segregate chain and state blockstores.

This paves the way for better object lifetime management.

Concretely, it makes it possible to:
- have different stores backing chain and state data.
- having the same datastore library, but using different parameters.
- attach different caching layers/policies to each class of data, e.g.
  sizing caches differently.
- specifying different retention policies for chain and state data.

This separation is important because:
- access patterns/frequency of chain and state data are different.
- state is derivable from chain, so one could never expunge the chain
  store, and only retain state objects reachable from the last finality
  in the state store.
This commit is contained in:
Raúl Kripalani
2021-02-28 22:49:44 +00:00
parent 853de3daf7
commit 3795cc2bd2
45 changed files with 630 additions and 370 deletions
+65
View File
@@ -0,0 +1,65 @@
package modules
import (
"context"
"io"
bstore "github.com/ipfs/go-ipfs-blockstore"
"go.uber.org/fx"
"golang.org/x/xerrors"
"github.com/filecoin-project/lotus/blockstore"
"github.com/filecoin-project/lotus/node/modules/dtypes"
"github.com/filecoin-project/lotus/node/modules/helpers"
"github.com/filecoin-project/lotus/node/repo"
)
// UniversalBlockstore returns a single universal blockstore that stores both
// chain data and state data.
func UniversalBlockstore(lc fx.Lifecycle, mctx helpers.MetricsCtx, r repo.LockedRepo) (dtypes.UniversalBlockstore, error) {
bs, err := r.Blockstore(helpers.LifecycleCtx(mctx, lc), repo.UniversalBlockstore)
if err != nil {
return nil, err
}
if c, ok := bs.(io.Closer); ok {
lc.Append(fx.Hook{
OnStop: func(_ context.Context) error {
return c.Close()
},
})
}
return bs, err
}
// StateBlockstore is a hook to overlay caches for state objects, or in the
// future, to segregate the universal blockstore into different physical state
// and chain stores.
func StateBlockstore(lc fx.Lifecycle, mctx helpers.MetricsCtx, bs dtypes.UniversalBlockstore) (dtypes.StateBlockstore, error) {
return bs, nil
}
// ChainBlockstore is a hook to overlay caches for state objects, or in the
// future, to segregate the universal blockstore into different physical state
// and chain stores.
func ChainBlockstore(lc fx.Lifecycle, mctx helpers.MetricsCtx, bs dtypes.UniversalBlockstore) (dtypes.ChainBlockstore, error) {
return bs, nil
}
func FallbackChainBlockstore(cbs dtypes.ChainBlockstore) dtypes.ChainBlockstore {
return &blockstore.FallbackStore{Blockstore: cbs}
}
func FallbackStateBlockstore(sbs dtypes.StateBlockstore) dtypes.StateBlockstore {
return &blockstore.FallbackStore{Blockstore: sbs}
}
func InitFallbackBlockstores(cbs dtypes.ChainBlockstore, sbs dtypes.StateBlockstore, rem dtypes.ChainBitswap) error {
for _, bs := range []bstore.Blockstore{cbs, sbs} {
if fbs, ok := bs.(*blockstore.FallbackStore); ok {
fbs.SetFallback(rem.GetBlock)
continue
}
return xerrors.Errorf("expected a FallbackStore")
}
return nil
}
+10 -106
View File
@@ -1,25 +1,18 @@
package modules
import (
"bytes"
"context"
"os"
"time"
"github.com/ipfs/go-bitswap"
"github.com/ipfs/go-bitswap/network"
"github.com/ipfs/go-blockservice"
"github.com/ipfs/go-datastore"
"github.com/ipld/go-car"
"github.com/libp2p/go-libp2p-core/host"
"github.com/libp2p/go-libp2p-core/routing"
pubsub "github.com/libp2p/go-libp2p-pubsub"
"go.uber.org/fx"
"golang.org/x/xerrors"
"github.com/filecoin-project/lotus/extern/sector-storage/ffiwrapper"
"github.com/filecoin-project/lotus/journal"
"github.com/filecoin-project/lotus/blockstore"
"github.com/filecoin-project/lotus/build"
"github.com/filecoin-project/lotus/chain"
@@ -29,14 +22,15 @@ import (
"github.com/filecoin-project/lotus/chain/messagepool"
"github.com/filecoin-project/lotus/chain/stmgr"
"github.com/filecoin-project/lotus/chain/store"
"github.com/filecoin-project/lotus/chain/types"
"github.com/filecoin-project/lotus/chain/vm"
"github.com/filecoin-project/lotus/extern/sector-storage/ffiwrapper"
"github.com/filecoin-project/lotus/journal"
"github.com/filecoin-project/lotus/node/modules/dtypes"
"github.com/filecoin-project/lotus/node/modules/helpers"
"github.com/filecoin-project/lotus/node/repo"
)
func ChainBitswap(mctx helpers.MetricsCtx, lc fx.Lifecycle, host host.Host, rt routing.Routing, bs dtypes.ChainBlockstore) dtypes.ChainBitswap {
// ChainBitswap uses a blockstore that bypasses all caches.
