2022-08-29 14:25:30 +00:00
|
|
|
// stm: #unit
|
2022-03-21 09:37:35 +00:00
|
|
|
package unixfs
|
integrate DAG store and CARv2 in deal-making (#6671)
This commit removes badger from the deal-making processes, and
moves to a new architecture with the dagstore as the cental
component on the miner-side, and CARv2s on the client-side.
Every deal that has been handed off to the sealing subsystem becomes
a shard in the dagstore. Shards are mounted via the LotusMount, which
teaches the dagstore how to load the related piece when serving
retrievals.
When the miner starts the Lotus for the first time with this patch,
we will perform a one-time migration of all active deals into the
dagstore. This is a lightweight process, and it consists simply
of registering the shards in the dagstore.
Shards are backed by the unsealed copy of the piece. This is currently
a CARv1. However, the dagstore keeps CARv2 indices for all pieces, so
when it's time to acquire a shard to serve a retrieval, the unsealed
CARv1 is joined with its index (safeguarded by the dagstore), to form
a read-only blockstore, thus taking the place of the monolithic
badger.
Data transfers have been adjusted to interface directly with CARv2 files.
On inbound transfers (client retrievals, miner storage deals), we stream
the received data into a CARv2 ReadWrite blockstore. On outbound transfers
(client storage deals, miner retrievals), we serve the data off a CARv2
ReadOnly blockstore.
Client-side imports are managed by the refactored *imports.Manager
component (when not using IPFS integration). Just like it before, we use
the go-filestore library to avoid duplicating the data from the original
file in the resulting UnixFS DAG (concretely the leaves). However, the
target of those imports are what we call "ref-CARv2s": CARv2 files placed
under the `$LOTUS_PATH/imports` directory, containing the intermediate
nodes in full, and the leaves as positional references to the original file
on disk.
Client-side retrievals are placed into CARv2 files in the location:
`$LOTUS_PATH/retrievals`.
A new set of `Dagstore*` JSON-RPC operations and `lotus-miner dagstore`
subcommands have been introduced on the miner-side to inspect and manage
the dagstore.
Despite moving to a CARv2-backed system, the IPFS integration has been
respected, and it continues to be possible to make storage deals with data
held in an IPFS node, and to perform retrievals directly into an IPFS node.
NOTE: because the "staging" and "client" Badger blockstores are no longer
used, existing imports on the client will be rendered useless. On startup,
Lotus will enumerate all imports and print WARN statements on the log for
each import that needs to be reimported. These log lines contain these
messages:
- import lacks carv2 path; import will not work; please reimport
- import has missing/broken carv2; please reimport
At the end, we will print a "sanity check completed" message indicating
the count of imports found, and how many were deemed broken.
Co-authored-by: Aarsh Shah <aarshkshah1992@gmail.com>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
Co-authored-by: Raúl Kripalani <raul@protocol.ai>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
2021-08-16 22:34:32 +00:00
|
|
|
|
|
|
|
import (
|
|
|
|
"bytes"
|
|
|
|
"context"
|
|
|
|
"io"
|
|
|
|
"io/ioutil"
|
|
|
|
"os"
|
|
|
|
"strings"
|
|
|
|
"testing"
|
|
|
|
|
|
|
|
"github.com/ipfs/go-blockservice"
|
|
|
|
"github.com/ipfs/go-cid"
|
|
|
|
offline "github.com/ipfs/go-ipfs-exchange-offline"
|
|
|
|
files "github.com/ipfs/go-ipfs-files"
|
|
|
|
"github.com/ipfs/go-merkledag"
|
|
|
|
unixfile "github.com/ipfs/go-unixfs/file"
|
|
|
|
carv2 "github.com/ipld/go-car/v2"
|
|
|
|
"github.com/ipld/go-car/v2/blockstore"
|
|
|
|
"github.com/stretchr/testify/require"
|
|
|
|
|
|
|
|
"github.com/filecoin-project/go-fil-markets/stores"
|
|
|
|
)
|
|
|
|
|
|
|
|
// This test uses a full "dense" CARv2, and not a filestore (positional mapping).
|
|
|
|
func TestRoundtripUnixFS_Dense(t *testing.T) {
|
feat: Add additional test annotations (#8272)
* Annotate api,proxy_util,blockstore_badger, policy tests
* Annotate splitstore: bsbadger / markset
* Annotate splitstore feature
* Annotate union/timed blockstore tests
* Annotate openrpc, diff_adt tests
* Annotate error,drand,events tests
* Annotate predicates_test
* Fix annotations
* Annotate tscache, gen tests
* Annotate fundmanager test
* Annotate repub and selection tests
* Annotate statetree_test
* Annotate forks_test
* Annotate searchwait_test.go
* Fix duplicated @@ symbols
* Annotate chain stmgr/store tests
* Annotate more (types) tests
* More tests annotated
* Annotate conformance chaos actor tests
* Annotate more integration tests
* Annotate journal system tests
* Annotate more tests.
