mirror of
https://source.quilibrium.com/quilibrium/ceremonyclient.git
synced 2024-11-10 18:25:17 +00:00
335 lines
8.3 KiB
Go
335 lines
8.3 KiB
Go
package itest
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"os"
|
|
"sync"
|
|
"sync/atomic"
|
|
"testing"
|
|
"time"
|
|
|
|
"github.com/libp2p/go-libp2p"
|
|
"github.com/libp2p/go-libp2p/core/network"
|
|
"github.com/libp2p/go-libp2p/core/peer"
|
|
rcmgr "github.com/libp2p/go-libp2p/p2p/host/resource-manager"
|
|
|
|
"github.com/stretchr/testify/require"
|
|
)
|
|
|
|
func makeRcmgrOption(t *testing.T, cfg rcmgr.ConcreteLimitConfig) func(int) libp2p.Option {
|
|
return func(i int) libp2p.Option {
|
|
var opts []rcmgr.Option
|
|
if os.Getenv("LIBP2P_TEST_RCMGR_TRACE") == "1" {
|
|
opts = append(opts, rcmgr.WithTrace(fmt.Sprintf("%s-%d.json.gz", t.Name(), i)))
|
|
}
|
|
|
|
mgr, err := rcmgr.NewResourceManager(rcmgr.NewFixedLimiter(cfg), opts...)
|
|
require.NoError(t, err)
|
|
return libp2p.ResourceManager(mgr)
|
|
}
|
|
}
|
|
|
|
func closeRcmgrs(echos []*Echo) {
|
|
for _, e := range echos {
|
|
e.Host.Network().ResourceManager().Close()
|
|
}
|
|
}
|
|
|
|
func waitForConnection(t *testing.T, src, dest *Echo) {
|
|
require.Eventually(t, func() bool {
|
|
return src.Host.Network().Connectedness(dest.Host.ID()) == network.Connected &&
|
|
dest.Host.Network().Connectedness(src.Host.ID()) == network.Connected
|
|
}, time.Second, 10*time.Millisecond)
|
|
}
|
|
|
|
func TestResourceManagerConnInbound(t *testing.T) {
|
|
// this test checks that we can not exceed the inbound conn limit at system level
|
|
// we specify: 1 conn per peer, 3 conns total, and we try to create 4 conns
|
|
cfg := rcmgr.PartialLimitConfig{
|
|
System: rcmgr.ResourceLimits{
|
|
ConnsInbound: 3,
|
|
ConnsOutbound: 1024,
|
|
Conns: 1024,
|
|
StreamsOutbound: rcmgr.Unlimited,
|
|
},
|
|
PeerDefault: rcmgr.ResourceLimits{
|
|
ConnsInbound: 1,
|
|
ConnsOutbound: 1,
|
|
Conns: 1,
|
|
},
|
|
}.Build(rcmgr.DefaultLimits.AutoScale())
|
|
|
|
echos := createEchos(t, 5, makeRcmgrOption(t, cfg))
|
|
defer closeEchos(echos)
|
|
defer closeRcmgrs(echos)
|
|
|
|
for i := 1; i < 4; i++ {
|
|
err := echos[i].Host.Connect(context.Background(), peer.AddrInfo{ID: echos[0].Host.ID()})
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
waitForConnection(t, echos[i], echos[0])
|
|
}
|
|
|
|
for i := 1; i < 4; i++ {
|
|
count := len(echos[i].Host.Network().ConnsToPeer(echos[0].Host.ID()))
|
|
if count != 1 {
|
|
t.Fatalf("expected %d connections to peer, got %d", 1, count)
|
|
}
|
|
}
|
|
|
|
err := echos[4].Host.Connect(context.Background(), peer.AddrInfo{ID: echos[0].Host.ID()})
|
|
if err == nil {
|
|
t.Fatal("expected ResourceManager to block incoming connection")
|
|
}
|
|
}
|
|
|
|
func TestResourceManagerConnOutbound(t *testing.T) {
|
|
// this test checks that we can not exceed the inbound conn limit at system level
|
|
// we specify: 1 conn per peer, 3 conns total, and we try to create 4 conns
|
|
cfg := rcmgr.PartialLimitConfig{
|
|
System: rcmgr.ResourceLimits{
|
|
ConnsInbound: 1024,
|
|
ConnsOutbound: 3,
|
|
Conns: 1024,
|
|
},
|
|
PeerDefault: rcmgr.ResourceLimits{
