DeBros/network
4
0
Fork 0
mirror of https://github.com/DeBrosOfficial/network.git synced 2025-12-13 00:38:49 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update configuration management and documentation. Added commands for generating and validating configuration files in the CLI, ensuring strict YAML decoding and improved error handling. Updated README to reflect new configuration paths and filesystem permissions. Enhanced Makefile to streamline node and gateway startup processes.

Browse Source
This commit is contained in:
anonpenguin23 2025-10-24 11:03:50 +03:00
parent 8a9deb50ec
commit 0717296822
No known key found for this signature in database
GPG Key ID: 1CBB1FE35AFBEE30
10 changed files with 1029 additions and 352 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
CHANGELOG.md
View File

@ -14,6 +14,7 @@ The format is based on [Keep a Changelog][keepachangelog] and adheres to [Semant
### Changed ### Changed
- Updated readme - Updated readme
- Where we read .yaml files from and where data is saved to ~/.debros
### Deprecated ### Deprecated

94
Makefile
View File

@ -21,7 +21,7 @@ test-e2e:
.PHONY: build clean test run-node run-node2 run-node3 run-example deps tidy fmt vet lint clear-ports .PHONY: build clean test run-node run-node2 run-node3 run-example deps tidy fmt vet lint clear-ports
VERSION := 0.51.4-beta VERSION := 0.51.5-beta
COMMIT ?= $(shell git rev-parse --short HEAD 2>/dev/null || echo unknown) COMMIT ?= $(shell git rev-parse --short HEAD 2>/dev/null || echo unknown)
DATE ?= $(shell date -u +%Y-%m-%dT%H:%M:%SZ) DATE ?= $(shell date -u +%Y-%m-%dT%H:%M:%SZ)
LDFLAGS := -X 'main.version=$(VERSION)' -X 'main.commit=$(COMMIT)' -X 'main.date=$(DATE)' LDFLAGS := -X 'main.version=$(VERSION)' -X 'main.commit=$(COMMIT)' -X 'main.date=$(DATE)'
@ -46,35 +46,35 @@ clean:
# Run bootstrap node (auto-selects identity and data dir) # Run bootstrap node (auto-selects identity and data dir)
run-node: run-node:
@echo "Starting bootstrap node with config..." @echo "Starting bootstrap node..."
go run ./cmd/node --config configs/bootstrap.yaml @echo "Config: ~/.debros/bootstrap.yaml"
@echo "Generate it with: network-cli config init --type bootstrap"
go run ./cmd/node --config node.yaml
# Run second node (regular) - requires join address of bootstrap node # Run second node (regular) - requires join address of bootstrap node
# Usage: make run-node2 JOINADDR=/ip4/127.0.0.1/tcp/5001 HTTP=5002 RAFT=7002 P2P=4002 # Usage: make run-node2 JOINADDR=/ip4/127.0.0.1/tcp/5001 HTTP=5002 RAFT=7002 P2P=4002
run-node2: run-node2:
@echo "Starting regular node2 with config..." @echo "Starting regular node (node.yaml)..."
go run ./cmd/node --config configs/node.yaml @echo "Config: ~/.debros/node.yaml"
@echo "Generate it with: network-cli config init --type node --join localhost:5001 --bootstrap-peers '<peer_multiaddr>'"
go run ./cmd/node --config node2.yaml
# Run third node (regular) - requires join address of bootstrap node # Run third node (regular) - requires join address of bootstrap node
# Usage: make run-node3 JOINADDR=/ip4/127.0.0.1/tcp/5001 HTTP=5003 RAFT=7003 P2P=4003 # Usage: make run-node3 JOINADDR=/ip4/127.0.0.1/tcp/5001 HTTP=5003 RAFT=7003 P2P=4003
run-node3: run-node3:
@echo "Starting regular node3 with config..." @echo "Starting regular node (node2.yaml)..."
go run ./cmd/node --config configs/node3.yaml @echo "Config: ~/.debros/node2.yaml"
@echo "Generate it with: network-cli config init --type node --name node2.yaml --join localhost:5001 --bootstrap-peers '<peer_multiaddr>'"
go run ./cmd/node --config node3.yaml
# Run gateway HTTP server # Run gateway HTTP server
# Usage examples: # Usage examples:
# make run-gateway # uses defaults (:8080, namespace=default) # make run-gateway # uses ~/.debros/gateway.yaml
# GATEWAY_ADDR=":8081" make run-gateway # override listen addr via env # Config generated with: network-cli config init --type gateway
# GATEWAY_NAMESPACE=myapp make run-gateway # set namespace
# GATEWAY_BOOTSTRAP_PEERS="/ip4/127.0.0.1/tcp/4001/p2p/<ID>" make run-gateway
# GATEWAY_REQUIRE_AUTH=1 GATEWAY_API_KEYS="key1:ns1,key2:ns2" make run-gateway
run-gateway: run-gateway:
@echo "Starting gateway HTTP server..." @echo "Starting gateway HTTP server..."
GATEWAY_ADDR=$(or $(ADDR),$(GATEWAY_ADDR)) \ @echo "Note: Config must be in ~/.debros/gateway.yaml"
GATEWAY_NAMESPACE=$(or $(NAMESPACE),$(GATEWAY_NAMESPACE)) \ @echo "Generate it with: network-cli config init --type gateway"
GATEWAY_BOOTSTRAP_PEERS=$(GATEWAY_BOOTSTRAP_PEERS) \
GATEWAY_REQUIRE_AUTH=$(GATEWAY_REQUIRE_AUTH) \
GATEWAY_API_KEYS=$(GATEWAY_API_KEYS) \
go run ./cmd/gateway go run ./cmd/gateway
# Run basic usage example # Run basic usage example
@ -155,15 +155,28 @@ dev-setup: deps
# Start development cluster (requires multiple terminals) # Start development cluster (requires multiple terminals)
dev-cluster: dev-cluster:
@echo "To start a development cluster, run these commands in separate terminals:" @echo "To start a development cluster with 3 nodes:"
@echo "1. make run-node # Start bootstrap node (uses configs/bootstrap.yaml)" @echo ""
@echo "2. make run-node2 # Start second node (uses configs/node.yaml)" @echo "1. Generate config files in ~/.debros:"
@echo "3. make run-node3 # Start third node (uses configs/node.yaml)" @echo " make build"
@echo "4. make run-example # Test basic functionality" @echo " ./bin/network-cli config init --type bootstrap"
@echo "5. make cli-health # Check network health" @echo " ./bin/network-cli config init --type node --name node.yaml --bootstrap-peers '<bootstrap_peer_multiaddr>'"
@echo "6. make cli-peers # List peers" @echo " ./bin/network-cli config init --type node --name node2.yaml --bootstrap-peers '<bootstrap_peer_multiaddr>'"
@echo "7. make cli-storage-test # Test storage" @echo ""
@echo "8. make cli-pubsub-test # Test messaging" @echo "2. Run in separate terminals:"
@echo " Terminal 1: make run-node # Start bootstrap node (bootstrap.yaml)"
@echo " Terminal 2: make run-node2 # Start node 1 (node.yaml)"
@echo " Terminal 3: make run-node3 # Start node 2 (node2.yaml)"
@echo " Terminal 4: make run-gateway # Start gateway"
@echo ""
@echo "3. Or run custom node with any config file:"
@echo " go run ./cmd/node --config custom-node.yaml"
@echo ""
@echo "4. Test:"
@echo " make cli-health # Check network health"
@echo " make cli-peers # List peers"
@echo " make cli-storage-test # Test storage"
@echo " make cli-pubsub-test # Test messaging"
# Full development workflow # Full development workflow
dev: clean build test dev: clean build test
@ -175,22 +188,43 @@ help:
@echo " build - Build all executables" @echo " build - Build all executables"
@echo " clean - Clean build artifacts" @echo " clean - Clean build artifacts"
@echo " test - Run tests" @echo " test - Run tests"
@echo ""
@echo "Configuration (NEW):"
@echo " First, generate config files in ~/.debros with:"
@echo " make build # Build CLI first"
@echo " ./bin/network-cli config init --type bootstrap # Generate bootstrap config"
@echo " ./bin/network-cli config init --type node --bootstrap-peers '<peer_multiaddr>'"
@echo " ./bin/network-cli config init --type gateway"
@echo ""
@echo "Network Targets (requires config files in ~/.debros):"
@echo " run-node - Start bootstrap node" @echo " run-node - Start bootstrap node"
@echo " run-node2 - Start second node (requires JOINADDR, optional HTTP/RAFT/P2P)" @echo " run-node2 - Start second node"
@echo " run-node3 - Start third node (requires JOINADDR, optional HTTP/RAFT/P2P)" @echo " run-node3 - Start third node"
@echo " run-gateway - Start HTTP gateway (flags via env: GATEWAY_ADDR, GATEWAY_NAMESPACE, GATEWAY_BOOTSTRAP_PEERS, GATEWAY_REQUIRE_AUTH, GATEWAY_API_KEYS)" @echo " run-gateway - Start HTTP gateway"
@echo " run-example - Run usage example" @echo " run-example - Run usage example"
@echo ""
@echo "Running Multiple Nodes:"
@echo " Nodes use --config flag to select which YAML file in ~/.debros to load:"