* Annotate gas,head buffer behaviors
* Fix markset annotations
* doc: test annotations for the markets dagstore wrapper
* Annotate miner_api test in dagstore
* Annotate more test files
* Remove bad annotations from fsrepo
* Annotate wdpost system
* Remove bad annotations
* Renamce "conformance" to "chaos_actor" tests
* doc: stm annotations for blockheader & election proof tests
* Annotate remaining "A" tests
* annotate: stm for error_test
* memrepo_test.go
* Annotate "b" file tests
* message_test.go
* doc: stm annotate for fsrepo_test
* Annotate "c" file tests
* Annotate "D" test files
* message_test.go
* doc: stm annotate for chain, node/config & client
* docs: stm annotate node_test
* Annotate u,v,wl tests
* doc: stm annotations for various test files
* Annotate "T" test files
* doc: stm annotate for proxy_util_test & policy_test
* doc: stm annotate for various tests
* doc: final few stm annotations
* Add mempool unit tests
* Add two more memPool Add tests
* Update submodules
* Add check function tests
* Add stm annotations, refactor test helper
* Annotate api,proxy_util,blockstore_badger, policy tests
* Annotate splitstore: bsbadger / markset
solving merge conflicts
* Annotate splitstore feature
* Annotate union/timed blockstore tests
* Annotate openrpc, diff_adt tests
* Annotate error,drand,events tests
* Annotate predicates_test
* Fix annotations
* Annotate tscache, gen tests
* Annotate fundmanager test
* Annotate statetree_test
* Annotate forks_test
* Annotate searchwait_test.go
* Fix duplicated @@ symbols
* Annotate chain stmgr/store tests
* Annotate more (types) tests
* More tests annotated
* Annotate conformance chaos actor tests
* Annotate more integration tests
* Annotate journal system tests
* Annotate more tests.
* Annotate gas,head buffer behaviors
solve merge conflict
* Fix markset annotations
* Annotate miner_api test in dagstore
* Annotate more test files
* doc: test annotations for the markets dagstore wrapper
* Annotate wdpost system
* Renamce "conformance" to "chaos_actor" tests
* Annotate remaining "A" tests
* doc: stm annotations for blockheader & election proof tests
* annotate: stm for error_test
* Annotate "b" file tests
* memrepo_test.go
* Annotate "c" file tests
* message_test.go
* Annotate "D" test files
* doc: stm annotate for fsrepo_test
* Annotate u,v,wl tests
* message_test.go
* doc: stm annotate for chain, node/config & client
* docs: stm annotate node_test
* Annotate "T" test files
* doc: stm annotations for various test files
* Add mempool unit tests
solve merge conflict
* doc: stm annotate for proxy_util_test & policy_test
* doc: stm annotate for various tests
* doc: final few stm annotations
* Add two more memPool Add tests
* Update submodules
* Add check function tests
solve conflict
* Add stm annotations, refactor test helper
solve merge conflict
* Change CLI test kinds to "unit"
* Fix double merged test
* Fix ccupgrade_test merge
* Fix lint issues
* Add stm annotation to types_Test
* Test vectors submodule
* Add file annotation to burn_test
Co-authored-by: Nikola Divic <divicnikola@gmail.com>
Co-authored-by: TheMenko <themenkoprojects@gmail.com>
2022-03-16 17:37:34 +00:00
|
|
|
//stm: @CLIENT_DATA_IMPORT_002
|
integrate DAG store and CARv2 in deal-making (#6671)
This commit removes badger from the deal-making processes, and
moves to a new architecture with the dagstore as the cental
component on the miner-side, and CARv2s on the client-side.
Every deal that has been handed off to the sealing subsystem becomes
a shard in the dagstore. Shards are mounted via the LotusMount, which
teaches the dagstore how to load the related piece when serving
retrievals.
