724bb44723
What: cmd/inspector/main.go: removes kad commands internal/testplanet/planet.go: Waits for contact chore to finish satellite/contact/nodesservice.go: creates an empty nodes service implementation satellite/contact/service.go: implements Local and FetchInfo methods & adds external address config value satellite/discovery/service.go: replaces kad.FetchInfo with contact.FetchInfo in Refresh() & removes Discover() satellite/peer.go: sets up contact service and endpoints storagenode/console/service.go: replaces nodeID with contact.Local() storagenode/contact/chore.go: replaces routing table with contact service storagenode/contact/nodesservice.go: creates empty implementation for ping and request info nodes service & implements RequestInfo method storagenode/contact/service.go: creates a service to return the local node and update its own capacity storagenode/monitor/monitor.go: uses contact service in place of routing table storagenode/operator.go: moves operatorconfig from kad into its own setup storagenode/peer.go: sets up contact service, chore, pingstats and endpoints satellite/overlay/config.go: changes NodeSelectionConfig.OnlineWindow default to 4hr to allow for accurate repair selection Removes kademlia setups in: cmd/storagenode/main.go cmd/storj-sim/network.go internal/testplane/planet.go internal/testplanet/satellite.go internal/testplanet/storagenode.go satellite/peer.go scripts/test-sim-backwards.sh scripts/testdata/satellite-config.yaml.lock storagenode/inspector/inspector.go storagenode/peer.go storagenode/storagenodedb/database.go Why: Replacing Kademlia Please describe the tests: • internal/testplanet/planet_test.go: TestBasic: assert that the storagenode can check in with the satellite without any errors TestContact: test that all nodes get inserted into both satellites' overlay cache during testplanet setup • satellite/contact/contact_test.go: TestFetchInfo: Tests that the FetchInfo method returns the correct info • storagenode/contact/contact_test.go: TestNodeInfoUpdated: tests that the contact chore updates the node information TestRequestInfoEndpoint: tests that the Request info endpoint returns the correct info Please describe the performance impact: Node discovery should be at least slightly more performant since each node connects directly to each satellite and no longer needs to wait for bootstrapping. It probably won't be faster in real time on start up since each node waits a random amount of time (less than 1 hr) to initialize its first connection (jitter).
297 lines
8.5 KiB
Go
297 lines
8.5 KiB
Go
// Copyright (C) 2019 Storj Labs, Inc.
|
|
// See LICENSE for copying information
|
|
|
|
package testplanet_test
|
|
|
|
import (
|
|
"bytes"
|
|
"context"
|
|
"fmt"
|
|
"path/filepath"
|
|
"testing"
|
|
"time"
|
|
|
|
"github.com/stretchr/testify/assert"
|
|
"github.com/stretchr/testify/require"
|
|
"golang.org/x/sync/errgroup"
|
|
|
|
"storj.io/storj/internal/memory"
|
|
"storj.io/storj/internal/testcontext"
|
|
"storj.io/storj/internal/testplanet"
|
|
"storj.io/storj/internal/testrand"
|
|
"storj.io/storj/pkg/pb"
|
|
"storj.io/storj/pkg/peertls/extensions"
|
|
"storj.io/storj/pkg/peertls/tlsopts"
|
|
"storj.io/storj/pkg/revocation"
|
|
"storj.io/storj/pkg/server"
|
|
"storj.io/storj/pkg/storj"
|
|
"storj.io/storj/satellite/overlay"
|
|
"storj.io/storj/uplink"
|
|
"storj.io/storj/uplink/metainfo"
|
|
)
|
|
|
|
func TestUplinksParallel(t *testing.T) {
|
|
t.Skip("flaky")
|
|
|
|
const uplinkCount = 3
|
|
const parallelCount = 2
|
|
|
|
testplanet.Run(t, testplanet.Config{
|
|
SatelliteCount: 1, StorageNodeCount: 6, UplinkCount: uplinkCount,
|
|
}, func(t *testing.T, ctx *testcontext.Context, planet *testplanet.Planet) {
|
|
satellite := planet.Satellites[0]
|
|
|
|
var group errgroup.Group
|
|
for i := range planet.Uplinks {
|
|
uplink := planet.Uplinks[i]
|
|
|
|
for p := 0; p < parallelCount; p++ {
|
|