@echo " go run ./cmd/node --config bootstrap.yaml"
@echo " go run ./cmd/node --config node.yaml"
@echo " go run ./cmd/node --config node2.yaml"
@echo " Generate configs with: ./bin/network-cli config init --name <filename.yaml>"
@echo ""
@echo "CLI Commands:"
@echo " run-cli - Run network CLI help" @echo " run-cli - Run network CLI help"
@echo " show-bootstrap - Show example bootstrap usage with flags"
@echo " cli-health - Check network health" @echo " cli-health - Check network health"
@echo " cli-peers - List network peers" @echo " cli-peers - List network peers"
@echo " cli-status - Get network status" @echo " cli-status - Get network status"
@echo " cli-storage-test - Test storage operations" @echo " cli-storage-test - Test storage operations"
@echo " cli-pubsub-test - Test pub/sub operations" @echo " cli-pubsub-test - Test pub/sub operations"
@echo ""
@echo "Development:"
@echo " test-multinode - Full multi-node test with 1 bootstrap + 2 nodes" @echo " test-multinode - Full multi-node test with 1 bootstrap + 2 nodes"
@echo " test-peer-discovery - Test peer discovery (requires running nodes)" @echo " test-peer-discovery - Test peer discovery (requires running nodes)"
@echo " test-replication - Test data replication (requires running nodes)" @echo " test-replication - Test data replication (requires running nodes)"
@echo " test-consensus - Test database consensus (requires running nodes)" @echo " test-consensus - Test database consensus (requires running nodes)"
@echo ""
@echo "Maintenance:"
@echo " deps - Download dependencies" @echo " deps - Download dependencies"
@echo " tidy - Tidy dependencies" @echo " tidy - Tidy dependencies"
@echo " fmt - Format code" @echo " fmt - Format code"

552
README.md
View File

@ -88,6 +88,75 @@ A robust, decentralized peer-to-peer network built in Go, providing distributed
- **5001:** RQLite HTTP API - **5001:** RQLite HTTP API
- **7001:** RQLite Raft consensus - **7001:** RQLite Raft consensus
### Filesystem Permissions
DeBros Network stores all configuration and data in `~/.debros/` directory. Ensure you have:
- **Read/Write access** to your home directory (`~`)
- **Available disk space**: At least 10GB for database and logs
- **No restrictive mount options**: The home directory must not be mounted read-only
- **Unix permissions**: Standard user permissions are sufficient (no root/sudo required)
#### Directory Structure
DeBros automatically creates the following directory structure:
```
~/.debros/
├── bootstrap.yaml # Bootstrap node config
├── node.yaml # Node config
├── gateway.yaml # Gateway config
├── bootstrap/ # Bootstrap node data (auto-created)
│ ├── rqlite/ # RQLite database files
│ │ ├── db.sqlite # Main database
│ │ ├── raft/ # Raft consensus data
│ │ └── rsnapshots/ # Raft snapshots
│ ├── peer.info # Node multiaddr (created at startup)
│ └── identity.key # Node private key (created at startup)
├── node/ # Node data (auto-created)
│ ├── rqlite/ # RQLite database files
│ ├── raft/ # Raft data
│ ├── peer.info # Node multiaddr (created at startup)
│ └── identity.key # Node private key (created at startup)
└── node2/ # Additional node configs (if running multiple)
└── rqlite/ # RQLite database files
```
**Files Created at Startup:**
- `identity.key` - LibP2P private key for the node (generated once, reused)
- `peer.info` - The node's multiaddr (e.g., `/ip4/0.0.0.0/tcp/4001/p2p/12D3KooW...`)
**Automatic Creation**: The node automatically creates all necessary data directories when started. You only need to ensure:
1. `~/.debros/` is writable
2. Sufficient disk space available
3. Correct config files exist
**Permission Check:**
```bash
# Verify home directory is writable
touch ~/test-write && rm ~/test-write && echo "✓ Home directory is writable"
# Check available disk space
df -h ~
```
**If you get permission errors:**
```
Error: Failed to create/access config directory
Please ensure:
1. Home directory is accessible
2. You have write permissions to home directory
3. Disk space is available
```
**Solution:**
- Ensure you're not running with overly restrictive umask: `umask` should show `0022` or similar
- Check home directory permissions: `ls -ld ~` should show your user as owner
- For sandboxed/containerized environments: Ensure `/home/<user>` is writable
--- ---
## Quick Start ## Quick Start
@ -110,7 +179,7 @@ make build
```bash ```bash
make run-node make run-node
# Or manually: # Or manually:
go run ./cmd/node --config configs/bootstrap.yaml go run ./cmd/node --config configs/node.yaml
``` ```
### 4. Start Additional Nodes ### 4. Start Additional Nodes
@ -178,212 +247,141 @@ sudo journalctl -u debros-node.service -f
## Configuration ## Configuration
### Example Configuration Files ### Configuration Files Location
#### `configs/bootstrap.yaml` All configuration files are stored in `~/.debros/` for both local development and production deployments:
```yaml - `~/.debros/node.yaml` - Node configuration
node: - `~/.debros/node.yaml` - Bootstrap node configuration
id: "" - `~/.debros/gateway.yaml` - Gateway configuration
type: "bootstrap"
listen_addresses:
- "/ip4/0.0.0.0/tcp/4001"
data_dir: "./data/bootstrap"
max_connections: 100
database: The system will **only** load config from `~/.debros/` and will error if required config files are missing.
data_dir: "./data/bootstrap/rqlite"
replication_factor: 3
shard_count: 16
max_database_size: 1073741824
backup_interval: 24h
rqlite_port: 5001
rqlite_raft_port: 7001
rqlite_join_address: ""
discovery: ### Generating Configuration Files
bootstrap_peers: []
discovery_interval: 15s
bootstrap_port: 4001
http_adv_address: "127.0.0.1"
raft_adv_address: ""
security: Use the `network-cli config init` command to generate configuration files:
enable_tls: false
private_key_file: ""
certificate_file: ""
logging: #### Generate a Node Config
level: "info"
format: "console" ```bash
output_file: "" # Generate basic node config with bootstrap peers
network-cli config init --type node --bootstrap-peers "/ip4/127.0.0.1/tcp/4001/p2p/QmXxx,/ip4/127.0.0.1/tcp/4002/p2p/QmYyy"
# With custom ports
network-cli config init --type node --name node2.yaml --listen-port 4002 --rqlite-http-port 5002 --rqlite-raft-port 7002 --join localhost:5001 --bootstrap-peers "/ip4/127.0.0.1/tcp/4001/p2p/QmXxx"
# Force overwrite existing config
network-cli config init --type node --force
``` ```
#### `configs/node.yaml` #### Generate a Bootstrap Node Config
```yaml ```bash
node: # Generate bootstrap node (no join address required)
id: "node2" network-cli config init --type bootstrap
type: "node"
listen_addresses:
- "/ip4/0.0.0.0/tcp/4002"
data_dir: "./data/node2"
max_connections: 50
database: # With custom ports
data_dir: "./data/node2/rqlite" network-cli config init --type bootstrap --listen-port 4001 --rqlite-http-port 5001 --rqlite-raft-port 7001
replication_factor: 3
shard_count: 16
max_database_size: 1073741824
backup_interval: 24h
rqlite_port: 5002
rqlite_raft_port: 7002
rqlite_join_address: "127.0.0.1:7001"
discovery:
bootstrap_peers:
- "/ip4/127.0.0.1/tcp/4001/p2p/<YOUR_BOOTSTRAP_PEER_ID>"
discovery_interval: 15s
bootstrap_port: 4002
http_adv_address: "127.0.0.1"
raft_adv_address: ""
security:
enable_tls: false
private_key_file: ""
certificate_file: ""
logging:
level: "info"
format: "console"
output_file: ""
``` ```
### YAML Reference #### Generate a Gateway Config
#### Node YAML (configs/node.yaml or configs/bootstrap.yaml) ```bash
# Generate gateway config
network-cli config init --type gateway
The .yaml files are required in order for the nodes and the gateway to run correctly. # With bootstrap peers
network-cli config init --type gateway --bootstrap-peers "/ip4/127.0.0.1/tcp/4001/p2p/QmXxx"
node:
- id (string) Optional node ID. Auto-generated if empty.
- type (string) "bootstrap" or "node". Default: "node".
- listen_addresses (string[]) LibP2P listen multiaddrs. Default: ["/ip4/0.0.0.0/tcp/4001"].