|
|
ConnsInbound: 1,
|
|
ConnsOutbound: 1,
|
|
Conns: 1,
|
|
},
|
|
}.Build(rcmgr.DefaultLimits.AutoScale())
|
|
echos := createEchos(t, 5, makeRcmgrOption(t, cfg))
|
|
defer closeEchos(echos)
|
|
defer closeRcmgrs(echos)
|
|
|
|
for i := 1; i < 4; i++ {
|
|
err := echos[0].Host.Connect(context.Background(), peer.AddrInfo{ID: echos[i].Host.ID()})
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
waitForConnection(t, echos[0], echos[i])
|
|
}
|
|
|
|
for i := 1; i < 4; i++ {
|
|
count := len(echos[i].Host.Network().ConnsToPeer(echos[0].Host.ID()))
|
|
if count != 1 {
|
|
t.Fatalf("expected %d connections to peer, got %d", 1, count)
|
|
}
|
|
}
|
|
|
|
err := echos[0].Host.Connect(context.Background(), peer.AddrInfo{ID: echos[4].Host.ID()})
|
|
if err == nil {
|
|
t.Fatal("expected ResourceManager to block incoming connection")
|
|
}
|
|
}
|
|
|
|
func TestResourceManagerServiceInbound(t *testing.T) {
|
|
// this test checks that we can not exceed the inbound stream limit at service level
|
|
// we specify: 3 streams for the service, and we try to create 4 streams
|
|
cfg := rcmgr.PartialLimitConfig{
|
|
ServiceDefault: rcmgr.ResourceLimits{
|
|
StreamsInbound: 3,
|
|
StreamsOutbound: 1024,
|
|
Streams: 1024,
|
|
},
|
|
}.Build(rcmgr.DefaultLimits.AutoScale())
|
|
echos := createEchos(t, 5, makeRcmgrOption(t, cfg))
|
|
defer closeEchos(echos)
|
|
defer closeRcmgrs(echos)
|
|
|
|
for i := 1; i < 5; i++ {
|
|
err := echos[i].Host.Connect(context.Background(), peer.AddrInfo{ID: echos[0].Host.ID()})
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
waitForConnection(t, echos[i], echos[0])
|
|
}
|
|
|
|
ready := make(chan struct{})
|
|
echos[0].BeforeDone(waitForChannel(ready, time.Minute))
|
|
|
|
var eg sync.WaitGroup
|
|
echos[0].Done(eg.Done)
|
|
|
|
var once sync.Once
|
|
var wg sync.WaitGroup
|
|
for i := 1; i < 5; i++ {
|
|
eg.Add(1)
|
|
wg.Add(1)
|
|
go func(i int) {
|
|
defer wg.Done()
|
|
|
|
err := echos[i].Echo(echos[0].Host.ID(), "hello libp2p")
|
|
if err != nil {
|
|
t.Log(err)
|
|
once.Do(func() {
|
|
close(ready)
|
|
})
|
|
}
|
|
}(i)
|
|
}
|
|
wg.Wait()
|
|
eg.Wait()
|
|
|
|
checkEchoStatus(t, echos[0], EchoStatus{
|
|
StreamsIn: 4,
|
|
EchosIn: 3,
|
|
EchosOut: 3,
|
|
ResourceServiceErrors: 1,
|
|
})
|
|
}
|
|
|
|
func TestResourceManagerServicePeerInbound(t *testing.T) {
|
|
// this test checks that we cannot exceed the per peer inbound stream limit at service level
|
|
// we specify: 2 streams per peer for echo, and we try to create 3 streams
|
|
cfg := rcmgr.DefaultLimits
|
|
cfg.AddServicePeerLimit(
|
|
EchoService,
|
|
rcmgr.BaseLimit{StreamsInbound: 2, StreamsOutbound: 1024, Streams: 1024, Memory: 9999999},
|
|
rcmgr.BaseLimitIncrease{},
|
|
)
|
|
limits := cfg.AutoScale()
|
|
|
|
echos := createEchos(t, 5, makeRcmgrOption(t, limits))
|
|
defer closeEchos(echos)
|
|
defer closeRcmgrs(echos)
|
|
|
|
for i := 1; i < 5; i++ {
|
|
err := echos[i].Host.Connect(context.Background(), peer.AddrInfo{ID: echos[0].Host.ID()})
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
waitForConnection(t, echos[i], echos[0])
|
|
}
|
|
|
|