When the miner starts the Lotus for the first time with this patch,
we will perform a one-time migration of all active deals into the
dagstore. This is a lightweight process, and it consists simply
of registering the shards in the dagstore.
Shards are backed by the unsealed copy of the piece. This is currently
a CARv1. However, the dagstore keeps CARv2 indices for all pieces, so
when it's time to acquire a shard to serve a retrieval, the unsealed
CARv1 is joined with its index (safeguarded by the dagstore), to form
a read-only blockstore, thus taking the place of the monolithic
badger.
Data transfers have been adjusted to interface directly with CARv2 files.
On inbound transfers (client retrievals, miner storage deals), we stream
the received data into a CARv2 ReadWrite blockstore. On outbound transfers
(client storage deals, miner retrievals), we serve the data off a CARv2
ReadOnly blockstore.
Client-side imports are managed by the refactored *imports.Manager
component (when not using IPFS integration). Just like it before, we use
the go-filestore library to avoid duplicating the data from the original
file in the resulting UnixFS DAG (concretely the leaves). However, the
target of those imports are what we call "ref-CARv2s": CARv2 files placed
under the `$LOTUS_PATH/imports` directory, containing the intermediate
nodes in full, and the leaves as positional references to the original file
on disk.
Client-side retrievals are placed into CARv2 files in the location:
`$LOTUS_PATH/retrievals`.
A new set of `Dagstore*` JSON-RPC operations and `lotus-miner dagstore`
subcommands have been introduced on the miner-side to inspect and manage
the dagstore.
Despite moving to a CARv2-backed system, the IPFS integration has been
respected, and it continues to be possible to make storage deals with data
held in an IPFS node, and to perform retrievals directly into an IPFS node.
NOTE: because the "staging" and "client" Badger blockstores are no longer
used, existing imports on the client will be rendered useless. On startup,
Lotus will enumerate all imports and print WARN statements on the log for
each import that needs to be reimported. These log lines contain these
messages:
- import lacks carv2 path; import will not work; please reimport
- import has missing/broken carv2; please reimport
At the end, we will print a "sanity check completed" message indicating
the count of imports found, and how many were deemed broken.
Co-authored-by: Aarsh Shah <aarshkshah1992@gmail.com>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
Co-authored-by: Raúl Kripalani <raul@protocol.ai>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
2021-08-16 22:34:32 +00:00
|
|
|
ctx := context.Background()
|
|
|
|
|
|
|
|
inputPath, inputContents := genInputFile(t)
|
|
|
|
defer os.Remove(inputPath) //nolint:errcheck
|
|
|
|
|
|
|
|
carv2File := newTmpFile(t)
|
|
|
|
defer os.Remove(carv2File) //nolint:errcheck
|
|
|
|
|
|
|
|
// import a file to a Unixfs DAG using a CARv2 read/write blockstore.
|
|
|
|
bs, err := blockstore.OpenReadWrite(carv2File, nil,
|
|
|
|
carv2.ZeroLengthSectionAsEOF(true),
|
|
|
|
blockstore.UseWholeCIDs(true))
|
|
|
|
require.NoError(t, err)
|
|
|
|
|
2022-03-21 09:37:35 +00:00
|
|
|
root, err := Build(ctx, bytes.NewBuffer(inputContents), bs, false)
|
integrate DAG store and CARv2 in deal-making (#6671)
This commit removes badger from the deal-making processes, and
moves to a new architecture with the dagstore as the cental
component on the miner-side, and CARv2s on the client-side.
Every deal that has been handed off to the sealing subsystem becomes
a shard in the dagstore. Shards are mounted via the LotusMount, which
teaches the dagstore how to load the related piece when serving
retrievals.
When the miner starts the Lotus for the first time with this patch,
we will perform a one-time migration of all active deals into the
dagstore. This is a lightweight process, and it consists simply
of registering the shards in the dagstore.
Shards are backed by the unsealed copy of the piece. This is currently
a CARv1. However, the dagstore keeps CARv2 indices for all pieces, so
when it's time to acquire a shard to serve a retrieval, the unsealed
CARv1 is joined with its index (safeguarded by the dagstore), to form
a read-only blockstore, thus taking the place of the monolithic
badger.
Data transfers have been adjusted to interface directly with CARv2 files.
