feat(p2p/discovery)!: Implement archival discovery + syncing of his…

…toric blocks and blobs (#3188)

This PR introduces the discovery of nodes advertising on the `archival`
topic for the purpose of being able to request blobs and blocks that are
older than the sampling window.

In order to do so, there are now 2 instances of discovery (one for
`full` nodes -- _all nodes running full block sync within the sampling
window_, and one for `archival` nodes -- _nodes that sync, store and
serve all historic data_) and 2 instances of peer managers (one for
`full` and one for `archival`).

**This PR is _breaking_ due to changes to metrics names.**

Please also note that this PR introduces the ability to perform archival
sync of blobs + blocks for **LIGHT AND FULL** nodes only regardless of
whether the full nodes have pruning enabled or not. **Bridge nodes that
have previously had pruning enabled will NOT be able to request historic
data from the consensus core node.** The bridge node must run in
archival mode (without `--experimental-pruning` enabled) in order to
access historic data.

**Bridge and Full nodes in archival mode (without
`--experimental-pruning`)**
```
2024-05-10T15:21:44.490+0200	INFO	share/discovery	discovery/manager.go:30	starting discovery	{"topic": "full"}
2024-05-10T15:21:44.491+0200	INFO	share/discovery	discovery/manager.go:33	advertising to topic	{"topic": "full"}
2024-05-10T15:21:44.491+0200	INFO	share/discovery	discovery/manager.go:30	starting discovery	{"topic": "archival"}
2024-05-10T15:21:44.491+0200	INFO	share/discovery	discovery/manager.go:33	advertising to topic	{"topic": "archival"}
```


**Bridge and Full nodes running `--experimental-pruning`**
```
2024-05-10T15:19:20.698+0200	INFO	share/discovery	discovery/manager.go:30	starting discovery	{"topic": "full"}
2024-05-10T15:19:20.698+0200	INFO	share/discovery	discovery/manager.go:33	advertising to topic	{"topic": "full"}
2024-05-10T15:19:20.698+0200	INFO	share/discovery	discovery/manager.go:30	starting discovery	{"topic": "archival"}
```

**Light nodes**
```
2024-05-10T15:20:42.440+0200	INFO	share/discovery	discovery/manager.go:30	starting discovery	{"topic": "full"}
2024-05-10T15:20:42.440+0200	INFO	share/discovery	discovery/manager.go:30	starting discovery	{"topic": "archival"}
```


-----------------------------------------------------------

**TODO**
- [x] test archival request routing per node type
- [x] renaming some things inside discovery away from being
`full`-node-centric
- [x] rebase
- [x] swamp test

For follow-up PR:
- [x] clarify / make disc metrics generic as there will now be 2
instances of discovery happening

Nice to have; 
- [ ] benchmarks for LN

das/daser.go

@@ -5,7 +5,6 @@ import (
 	"errors"
 	"fmt"
 	"sync/atomic"
-	"time"
 
 	"github.com/ipfs/go-datastore"
 	logging "github.com/ipfs/go-log/v2"
@@ -14,6 +13,7 @@ import (
 	libhead "github.com/celestiaorg/go-header"
 
 	"github.com/celestiaorg/celestia-node/header"
+	"github.com/celestiaorg/celestia-node/pruner"
 	"github.com/celestiaorg/celestia-node/share"
 	"github.com/celestiaorg/celestia-node/share/eds/byzantine"
 	"github.com/celestiaorg/celestia-node/share/p2p/shrexsub"
@@ -162,7 +162,7 @@ func (d *DASer) Stop(ctx context.Context) error {
 func (d *DASer) sample(ctx context.Context, h *header.ExtendedHeader) error {
 	// short-circuit if pruning is enabled and the header is outside the
 	// availability window
-	if !d.isWithinSamplingWindow(h) {
+	if !pruner.IsWithinAvailabilityWindow(h.Time(), d.params.samplingWindow) {
 		log.Debugw("skipping header outside sampling window", "height", h.Height(),
 			"time", h.Time())
 		return errOutsideSamplingWindow
@@ -183,14 +183,6 @@ func (d *DASer) sample(ctx context.Context, h *header.ExtendedHeader) error {
 	return nil
 }
 
-func (d *DASer) isWithinSamplingWindow(eh *header.ExtendedHeader) bool {
-	// if sampling window is not set, then all headers are within the window
-	if d.params.samplingWindow.Duration() == 0 {
-		return true
-	}
-	return time.Since(eh.Time()) <= d.params.samplingWindow.Duration()
-}
-
 // SamplingStats returns the current statistics over the DA sampling process.
 func (d *DASer) SamplingStats(ctx context.Context) (SamplingStats, error) {
 	return d.sampler.stats(ctx)

das/daser_test.go

@@ -280,7 +280,11 @@ func TestDASer_SamplingWindow(t *testing.T) {
 			eh := headertest.RandExtendedHeader(t)
 			eh.RawHeader.Time = tt.timestamp
 
-			assert.Equal(t, tt.withinWindow, daser.isWithinSamplingWindow(eh))
+			assert.Equal(
+				t,
+				tt.withinWindow,
+				pruner.IsWithinAvailabilityWindow(eh.Time(), daser.params.samplingWindow),
+			)
 		})
 	}
 }

nodebuilder/node.go

@@ -30,6 +30,7 @@ import (
 	"github.com/celestiaorg/celestia-node/nodebuilder/p2p"
 	"github.com/celestiaorg/celestia-node/nodebuilder/share"
 	"github.com/celestiaorg/celestia-node/nodebuilder/state"
+	"github.com/celestiaorg/celestia-node/share/eds"
 )
 
 var (
@@ -56,6 +57,9 @@ type Node struct {
 	RPCServer     *rpc.Server     // not optional
 	GatewayServer *gateway.Server `optional:"true"`
 