suffix := fmt.Sprintf("-%d-%d", i, p)
|
|
group.Go(func() error {
|
|
data := testrand.Bytes(memory.Size(100+testrand.Intn(500)) * memory.KiB)
|
|
|
|
err := uplink.Upload(ctx, satellite, "testbucket"+suffix, "test/path"+suffix, data)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
downloaded, err := uplink.Download(ctx, satellite, "testbucket"+suffix, "test/path"+suffix)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
if !bytes.Equal(data, downloaded) {
|
|
return fmt.Errorf("upload != download data: %s", suffix)
|
|
}
|
|
|
|
return nil
|
|
})
|
|
}
|
|
}
|
|
err := group.Wait()
|
|
require.NoError(t, err)
|
|
})
|
|
}
|
|
|
|
func TestDownloadWithSomeNodesOffline(t *testing.T) {
|
|
testplanet.Run(t, testplanet.Config{
|
|
SatelliteCount: 1, StorageNodeCount: 5, UplinkCount: 1,
|
|
}, func(t *testing.T, ctx *testcontext.Context, planet *testplanet.Planet) {
|
|
// first, upload some remote data
|
|
ul := planet.Uplinks[0]
|
|
satellite := planet.Satellites[0]
|
|
|
|
testData := testrand.Bytes(memory.MiB)
|
|
|
|
err := ul.UploadWithConfig(ctx, satellite, &uplink.RSConfig{
|
|
MinThreshold: 2,
|
|
RepairThreshold: 3,
|
|
SuccessThreshold: 4,
|
|
MaxThreshold: 5,
|
|
}, "testbucket", "test/path", testData)
|
|
require.NoError(t, err)
|
|
|
|
// get a remote segment from pointerdb
|
|
pdb := satellite.Metainfo.Service
|
|
listResponse, _, err := pdb.List(ctx, "", "", "", true, 0, 0)
|
|
require.NoError(t, err)
|
|
|
|
var path string
|
|
var pointer *pb.Pointer
|
|
for _, v := range listResponse {
|
|
path = v.GetPath()
|
|
pointer, err = pdb.Get(ctx, path)
|
|
require.NoError(t, err)
|
|
if pointer.GetType() == pb.Pointer_REMOTE {
|
|
break
|
|
}
|
|
}
|
|
|
|
// calculate how many storagenodes to kill
|
|
redundancy := pointer.GetRemote().GetRedundancy()
|
|
remotePieces := pointer.GetRemote().GetRemotePieces()
|
|
minReq := redundancy.GetMinReq()
|
|
numPieces := len(remotePieces)
|
|
toKill := numPieces - int(minReq)
|
|
|
|
nodesToKill := make(map[storj.NodeID]bool)
|
|
for i, piece := range remotePieces {
|
|
if i >= toKill {
|
|
continue
|
|
}
|
|
nodesToKill[piece.NodeId] = true
|
|
}
|
|
|
|
for _, node := range planet.StorageNodes {
|
|
if nodesToKill[node.ID()] {
|
|
err = planet.StopPeer(node)
|
|
require.NoError(t, err)
|
|
|
|
// mark node as offline in overlay
|
|
info := overlay.NodeCheckInInfo{
|
|
NodeID: node.ID(),
|
|
IsUp: false,
|
|
Address: &pb.NodeAddress{
|
|
Address: "1.2.3.4",
|
|
},
|
|
Version: &pb.NodeVersion{
|
|
Version: "v0.0.0",
|
|
CommitHash: "",
|
|
Timestamp: time.Time{},
|
|
Release: false,
|
|
},
|
|
}
|
|
err = satellite.Overlay.Service.UpdateCheckIn(ctx, info)
|
|
require.NoError(t, err)
|
|
}
|
|
}
|
|
// confirm that we marked the correct number of storage nodes as offline
|
|
nodes, err := satellite.Overlay.DB.SelectStorageNodes(ctx, len(planet.StorageNodes), &overlay.NodeCriteria{})
|
|
require.NoError(t, err)
|
|
require.Len(t, nodes, len(planet.StorageNodes)-len(nodesToKill))
|
|
|
|
// we should be able to download data without any of the original nodes
|
|
newData, err := ul.Download(ctx, satellite, "testbucket", "test/path")
|
|
require.NoError(t, err)
|
|
require.Equal(t, testData, newData)
|
|
})
|
|
}
|
|
|
|
type piecestoreMock struct {
|
|
}
|
|
|
|
func (mock *piecestoreMock) Upload(server pb.Piecestore_UploadServer) error {
|
|
return nil
|
|
}
|
|
func (mock *piecestoreMock) Download(server pb.Piecestore_DownloadServer) error {
|
|
timoutTicker := time.NewTicker(30 * time.Second)
|
|
defer timoutTicker.Stop()
|
|
|
|
select {
|
|
case <-timoutTicker.C:
|
|
return nil
|
|
case <-server.Context().Done():
|
|
return nil
|
|
}
|
|
}
|
|
func (mock *piecestoreMock) Delete(ctx context.Context, delete *pb.PieceDeleteRequest) (_ *pb.PieceDeleteResponse, err error) {
|
|
return nil, nil
|
|
}
|
|
func (mock *piecestoreMock) Retain(ctx context.Context, retain *pb.RetainRequest) (_ *pb.RetainResponse, err error) {
