1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
|
// Copyright (C) 2019-2021 Algorand, Inc.
// This file is part of go-algorand
//
// go-algorand is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as
// published by the Free Software Foundation, either version 3 of the
// License, or (at your option) any later version.
//
// go-algorand is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with go-algorand. If not, see <https://www.gnu.org/licenses/>.
package catchup
import (
"fmt"
"net/http"
"os/exec"
"path/filepath"
"runtime"
"syscall"
"testing"
"time"
"github.com/algorand/go-deadlock"
"github.com/stretchr/testify/require"
"github.com/algorand/go-algorand/config"
algodclient "github.com/algorand/go-algorand/daemon/algod/api/client"
"github.com/algorand/go-algorand/logging"
"github.com/algorand/go-algorand/nodecontrol"
"github.com/algorand/go-algorand/protocol"
"github.com/algorand/go-algorand/test/framework/fixtures"
"github.com/algorand/go-algorand/test/partitiontest"
)
type nodeExitErrorCollector struct {
errors []error
messages []string
mu deadlock.Mutex
t fixtures.TestingTB
}
func (ec *nodeExitErrorCollector) nodeExitWithError(nc *nodecontrol.NodeController, err error) {
if err == nil {
return
}
exitError, ok := err.(*exec.ExitError)
if !ok {
if err != nil {
ec.mu.Lock()
ec.errors = append(ec.errors, err)
ec.messages = append(ec.messages, "Node at %s has terminated with an error", nc.GetDataDir())
ec.mu.Unlock()
}
return
}
ws := exitError.Sys().(syscall.WaitStatus)
exitCode := ws.ExitStatus()
if err != nil {
ec.mu.Lock()
ec.errors = append(ec.errors, err)
ec.messages = append(ec.messages, fmt.Sprintf("Node at %s has terminated with error code %d", nc.GetDataDir(), exitCode))
ec.mu.Unlock()
}
}
func (ec *nodeExitErrorCollector) Print() {
ec.mu.Lock()
defer ec.mu.Unlock()
for i, err := range ec.errors {
require.NoError(ec.t, err, ec.messages[i])
}
}
func TestBasicCatchpointCatchup(t *testing.T) {
partitiontest.PartitionTest(t)
if testing.Short() {
t.Skip()
}
a := require.New(fixtures.SynchronizedTest(t))
log := logging.TestingLog(t)
// Overview of this test:
// Start a two-node network (primary has 100%, secondary has 0%)
// Nodes are having a consensus allowing balances history of 32 rounds and transaction history of 33 rounds.
// Let it run for 37 rounds.
// create a web proxy, and connect it to the primary node, blocking all requests for round #1. ( and allowing everything else )
// start a secondary node, and instuct it to catchpoint catchup from the proxy. ( which would be for round 36 )
// wait until the clone node cought up, skipping the "impossible" hole of round #1.
consensus := make(config.ConsensusProtocols)
const consensusCatchpointCatchupTestProtocol = protocol.ConsensusVersion("catchpointtestingprotocol")
catchpointCatchupProtocol := config.Consensus[protocol.ConsensusCurrentVersion]
catchpointCatchupProtocol.ApprovedUpgrades = map[protocol.ConsensusVersion]uint64{}
// MaxBalLookback = 2 x SeedRefreshInterval x SeedLookback
// ref. https://github.com/algorandfoundation/specs/blob/master/dev/abft.md
catchpointCatchupProtocol.SeedLookback = 2
catchpointCatchupProtocol.SeedRefreshInterval = 8
catchpointCatchupProtocol.MaxBalLookback = 2 * catchpointCatchupProtocol.SeedLookback * catchpointCatchupProtocol.SeedRefreshInterval // 32
catchpointCatchupProtocol.MaxTxnLife = 33
if runtime.GOARCH == "amd64" {
// amd64 platforms are generally quite capable, so accelerate the round times to make the test run faster.