On inbound transfers (client retrievals, miner storage deals), we stream
the received data into a CARv2 ReadWrite blockstore. On outbound transfers
(client storage deals, miner retrievals), we serve the data off a CARv2
ReadOnly blockstore.
Client-side imports are managed by the refactored *imports.Manager
component (when not using IPFS integration). Just like it before, we use
the go-filestore library to avoid duplicating the data from the original
file in the resulting UnixFS DAG (concretely the leaves). However, the
target of those imports are what we call "ref-CARv2s": CARv2 files placed
under the `$LOTUS_PATH/imports` directory, containing the intermediate
nodes in full, and the leaves as positional references to the original file
on disk.
Client-side retrievals are placed into CARv2 files in the location:
`$LOTUS_PATH/retrievals`.
A new set of `Dagstore*` JSON-RPC operations and `lotus-miner dagstore`
subcommands have been introduced on the miner-side to inspect and manage
the dagstore.
Despite moving to a CARv2-backed system, the IPFS integration has been
respected, and it continues to be possible to make storage deals with data
held in an IPFS node, and to perform retrievals directly into an IPFS node.
NOTE: because the "staging" and "client" Badger blockstores are no longer
used, existing imports on the client will be rendered useless. On startup,
Lotus will enumerate all imports and print WARN statements on the log for
each import that needs to be reimported. These log lines contain these
messages:
- import lacks carv2 path; import will not work; please reimport
- import has missing/broken carv2; please reimport
At the end, we will print a "sanity check completed" message indicating
the count of imports found, and how many were deemed broken.
Co-authored-by: Aarsh Shah <aarshkshah1992@gmail.com>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
Co-authored-by: Raúl Kripalani <raul@protocol.ai>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
2021-08-16 22:34:32 +00:00
|
|
|
require.NoError(t, err)
|
|
|
|
require.NotEqual(t, cid.Undef, root)
|
|
|
|
require.NoError(t, bs.Finalize())
|
|
|
|
|
|
|
|
// reconstruct the file.
|
|
|
|
readOnly, err := blockstore.OpenReadOnly(carv2File,
|
|
|
|
carv2.ZeroLengthSectionAsEOF(true),
|
|
|
|
blockstore.UseWholeCIDs(true))
|
|
|
|
require.NoError(t, err)
|
|
|
|
defer readOnly.Close() //nolint:errcheck
|
|
|
|
|
|
|
|
dags := merkledag.NewDAGService(blockservice.New(readOnly, offline.Exchange(readOnly)))
|
|
|
|
|
|
|
|
nd, err := dags.Get(ctx, root)
|
|
|
|
require.NoError(t, err)
|
|
|
|
|
|
|
|
file, err := unixfile.NewUnixfsFile(ctx, dags, nd)
|
|
|
|
require.NoError(t, err)
|
|
|
|
|
|
|
|
tmpOutput := newTmpFile(t)
|
|
|
|
defer os.Remove(tmpOutput) //nolint:errcheck
|
|
|
|
require.NoError(t, files.WriteTo(file, tmpOutput))
|
|
|
|
|
|
|
|
// ensure contents of the initial input file and the output file are identical.
|
|
|
|
fo, err := os.Open(tmpOutput)
|
|
|
|
require.NoError(t, err)
|
|
|
|
bz2, err := ioutil.ReadAll(fo)
|
|
|
|
require.NoError(t, err)
|
|
|
|
require.NoError(t, fo.Close())
|
|
|
|
require.Equal(t, inputContents, bz2)
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestRoundtripUnixFS_Filestore(t *testing.T) {
|
feat: Add additional test annotations (#8272)
* Annotate api,proxy_util,blockstore_badger, policy tests
* Annotate splitstore: bsbadger / markset
* Annotate splitstore feature
* Annotate union/timed blockstore tests
* Annotate openrpc, diff_adt tests
* Annotate error,drand,events tests
* Annotate predicates_test
* Fix annotations
* Annotate tscache, gen tests
* Annotate fundmanager test
* Annotate repub and selection tests
* Annotate statetree_test
* Annotate forks_test
* Annotate searchwait_test.go
* Fix duplicated @@ symbols
* Annotate chain stmgr/store tests
* Annotate more (types) tests
* More tests annotated
* Annotate conformance chaos actor tests
* Annotate more integration tests
* Annotate journal system tests
* Annotate more tests.
