application identifiers
- Keep prefixes reasonably short for efficiency
### 5. Error Handling
- Configure appropriate timeouts
- Monitor retry patterns
- Set up alerting for configuration errors
## Configuration Testing
### Validation Script
```bash
#!/bin/bash
# Test FoundationDB configuration
# Check cluster file
if [ ! -f "$WEED_FOUNDATIONDB_CLUSTER_FILE" ]; then
echo "ERROR: Cluster file not found: $WEED_FOUNDATIONDB_CLUSTER_FILE"
exit 1
fi
# Test connection
fdbcli -C "$WEED_FOUNDATIONDB_CLUSTER_FILE" --exec 'status' > /dev/null
if [ $? -ne 0 ]; then
echo "ERROR: Cannot connect to FoundationDB cluster"
exit 1
fi
echo "Configuration validation passed"
```
### Integration Testing
```bash
# Test configuration with SeaweedFS
cd test/foundationdb
make check-env
make test-unit
```