catchpointCatchupProtocol.AgreementFilterTimeoutPeriod0 = 1 * time.Second
catchpointCatchupProtocol.AgreementFilterTimeout = 1 * time.Second
}
consensus[consensusCatchpointCatchupTestProtocol] = catchpointCatchupProtocol
var fixture fixtures.RestClientFixture
fixture.SetConsensus(consensus)
errorsCollector := nodeExitErrorCollector{t: fixtures.SynchronizedTest(t)}
defer errorsCollector.Print()
fixture.SetupNoStart(t, filepath.Join("nettemplates", "CatchpointCatchupTestNetwork.json"))
// Get primary node
primaryNode, err := fixture.GetNodeController("Primary")
a.NoError(err)
// Get secondary node
secondNode, err := fixture.GetNodeController("Node")
a.NoError(err)
// prepare it's configuration file to set it to generate a catchpoint every 4 rounds.
cfg, err := config.LoadConfigFromDisk(primaryNode.GetDataDir())
a.NoError(err)
cfg.CatchpointInterval = 4
cfg.SaveToDisk(primaryNode.GetDataDir())
cfg.Archival = false
cfg.NetAddress = ""
cfg.EnableLedgerService = false
cfg.EnableBlockService = false
cfg.SaveToDisk(secondNode.GetDataDir())
// start the primary node
_, err = primaryNode.StartAlgod(nodecontrol.AlgodStartArgs{
PeerAddress: "",
ListenIP: "",
RedirectOutput: true,
RunUnderHost: false,
TelemetryOverride: "",
ExitErrorCallback: errorsCollector.nodeExitWithError,
})
a.NoError(err)
// Let the network make some progress
currentRound := uint64(1)
targetRound := uint64(37)
primaryNodeRestClient := fixture.GetAlgodClientForController(primaryNode)
primaryNodeRestClient.SetAPIVersionAffinity(algodclient.APIVersionV2)
log.Infof("Building ledger history..")
for {
err = fixture.ClientWaitForRound(primaryNodeRestClient, currentRound, 45000*time.Millisecond)
a.NoError(err)
if targetRound <= currentRound {
break
}
currentRound++
}
log.Infof("done building!\n")
primaryListeningAddress, err := primaryNode.GetListeningAddress()
a.NoError(err)
wp, err := fixtures.MakeWebProxy(primaryListeningAddress, func(response http.ResponseWriter, request *http.Request, next http.HandlerFunc) {
// prevent requests for block #2 to go through.
if request.URL.String() == "/v1/test-v1/block/2" {
response.WriteHeader(http.StatusBadRequest)
return
}
next(response, request)
})
a.NoError(err)
defer wp.Close()
log.Infof("web proxy listens at %s\n", wp.GetListenAddress())
// start the second node
_, err = secondNode.StartAlgod(nodecontrol.AlgodStartArgs{
PeerAddress: wp.GetListenAddress(),
ListenIP: "",
RedirectOutput: true,
RunUnderHost: false,
TelemetryOverride: "",
ExitErrorCallback: errorsCollector.nodeExitWithError,
})
a.NoError(err)
// wait until node is caught up.
secondNodeRestClient := fixture.GetAlgodClientForController(secondNode)
currentRound = uint64(1)
targetRound = uint64(1)
log.Infof("Second node catching up to round 1")
for {
err = fixture.ClientWaitForRound(secondNodeRestClient, currentRound, 10*time.Second)
a.NoError(err)
if targetRound <= currentRound {
break
}
currentRound++
}
log.Infof(" - done catching up!\n")
primaryNodeStatus, err := primaryNodeRestClient.Status()
a.NoError(err)
a.NotNil(primaryNodeStatus.LastCatchpoint)
log.Infof("primary node latest catchpoint - %s!\n", *primaryNodeStatus.LastCatchpoint)
secondNodeRestClient.Catchup(*primaryNodeStatus.LastCatchpoint)
currentRound = primaryNodeStatus.LastRound
targetRound = currentRound + 1
log.Infof("Second node catching up to round 36")
for {
err = fixture.ClientWaitForRound(secondNodeRestClient, currentRound, 10*time.Second)
a.NoError(err)
if targetRound <= currentRound {
break
}
currentRound++
}
log.Infof("done catching up!\n")
secondNode.StopAlgod()
primaryNode.StopAlgod()
}
|