- data_dir (string) Data directory. Default: "./data".
- max_connections (int) Max peer connections. Default: 50.
database:
- data_dir (string) Directory for database files. Default: "./data/db".
- replication_factor (int) Number of replicas. Default: 3.
- shard_count (int) Shards for data distribution. Default: 16.
- max_database_size (int64 bytes) Max DB size. Default: 1073741824 (1GB).
- backup_interval (duration) e.g., "24h". Default: 24h.
- rqlite_port (int) RQLite HTTP API port. Default: 5001.
- rqlite_raft_port (int) RQLite Raft port. Default: 7001.
- rqlite_join_address (string) Raft address of an existing RQLite node to join (host:port format). Empty for bootstrap.
discovery:
- bootstrap_peers (string[]) List of LibP2P multiaddrs of bootstrap peers.
- discovery_interval (duration) How often to announce/discover peers. Default: 15s.
- bootstrap_port (int) Default port for bootstrap nodes. Default: 4001.
- http_adv_address (string) Advertised HTTP address for RQLite (host:port).
- raft_adv_address (string) Advertised Raft address (host:port).
- node_namespace (string) Namespace for node identifiers. Default: "default".
security:
- enable_tls (bool) Enable TLS for externally exposed services. Default: false.
- private_key_file (string) Path to TLS private key (if TLS enabled).
- certificate_file (string) Path to TLS certificate (if TLS enabled).
logging:
- level (string) one of "debug", "info", "warn", "error". Default: "info".
- format (string) "json" or "console". Default: "console".
- output_file (string) Empty for stdout; otherwise path to log file.
Precedence (node): Flags > YAML > Defaults.
Example node.yaml
```yaml
node:
id: "node2"
type: "node"
listen_addresses:
- "/ip4/0.0.0.0/tcp/4002"
data_dir: "./data/node2"
max_connections: 50
database:
data_dir: "./data/node2/rqlite"
replication_factor: 3
shard_count: 16
max_database_size: 1073741824
backup_interval: 24h
rqlite_port: 5002
rqlite_raft_port: 7002
rqlite_join_address: "127.0.0.1:7001"
discovery:
bootstrap_peers:
- "/ip4/127.0.0.1/tcp/4001/p2p/<YOUR_BOOTSTRAP_PEER_ID>"
discovery_interval: 15s
bootstrap_port: 4001
http_adv_address: "127.0.0.1"
raft_adv_address: ""
node_namespace: "default"
security:
enable_tls: false
private_key_file: ""
certificate_file: ""
logging:
level: "info"
format: "console"
output_file: ""
``` ```
#### Gateway YAML (configs/gateway.yaml) ### Running Multiple Nodes on the Same Machine
- listen_addr (string) HTTP listen address, e.g., ":6001". Default: ":6001". You can run multiple nodes on a single machine by creating separate configuration files and using the `--config` flag:
- client_namespace (string) Namespace used by the gateway client. Default: "default".
- bootstrap_peers (string[]) List of bootstrap peer multiaddrs. Default: empty.
Precedence (gateway): Flags > Environment Variables > YAML > Defaults. #### Create Multiple Node Configs
Environment variables:
- GATEWAY_ADDR ```bash
- GATEWAY_NAMESPACE # Node 1
- GATEWAY_BOOTSTRAP_PEERS (comma-separated) ./bin/network-cli config init --type node --name node1.yaml \
--listen-port 4001 --rqlite-http-port 5001 --rqlite-raft-port 7001 \
--bootstrap-peers "/ip4/127.0.0.1/tcp/4001/p2p/<BOOTSTRAP_ID>"
Example gateway.yaml # Node 2
./bin/network-cli config init --type node --name node2.yaml \
--listen-port 4002 --rqlite-http-port 5002 --rqlite-raft-port 7002 \
--join localhost:5001 \
--bootstrap-peers "/ip4/127.0.0.1/tcp/4001/p2p/<BOOTSTRAP_ID>"
```yaml # Node 3
listen_addr: ":6001" ./bin/network-cli config init --type node --name node3.yaml \
client_namespace: "default" --listen-port 4003 --rqlite-http-port 5003 --rqlite-raft-port 7003 \
bootstrap_peers: --join localhost:5001 \
- "<YOUR_BOOTSTRAP_PEER_ID_MULTIADDR>" --bootstrap-peers "/ip4/127.0.0.1/tcp/4001/p2p/<BOOTSTRAP_ID>"
``` ```
### Flags & Environment Variables #### Run Multiple Nodes in Separate Terminals
- **Flags**: Override config at startup (`--data`, `--p2p-port`, `--rqlite-http-port`, etc.) ```bash
- **Env Vars**: Override config and flags (`NODE_ID`, `RQLITE_PORT`, `BOOTSTRAP_PEERS`, etc.) # Terminal 1 - Bootstrap node
- **Precedence (gateway)**: Flags > Env Vars > YAML > Defaults go run ./cmd/node --config bootstrap.yaml
- **Precedence (node)**: Flags > YAML > Defaults
### Bootstrap & Database Endpoints # Terminal 2 - Node 1
go run ./cmd/node --config node1.yaml
- **Bootstrap peers**: Set in config or via `BOOTSTRAP_PEERS` env var. # Terminal 3 - Node 2
- **Database endpoints**: Set in config or via `RQLITE_NODES` env var. go run ./cmd/node --config node2.yaml
- **Development mode**: Use `NETWORK_DEV_LOCAL=1` for localhost defaults.
### Configuration Validation # Terminal 4 - Node 3
go run ./cmd/node --config node3.yaml
```
#### Or Use Makefile Targets
```bash
# Terminal 1
make run-node # Runs: go run ./cmd/node --config bootstrap.yaml
# Terminal 2
make run-node2 # Runs: go run ./cmd/node --config node.yaml
# Terminal 3
make run-node3 # Runs: go run ./cmd/node --config node2.yaml
```
#### Key Points for Multiple Nodes
- **Each node needs unique ports**: P2P port, RQLite HTTP port, and RQLite Raft port must all be different
- **Join address**: Non-bootstrap nodes need `rqlite_join_address` pointing to the bootstrap or an existing node
- **Bootstrap peers**: All nodes need the bootstrap node's multiaddr in `discovery.bootstrap_peers`
- **Config files**: Store all configs in `~/.debros/` with different filenames
- **--config flag**: Specify which config file to load (defaults to `node.yaml`)
⚠️ **Common Mistake - Same Ports:**
If all nodes use the same ports (e.g., 5001, 7001), they will try to bind to the same addresses and fail to communicate. Verify each node has unique ports:
```bash
# Bootstrap
grep "rqlite_port\|rqlite_raft_port" ~/.debros/bootstrap.yaml
# Should show: rqlite_port: 5001, rqlite_raft_port: 7001
# Node 2
grep "rqlite_port\|rqlite_raft_port" ~/.debros/node.yaml
# Should show: rqlite_port: 5002, rqlite_raft_port: 7002
# Node 3
grep "rqlite_port\|rqlite_raft_port" ~/.debros/node2.yaml
# Should show: rqlite_port: 5003, rqlite_raft_port: 7003
```
If ports are wrong, regenerate the config with `--force`:
```bash
./bin/network-cli config init --type node --name node.yaml \
--listen-port 4002 --rqlite-http-port 5002 --rqlite-raft-port 7002 \
--join localhost:5001 --bootstrap-peers '<bootstrap_multiaddr>' --force
```
### Validating Configuration
DeBros Network performs strict validation of all configuration files at startup. This ensures invalid configurations are caught immediately rather than causing silent failures later. DeBros Network performs strict validation of all configuration files at startup. This ensures invalid configurations are caught immediately rather than causing silent failures later.
@ -757,4 +755,196 @@ GET /v1/pubsub/topics # List active topics
- **PubSub** - **PubSub**
- WS Subscribe: `GET /v1/pubsub/ws?topic=<topic>` - WS Subscribe: `GET /v1/pubsub/ws?topic=<topic>`
- Publish: `POST /v1/pubsub/publish` `{topic, data_base64}` → `{status:"ok"}` - Publish: `POST /v1/pubsub/publish` `{topic, data_base64}` → `{status:"ok"}`
- Topics: `GET /v1/pubsub/topics` → ` - Topics: `GET /v1/pubsub/topics` → `{topics:[...]}`
---
## Troubleshooting
### Configuration & Permissions
**Error: "Failed to create/access config directory"**
This happens when DeBros cannot access or create `~/.debros/` directory.