|
|
return nil, nil
|
|
}
|
|
|
|
func TestDownloadFromUnresponsiveNode(t *testing.T) {
|
|
testplanet.Run(t, testplanet.Config{
|
|
SatelliteCount: 1, StorageNodeCount: 5, UplinkCount: 1,
|
|
}, func(t *testing.T, ctx *testcontext.Context, planet *testplanet.Planet) {
|
|
expectedData := testrand.Bytes(memory.MiB)
|
|
|
|
err := planet.Uplinks[0].UploadWithConfig(ctx, planet.Satellites[0], &uplink.RSConfig{
|
|
MinThreshold: 2,
|
|
RepairThreshold: 3,
|
|
SuccessThreshold: 4,
|
|
MaxThreshold: 5,
|
|
}, "testbucket", "test/path", expectedData)
|
|
require.NoError(t, err)
|
|
|
|
// get a remote segment from pointerdb
|
|
pdb := planet.Satellites[0].Metainfo.Service
|
|
listResponse, _, err := pdb.List(ctx, "", "", "", true, 0, 0)
|
|
require.NoError(t, err)
|
|
|
|
var path string
|
|
var pointer *pb.Pointer
|
|
for _, v := range listResponse {
|
|
path = v.GetPath()
|
|
pointer, err = pdb.Get(ctx, path)
|
|
require.NoError(t, err)
|
|
if pointer.GetType() == pb.Pointer_REMOTE {
|
|
break
|
|
}
|
|
}
|
|
|
|
stopped := false
|
|
// choose used storage node and replace it with fake listener
|
|
unresponsiveNode := pointer.Remote.RemotePieces[0].NodeId
|
|
for _, storageNode := range planet.StorageNodes {
|
|
if storageNode.ID() == unresponsiveNode {
|
|
err = planet.StopPeer(storageNode)
|
|
require.NoError(t, err)
|
|
|
|
wl, err := planet.WriteWhitelist(storj.LatestIDVersion())
|
|
require.NoError(t, err)
|
|
tlscfg := tlsopts.Config{
|
|
RevocationDBURL: "bolt://" + filepath.Join(ctx.Dir("fakestoragenode"), "revocation.db"),
|
|
UsePeerCAWhitelist: true,
|
|
PeerCAWhitelistPath: wl,
|
|
PeerIDVersions: "*",
|
|
Extensions: extensions.Config{
|
|
Revocation: false,
|
|
WhitelistSignedLeaf: false,
|
|
},
|
|
}
|
|
|
|
revocationDB, err := revocation.NewDBFromCfg(tlscfg)
|
|
require.NoError(t, err)
|
|
|
|
options, err := tlsopts.NewOptions(storageNode.Identity, tlscfg, revocationDB)
|
|
require.NoError(t, err)
|
|
|
|
server, err := server.New(storageNode.Log.Named("mock-server"), options, storageNode.Addr(), storageNode.PrivateAddr(), nil)
|
|
require.NoError(t, err)
|
|
pb.RegisterPiecestoreServer(server.GRPC(), &piecestoreMock{})
|
|
go func() {
|
|
// TODO: get goroutine under control
|
|
err := server.Run(ctx)
|
|
require.NoError(t, err)
|
|
|
|
err = revocationDB.Close()
|
|
require.NoError(t, err)
|
|
}()
|
|
stopped = true
|
|
break
|
|
}
|
|
}
|
|
assert.True(t, stopped, "no storage node was altered")
|
|
|
|
data, err := planet.Uplinks[0].Download(ctx, planet.Satellites[0], "testbucket", "test/path")
|
|
assert.NoError(t, err)
|
|
|
|
assert.Equal(t, expectedData, data)
|
|
})
|
|
}
|
|
|
|
func TestDeleteWithOfflineStoragenode(t *testing.T) {
|
|
testplanet.Run(t, testplanet.Config{
|
|
SatelliteCount: 1, StorageNodeCount: 6, UplinkCount: 1,
|
|
}, func(t *testing.T, ctx *testcontext.Context, planet *testplanet.Planet) {
|
|
expectedData := testrand.Bytes(5 * memory.MiB)
|
|
|
|
config := planet.Uplinks[0].GetConfig(planet.Satellites[0])
|
|
config.Client.SegmentSize = 1 * memory.MiB
|
|
err := planet.Uplinks[0].UploadWithClientConfig(ctx, planet.Satellites[0], config, "test-bucket", "test-file", expectedData)
|
|
require.NoError(t, err)
|
|
|
|
for _, node := range planet.StorageNodes {
|
|
err = planet.StopPeer(node)
|
|
require.NoError(t, err)
|
|
}
|
|
|
|
err = planet.Uplinks[0].Delete(ctx, planet.Satellites[0], "test-bucket", "test-file")
|
|
require.Error(t, err)
|
|
|
|
key := planet.Uplinks[0].APIKey[planet.Satellites[0].ID()]
|
|
|
|
metainfoClient, err := planet.Uplinks[0].DialMetainfo(ctx, planet.Satellites[0], key)
|
|
require.NoError(t, err)
|
|
defer ctx.Check(metainfoClient.Close)
|
|
|
|
objects, _, err := metainfoClient.ListObjects(ctx, metainfo.ListObjectsParams{
|
|
Bucket: []byte("test-bucket"),
|
|
})
|
|
require.NoError(t, err)
|
|
require.Equal(t, 0, len(objects))
|
|
})
|
|
}
|