* Annotate gas,head buffer behaviors
* Fix markset annotations
* doc: test annotations for the markets dagstore wrapper
* Annotate miner_api test in dagstore
* Annotate more test files
* Remove bad annotations from fsrepo
* Annotate wdpost system
* Remove bad annotations
* Renamce "conformance" to "chaos_actor" tests
* doc: stm annotations for blockheader & election proof tests
* Annotate remaining "A" tests
* annotate: stm for error_test
* memrepo_test.go
* Annotate "b" file tests
* message_test.go
* doc: stm annotate for fsrepo_test
* Annotate "c" file tests
* Annotate "D" test files
* message_test.go
* doc: stm annotate for chain, node/config & client
* docs: stm annotate node_test
* Annotate u,v,wl tests
* doc: stm annotations for various test files
* Annotate "T" test files
* doc: stm annotate for proxy_util_test & policy_test
* doc: stm annotate for various tests
* doc: final few stm annotations
* Add mempool unit tests
* Add two more memPool Add tests
* Update submodules
* Add check function tests
* Add stm annotations, refactor test helper
* Annotate api,proxy_util,blockstore_badger, policy tests
* Annotate splitstore: bsbadger / markset
solving merge conflicts
* Annotate splitstore feature
* Annotate union/timed blockstore tests
* Annotate openrpc, diff_adt tests
* Annotate error,drand,events tests
* Annotate predicates_test
* Fix annotations
* Annotate tscache, gen tests
* Annotate fundmanager test
* Annotate statetree_test
* Annotate forks_test
* Annotate searchwait_test.go
* Fix duplicated @@ symbols
* Annotate chain stmgr/store tests
* Annotate more (types) tests
* More tests annotated
* Annotate conformance chaos actor tests
* Annotate more integration tests
* Annotate journal system tests
* Annotate more tests.
* Annotate gas,head buffer behaviors
solve merge conflict
* Fix markset annotations
* Annotate miner_api test in dagstore
* Annotate more test files
* doc: test annotations for the markets dagstore wrapper
* Annotate wdpost system
* Renamce "conformance" to "chaos_actor" tests
* Annotate remaining "A" tests
* doc: stm annotations for blockheader & election proof tests
* annotate: stm for error_test
* Annotate "b" file tests
* memrepo_test.go
* Annotate "c" file tests
* message_test.go
* Annotate "D" test files
* doc: stm annotate for fsrepo_test
* Annotate u,v,wl tests
* message_test.go
* doc: stm annotate for chain, node/config & client
* docs: stm annotate node_test
* Annotate "T" test files
* doc: stm annotations for various test files
* Add mempool unit tests
solve merge conflict
* doc: stm annotate for proxy_util_test & policy_test
* doc: stm annotate for various tests
* doc: final few stm annotations
* Add two more memPool Add tests
* Update submodules
* Add check function tests
solve conflict
* Add stm annotations, refactor test helper
solve merge conflict
* Change CLI test kinds to "unit"
* Fix double merged test
* Fix ccupgrade_test merge
* Fix lint issues
* Add stm annotation to types_Test
* Test vectors submodule
* Add file annotation to burn_test
Co-authored-by: Nikola Divic <divicnikola@gmail.com>
Co-authored-by: TheMenko <themenkoprojects@gmail.com>
2022-03-16 17:37:34 +00:00
|
|
|
//stm: @CLIENT_DATA_IMPORT_001
|
integrate DAG store and CARv2 in deal-making (#6671)
This commit removes badger from the deal-making processes, and
moves to a new architecture with the dagstore as the cental
component on the miner-side, and CARv2s on the client-side.
Every deal that has been handed off to the sealing subsystem becomes
a shard in the dagstore. Shards are mounted via the LotusMount, which
teaches the dagstore how to load the related piece when serving
retrievals.
When the miner starts the Lotus for the first time with this patch,
we will perform a one-time migration of all active deals into the
dagstore. This is a lightweight process, and it consists simply
of registering the shards in the dagstore.
Shards are backed by the unsealed copy of the piece. This is currently
a CARv1. However, the dagstore keeps CARv2 indices for all pieces, so
when it's time to acquire a shard to serve a retrieval, the unsealed
CARv1 is joined with its index (safeguarded by the dagstore), to form
a read-only blockstore, thus taking the place of the monolithic
badger.