func ChainBitswap(mctx helpers.MetricsCtx, lc fx.Lifecycle, host host.Host, rt routing.Routing, bs dtypes.ExposedBlockstore) dtypes.ChainBitswap {
// prefix protocol for chain bitswap
// (so bitswap uses /chain/ipfs/bitswap/1.0.0 internally for chain sync stuff)
bitswapNetwork := network.NewFromIpfsHost(host, rt, network.Prefix("/chain"))
@@ -60,6 +54,10 @@ func ChainBitswap(mctx helpers.MetricsCtx, lc fx.Lifecycle, host host.Host, rt r
return exch
}
func ChainBlockService(bs dtypes.ExposedBlockstore, rem dtypes.ChainBitswap) dtypes.ChainBlockService {
return blockservice.New(bs, rem)
}
func MessagePool(lc fx.Lifecycle, sm *stmgr.StateManager, ps *pubsub.PubSub, ds dtypes.MetadataDS, nn dtypes.NetworkName, j journal.Journal) (*messagepool.MessagePool, error) {
mpp := messagepool.NewProvider(sm, ps)
mp, err := messagepool.New(mpp, ds, nn, j)
@@ -74,43 +72,8 @@ func MessagePool(lc fx.Lifecycle, sm *stmgr.StateManager, ps *pubsub.PubSub, ds
return mp, nil
}
func ChainRawBlockstore(lc fx.Lifecycle, mctx helpers.MetricsCtx, r repo.LockedRepo) (dtypes.ChainRawBlockstore, error) {
bs, err := r.Blockstore(helpers.LifecycleCtx(mctx, lc), repo.BlockstoreChain)
if err != nil {
return nil, err
}
// TODO potentially replace this cached blockstore by a CBOR cache.
cbs, err := blockstore.CachedBlockstore(helpers.LifecycleCtx(mctx, lc), bs, blockstore.DefaultCacheOpts())
if err != nil {
return nil, err
}
return cbs, nil
}
func ChainBlockService(bs dtypes.ChainRawBlockstore, rem dtypes.ChainBitswap) dtypes.ChainBlockService {
return blockservice.New(bs, rem)
}
func FallbackChainBlockstore(rbs dtypes.ChainRawBlockstore) dtypes.ChainBlockstore {
return &blockstore.FallbackStore{
Blockstore: rbs,
}
}
func SetupFallbackBlockstore(cbs dtypes.ChainBlockstore, rem dtypes.ChainBitswap) error {
fbs, ok := cbs.(*blockstore.FallbackStore)
if !ok {
return xerrors.Errorf("expected a FallbackStore")
}
fbs.SetFallback(rem.GetBlock)
return nil
}
func ChainStore(lc fx.Lifecycle, bs dtypes.ChainBlockstore, lbs dtypes.ChainRawBlockstore, ds dtypes.MetadataDS, syscalls vm.SyscallBuilder, j journal.Journal) *store.ChainStore {
chain := store.NewChainStore(bs, lbs, ds, syscalls, j)
func ChainStore(lc fx.Lifecycle, cbs dtypes.ChainBlockstore, sbs dtypes.StateBlockstore, ds dtypes.MetadataDS, syscalls vm.SyscallBuilder, j journal.Journal) *store.ChainStore {
chain := store.NewChainStore(cbs, sbs, ds, syscalls, j)
if err := chain.Load(); err != nil {
log.Warnf("loading chain state from disk: %s", err)
@@ -125,65 +88,6 @@ func ChainStore(lc fx.Lifecycle, bs dtypes.ChainBlockstore, lbs dtypes.ChainRawB
return chain
}
func ErrorGenesis() Genesis {
return func() (header *types.BlockHeader, e error) {
return nil, xerrors.New("No genesis block provided, provide the file with 'lotus daemon --genesis=[genesis file]'")
}
}
func LoadGenesis(genBytes []byte) func(dtypes.ChainBlockstore) Genesis {
return func(bs dtypes.ChainBlockstore) Genesis {
return func() (header *types.BlockHeader, e error) {
c, err := car.LoadCar(bs, bytes.NewReader(genBytes))
if err != nil {
return nil, xerrors.Errorf("loading genesis car file failed: %w", err)
}
if len(c.Roots) != 1 {
return nil, xerrors.New("expected genesis file to have one root")
}
root, err := bs.Get(c.Roots[0])
if err != nil {
return nil, err
}
h, err := types.DecodeBlock(root.RawData())
if err != nil {
return nil, xerrors.Errorf("decoding block failed: %w", err)
}
return h, nil
}
}
}
func DoSetGenesis(_ dtypes.AfterGenesisSet) {}
func SetGenesis(cs *store.ChainStore, g Genesis) (dtypes.AfterGenesisSet, error) {
genFromRepo, err := cs.GetGenesis()
if err == nil {
if os.Getenv("LOTUS_SKIP_GENESIS_CHECK") != "_yes_" {
expectedGenesis, err := g()
if err != nil {
return dtypes.AfterGenesisSet{}, xerrors.Errorf("getting expected genesis failed: %w", err)
}
if genFromRepo.Cid() != expectedGenesis.Cid() {
return dtypes.AfterGenesisSet{}, xerrors.Errorf("genesis in the repo is not the one expected by this version of Lotus!")