+	// block store
+	EDSStore *eds.Store `optional:"true"`
+
 	// p2p components
 	Host         host.Host
 	ConnGater    *conngater.BasicConnectionGater

nodebuilder/pruner/module.go

@@ -15,31 +15,13 @@ import (
 )
 
 func ConstructModule(tp node.Type, cfg *Config) fx.Option {
-	if !cfg.EnableService {
-		switch tp {
-		case node.Light:
-			// light nodes are still subject to sampling within window
-			// even if pruning is not enabled.
-			return fx.Options(
-				fx.Supply(light.Window),
-			)
-		case node.Full:
-			return fx.Options(
-				fx.Supply(archival.Window),
-			)
-		case node.Bridge:
-			return fx.Options(
-				fx.Supply(archival.Window),
-				fx.Provide(func() []core.Option {
-					return []core.Option{}
-				}),
-			)
-		default:
-			panic("unknown node type")
-		}
-	}
-
 	baseComponents := fx.Options(
+		fx.Supply(cfg),
+		availWindow(tp, cfg.EnableService),
+	)
+
+	fullAndBridgeComponents := fx.Options(
+		baseComponents,
 		fx.Provide(fx.Annotate(
 			newPrunerService,
 			fx.OnStart(func(ctx context.Context, p *pruner.Service) error {
@@ -52,34 +34,61 @@ func ConstructModule(tp node.Type, cfg *Config) fx.Option {
 		// This is necessary to invoke the pruner service as independent thanks to a
 		// quirk in FX.
 		fx.Invoke(func(_ *pruner.Service) {}),
+		fx.Provide(func(store *eds.Store) pruner.Pruner {
+			return full.NewPruner(store)
+		}),
 	)
 
 	switch tp {
-	case node.Full:
+	// TODO: Eventually, light nodes will be capable of pruning samples
+	//  in which case, this can be enabled.
+	case node.Light:
 		return fx.Module("prune",
 			baseComponents,
-			fx.Provide(func(store *eds.Store) pruner.Pruner {
-				return full.NewPruner(store)
-			}),
-			fx.Supply(full.Window),
+		)
+	case node.Full:
+		opts := baseComponents
+		if cfg.EnableService {
+			opts = fullAndBridgeComponents
+		}
+		return fx.Module("prune",
+			opts,
 		)
 	case node.Bridge:
+		if cfg.EnableService {
+			return fx.Module("prune",
+				fullAndBridgeComponents,
+				fx.Provide(func(window pruner.AvailabilityWindow) []core.Option {
+					return []core.Option{core.WithAvailabilityWindow(window)}
+				}),
+			)
+		}
 		return fx.Module("prune",
 			baseComponents,
-			fx.Provide(func(store *eds.Store) pruner.Pruner {
-				return full.NewPruner(store)
-			}),
-			fx.Supply(full.Window),
-			fx.Provide(func(window pruner.AvailabilityWindow) []core.Option {
-				return []core.Option{core.WithAvailabilityWindow(window)}
+			fx.Provide(func() []core.Option {
+				return []core.Option{}
 			}),
 		)
-	// TODO: Eventually, light nodes will be capable of pruning samples
-	//  in which case, this can be enabled.
+	default:
+		panic("unknown node type")
+	}
+}
+
+func availWindow(tp node.Type, pruneEnabled bool) fx.Option {
+	switch tp {
 	case node.Light:
-		return fx.Module("prune",
-			fx.Supply(light.Window),
-		)
+		// light nodes are still subject to sampling within window
+		// even if pruning is not enabled.
+		return fx.Provide(func() pruner.AvailabilityWindow {
+			return light.Window
+		})
+	case node.Full, node.Bridge:
+		return fx.Provide(func() pruner.AvailabilityWindow {
+			if pruneEnabled {
+				return full.Window
+			}
+			return archival.Window
+		})
 	default:
 		panic("unknown node type")
 	}

nodebuilder/share/constructors.go

@@ -6,42 +6,15 @@ import (
 
 	"github.com/filecoin-project/dagstore"
 	"github.com/ipfs/boxo/blockservice"
-	"github.com/libp2p/go-libp2p/core/host"
-	"github.com/libp2p/go-libp2p/core/routing"
-	routingdisc "github.com/libp2p/go-libp2p/p2p/discovery/routing"
 
 	"github.com/celestiaorg/celestia-app/pkg/da"
 
 	"github.com/celestiaorg/celestia-node/share"
 	"github.com/celestiaorg/celestia-node/share/eds"
 	"github.com/celestiaorg/celestia-node/share/getters"
 	"github.com/celestiaorg/celestia-node/share/ipld"
-	disc "github.com/celestiaorg/celestia-node/share/p2p/discovery"
-	"github.com/celestiaorg/celestia-node/share/p2p/peers"
 )
 
-const (
-	// fullNodesTag is the tag used to identify full nodes in the discovery service.
-	fullNodesTag = "full"
-)
-
-func newDiscovery(cfg *disc.Parameters,
-) func(routing.ContentRouting, host.Host, *peers.Manager) (*disc.Discovery, error) {
-	return func(
-		r routing.ContentRouting,
-		h host.Host,
-		manager *peers.Manager,
-	) (*disc.Discovery, error) {
-		return disc.NewDiscovery(
-			cfg,
-			h,
-			routingdisc.NewRoutingDiscovery(r),
-			fullNodesTag,
-			disc.WithOnPeersUpdate(manager.UpdateNodePool),
-		)
-	}
-}
-
 func newShareModule(getter share.Getter, avail share.Availability) Module {
 	return &module{getter, avail}
 }