First efficient ranged-export implementation by @frisst

This first commit contains the first and second implementation stabs (after
primary review by @hsanjuan), using a stack for task buffering.

Known issues: ctrl-c (context cancellation) results in the export code getting
deadlocked. Duplicate blocks in exports. Duplicate block reads from store.

Original commit messages:

works

works against mainnet and calibnet

feat: add internal export api method

- will hopfully make things faster by not streaming the export over the json rpc api

polish: better file nameing

fix: potential race in marking cids as seen

chore: improve logging

feat: front export with cache

fix: give hector a good channel buffer on this shit

docsgen
This commit is contained in:
frrist
2023-02-14 15:41:10 +01:00
committed by Hector Sanjuan
parent 61f29a84b5
commit 21efd481d8
13 changed files with 992 additions and 16 deletions
+4
View File
@@ -173,6 +173,10 @@ type FullNode interface {
// If oldmsgskip is set, messages from before the requested roots are also not included.
ChainExport(ctx context.Context, nroots abi.ChainEpoch, oldmsgskip bool, tsk types.TipSetKey) (<-chan []byte, error) //perm:read
ChainExportRange(ctx context.Context, head, tail types.TipSetKey, cfg *ChainExportConfig) (<-chan []byte, error) //perm:read
ChainExportRangeInternal(ctx context.Context, head, tail types.TipSetKey, cfg *ChainExportConfig) error //perm:read
// ChainPrune prunes the stored chain state and garbage collects; only supported if you
// are using the splitstore
ChainPrune(ctx context.Context, opts PruneOpts) error //perm:admin
+29
View File
@@ -155,6 +155,35 @@ func (mr *MockFullNodeMockRecorder) ChainExport(arg0, arg1, arg2, arg3 interface
return mr.mock.ctrl.RecordCallWithMethodType(mr.mock, "ChainExport", reflect.TypeOf((*MockFullNode)(nil).ChainExport), arg0, arg1, arg2, arg3)
}
// ChainExportRange mocks base method.
func (m *MockFullNode) ChainExportRange(arg0 context.Context, arg1, arg2 types.TipSetKey, arg3 *api.ChainExportConfig) (<-chan []byte, error) {
m.ctrl.T.Helper()
ret := m.ctrl.Call(m, "ChainExportRange", arg0, arg1, arg2, arg3)
ret0, _ := ret[0].(<-chan []byte)
ret1, _ := ret[1].(error)
return ret0, ret1
}
// ChainExportRange indicates an expected call of ChainExportRange.
func (mr *MockFullNodeMockRecorder) ChainExportRange(arg0, arg1, arg2, arg3 interface{}) *gomock.Call {
mr.mock.ctrl.T.Helper()
return mr.mock.ctrl.RecordCallWithMethodType(mr.mock, "ChainExportRange", reflect.TypeOf((*MockFullNode)(nil).ChainExportRange), arg0, arg1, arg2, arg3)
}
// ChainExportRangeInternal mocks base method.
func (m *MockFullNode) ChainExportRangeInternal(arg0 context.Context, arg1, arg2 types.TipSetKey, arg3 *api.ChainExportConfig) error {
m.ctrl.T.Helper()
ret := m.ctrl.Call(m, "ChainExportRangeInternal", arg0, arg1, arg2, arg3)
ret0, _ := ret[0].(error)
return ret0
}
// ChainExportRangeInternal indicates an expected call of ChainExportRangeInternal.
func (mr *MockFullNodeMockRecorder) ChainExportRangeInternal(arg0, arg1, arg2, arg3 interface{}) *gomock.Call {
mr.mock.ctrl.T.Helper()
return mr.mock.ctrl.RecordCallWithMethodType(mr.mock, "ChainExportRangeInternal", reflect.TypeOf((*MockFullNode)(nil).ChainExportRangeInternal), arg0, arg1, arg2, arg3)
}
// ChainGetBlock mocks base method.
func (m *MockFullNode) ChainGetBlock(arg0 context.Context, arg1 cid.Cid) (*types.BlockHeader, error) {
m.ctrl.T.Helper()
+34 -10
View File
@@ -7,15 +7,6 @@ import (
"encoding/json"
"time"
"github.com/google/uuid"
"github.com/ipfs/go-cid"
blocks "github.com/ipfs/go-libipfs/blocks"
"github.com/libp2p/go-libp2p/core/metrics"
"github.com/libp2p/go-libp2p/core/network"
"github.com/libp2p/go-libp2p/core/peer"
"github.com/libp2p/go-libp2p/core/protocol"
"golang.org/x/xerrors"
"github.com/filecoin-project/go-address"
"github.com/filecoin-project/go-bitfield"
datatransfer "github.com/filecoin-project/go-data-transfer"
@@ -32,7 +23,6 @@ import (
"github.com/filecoin-project/go-state-types/dline"
abinetwork "github.com/filecoin-project/go-state-types/network"