**Causes:**
1. Home directory is not writable
2. Home directory doesn't exist
3. Filesystem is read-only (sandboxed/containerized environment)
4. Permission denied (running with wrong user/umask)
**Solutions:**
```bash
# Check home directory exists and is writable
ls -ld ~
touch ~/test-write && rm ~/test-write
# Check umask (should be 0022 or 0002)
umask
# If umask is too restrictive, change it
umask 0022
# Check disk space
df -h ~
# For containerized environments, ensure /home/<user> is mounted with write permissions
docker run -v /home:/home --user $(id -u):$(id -g) debros-network
```
**Error: "Config file not found at ~/.debros/node.yaml"**
The node requires a config file to exist before starting.
**Solution:**
Generate config files first:
```bash
# Build CLI
make build
# Generate configs
./bin/network-cli config init --type bootstrap
./bin/network-cli config init --type node --bootstrap-peers '<peer_multiaddr>'
./bin/network-cli config init --type gateway
```
### Node Startup Issues
**Error: "node.data_dir: parent directory not writable"**
The data directory parent is not accessible.
**Solution:**
Ensure `~/.debros` is writable and has at least 10GB free space:
```bash
# Check permissions
ls -ld ~/.debros
# Check available space
df -h ~/.debros
# Recreate if corrupted
rm -rf ~/.debros
./bin/network-cli config init --type bootstrap
```
**Error: "failed to create data directory"**
The node cannot create its data directory in `~/.debros`.
**Causes:**
1. `~/.debros` is not writable
2. Parent directory path in config uses `~` which isn't expanded properly
3. Disk is full
**Solutions:**
```bash
# Check ~/.debros exists and is writable
mkdir -p ~/.debros
ls -ld ~/.debros
# Verify data_dir in config uses ~ (e.g., ~/.debros/node)
cat ~/.debros/node.yaml | grep data_dir
# Check disk space
df -h ~
# Ensure user owns ~/.debros
chown -R $(whoami) ~/.debros
# Retry node startup
make run-node
```
**Error: "stat ~/.debros: no such file or directory"**
**Port Already in Use**
If you get "address already in use" errors:
```bash
# Find processes using ports
lsof -i :4001 # P2P port
lsof -i :5001 # RQLite HTTP
lsof -i :7001 # RQLite Raft
# Kill if needed
kill -9 <PID>
# Or use different ports in config
./bin/network-cli config init --type node --listen-port 4002 --rqlite-http-port 5002 --rqlite-raft-port 7002
```
### Common Configuration Errors
**Error: "discovery.bootstrap_peers: required for node type"**
Nodes (non-bootstrap) must specify bootstrap peers to discover the network.
**Solution:**
Generate node config with bootstrap peers:
```bash
./bin/network-cli config init --type node --bootstrap-peers '/ip4/127.0.0.1/tcp/4001/p2p/12D3KooW...'
```
**Error: "database.rqlite_join_address: required for node type"**
Non-bootstrap nodes must specify which node to join in the Raft cluster.
**Solution:**
Generate config with join address:
```bash
./bin/network-cli config init --type node --join localhost:5001
```
**Error: "database.rqlite_raft_port: must differ from database.rqlite_port"**
HTTP and Raft ports cannot be the same.
**Solution:**
Use different ports (RQLite HTTP and Raft must be on different ports):
```bash
./bin/network-cli config init --type node \
--rqlite-http-port 5001 \
--rqlite-raft-port 7001
```
### Peer Discovery Issues
If nodes can't find each other:
1. **Verify bootstrap node is running:**
```bash
./bin/network-cli health
./bin/network-cli peers
```
2. **Check bootstrap peer multiaddr is correct:**
```bash
cat ~/.debros/bootstrap/peer.info # On bootstrap node
# Should match value in other nodes' discovery.bootstrap_peers
```
3. **Ensure all nodes have same bootstrap peers in config**
4. **Check firewall/network:**
```bash
# Verify P2P port is open
nc -zv 127.0.0.1 4001
```
---
## License

332
cmd/cli/main.go
View File

@ -7,12 +7,14 @@ import (
"log" "log"
"os" "os"
"os/exec" "os/exec"
"path/filepath"
"strconv" "strconv"
"strings" "strings"
"time" "time"
"github.com/DeBrosOfficial/network/pkg/auth" "github.com/DeBrosOfficial/network/pkg/auth"
"github.com/DeBrosOfficial/network/pkg/client" "github.com/DeBrosOfficial/network/pkg/client"
"github.com/DeBrosOfficial/network/pkg/config"
"github.com/libp2p/go-libp2p/core/crypto" "github.com/libp2p/go-libp2p/core/crypto"
"github.com/libp2p/go-libp2p/core/peer" "github.com/libp2p/go-libp2p/core/peer"
) )
@ -78,6 +80,8 @@ func main() {
handlePeerID() handlePeerID()
case "auth": case "auth":
handleAuth(args) handleAuth(args)
case "config":
handleConfig(args)
case "help", "--help", "-h": case "help", "--help", "-h":
showHelp() showHelp()
@ -576,6 +580,333 @@ func isPrintableText(s string) bool {
return len(s) > 0 && float64(printableCount)/float64(len(s)) > 0.8 return len(s) > 0 && float64(printableCount)/float64(len(s)) > 0.8
} }
func handleConfig(args []string) {
if len(args) == 0 {
showConfigHelp()
return
}
subcommand := args[0]
subargs := args[1:]
switch subcommand {
case "init":
handleConfigInit(subargs)
case "validate":
handleConfigValidate(subargs)
case "help":
showConfigHelp()
default:
fmt.Fprintf(os.Stderr, "Unknown config subcommand: %s\n", subcommand)
showConfigHelp()
os.Exit(1)
}
}
func showConfigHelp() {
fmt.Printf("Config Management Commands\n\n")
fmt.Printf("Usage: network-cli config <subcommand> [options]\n\n")
fmt.Printf("Subcommands:\n")
fmt.Printf(" init - Generate configuration files in ~/.debros\n")
fmt.Printf(" validate --name <file> - Validate a config file\n\n")
fmt.Printf("Init Options:\n")
fmt.Printf(" --type <type> - Config type: node, bootstrap, gateway (default: node)\n")
fmt.Printf(" --name <file> - Output filename (default: node.yaml)\n")
fmt.Printf(" --id <id> - Node ID for bootstrap peers\n")
fmt.Printf(" --listen-port <port> - LibP2P listen port (default: 4001)\n")
fmt.Printf(" --rqlite-http-port <port> - RQLite HTTP port (default: 5001)\n")
fmt.Printf(" --rqlite-raft-port <port> - RQLite Raft port (default: 7001)\n")
fmt.Printf(" --join <host:port> - RQLite address to join (required for non-bootstrap)\n")
fmt.Printf(" --bootstrap-peers <peers> - Comma-separated bootstrap peer multiaddrs\n")
fmt.Printf(" --force - Overwrite existing config\n\n")
fmt.Printf("Examples:\n")
fmt.Printf(" network-cli config init\n")
fmt.Printf(" network-cli config init --type node --bootstrap-peers /ip4/127.0.0.1/tcp/4001/p2p/QmXxx,/ip4/127.0.0.1/tcp/4002/p2p/QmYyy\n")
fmt.Printf(" network-cli config init --type bootstrap\n")
fmt.Printf(" network-cli config init --type gateway\n")
fmt.Printf(" network-cli config validate --name node.yaml\n")
}
func handleConfigInit(args []string) {
// Parse flags
var (
cfgType = "node"
name = "node.yaml"
id string
listenPort = 4001
rqliteHTTPPort = 5001
rqliteRaftPort = 7001
joinAddr string
bootstrapPeers string
force bool
)
for i := 0; i < len(args); i++ {
switch args[i] {
case "--type":
if i+1 < len(args) {
cfgType = args[i+1]
i++