}
}
return dtypes.AfterGenesisSet{}, nil // already set, noop
}
if err != datastore.ErrNotFound {
return dtypes.AfterGenesisSet{}, xerrors.Errorf("getting genesis block failed: %w", err)
}
genesis, err := g()
if err != nil {
return dtypes.AfterGenesisSet{}, xerrors.Errorf("genesis func failed: %w", err)
}
return dtypes.AfterGenesisSet{}, cs.SetGenesis(genesis)
}
func NetworkName(mctx helpers.MetricsCtx, lc fx.Lifecycle, cs *store.ChainStore, us stmgr.UpgradeSchedule, _ dtypes.AfterGenesisSet) (dtypes.NetworkName, error) {
if !build.Devnet {
return "testnetnet", nil
+1 -1
View File
@@ -83,7 +83,7 @@ func ClientMultiDatastore(lc fx.Lifecycle, mctx helpers.MetricsCtx, r repo.Locke
ctx := helpers.LifecycleCtx(mctx, lc)
ds, err := r.Datastore(ctx, "/client")
if err != nil {
return nil, xerrors.Errorf("getting datastore out of reop: %w", err)
return nil, xerrors.Errorf("getting datastore out of repo: %w", err)
}
mds, err := multistore.NewMultiDstore(ds)
+28 -6
View File
@@ -19,19 +19,41 @@ import (
"github.com/filecoin-project/lotus/node/repo/retrievalstoremgr"
)
// MetadataDS stores metadata
// dy default it's namespaced under /metadata in main repo datastore
// MetadataDS stores metadata. By default it's namespaced under /metadata in
// main repo datastore.
type MetadataDS datastore.Batching
type ChainRawBlockstore blockstore.Blockstore
type ChainBlockstore blockstore.Blockstore // optionally bitswap backed
type (
// UniversalBlockstore is the cold blockstore.
UniversalBlockstore blockstore.Blockstore
// ChainBlockstore is a blockstore to store chain data (tipsets, blocks,
// messages). It is physically backed by the BareMonolithBlockstore, but it
// has a cache on top that is specially tuned for chain data access
// patterns.
ChainBlockstore blockstore.Blockstore
// StateBlockstore is a blockstore to store state data (state tree). It is
// physically backed by the BareMonolithBlockstore, but it has a cache on
// top that is specially tuned for state data access patterns.
StateBlockstore blockstore.Blockstore
// ExposedBlockstore is a blockstore that interfaces directly with the
// network or with users, from which queries are served, and where incoming
// data is deposited. For security reasons, this store is disconnected from
// any internal caches. If blocks are added to this store in a way that
// could render caches dirty (e.g. a block is added when an existence cache
// holds a 'false' for that block), the process should signal so by calling
// blockstore.AllCaches.Dirty(cid).