Data transfers have been adjusted to interface directly with CARv2 files.
On inbound transfers (client retrievals, miner storage deals), we stream
the received data into a CARv2 ReadWrite blockstore. On outbound transfers
(client storage deals, miner retrievals), we serve the data off a CARv2
ReadOnly blockstore.
Client-side imports are managed by the refactored *imports.Manager
component (when not using IPFS integration). Just like it before, we use
the go-filestore library to avoid duplicating the data from the original
file in the resulting UnixFS DAG (concretely the leaves). However, the
target of those imports are what we call "ref-CARv2s": CARv2 files placed
under the `$LOTUS_PATH/imports` directory, containing the intermediate
nodes in full, and the leaves as positional references to the original file
on disk.
Client-side retrievals are placed into CARv2 files in the location:
`$LOTUS_PATH/retrievals`.
A new set of `Dagstore*` JSON-RPC operations and `lotus-miner dagstore`
subcommands have been introduced on the miner-side to inspect and manage
the dagstore.
Despite moving to a CARv2-backed system, the IPFS integration has been
respected, and it continues to be possible to make storage deals with data
held in an IPFS node, and to perform retrievals directly into an IPFS node.
NOTE: because the "staging" and "client" Badger blockstores are no longer
used, existing imports on the client will be rendered useless. On startup,
Lotus will enumerate all imports and print WARN statements on the log for
each import that needs to be reimported. These log lines contain these
messages:
- import lacks carv2 path; import will not work; please reimport
- import has missing/broken carv2; please reimport
At the end, we will print a "sanity check completed" message indicating
the count of imports found, and how many were deemed broken.
Co-authored-by: Aarsh Shah <aarshkshah1992@gmail.com>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
Co-authored-by: Raúl Kripalani <raul@protocol.ai>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
2021-08-16 22:34:32 +00:00
|
|
|
ctx := context.Background()
|
|
|
|
|
|
|
|
inputPath, inputContents := genInputFile(t)
|
|
|
|
defer os.Remove(inputPath) //nolint:errcheck
|
|
|
|
|
|
|
|
dst := newTmpFile(t)
|
|
|
|
defer os.Remove(dst) //nolint:errcheck
|
|
|
|
|
2022-03-21 09:37:35 +00:00
|
|
|
root, err := CreateFilestore(ctx, inputPath, dst)
|
integrate DAG store and CARv2 in deal-making (#6671)
This commit removes badger from the deal-making processes, and
moves to a new architecture with the dagstore as the cental
component on the miner-side, and CARv2s on the client-side.
Every deal that has been handed off to the sealing subsystem becomes
a shard in the dagstore. Shards are mounted via the LotusMount, which
teaches the dagstore how to load the related piece when serving
retrievals.
When the miner starts the Lotus for the first time with this patch,
we will perform a one-time migration of all active deals into the
dagstore. This is a lightweight process, and it consists simply
of registering the shards in the dagstore.
Shards are backed by the unsealed copy of the piece. This is currently
a CARv1. However, the dagstore keeps CARv2 indices for all pieces, so
when it's time to acquire a shard to serve a retrieval, the unsealed
CARv1 is joined with its index (safeguarded by the dagstore), to form
a read-only blockstore, thus taking the place of the monolithic
badger.
Data transfers have been adjusted to interface directly with CARv2 files.
On inbound transfers (client retrievals, miner storage deals), we stream
the received data into a CARv2 ReadWrite blockstore. On outbound transfers
(client storage deals, miner retrievals), we serve the data off a CARv2
ReadOnly blockstore.
Client-side imports are managed by the refactored *imports.Manager
component (when not using IPFS integration). Just like it before, we use
the go-filestore library to avoid duplicating the data from the original
file in the resulting UnixFS DAG (concretely the leaves). However, the
target of those imports are what we call "ref-CARv2s": CARv2 files placed
under the `$LOTUS_PATH/imports` directory, containing the intermediate
nodes in full, and the leaves as positional references to the original file
on disk.
Client-side retrievals are placed into CARv2 files in the location:
`$LOTUS_PATH/retrievals`.
A new set of `Dagstore*` JSON-RPC operations and `lotus-miner dagstore`
subcommands have been introduced on the miner-side to inspect and manage
the dagstore.
Despite moving to a CARv2-backed system, the IPFS integration has been
respected, and it continues to be possible to make storage deals with data
held in an IPFS node, and to perform retrievals directly into an IPFS node.