}
case "--name":
if i+1 < len(args) {
name = args[i+1]
i++
}
case "--id":
if i+1 < len(args) {
id = args[i+1]
i++
}
case "--listen-port":
if i+1 < len(args) {
if p, err := strconv.Atoi(args[i+1]); err == nil {
listenPort = p
}
i++
}
case "--rqlite-http-port":
if i+1 < len(args) {
if p, err := strconv.Atoi(args[i+1]); err == nil {
rqliteHTTPPort = p
}
i++
}
case "--rqlite-raft-port":
if i+1 < len(args) {
if p, err := strconv.Atoi(args[i+1]); err == nil {
rqliteRaftPort = p
}
i++
}
case "--join":
if i+1 < len(args) {
joinAddr = args[i+1]
i++
}
case "--bootstrap-peers":
if i+1 < len(args) {
bootstrapPeers = args[i+1]
i++
}
case "--force":
force = true
}
}
// Validate type
if cfgType != "node" && cfgType != "bootstrap" && cfgType != "gateway" {
fmt.Fprintf(os.Stderr, "Invalid --type: %s (expected: node, bootstrap, or gateway)\n", cfgType)
os.Exit(1)
}
// Ensure config directory exists
configDir, err := config.EnsureConfigDir()
if err != nil {
fmt.Fprintf(os.Stderr, "Failed to ensure config directory: %v\n", err)
os.Exit(1)
}
configPath := filepath.Join(configDir, name)
// Check if file exists
if !force {
if _, err := os.Stat(configPath); err == nil {
fmt.Fprintf(os.Stderr, "Config file already exists at %s (use --force to overwrite)\n", configPath)
os.Exit(1)
}
}
// Generate config based on type
var configContent string
switch cfgType {
case "node":
configContent = generateNodeConfig(name, id, listenPort, rqliteHTTPPort, rqliteRaftPort, joinAddr, bootstrapPeers)
case "bootstrap":
configContent = generateBootstrapConfig(name, id, listenPort, rqliteHTTPPort, rqliteRaftPort)
case "gateway":
configContent = generateGatewayConfig()
}
// Write config file
if err := os.WriteFile(configPath, []byte(configContent), 0644); err != nil {
fmt.Fprintf(os.Stderr, "Failed to write config file: %v\n", err)
os.Exit(1)
}
fmt.Printf("✅ Configuration file created: %s\n", configPath)
fmt.Printf(" Type: %s\n", cfgType)
fmt.Printf("\nYou can now start the %s using the generated config.\n", cfgType)
}
func handleConfigValidate(args []string) {
var name string
for i := 0; i < len(args); i++ {
if args[i] == "--name" && i+1 < len(args) {
name = args[i+1]
i++
}
}
if name == "" {
fmt.Fprintf(os.Stderr, "Missing --name flag\n")
showConfigHelp()
os.Exit(1)
}
configDir, err := config.ConfigDir()
if err != nil {
fmt.Fprintf(os.Stderr, "Failed to get config directory: %v\n", err)
os.Exit(1)
}
configPath := filepath.Join(configDir, name)
file, err := os.Open(configPath)
if err != nil {
fmt.Fprintf(os.Stderr, "Failed to open config file: %v\n", err)
os.Exit(1)
}
defer file.Close()
var cfg config.Config
if err := config.DecodeStrict(file, &cfg); err != nil {
fmt.Fprintf(os.Stderr, "Failed to parse config: %v\n", err)
os.Exit(1)
}
// Run validation
errs := cfg.Validate()
if len(errs) > 0 {
fmt.Fprintf(os.Stderr, "\n❌ Configuration errors (%d):\n", len(errs))
for _, err := range errs {
fmt.Fprintf(os.Stderr, " - %s\n", err)
}
os.Exit(1)
}
fmt.Printf("✅ Config is valid: %s\n", configPath)
}
func generateNodeConfig(name, id string, listenPort, rqliteHTTPPort, rqliteRaftPort int, joinAddr, bootstrapPeers string) string {
nodeID := id
if nodeID == "" {
nodeID = fmt.Sprintf("node-%d", time.Now().Unix())
}
// Parse bootstrap peers
var peers []string
if bootstrapPeers != "" {
for _, p := range strings.Split(bootstrapPeers, ",") {
if p = strings.TrimSpace(p); p != "" {
peers = append(peers, p)
}
}
}
// Construct data_dir from name stem (remove .yaml)
dataDir := strings.TrimSuffix(name, ".yaml")
dataDir = filepath.Join(os.ExpandEnv("~"), ".debros", dataDir)
var peersYAML strings.Builder
if len(peers) == 0 {
peersYAML.WriteString(" bootstrap_peers: []")
} else {
peersYAML.WriteString(" bootstrap_peers:\n")
for _, p := range peers {
fmt.Fprintf(&peersYAML, " - \"%s\"\n", p)
}
}
if joinAddr == "" {
joinAddr = "localhost:5001"
}
return fmt.Sprintf(`node:
id: "%s"
type: "node"
listen_addresses:
- "/ip4/0.0.0.0/tcp/%d"
data_dir: "%s"
max_connections: 50
database:
data_dir: "%s/rqlite"
replication_factor: 3
shard_count: 16
max_database_size: 1073741824
backup_interval: "24h"
rqlite_port: %d
rqlite_raft_port: %d
rqlite_join_address: "%s"
discovery:
%s
discovery_interval: "15s"
bootstrap_port: %d
http_adv_address: "127.0.0.1:%d"
raft_adv_address: "127.0.0.1:%d"
node_namespace: "default"
security:
enable_tls: false
logging:
level: "info"
format: "console"
`, nodeID, listenPort, dataDir, dataDir, rqliteHTTPPort, rqliteRaftPort, joinAddr, peersYAML.String(), 4001, rqliteHTTPPort, rqliteRaftPort)
}
func generateBootstrapConfig(name, id string, listenPort, rqliteHTTPPort, rqliteRaftPort int) string {
nodeID := id
if nodeID == "" {
nodeID = "bootstrap"
}
dataDir := filepath.Join(os.ExpandEnv("~"), ".debros", "bootstrap")
return fmt.Sprintf(`node:
id: "%s"
type: "bootstrap"
listen_addresses:
- "/ip4/0.0.0.0/tcp/%d"
data_dir: "%s"
max_connections: 50
database:
data_dir: "%s/rqlite"
replication_factor: 3
shard_count: 16
max_database_size: 1073741824
backup_interval: "24h"
rqlite_port: %d
rqlite_raft_port: %d
rqlite_join_address: ""
discovery:
bootstrap_peers: []
discovery_interval: "15s"
bootstrap_port: %d
http_adv_address: "127.0.0.1:%d"
raft_adv_address: "127.0.0.1:%d"
node_namespace: "default"
security:
enable_tls: false
logging:
level: "info"
format: "console"
`, nodeID, listenPort, dataDir, dataDir, rqliteHTTPPort, rqliteRaftPort, 4001, rqliteHTTPPort, rqliteRaftPort)
}
func generateGatewayConfig() string {
return `listen_addr: ":6001"
client_namespace: "default"
rqlite_dsn: ""
bootstrap_peers: []
`
}
func showHelp() { func showHelp() {
fmt.Printf("Network CLI - Distributed P2P Network Management Tool\n\n") fmt.Printf("Network CLI - Distributed P2P Network Management Tool\n\n")
fmt.Printf("Usage: network-cli <command> [args...]\n\n") fmt.Printf("Usage: network-cli <command> [args...]\n\n")
@ -591,6 +922,7 @@ func showHelp() {
fmt.Printf(" pubsub subscribe <topic> [duration] 🔐 Subscribe to topic\n") fmt.Printf(" pubsub subscribe <topic> [duration] 🔐 Subscribe to topic\n")
fmt.Printf(" pubsub topics 🔐 List topics\n") fmt.Printf(" pubsub topics 🔐 List topics\n")
fmt.Printf(" connect <peer_address> - Connect to peer\n") fmt.Printf(" connect <peer_address> - Connect to peer\n")
fmt.Printf(" config - Show current configuration\n")
fmt.Printf(" help - Show this help\n\n") fmt.Printf(" help - Show this help\n\n")
fmt.Printf("Global Flags:\n") fmt.Printf("Global Flags:\n")

130
cmd/gateway/config.go
View File

@ -1,7 +1,6 @@
package main package main
import ( import (
"flag"
"fmt" "fmt"
"os" "os"
"strings" "strings"
@ -38,10 +37,43 @@ func getEnvBoolDefault(key string, def bool) bool {
} }
} }
// parseGatewayConfig loads optional configs/gateway.yaml then applies env and flags. // parseGatewayConfig loads gateway.yaml from ~/.debros exclusively.
// Priority: flags > env > yaml > defaults.