echos[0].BeforeDone(waitForBarrier(4, time.Minute))
|
|
|
|
var eg sync.WaitGroup
|
|
echos[0].Done(eg.Done)
|
|
|
|
var wg sync.WaitGroup
|
|
for i := 1; i < 5; i++ {
|
|
eg.Add(1)
|
|
wg.Add(1)
|
|
go func(i int) {
|
|
defer wg.Done()
|
|
|
|
err := echos[i].Echo(echos[0].Host.ID(), "hello libp2p")
|
|
if err != nil {
|
|
t.Log(err)
|
|
}
|
|
}(i)
|
|
}
|
|
wg.Wait()
|
|
eg.Wait()
|
|
|
|
checkEchoStatus(t, echos[0], EchoStatus{
|
|
StreamsIn: 4,
|
|
EchosIn: 4,
|
|
EchosOut: 4,
|
|
ResourceServiceErrors: 0,
|
|
})
|
|
|
|
ready := make(chan struct{})
|
|
echos[0].BeforeDone(waitForChannel(ready, time.Minute))
|
|
|
|
var once sync.Once
|
|
for i := 0; i < 3; i++ {
|
|
eg.Add(1)
|
|
wg.Add(1)
|
|
go func() {
|
|
defer wg.Done()
|
|
|
|
err := echos[2].Echo(echos[0].Host.ID(), "hello libp2p")
|
|
if err != nil {
|
|
t.Log(err)
|
|
once.Do(func() {
|
|
close(ready)
|
|
})
|
|
}
|
|
}()
|
|
}
|
|
wg.Wait()
|
|
eg.Wait()
|
|
|
|
checkEchoStatus(t, echos[0], EchoStatus{
|
|
StreamsIn: 7,
|
|
EchosIn: 6,
|
|
EchosOut: 6,
|
|
ResourceServiceErrors: 1,
|
|
})
|
|
}
|
|
|
|
func waitForBarrier(count int32, timeout time.Duration) func() error {
|
|
ready := make(chan struct{})
|
|
var wait atomic.Int32
|
|
wait.Store(count)
|
|
return func() error {
|
|
if wait.Add(-1) == 0 {
|
|
close(ready)
|
|
}
|
|
|
|
select {
|
|
case <-ready:
|
|
return nil
|
|
case <-time.After(timeout):
|
|
return fmt.Errorf("timeout")
|
|
}
|
|
}
|
|
}
|
|
|
|
func waitForChannel(ready chan struct{}, timeout time.Duration) func() error {
|
|
return func() error {
|
|
select {
|
|
case <-ready:
|
|
return nil
|
|
case <-time.After(timeout):
|
|
return fmt.Errorf("timeout")
|
|
}
|
|
}
|
|
}
|
|
|
|
func TestReadmeExample(t *testing.T) {
|
|
// Start with the default scaling limits.
|
|
scalingLimits := rcmgr.DefaultLimits
|
|
|
|
// Add limits around included libp2p protocols
|
|
libp2p.SetDefaultServiceLimits(&scalingLimits)
|
|
|
|
// Turn the scaling limits into a concrete set of limits using `.AutoScale`. This
|
|
// scales the limits proportional to your system memory.
|
|
scaledDefaultLimits := scalingLimits.AutoScale()
|
|
|
|
// Tweak certain settings
|
|
cfg := rcmgr.PartialLimitConfig{
|
|
System: rcmgr.ResourceLimits{
|
|
// Allow unlimited outbound streams
|
|
StreamsOutbound: rcmgr.Unlimited,
|
|
},
|
|
// Everything else is default. The exact values will come from `scaledDefaultLimits` above.
|
|
}
|
|
|
|
// Create our limits by using our cfg and replacing the default values with values from `scaledDefaultLimits`
|
|
limits := cfg.Build(scaledDefaultLimits)
|
|
|
|
// The resource manager expects a limiter, se we create one from our limits.
|
|
limiter := rcmgr.NewFixedLimiter(limits)
|
|
|
|
// Metrics are enabled by default. If you want to disable metrics, use the
|
|
// WithMetricsDisabled option
|
|
// Initialize the resource manager
|
|
rm, err := rcmgr.NewResourceManager(limiter, rcmgr.WithMetricsDisabled())
|
|
if err != nil {
|
|
panic(err)
|
|
}
|
|
|
|
// Create a libp2p host
|
|
host, err := libp2p.New(libp2p.ResourceManager(rm))
|
|
if err != nil {
|
|
panic(err)
|
|
}
|
|
host.Close()
|
|
}
|