"github.com/filecoin-project/go-state-types/proof"
apitypes "github.com/filecoin-project/lotus/api/types"
builtinactors "github.com/filecoin-project/lotus/chain/actors/builtin"
lminer "github.com/filecoin-project/lotus/chain/actors/builtin/miner"
@@ -45,6 +35,14 @@ import (
"github.com/filecoin-project/lotus/storage/sealer/fsutil"
"github.com/filecoin-project/lotus/storage/sealer/sealtasks"
"github.com/filecoin-project/lotus/storage/sealer/storiface"
"github.com/google/uuid"
"github.com/ipfs/go-cid"
blocks "github.com/ipfs/go-libipfs/blocks"
"github.com/libp2p/go-libp2p/core/metrics"
"github.com/libp2p/go-libp2p/core/network"
"github.com/libp2p/go-libp2p/core/peer"
"github.com/libp2p/go-libp2p/core/protocol"
"golang.org/x/xerrors"
)
var ErrNotSupported = xerrors.New("method not supported")
@@ -140,6 +138,10 @@ type FullNodeMethods struct {
ChainExport func(p0 context.Context, p1 abi.ChainEpoch, p2 bool, p3 types.TipSetKey) (<-chan []byte, error) `perm:"read"`
ChainExportRange func(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *ChainExportConfig) (<-chan []byte, error) `perm:"read"`
ChainExportRangeInternal func(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *ChainExportConfig) error `perm:"read"`
ChainGetBlock func(p0 context.Context, p1 cid.Cid) (*types.BlockHeader, error) `perm:"read"`
ChainGetBlockMessages func(p0 context.Context, p1 cid.Cid) (*BlockMessages, error) `perm:"read"`
@@ -1435,6 +1437,28 @@ func (s *FullNodeStub) ChainExport(p0 context.Context, p1 abi.ChainEpoch, p2 boo
return nil, ErrNotSupported
}
func (s *FullNodeStruct) ChainExportRange(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *ChainExportConfig) (<-chan []byte, error) {
if s.Internal.ChainExportRange == nil {
return nil, ErrNotSupported
}
return s.Internal.ChainExportRange(p0, p1, p2, p3)
}
func (s *FullNodeStub) ChainExportRange(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *ChainExportConfig) (<-chan []byte, error) {
return nil, ErrNotSupported
}
func (s *FullNodeStruct) ChainExportRangeInternal(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *ChainExportConfig) error {
if s.Internal.ChainExportRangeInternal == nil {
return ErrNotSupported
}
return s.Internal.ChainExportRangeInternal(p0, p1, p2, p3)
}
func (s *FullNodeStub) ChainExportRangeInternal(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *ChainExportConfig) error {
return ErrNotSupported
}
func (s *FullNodeStruct) ChainGetBlock(p0 context.Context, p1 cid.Cid) (*types.BlockHeader, error) {
if s.Internal.ChainGetBlock == nil {
return nil, ErrNotSupported
+9
View File
@@ -398,3 +398,12 @@ func (m *MsgUuidMapType) UnmarshalJSON(b []byte) error {
}
return nil
}
type ChainExportConfig struct {
WriteBufferSize int
Workers int64
CacheSize int
IncludeMessages bool
IncludeReceipts bool
IncludeStateRoots bool
}
+4
View File
@@ -161,6 +161,10 @@ type FullNode interface {
// If oldmsgskip is set, messages from before the requested roots are also not included.
ChainExport(ctx context.Context, nroots abi.ChainEpoch, oldmsgskip bool, tsk types.TipSetKey) (<-chan []byte, error) //perm:read
ChainExportRange(ctx context.Context, head, tail types.TipSetKey, cfg *api.ChainExportConfig) (<-chan []byte, error) //perm:read
ChainExportRangeInternal(ctx context.Context, head, tail types.TipSetKey, cfg *api.ChainExportConfig) error //perm:read
// MethodGroup: Beacon
// The Beacon method group contains methods for interacting with the random beacon (DRAND)
+30 -6
View File
@@ -5,11 +5,6 @@ package v0api
import (
"context"
"github.com/ipfs/go-cid"
blocks "github.com/ipfs/go-libipfs/blocks"
"github.com/libp2p/go-libp2p/core/peer"
"golang.org/x/xerrors"
"github.com/filecoin-project/go-address"
"github.com/filecoin-project/go-bitfield"
datatransfer "github.com/filecoin-project/go-data-transfer"
@@ -22,7 +17,6 @@ import (
"github.com/filecoin-project/go-state-types/crypto"
"github.com/filecoin-project/go-state-types/dline"
abinetwork "github.com/filecoin-project/go-state-types/network"
"github.com/filecoin-project/lotus/api"
apitypes "github.com/filecoin-project/lotus/api/types"