func parseGatewayConfig(logger *logging.ColoredLogger) *gateway.Config { func parseGatewayConfig(logger *logging.ColoredLogger) *gateway.Config {
// Base defaults // Determine config path
configPath, err := config.DefaultPath("gateway.yaml")
if err != nil {
logger.ComponentError(logging.ComponentGeneral, "Failed to determine config path", zap.Error(err))
fmt.Fprintf(os.Stderr, "Configuration error: %v\n", err)
os.Exit(1)
}
// Load YAML
type yamlCfg struct {
ListenAddr string `yaml:"listen_addr"`
ClientNamespace string `yaml:"client_namespace"`
RQLiteDSN string `yaml:"rqlite_dsn"`
BootstrapPeers []string `yaml:"bootstrap_peers"`
}
data, err := os.ReadFile(configPath)
if err != nil {
logger.ComponentError(logging.ComponentGeneral, "Config file not found",
zap.String("path", configPath),
zap.Error(err))
fmt.Fprintf(os.Stderr, "\nConfig file not found at %s\n", configPath)
fmt.Fprintf(os.Stderr, "Generate it using: network-cli config init --type gateway\n")
os.Exit(1)
}
var y yamlCfg
// Use strict YAML decoding to reject unknown fields
if err := config.DecodeStrict(strings.NewReader(string(data)), &y); err != nil {
logger.ComponentError(logging.ComponentGeneral, "Failed to parse gateway config", zap.Error(err))
fmt.Fprintf(os.Stderr, "Configuration parse error: %v\n", err)
os.Exit(1)
}
// Build config from YAML
cfg := &gateway.Config{ cfg := &gateway.Config{
ListenAddr: ":6001", ListenAddr: ":6001",
ClientNamespace: "default", ClientNamespace: "default",
@ -49,93 +81,26 @@ func parseGatewayConfig(logger *logging.ColoredLogger) *gateway.Config {
RQLiteDSN: "", RQLiteDSN: "",
} }
// 1) YAML (optional) if v := strings.TrimSpace(y.ListenAddr); v != "" {
{
type yamlCfg struct {
ListenAddr string `yaml:"listen_addr"`
ClientNamespace string `yaml:"client_namespace"`
RQLiteDSN string `yaml:"rqlite_dsn"`
BootstrapPeers []string `yaml:"bootstrap_peers"`
}
const path = "configs/gateway.yaml"
if data, err := os.ReadFile(path); err == nil {
var y yamlCfg
// Use strict YAML decoding to reject unknown fields
if err := config.DecodeStrict(strings.NewReader(string(data)), &y); err != nil {
logger.ComponentError(logging.ComponentGeneral, "failed to parse configs/gateway.yaml", zap.Error(err))
fmt.Fprintf(os.Stderr, "Configuration load error: %v\n", err)
os.Exit(1)
}
if v := strings.TrimSpace(y.ListenAddr); v != "" {
cfg.ListenAddr = v
}
if v := strings.TrimSpace(y.ClientNamespace); v != "" {
cfg.ClientNamespace = v
}
if v := strings.TrimSpace(y.RQLiteDSN); v != "" {
cfg.RQLiteDSN = v
}
if len(y.BootstrapPeers) > 0 {
var bp []string
for _, p := range y.BootstrapPeers {
p = strings.TrimSpace(p)
if p != "" {
bp = append(bp, p)
}
}
if len(bp) > 0 {
cfg.BootstrapPeers = bp
}
}
}
}
// 2) Env overrides
if v := strings.TrimSpace(os.Getenv("GATEWAY_ADDR")); v != "" {
cfg.ListenAddr = v cfg.ListenAddr = v
} }
if v := strings.TrimSpace(os.Getenv("GATEWAY_NAMESPACE")); v != "" { if v := strings.TrimSpace(y.ClientNamespace); v != "" {
cfg.ClientNamespace = v cfg.ClientNamespace = v
} }
if v := strings.TrimSpace(os.Getenv("GATEWAY_RQLITE_DSN")); v != "" { if v := strings.TrimSpace(y.RQLiteDSN); v != "" {
cfg.RQLiteDSN = v cfg.RQLiteDSN = v
} }
if v := strings.TrimSpace(os.Getenv("GATEWAY_BOOTSTRAP_PEERS")); v != "" { if len(y.BootstrapPeers) > 0 {
parts := strings.Split(v, ",")
var bp []string var bp []string
for _, part := range parts { for _, p := range y.BootstrapPeers {
s := strings.TrimSpace(part) p = strings.TrimSpace(p)
if s != "" { if p != "" {
bp = append(bp, s) bp = append(bp, p)
} }
} }
cfg.BootstrapPeers = bp if len(bp) > 0 {
} cfg.BootstrapPeers = bp
// 3) Flags (override env)
addr := flag.String("addr", "", "HTTP listen address (e.g., :6001)")
ns := flag.String("namespace", "", "Client namespace for scoping resources")
peers := flag.String("bootstrap-peers", "", "Comma-separated bootstrap peers for network client")
// Do not call flag.Parse() elsewhere to avoid double-parsing
flag.Parse()
if a := strings.TrimSpace(*addr); a != "" {
cfg.ListenAddr = a
}
if n := strings.TrimSpace(*ns); n != "" {
cfg.ClientNamespace = n
}
if p := strings.TrimSpace(*peers); p != "" {
parts := strings.Split(p, ",")
var bp []string
for _, part := range parts {
s := strings.TrimSpace(part)
if s != "" {
bp = append(bp, s)
}
} }
cfg.BootstrapPeers = bp
} }
// Validate configuration // Validate configuration
@ -148,7 +113,8 @@ func parseGatewayConfig(logger *logging.ColoredLogger) *gateway.Config {
os.Exit(1) os.Exit(1)
} }
logger.ComponentInfo(logging.ComponentGeneral, "Loaded gateway configuration", logger.ComponentInfo(logging.ComponentGeneral, "Loaded gateway configuration from YAML",
zap.String("path", configPath),
zap.String("addr", cfg.ListenAddr), zap.String("addr", cfg.ListenAddr),
zap.String("namespace", cfg.ClientNamespace), zap.String("namespace", cfg.ClientNamespace),
zap.Int("bootstrap_peer_count", len(cfg.BootstrapPeers)), zap.Int("bootstrap_peer_count", len(cfg.BootstrapPeers)),

165
cmd/node/main.go
View File

@ -7,6 +7,8 @@ import (
"os" "os"
"os/signal" "os/signal"
"path/filepath" "path/filepath"
"strconv"
"strings"
"syscall" "syscall"
"github.com/DeBrosOfficial/network/pkg/config" "github.com/DeBrosOfficial/network/pkg/config"
@ -29,15 +31,8 @@ func setup_logger(component logging.Component) (logger *logging.ColoredLogger) {
} }
// parse_flags parses command-line flags and returns them. // parse_flags parses command-line flags and returns them.
func parse_flags() (configPath, dataDir, nodeID *string, p2pPort, rqlHTTP, rqlRaft *int, rqlJoinAddr, advAddr *string, help *bool) { func parse_flags() (configName *string, help *bool) {
configPath = flag.String("config", "", "Path to config YAML file (overrides defaults)") configName = flag.String("config", "node.yaml", "Config filename in ~/.debros (default: node.yaml)")
dataDir = flag.String("data", "", "Data directory (auto-detected if not provided)")
nodeID = flag.String("id", "", "Node identifier (for running multiple local nodes)")
p2pPort = flag.Int("p2p-port", 4001, "LibP2P listen port")
rqlHTTP = flag.Int("rqlite-http-port", 5001, "RQLite HTTP API port")
rqlRaft = flag.Int("rqlite-raft-port", 7001, "RQLite Raft port")
rqlJoinAddr = flag.String("rqlite-join-address", "", "RQLite address to join (e.g., /ip4/)")
advAddr = flag.String("adv-addr", "127.0.0.1", "Default Advertise address for rqlite and rafts")
help = flag.Bool("help", false, "Show help") help = flag.Bool("help", false, "Show help")
flag.Parse() flag.Parse()
@ -67,18 +62,39 @@ func check_if_should_open_help(help *bool) {
} }
} }
// select_data_dir selects the data directory for the node // select_data_dir validates that we can load the config from ~/.debros
// If none of (hasConfigFile, nodeID, dataDir) are present, throw an error and do not start func select_data_dir_check(configName *string) {
func select_data_dir(dataDir *string, nodeID *string, hasConfigFile bool) {
logger := setup_logger(logging.ComponentNode) logger := setup_logger(logging.ComponentNode)
if !hasConfigFile && (*nodeID == "" || nodeID == nil) && (*dataDir == "" || dataDir == nil) { // Ensure config directory exists and is writable
logger.Error("No config file, node ID, or data directory specified. Please provide at least one. Refusing to start.") _, err := config.EnsureConfigDir()
if err != nil {
logger.Error("Failed to ensure config directory", zap.Error(err))
fmt.Fprintf(os.Stderr, "\n❌ Configuration Error:\n")
fmt.Fprintf(os.Stderr, "Failed to create/access config directory: %v\n", err)
fmt.Fprintf(os.Stderr, "\nPlease ensure:\n")
fmt.Fprintf(os.Stderr, " 1. Home directory is accessible: %s\n", os.ExpandEnv("~"))
fmt.Fprintf(os.Stderr, " 2. You have write permissions to home directory\n")
fmt.Fprintf(os.Stderr, " 3. Disk space is available\n")
os.Exit(1) os.Exit(1)
} }
if *dataDir != "" { configPath, err := config.DefaultPath(*configName)
logger.Info("Data directory selected: %s", zap.String("dataDir", *dataDir)) if err != nil {
logger.Error("Failed to determine config path", zap.Error(err))
os.Exit(1)
}
if _, err := os.Stat(configPath); err != nil {
logger.Error("Config file not found",
zap.String("path", configPath),
zap.Error(err))
fmt.Fprintf(os.Stderr, "\n❌ Configuration Error:\n")
fmt.Fprintf(os.Stderr, "Config file not found at %s\n", configPath)
fmt.Fprintf(os.Stderr, "\nGenerate it with one of:\n")
fmt.Fprintf(os.Stderr, " network-cli config init --type bootstrap\n")
fmt.Fprintf(os.Stderr, " network-cli config init --type node --bootstrap-peers '<peer_multiaddr>'\n")
os.Exit(1)
} }
} }
@ -97,9 +113,21 @@ func startNode(ctx context.Context, cfg *config.Config, port int) error {
return err return err
} }
// Expand data directory path for peer.info file
dataDir := os.ExpandEnv(cfg.Node.DataDir)
if strings.HasPrefix(dataDir, "~") {
home, err := os.UserHomeDir()
if err != nil {
logger.Error("failed to determine home directory: %v", zap.Error(err))
dataDir = cfg.Node.DataDir
} else {
dataDir = filepath.Join(home, dataDir[1:])
}
}
// Save the peer ID to a file for CLI access (especially useful for bootstrap) // Save the peer ID to a file for CLI access (especially useful for bootstrap)
peerID := n.GetPeerID() peerID := n.GetPeerID()
peerInfoFile := filepath.Join(cfg.Node.DataDir, "peer.info") peerInfoFile := filepath.Join(dataDir, "peer.info")
peerMultiaddr := fmt.Sprintf("/ip4/0.0.0.0/tcp/%d/p2p/%s", port, peerID) peerMultiaddr := fmt.Sprintf("/ip4/0.0.0.0/tcp/%d/p2p/%s", port, peerID)
if err := os.WriteFile(peerInfoFile, []byte(peerMultiaddr), 0644); err != nil { if err := os.WriteFile(peerInfoFile, []byte(peerMultiaddr), 0644); err != nil {
@ -168,54 +196,107 @@ func printValidationErrors(errs []error) {
os.Exit(1) os.Exit(1)
} }
// ensureDataDirectories ensures that all necessary data directories exist and have correct permissions.