NOTE: because the "staging" and "client" Badger blockstores are no longer
used, existing imports on the client will be rendered useless. On startup,
Lotus will enumerate all imports and print WARN statements on the log for
each import that needs to be reimported. These log lines contain these
messages:
- import lacks carv2 path; import will not work; please reimport
- import has missing/broken carv2; please reimport
At the end, we will print a "sanity check completed" message indicating
the count of imports found, and how many were deemed broken.
Co-authored-by: Aarsh Shah <aarshkshah1992@gmail.com>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
Co-authored-by: Raúl Kripalani <raul@protocol.ai>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
2021-08-16 22:34:32 +00:00
|
|
|
require.NoError(t, err)
|
|
|
|
require.NotEqual(t, cid.Undef, root)
|
|
|
|
|
|
|
|
// convert the CARv2 to a normal file again and ensure the contents match
|
|
|
|
fs, err := stores.ReadOnlyFilestore(dst)
|
|
|
|
require.NoError(t, err)
|
|
|
|
defer fs.Close() //nolint:errcheck
|
|
|
|
|
|
|
|
dags := merkledag.NewDAGService(blockservice.New(fs, offline.Exchange(fs)))
|
|
|
|
|
|
|
|
nd, err := dags.Get(ctx, root)
|
|
|
|
require.NoError(t, err)
|
|
|
|
|
|
|
|
file, err := unixfile.NewUnixfsFile(ctx, dags, nd)
|
|
|
|
require.NoError(t, err)
|
|
|
|
|
|
|
|
tmpOutput := newTmpFile(t)
|
|
|
|
defer os.Remove(tmpOutput) //nolint:errcheck
|
|
|
|
require.NoError(t, files.WriteTo(file, tmpOutput))
|
|
|
|
|
|
|
|
// ensure contents of the initial input file and the output file are identical.
|
|
|
|
fo, err := os.Open(tmpOutput)
|
|
|
|
require.NoError(t, err)
|
|
|
|
bz2, err := ioutil.ReadAll(fo)
|
|
|
|
require.NoError(t, err)
|
|
|
|
require.NoError(t, fo.Close())
|
|
|
|
require.Equal(t, inputContents, bz2)
|
|
|
|
}
|
|
|
|
|
2022-06-29 17:07:29 +00:00
|
|
|
// creates a new tempdir each time, guaranteeing uniqueness
|
integrate DAG store and CARv2 in deal-making (#6671)
This commit removes badger from the deal-making processes, and
moves to a new architecture with the dagstore as the cental
component on the miner-side, and CARv2s on the client-side.
Every deal that has been handed off to the sealing subsystem becomes
a shard in the dagstore. Shards are mounted via the LotusMount, which
teaches the dagstore how to load the related piece when serving
retrievals.
When the miner starts the Lotus for the first time with this patch,
we will perform a one-time migration of all active deals into the
dagstore. This is a lightweight process, and it consists simply
of registering the shards in the dagstore.
Shards are backed by the unsealed copy of the piece. This is currently
a CARv1. However, the dagstore keeps CARv2 indices for all pieces, so
when it's time to acquire a shard to serve a retrieval, the unsealed
CARv1 is joined with its index (safeguarded by the dagstore), to form
a read-only blockstore, thus taking the place of the monolithic
badger.
Data transfers have been adjusted to interface directly with CARv2 files.
On inbound transfers (client retrievals, miner storage deals), we stream
the received data into a CARv2 ReadWrite blockstore. On outbound transfers
(client storage deals, miner retrievals), we serve the data off a CARv2
ReadOnly blockstore.
Client-side imports are managed by the refactored *imports.Manager
component (when not using IPFS integration). Just like it before, we use
the go-filestore library to avoid duplicating the data from the original
file in the resulting UnixFS DAG (concretely the leaves). However, the
target of those imports are what we call "ref-CARv2s": CARv2 files placed
under the `$LOTUS_PATH/imports` directory, containing the intermediate
nodes in full, and the leaves as positional references to the original file
on disk.
Client-side retrievals are placed into CARv2 files in the location:
`$LOTUS_PATH/retrievals`.
A new set of `Dagstore*` JSON-RPC operations and `lotus-miner dagstore`
subcommands have been introduced on the miner-side to inspect and manage
the dagstore.