lminer "github.com/filecoin-project/lotus/chain/actors/builtin/miner"
@@ -30,6 +24,10 @@ import (
marketevents "github.com/filecoin-project/lotus/markets/loggers"
"github.com/filecoin-project/lotus/node/modules/dtypes"
"github.com/filecoin-project/lotus/node/repo/imports"
"github.com/ipfs/go-cid"
blocks "github.com/ipfs/go-libipfs/blocks"
"github.com/libp2p/go-libp2p/core/peer"
"golang.org/x/xerrors"
)
var ErrNotSupported = xerrors.New("method not supported")
@@ -49,6 +47,10 @@ type FullNodeMethods struct {
ChainExport func(p0 context.Context, p1 abi.ChainEpoch, p2 bool, p3 types.TipSetKey) (<-chan []byte, error) `perm:"read"`
ChainExportRange func(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *api.ChainExportConfig) (<-chan []byte, error) `perm:"read"`
ChainExportRangeInternal func(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *api.ChainExportConfig) error `perm:"read"`
ChainGetBlock func(p0 context.Context, p1 cid.Cid) (*types.BlockHeader, error) `perm:"read"`
ChainGetBlockMessages func(p0 context.Context, p1 cid.Cid) (*api.BlockMessages, error) `perm:"read"`
@@ -532,6 +534,28 @@ func (s *FullNodeStub) ChainExport(p0 context.Context, p1 abi.ChainEpoch, p2 boo
return nil, ErrNotSupported
}
func (s *FullNodeStruct) ChainExportRange(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *api.ChainExportConfig) (<-chan []byte, error) {
if s.Internal.ChainExportRange == nil {
return nil, ErrNotSupported
}
return s.Internal.ChainExportRange(p0, p1, p2, p3)
}
func (s *FullNodeStub) ChainExportRange(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *api.ChainExportConfig) (<-chan []byte, error) {
return nil, ErrNotSupported
}
func (s *FullNodeStruct) ChainExportRangeInternal(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *api.ChainExportConfig) error {
if s.Internal.ChainExportRangeInternal == nil {
return ErrNotSupported
}
return s.Internal.ChainExportRangeInternal(p0, p1, p2, p3)
}
func (s *FullNodeStub) ChainExportRangeInternal(p0 context.Context, p1 types.TipSetKey, p2 types.TipSetKey, p3 *api.ChainExportConfig) error {
return ErrNotSupported
}
func (s *FullNodeStruct) ChainGetBlock(p0 context.Context, p1 cid.Cid) (*types.BlockHeader, error) {
if s.Internal.ChainGetBlock == nil {
return nil, ErrNotSupported
+29
View File
@@ -141,6 +141,35 @@ func (mr *MockFullNodeMockRecorder) ChainExport(arg0, arg1, arg2, arg3 interface
return mr.mock.ctrl.RecordCallWithMethodType(mr.mock, "ChainExport", reflect.TypeOf((*MockFullNode)(nil).ChainExport), arg0, arg1, arg2, arg3)
}
// ChainExportRange mocks base method.
func (m *MockFullNode) ChainExportRange(arg0 context.Context, arg1, arg2 types.TipSetKey, arg3 *api.ChainExportConfig) (<-chan []byte, error) {
m.ctrl.T.Helper()
ret := m.ctrl.Call(m, "ChainExportRange", arg0, arg1, arg2, arg3)
ret0, _ := ret[0].(<-chan []byte)
ret1, _ := ret[1].(error)
return ret0, ret1
}
// ChainExportRange indicates an expected call of ChainExportRange.
func (mr *MockFullNodeMockRecorder) ChainExportRange(arg0, arg1, arg2, arg3 interface{}) *gomock.Call {
mr.mock.ctrl.T.Helper()
return mr.mock.ctrl.RecordCallWithMethodType(mr.mock, "ChainExportRange", reflect.TypeOf((*MockFullNode)(nil).ChainExportRange), arg0, arg1, arg2, arg3)
}
// ChainExportRangeInternal mocks base method.
func (m *MockFullNode) ChainExportRangeInternal(arg0 context.Context, arg1, arg2 types.TipSetKey, arg3 *api.ChainExportConfig) error {
m.ctrl.T.Helper()
ret := m.ctrl.Call(m, "ChainExportRangeInternal", arg0, arg1, arg2, arg3)
ret0, _ := ret[0].(error)
return ret0
}
// ChainExportRangeInternal indicates an expected call of ChainExportRangeInternal.
func (mr *MockFullNodeMockRecorder) ChainExportRangeInternal(arg0, arg1, arg2, arg3 interface{}) *gomock.Call {
mr.mock.ctrl.T.Helper()
return mr.mock.ctrl.RecordCallWithMethodType(mr.mock, "ChainExportRangeInternal", reflect.TypeOf((*MockFullNode)(nil).ChainExportRangeInternal), arg0, arg1, arg2, arg3)
}
// ChainGetBlock mocks base method.
func (m *MockFullNode) ChainGetBlock(arg0 context.Context, arg1 cid.Cid) (*types.BlockHeader, error) {
m.ctrl.T.Helper()