func ensureDataDirectories(cfg *config.Config, logger *logging.ColoredLogger) error {
// Expand ~ in data_dir path
dataDir := os.ExpandEnv(cfg.Node.DataDir)
if strings.HasPrefix(dataDir, "~") {
home, err := os.UserHomeDir()
if err != nil {
return fmt.Errorf("failed to determine home directory: %w", err)
}
dataDir = filepath.Join(home, dataDir[1:])
}
// Ensure Node.DataDir exists and is writable
if err := os.MkdirAll(dataDir, 0755); err != nil {
return fmt.Errorf("failed to create data directory %s: %w", dataDir, err)
}
logger.ComponentInfo(logging.ComponentNode, "Data directory created/verified", zap.String("path", dataDir))
// Ensure RQLite data directory exists
rqliteDir := filepath.Join(dataDir, "rqlite")
if err := os.MkdirAll(rqliteDir, 0755); err != nil {
return fmt.Errorf("failed to create rqlite data directory: %w", err)
}
logger.ComponentInfo(logging.ComponentNode, "RQLite data directory created/verified", zap.String("path", rqliteDir))
return nil
}
func main() { func main() {
logger := setup_logger(logging.ComponentNode) logger := setup_logger(logging.ComponentNode)
// Parse command-line flags // Parse command-line flags
configPath, dataDir, nodeID, p2pPort, rqlHTTP, rqlRaft, rqlJoinAddr, advAddr, help := parse_flags() configName, help := parse_flags()
check_if_should_open_help(help) check_if_should_open_help(help)
select_data_dir(dataDir, nodeID, *configPath != "")
// Load configuration // Check if config file exists
var cfg *config.Config select_data_dir_check(configName)
if *configPath != "" {
// Load from YAML with strict decoding // Load configuration from ~/.debros/node.yaml
var err error configPath, err := config.DefaultPath(*configName)
cfg, err = LoadConfigFromYAML(*configPath) if err != nil {
if err != nil { logger.Error("Failed to determine config path", zap.Error(err))
logger.Error("Failed to load config from YAML", zap.Error(err)) fmt.Fprintf(os.Stderr, "Configuration error: %v\n", err)
fmt.Fprintf(os.Stderr, "Configuration load error: %v\n", err) os.Exit(1)
os.Exit(1)
}
logger.ComponentInfo(logging.ComponentNode, "Configuration loaded from YAML file", zap.String("path", *configPath))
} else {
// Use default configuration
cfg = config.DefaultConfig()
logger.ComponentInfo(logging.ComponentNode, "Default configuration loaded successfully")
} }
// Apply command-line flag overrides var cfg *config.Config
apply_flag_overrides(cfg, p2pPort, rqlHTTP, rqlRaft, rqlJoinAddr, advAddr, dataDir) var cfgErr error
logger.ComponentInfo(logging.ComponentNode, "Command line arguments applied to configuration") cfg, cfgErr = LoadConfigFromYAML(configPath)
if cfgErr != nil {
logger.Error("Failed to load config from YAML", zap.Error(cfgErr))
fmt.Fprintf(os.Stderr, "Configuration load error: %v\n", cfgErr)
os.Exit(1)
}
logger.ComponentInfo(logging.ComponentNode, "Configuration loaded from YAML file", zap.String("path", configPath))
// Set default advertised addresses if empty
if cfg.Discovery.HttpAdvAddress == "" {
cfg.Discovery.HttpAdvAddress = fmt.Sprintf("127.0.0.1:%d", cfg.Database.RQLitePort)
}
if cfg.Discovery.RaftAdvAddress == "" {
cfg.Discovery.RaftAdvAddress = fmt.Sprintf("127.0.0.1:%d", cfg.Database.RQLiteRaftPort)
}
// Validate configuration // Validate configuration
if errs := cfg.Validate(); len(errs) > 0 { if errs := cfg.Validate(); len(errs) > 0 {
printValidationErrors(errs) printValidationErrors(errs)
} }
// LibP2P uses configurable port (default 4001); RQLite uses 5001 (HTTP) and 7001 (Raft) // Expand and create data directories
port := *p2pPort if err := ensureDataDirectories(cfg, logger); err != nil {
logger.Error("Failed to create data directories", zap.Error(err))
fmt.Fprintf(os.Stderr, "\n❌ Data Directory Error:\n")
fmt.Fprintf(os.Stderr, "%v\n", err)
os.Exit(1)
}
logger.ComponentInfo(logging.ComponentNode, "Node configuration summary", logger.ComponentInfo(logging.ComponentNode, "Node configuration summary",
zap.Strings("listen_addresses", cfg.Node.ListenAddresses), zap.Strings("listen_addresses", cfg.Node.ListenAddresses),
zap.Int("rqlite_http_port", cfg.Database.RQLitePort), zap.Int("rqlite_http_port", cfg.Database.RQLitePort),
zap.Int("rqlite_raft_port", cfg.Database.RQLiteRaftPort), zap.Int("rqlite_raft_port", cfg.Database.RQLiteRaftPort),
zap.Int("p2p_port", port),
zap.Strings("bootstrap_peers", cfg.Discovery.BootstrapPeers), zap.Strings("bootstrap_peers", cfg.Discovery.BootstrapPeers),
zap.String("rqlite_join_address", cfg.Database.RQLiteJoinAddress), zap.String("rqlite_join_address", cfg.Database.RQLiteJoinAddress),
zap.String("data_directory", cfg.Node.DataDir)) zap.String("data_directory", cfg.Node.DataDir))
// Extract P2P port from listen addresses
p2pPort := 4001 // default
if len(cfg.Node.ListenAddresses) > 0 {
// Parse port from multiaddr like "/ip4/0.0.0.0/tcp/4001"
parts := strings.Split(cfg.Node.ListenAddresses[0], "/")
for i, part := range parts {
if part == "tcp" && i+1 < len(parts) {
if port, err := strconv.Atoi(parts[i+1]); err == nil {
p2pPort = port
break
}
}
}
}
// Create context for graceful shutdown // Create context for graceful shutdown
ctx, cancel := context.WithCancel(context.Background()) ctx, cancel := context.WithCancel(context.Background())
defer cancel() defer cancel()
@ -224,7 +305,7 @@ func main() {
errChan := make(chan error, 1) errChan := make(chan error, 1)
doneChan := make(chan struct{}) doneChan := make(chan struct{})
go func() { go func() {
if err := startNode(ctx, cfg, port); err != nil { if err := startNode(ctx, cfg, p2pPort); err != nil {
errChan <- err errChan <- err
} }
close(doneChan) close(doneChan)

8
pkg/config/config.go
View File

@ -108,13 +108,7 @@ func DefaultConfig() *Config {
RQLiteJoinAddress: "", // Empty for bootstrap node RQLiteJoinAddress: "", // Empty for bootstrap node
}, },
Discovery: DiscoveryConfig{ Discovery: DiscoveryConfig{
BootstrapPeers: []string{ BootstrapPeers: []string{},
"/ip4/127.0.0.1/tcp/4001/p2p/12D3KooWHbcFcrGPXKUrHcxvd8MXEeUzRYyvY8fQcpEBxncSUwhj",
// "/ip4/217.76.54.178/tcp/4001/p2p/12D3KooWKZnirPwNT4URtNSWK45f6vLkEs4xyUZ792F8Uj1oYnm1",
// "/ip4/51.83.128.181/tcp/4001/p2p/12D3KooWBn2Zf1R8v9pEfmz7hDZ5b3oADxfejA3zJBYzKRCzgvhR",
// "/ip4/155.133.27.199/tcp/4001/p2p/12D3KooWC69SBzM5QUgrLrfLWUykE8au32X5LwT7zwv9bixrQPm1",
// "/ip4/217.76.56.2/tcp/4001/p2p/12D3KooWEiqJHvznxqJ5p2y8mUs6Ky6dfU1xTYFQbyKRCABfcZz4",
},
BootstrapPort: 4001, // Default LibP2P port BootstrapPort: 4001, // Default LibP2P port
DiscoveryInterval: time.Second * 15, // Back to 15 seconds for testing DiscoveryInterval: time.Second * 15, // Back to 15 seconds for testing
HttpAdvAddress: "", HttpAdvAddress: "",

38
pkg/config/paths.go Normal file
View File

@ -0,0 +1,38 @@
package config
import (
"fmt"
"os"
"path/filepath"
)
// ConfigDir returns the path to the DeBros config directory (~/.debros).