Despite moving to a CARv2-backed system, the IPFS integration has been
respected, and it continues to be possible to make storage deals with data
held in an IPFS node, and to perform retrievals directly into an IPFS node.
NOTE: because the "staging" and "client" Badger blockstores are no longer
used, existing imports on the client will be rendered useless. On startup,
Lotus will enumerate all imports and print WARN statements on the log for
each import that needs to be reimported. These log lines contain these
messages:
- import lacks carv2 path; import will not work; please reimport
- import has missing/broken carv2; please reimport
At the end, we will print a "sanity check completed" message indicating
the count of imports found, and how many were deemed broken.
Co-authored-by: Aarsh Shah <aarshkshah1992@gmail.com>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
Co-authored-by: Raúl Kripalani <raul@protocol.ai>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
2021-08-16 22:34:32 +00:00
|
|
|
func newTmpFile(t *testing.T) string {
|
2022-06-29 17:07:29 +00:00
|
|
|
return t.TempDir() + string(os.PathSeparator) + "tmp"
|
integrate DAG store and CARv2 in deal-making (#6671)
This commit removes badger from the deal-making processes, and
moves to a new architecture with the dagstore as the cental
component on the miner-side, and CARv2s on the client-side.
Every deal that has been handed off to the sealing subsystem becomes
a shard in the dagstore. Shards are mounted via the LotusMount, which
teaches the dagstore how to load the related piece when serving
retrievals.
When the miner starts the Lotus for the first time with this patch,
we will perform a one-time migration of all active deals into the
dagstore. This is a lightweight process, and it consists simply
of registering the shards in the dagstore.
Shards are backed by the unsealed copy of the piece. This is currently
a CARv1. However, the dagstore keeps CARv2 indices for all pieces, so
when it's time to acquire a shard to serve a retrieval, the unsealed
CARv1 is joined with its index (safeguarded by the dagstore), to form
a read-only blockstore, thus taking the place of the monolithic
badger.
Data transfers have been adjusted to interface directly with CARv2 files.
On inbound transfers (client retrievals, miner storage deals), we stream
the received data into a CARv2 ReadWrite blockstore. On outbound transfers
(client storage deals, miner retrievals), we serve the data off a CARv2
ReadOnly blockstore.
Client-side imports are managed by the refactored *imports.Manager
component (when not using IPFS integration). Just like it before, we use
the go-filestore library to avoid duplicating the data from the original
file in the resulting UnixFS DAG (concretely the leaves). However, the
target of those imports are what we call "ref-CARv2s": CARv2 files placed
under the `$LOTUS_PATH/imports` directory, containing the intermediate
nodes in full, and the leaves as positional references to the original file
on disk.
Client-side retrievals are placed into CARv2 files in the location:
`$LOTUS_PATH/retrievals`.
A new set of `Dagstore*` JSON-RPC operations and `lotus-miner dagstore`
subcommands have been introduced on the miner-side to inspect and manage
the dagstore.
Despite moving to a CARv2-backed system, the IPFS integration has been
respected, and it continues to be possible to make storage deals with data
held in an IPFS node, and to perform retrievals directly into an IPFS node.
NOTE: because the "staging" and "client" Badger blockstores are no longer
used, existing imports on the client will be rendered useless. On startup,
Lotus will enumerate all imports and print WARN statements on the log for
each import that needs to be reimported. These log lines contain these
messages:
- import lacks carv2 path; import will not work; please reimport
- import has missing/broken carv2; please reimport
At the end, we will print a "sanity check completed" message indicating
the count of imports found, and how many were deemed broken.
Co-authored-by: Aarsh Shah <aarshkshah1992@gmail.com>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
Co-authored-by: Raúl Kripalani <raul@protocol.ai>
Co-authored-by: Dirk McCormick <dirkmdev@gmail.com>
2021-08-16 22:34:32 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
func genInputFile(t *testing.T) (filepath string, contents []byte) {
|
|
|
|
s := strings.Repeat("abcde", 100)
|
|
|
|
tmp, err := os.CreateTemp("", "")
|
|
|
|
require.NoError(t, err)
|
|
|
|
_, err = io.Copy(tmp, strings.NewReader(s))
|
|
|
|
require.NoError(t, err)
|
|
|
|
require.NoError(t, tmp.Close())
|
|
|
|
return tmp.Name(), []byte(s)
|
|
|
|
}
|