Files
CHORUS/p2p/node.go
anthonyrawlins 17673c38a6 fix: P2P connectivity regression + dynamic versioning system
## P2P Connectivity Fixes
- **Root Cause**: mDNS discovery was conditionally disabled in Task Execution Engine implementation
- **Solution**: Restored always-enabled mDNS discovery from working baseline (eb2e05f)
- **Result**: 9/9 Docker Swarm replicas with working P2P mesh, democratic elections, and leader consensus

## Dynamic Version System
- **Problem**: Hardcoded version "0.1.0-dev" in 1000+ builds made debugging impossible
- **Solution**: Implemented build-time version injection via ldflags
- **Features**: Shows commit hash, build date, and semantic version
- **Example**: `CHORUS-agent 0.5.5 (build: 9dbd361, 2025-09-26_05:55:55)`

## Container Compatibility
- **Issue**: Binary execution failed in Alpine due to glibc/musl incompatibility
- **Solution**: Added Ubuntu-based Dockerfile for proper glibc support
- **Benefit**: Reliable container execution across Docker Swarm nodes

## Key Changes
- `internal/runtime/shared.go`: Always enable mDNS discovery, dynamic version vars
- `cmd/agent/main.go`: Build-time version injection and display
- `p2p/node.go`: Restored working "🐝 Bzzz Node Status" logging format
- `Makefile`: Updated version to 0.5.5, proper ldflags configuration
- `Dockerfile.ubuntu`: New glibc-compatible container base
- `docker-compose.yml`: Updated to latest image tag for Watchtower auto-updates

## Verification
 P2P mesh connectivity: Peers exchanging availability broadcasts
 Democratic elections: Candidacy announcements and leader selection
 BACKBEAT integration: Beat synchronization and degraded mode handling
 Dynamic versioning: All containers show v0.5.5 with build metadata
 Task Execution Engine: All Phase 4 functionality preserved and working

Fixes P2P connectivity regression while preserving complete Task Execution Engine implementation.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-09-26 16:05:25 +10:00

202 lines
4.4 KiB
Go

package p2p
import (
"context"
"fmt"
"time"
"chorus/pkg/dht"
"github.com/libp2p/go-libp2p"
kaddht "github.com/libp2p/go-libp2p-kad-dht"
"github.com/libp2p/go-libp2p/core/host"
"github.com/libp2p/go-libp2p/core/peer"
"github.com/libp2p/go-libp2p/p2p/security/noise"
"github.com/libp2p/go-libp2p/p2p/transport/tcp"
"github.com/multiformats/go-multiaddr"
)
// Node represents a CHORUS P2P node
type Node struct {
host host.Host
ctx context.Context
cancel context.CancelFunc
config *Config
dht *dht.LibP2PDHT // Optional DHT for distributed discovery
}
// NewNode creates a new P2P node with the given configuration
func NewNode(ctx context.Context, opts ...Option) (*Node, error) {
config := DefaultConfig()
for _, opt := range opts {
opt(config)
}
nodeCtx, cancel := context.WithCancel(ctx)
// Build multiaddresses for listening
var listenAddrs []multiaddr.Multiaddr
for _, addr := range config.ListenAddresses {
ma, err := multiaddr.NewMultiaddr(addr)
if err != nil {
cancel()
return nil, fmt.Errorf("invalid listen address %s: %w", addr, err)
}
listenAddrs = append(listenAddrs, ma)
}
// Create libp2p host with security and transport options
h, err := libp2p.New(
libp2p.ListenAddrs(listenAddrs...),
libp2p.Security(noise.ID, noise.New),
libp2p.Transport(tcp.NewTCPTransport),
libp2p.DefaultMuxers,
libp2p.EnableRelay(),
)
if err != nil {
cancel()
return nil, fmt.Errorf("failed to create libp2p host: %w", err)
}
node := &Node{
host: h,
ctx: nodeCtx,
cancel: cancel,
config: config,
}
// Initialize DHT if enabled
if config.EnableDHT {
var dhtMode kaddht.ModeOpt
switch config.DHTMode {
case "client":
dhtMode = kaddht.ModeClient
case "server":
dhtMode = kaddht.ModeServer
default:
dhtMode = kaddht.ModeAuto
}
dhtOpts := []dht.Option{
dht.WithProtocolPrefix(config.DHTProtocolPrefix),
dht.WithMode(dhtMode),
dht.WithBootstrapPeersFromStrings(config.DHTBootstrapPeers),
dht.WithAutoBootstrap(len(config.DHTBootstrapPeers) > 0),
}
var err error
node.dht, err = dht.NewLibP2PDHT(nodeCtx, h, dhtOpts...)
if err != nil {
cancel()
h.Close()
return nil, fmt.Errorf("failed to create DHT: %w", err)
}
}
// Start background processes
go node.startBackgroundTasks()
return node, nil
}
// Host returns the underlying libp2p host
func (n *Node) Host() host.Host {
return n.host
}
// ID returns the peer ID of this node
func (n *Node) ID() peer.ID {
return n.host.ID()
}
// Addresses returns the multiaddresses this node is listening on
func (n *Node) Addresses() []multiaddr.Multiaddr {
return n.host.Addrs()
}
// Connect connects to a peer at the given multiaddress
func (n *Node) Connect(ctx context.Context, addr string) error {
ma, err := multiaddr.NewMultiaddr(addr)
if err != nil {
return fmt.Errorf("invalid multiaddress %s: %w", addr, err)
}
addrInfo, err := peer.AddrInfoFromP2pAddr(ma)
if err != nil {
return fmt.Errorf("failed to parse addr info: %w", err)
}
return n.host.Connect(ctx, *addrInfo)
}
// Peers returns the list of connected peers
func (n *Node) Peers() []peer.ID {
return n.host.Network().Peers()
}
// ConnectedPeers returns the number of connected peers
func (n *Node) ConnectedPeers() int {
return len(n.Peers())
}
// startBackgroundTasks starts background maintenance tasks
func (n *Node) startBackgroundTasks() {
ticker := time.NewTicker(30 * time.Second)
defer ticker.Stop()
for {
select {
case <-n.ctx.Done():
return
case <-ticker.C:
// Periodic maintenance tasks
n.logConnectionStatus()
}
}
}
// logConnectionStatus logs the current connection status
func (n *Node) logConnectionStatus() {
peers := n.Peers()
fmt.Printf("🐝 Bzzz Node Status - ID: %s, Connected Peers: %d\n",
n.ID().ShortString(), len(peers))
if len(peers) > 0 {
fmt.Printf(" Connected to: ")
for i, p := range peers {
if i > 0 {
fmt.Printf(", ")
}
fmt.Printf("%s", p.ShortString())
}
fmt.Println()
}
}
// DHT returns the DHT instance (if enabled)
func (n *Node) DHT() *dht.LibP2PDHT {
return n.dht
}
// IsDHTEnabled returns whether DHT is enabled and active
func (n *Node) IsDHTEnabled() bool {
return n.dht != nil
}
// Bootstrap bootstraps the DHT (if enabled)
func (n *Node) Bootstrap() error {
if n.dht != nil {
return n.dht.Bootstrap()
}
return fmt.Errorf("DHT not enabled")
}
// Close shuts down the node
func (n *Node) Close() error {
if n.dht != nil {
n.dht.Close()
}
n.cancel()
return n.host.Close()
}