func ConfigDir() (string, error) {
home, err := os.UserHomeDir()
if err != nil {
return "", fmt.Errorf("failed to determine home directory: %w", err)
}
return filepath.Join(home, ".debros"), nil
}
// EnsureConfigDir creates the config directory if it does not exist.
func EnsureConfigDir() (string, error) {
dir, err := ConfigDir()
if err != nil {
return "", err
}
if err := os.MkdirAll(dir, 0700); err != nil {
return "", fmt.Errorf("failed to create config directory %s: %w", dir, err)
}
return dir, nil
}
// DefaultPath returns the path to the config file for the given component name.
// component should be e.g., "node.yaml", "bootstrap.yaml", "gateway.yaml"
func DefaultPath(component string) (string, error) {
dir, err := ConfigDir()
if err != nil {
return "", err
}
return filepath.Join(dir, component), nil
}

31
pkg/config/validate.go
View File

@ -456,25 +456,46 @@ func validateDataDir(path string) error {
return fmt.Errorf("must not be empty") return fmt.Errorf("must not be empty")
} }
if info, err := os.Stat(path); err == nil { // Expand ~ to home directory
expandedPath := os.ExpandEnv(path)
if strings.HasPrefix(expandedPath, "~") {
home, err := os.UserHomeDir()
if err != nil {
return fmt.Errorf("cannot determine home directory: %v", err)
}
expandedPath = filepath.Join(home, expandedPath[1:])
}
if info, err := os.Stat(expandedPath); err == nil {
// Directory exists; check if it's a directory and writable // Directory exists; check if it's a directory and writable
if !info.IsDir() { if !info.IsDir() {
return fmt.Errorf("path exists but is not a directory") return fmt.Errorf("path exists but is not a directory")
} }
// Try to write a test file to check permissions // Try to write a test file to check permissions
testFile := filepath.Join(path, ".write_test") testFile := filepath.Join(expandedPath, ".write_test")
if err := os.WriteFile(testFile, []byte(""), 0644); err != nil { if err := os.WriteFile(testFile, []byte(""), 0644); err != nil {
return fmt.Errorf("directory not writable: %v", err) return fmt.Errorf("directory not writable: %v", err)
} }
os.Remove(testFile) os.Remove(testFile)
} else if os.IsNotExist(err) { } else if os.IsNotExist(err) {
// Directory doesn't exist; check if parent is writable // Directory doesn't exist; check if parent is writable
parent := filepath.Dir(path) parent := filepath.Dir(expandedPath)
if parent == "" || parent == "." { if parent == "" || parent == "." {
parent = "." parent = "."
} }
if err := validateDirWritable(parent); err != nil { // Allow parent not existing - it will be created at runtime
return fmt.Errorf("parent directory not writable: %v", err) if info, err := os.Stat(parent); err != nil {
if !os.IsNotExist(err) {
return fmt.Errorf("parent directory not accessible: %v", err)
}
// Parent doesn't exist either - that's ok, will be created
} else if !info.IsDir() {
return fmt.Errorf("parent path is not a directory")
} else {
// Parent exists, check if writable
if err := validateDirWritable(parent); err != nil {
return fmt.Errorf("parent directory not writable: %v", err)
}
} }
} else { } else {
return fmt.Errorf("cannot access path: %v", err) return fmt.Errorf("cannot access path: %v", err)

30
pkg/rqlite/rqlite.go
View File

@ -69,8 +69,18 @@ func NewRQLiteManager(cfg *config.DatabaseConfig, discoveryCfg *config.Discovery
// Start starts the RQLite node // Start starts the RQLite node
func (r *RQLiteManager) Start(ctx context.Context) error { func (r *RQLiteManager) Start(ctx context.Context) error {
// Expand ~ in data directory path
dataDir := os.ExpandEnv(r.dataDir)
if strings.HasPrefix(dataDir, "~") {
home, err := os.UserHomeDir()
if err != nil {
return fmt.Errorf("failed to determine home directory: %w", err)
}
dataDir = filepath.Join(home, dataDir[1:])
}
// Create data directory // Create data directory
rqliteDataDir := filepath.Join(r.dataDir, "rqlite") rqliteDataDir := filepath.Join(dataDir, "rqlite")
if err := os.MkdirAll(rqliteDataDir, 0755); err != nil { if err := os.MkdirAll(rqliteDataDir, 0755); err != nil {
return fmt.Errorf("failed to create RQLite data directory: %w", err) return fmt.Errorf("failed to create RQLite data directory: %w", err)
} }
@ -100,7 +110,7 @@ func (r *RQLiteManager) Start(ctx context.Context) error {
} }
// Wait for join target to become reachable to avoid forming a separate cluster (wait indefinitely) // Wait for join target to become reachable to avoid forming a separate cluster (wait indefinitely)
if err := r.waitForJoinTarget(ctx, joinArg, 0); err != nil { if err := r.waitForJoinTarget(ctx, r.config.RQLiteJoinAddress, 0); err != nil {
r.logger.Warn("Join target did not become reachable within timeout; will still attempt to join", r.logger.Warn("Join target did not become reachable within timeout; will still attempt to join",
zap.String("join_address", r.config.RQLiteJoinAddress), zap.String("join_address", r.config.RQLiteJoinAddress),
zap.Error(err)) zap.Error(err))
@ -126,7 +136,7 @@ func (r *RQLiteManager) Start(ctx context.Context) error {
// Start RQLite process (not bound to ctx for graceful Stop handling) // Start RQLite process (not bound to ctx for graceful Stop handling)
r.cmd = exec.Command("rqlited", args...) r.cmd = exec.Command("rqlited", args...)
// Uncomment if you want to see the stdout/stderr of the RQLite process // Enable debug logging of RQLite process to help diagnose issues
// r.cmd.Stdout = os.Stdout // r.cmd.Stdout = os.Stdout
// r.cmd.Stderr = os.Stderr // r.cmd.Stderr = os.Stderr
@ -166,7 +176,15 @@ func (r *RQLiteManager) Start(ctx context.Context) error {
} }
} else { } else {
r.logger.Info("Waiting for RQLite SQL availability (leader discovery)") r.logger.Info("Waiting for RQLite SQL availability (leader discovery)")
if err := r.waitForSQLAvailable(ctx); err != nil { // For joining nodes, wait longer for SQL availability
sqlCtx := ctx
if _, hasDeadline := ctx.Deadline(); !hasDeadline {
// If no deadline in context, create one for SQL availability check
var cancel context.CancelFunc
sqlCtx, cancel = context.WithTimeout(context.Background(), 2*time.Minute)
defer cancel()
}
if err := r.waitForSQLAvailable(sqlCtx); err != nil {
if r.cmd != nil && r.cmd.Process != nil { if r.cmd != nil && r.cmd.Process != nil {
_ = r.cmd.Process.Kill() _ = r.cmd.Process.Kill()
} }
@ -207,7 +225,9 @@ func (r *RQLiteManager) waitForReady(ctx context.Context) error {
url := fmt.Sprintf("http://localhost:%d/status", r.config.RQLitePort) url := fmt.Sprintf("http://localhost:%d/status", r.config.RQLitePort)
client := &http.Client{Timeout: 2 * time.Second} client := &http.Client{Timeout: 2 * time.Second}
for i := 0; i < 30; i++ { // Give joining nodes more time (120 seconds vs 30)
maxAttempts := 30
for i := 0; i < maxAttempts; i++ {
select { select {
case <-ctx.Done(): case <-ctx.Done():
return ctx.Err() return ctx.Err()