ExposedBlockstore blockstore.Blockstore
)
type ChainBitswap exchange.Interface
type ChainBlockService bserv.BlockService
type ClientMultiDstore *multistore.MultiStore
type ClientImportMgr *importmgr.Mgr
type ClientBlockstore blockstore.Blockstore
type ClientBlockstore blockstore.BasicBlockstore
type ClientDealStore *statestore.StateStore
type ClientRequestValidator *requestvalidation.UnifiedRequestValidator
type ClientDatastore datastore.Batching
@@ -50,6 +72,6 @@ type ProviderRequestValidator *requestvalidation.UnifiedRequestValidator
type ProviderDataTransfer datatransfer.Manager
type StagingDAG format.DAGService
type StagingBlockstore blockstore.Blockstore
type StagingBlockstore blockstore.BasicBlockstore
type StagingGraphsync graphsync.GraphExchange
type StagingMultiDstore *multistore.MultiStore
+73
View File
@@ -0,0 +1,73 @@
package modules
import (
"bytes"
"os"
"github.com/ipfs/go-datastore"
"github.com/ipld/go-car"
"golang.org/x/xerrors"
"github.com/filecoin-project/lotus/chain/store"
"github.com/filecoin-project/lotus/chain/types"
"github.com/filecoin-project/lotus/node/modules/dtypes"
)
func ErrorGenesis() Genesis {
return func() (header *types.BlockHeader, e error) {
return nil, xerrors.New("No genesis block provided, provide the file with 'lotus daemon --genesis=[genesis file]'")
}
}
func LoadGenesis(genBytes []byte) func(dtypes.ChainBlockstore) Genesis {
return func(bs dtypes.ChainBlockstore) Genesis {
return func() (header *types.BlockHeader, e error) {
c, err := car.LoadCar(bs, bytes.NewReader(genBytes))
if err != nil {
return nil, xerrors.Errorf("loading genesis car file failed: %w", err)
}
if len(c.Roots) != 1 {
return nil, xerrors.New("expected genesis file to have one root")
}
root, err := bs.Get(c.Roots[0])
if err != nil {
return nil, err
}
h, err := types.DecodeBlock(root.RawData())
if err != nil {
return nil, xerrors.Errorf("decoding block failed: %w", err)
}
return h, nil
}
}
}
func DoSetGenesis(_ dtypes.AfterGenesisSet) {}
func SetGenesis(cs *store.ChainStore, g Genesis) (dtypes.AfterGenesisSet, error) {
genFromRepo, err := cs.GetGenesis()
if err == nil {
if os.Getenv("LOTUS_SKIP_GENESIS_CHECK") != "_yes_" {
expectedGenesis, err := g()
if err != nil {
return dtypes.AfterGenesisSet{}, xerrors.Errorf("getting expected genesis failed: %w", err)
}
if genFromRepo.Cid() != expectedGenesis.Cid() {
return dtypes.AfterGenesisSet{}, xerrors.Errorf("genesis in the repo is not the one expected by this version of Lotus!")
}
}
return dtypes.AfterGenesisSet{}, nil // already set, noop
}
if err != datastore.ErrNotFound {
return dtypes.AfterGenesisSet{}, xerrors.Errorf("getting genesis block failed: %w", err)
}
genesis, err := g()
if err != nil {
return dtypes.AfterGenesisSet{}, xerrors.Errorf("genesis func failed: %w", err)
}
return dtypes.AfterGenesisSet{}, cs.SetGenesis(genesis)
}
+2 -2
View File
@@ -14,8 +14,8 @@ import (
)
// Graphsync creates a graphsync instance from the given loader and storer
func Graphsync(parallelTransfers uint64) func(mctx helpers.MetricsCtx, lc fx.Lifecycle, r repo.LockedRepo, clientBs dtypes.ClientBlockstore, chainBs dtypes.ChainBlockstore, h host.Host) (dtypes.Graphsync, error) {
return func(mctx helpers.MetricsCtx, lc fx.Lifecycle, r repo.LockedRepo, clientBs dtypes.ClientBlockstore, chainBs dtypes.ChainBlockstore, h host.Host) (dtypes.Graphsync, error) {
func Graphsync(parallelTransfers uint64) func(mctx helpers.MetricsCtx, lc fx.Lifecycle, r repo.LockedRepo, clientBs dtypes.ClientBlockstore, chainBs dtypes.ExposedBlockstore, h host.Host) (dtypes.Graphsync, error) {
return func(mctx helpers.MetricsCtx, lc fx.Lifecycle, r repo.LockedRepo, clientBs dtypes.ClientBlockstore, chainBs dtypes.ExposedBlockstore, h host.Host) (dtypes.Graphsync, error) {
graphsyncNetwork := gsnet.NewFromLibp2pHost(h)
loader := storeutil.LoaderForBlockstore(clientBs)
storer := storeutil.StorerForBlockstore(clientBs)
+1 -1
View File
@@ -18,7 +18,7 @@ import (
func IpfsClientBlockstore(ipfsMaddr string, onlineMode bool) func(helpers.MetricsCtx, fx.Lifecycle, dtypes.ClientImportMgr) (dtypes.ClientBlockstore, error) {
return func(mctx helpers.MetricsCtx, lc fx.Lifecycle, localStore dtypes.ClientImportMgr) (dtypes.ClientBlockstore, error) {
var err error
var ipfsbs blockstore.Blockstore
var ipfsbs blockstore.BasicBlockstore
if ipfsMaddr != "" {
var ma multiaddr.Multiaddr
ma, err = multiaddr.NewMultiaddr(ipfsMaddr)