crowdsec/pkg/time/rate/rate_test.go
2022-05-17 12:14:59 +02:00

483 lines
12 KiB
Go

// Copyright 2015 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
//go:build go1.7
// +build go1.7
package rate
import (
"context"
"math"
"runtime"
"sync"
"sync/atomic"
"testing"
"time"
)
func TestLimit(t *testing.T) {
if Limit(10) == Inf {
t.Errorf("Limit(10) == Inf should be false")
}
}
func closeEnough(a, b Limit) bool {
return (math.Abs(float64(a)/float64(b)) - 1.0) < 1e-9
}
func TestEvery(t *testing.T) {
cases := []struct {
interval time.Duration
lim Limit
}{
{0, Inf},
{-1, Inf},
{1 * time.Nanosecond, Limit(1e9)},
{1 * time.Microsecond, Limit(1e6)},
{1 * time.Millisecond, Limit(1e3)},
{10 * time.Millisecond, Limit(100)},
{100 * time.Millisecond, Limit(10)},
{1 * time.Second, Limit(1)},
{2 * time.Second, Limit(0.5)},
{time.Duration(2.5 * float64(time.Second)), Limit(0.4)},
{4 * time.Second, Limit(0.25)},
{10 * time.Second, Limit(0.1)},
{time.Duration(math.MaxInt64), Limit(1e9 / float64(math.MaxInt64))},
}
for _, tc := range cases {
lim := Every(tc.interval)
if !closeEnough(lim, tc.lim) {
t.Errorf("Every(%v) = %v want %v", tc.interval, lim, tc.lim)
}
}
}
const (
d = 100 * time.Millisecond
)
var (
t0 = time.Now()
t1 = t0.Add(time.Duration(1) * d)
t2 = t0.Add(time.Duration(2) * d)
t3 = t0.Add(time.Duration(3) * d)
t4 = t0.Add(time.Duration(4) * d)
t5 = t0.Add(time.Duration(5) * d)
t9 = t0.Add(time.Duration(9) * d)
)
type allow struct {
t time.Time
n int
ok bool
}
func run(t *testing.T, lim *Limiter, allows []allow) {
for i, allow := range allows {
ok := lim.AllowN(allow.t, allow.n)
if ok != allow.ok {
t.Errorf("step %d: lim.AllowN(%v, %v) = %v want %v",
i, allow.t, allow.n, ok, allow.ok)
}
}
}
func TestLimiterBurst1(t *testing.T) {
run(t, NewLimiter(10, 1), []allow{
{t0, 1, true},
{t0, 1, false},
{t0, 1, false},
{t1, 1, true},
{t1, 1, false},
{t1, 1, false},
{t2, 2, false}, // burst size is 1, so n=2 always fails
{t2, 1, true},
{t2, 1, false},
})
}
func TestLimiterBurst3(t *testing.T) {
run(t, NewLimiter(10, 3), []allow{
{t0, 2, true},
{t0, 2, false},
{t0, 1, true},
{t0, 1, false},
{t1, 4, false},
{t2, 1, true},
{t3, 1, true},
{t4, 1, true},
{t4, 1, true},
{t4, 1, false},
{t4, 1, false},
{t9, 3, true},
{t9, 0, true},
})
}
func TestLimiterJumpBackwards(t *testing.T) {
run(t, NewLimiter(10, 3), []allow{
{t1, 1, true}, // start at t1
{t0, 1, true}, // jump back to t0, two tokens remain
{t0, 1, true},
{t0, 1, false},
{t0, 1, false},
{t1, 1, true}, // got a token
{t1, 1, false},
{t1, 1, false},
{t2, 1, true}, // got another token
{t2, 1, false},
{t2, 1, false},
})
}
// Ensure that tokensFromDuration doesn't produce
// rounding errors by truncating nanoseconds.
// See golang.org/issues/34861.
func TestLimiter_noTruncationErrors(t *testing.T) {
if !NewLimiter(0.7692307692307693, 1).Allow() {
t.Fatal("expected true")
}
}
func TestSimultaneousRequests(t *testing.T) {
const (
limit = 1
burst = 5
numRequests = 15
)
var (
wg sync.WaitGroup
numOK = uint32(0)
)
// Very slow replenishing bucket.
lim := NewLimiter(limit, burst)
// Tries to take a token, atomically updates the counter and decreases the wait
// group counter.
f := func() {
defer wg.Done()
if ok := lim.Allow(); ok {
atomic.AddUint32(&numOK, 1)
}
}
wg.Add(numRequests)
for i := 0; i < numRequests; i++ {
go f()
}
wg.Wait()
if numOK != burst {
t.Errorf("numOK = %d, want %d", numOK, burst)
}
}
func TestLongRunningQPS(t *testing.T) {
if testing.Short() {
t.Skip("skipping in short mode")
}
if runtime.GOOS == "openbsd" {
t.Skip("low resolution time.Sleep invalidates test (golang.org/issue/14183)")
return
}
if runtime.GOOS == "windows" {
t.Skip("test is unreliable on windows")
}
// The test runs for a few seconds executing many requests and then checks
// that overall number of requests is reasonable.
const (
limit = 100
burst = 100
)
var numOK = int32(0)
lim := NewLimiter(limit, burst)
var wg sync.WaitGroup
f := func() {
if ok := lim.Allow(); ok {
atomic.AddInt32(&numOK, 1)
}
wg.Done()
}
start := time.Now()
end := start.Add(5 * time.Second)
for time.Now().Before(end) {
wg.Add(1)
go f()
// This will still offer ~500 requests per second, but won't consume
// outrageous amount of CPU.
time.Sleep(2 * time.Millisecond)
}
wg.Wait()
elapsed := time.Since(start)
ideal := burst + (limit * float64(elapsed) / float64(time.Second))
// We should never get more requests than allowed.
if want := int32(ideal + 1); numOK > want {
t.Errorf("numOK = %d, want %d (ideal %f)", numOK, want, ideal)
}
// We should get very close to the number of requests allowed.
if want := int32(0.999 * ideal); numOK < want {
t.Errorf("numOK = %d, want %d (ideal %f)", numOK, want, ideal)
}
}
type request struct {
t time.Time
n int
act time.Time
ok bool
}
// dFromDuration converts a duration to a multiple of the global constant d
func dFromDuration(dur time.Duration) int {
// Adding a millisecond to be swallowed by the integer division
// because we don't care about small inaccuracies
return int((dur + time.Millisecond) / d)
}
// dSince returns multiples of d since t0
func dSince(t time.Time) int {
return dFromDuration(t.Sub(t0))
}
func runReserve(t *testing.T, lim *Limiter, req request) *Reservation {
return runReserveMax(t, lim, req, InfDuration)
}
func runReserveMax(t *testing.T, lim *Limiter, req request, maxReserve time.Duration) *Reservation {
r := lim.reserveN(req.t, req.n, maxReserve)
if r.ok && (dSince(r.timeToAct) != dSince(req.act)) || r.ok != req.ok {
t.Errorf("lim.reserveN(t%d, %v, %v) = (t%d, %v) want (t%d, %v)",
dSince(req.t), req.n, maxReserve, dSince(r.timeToAct), r.ok, dSince(req.act), req.ok)
}
return &r
}
func TestSimpleReserve(t *testing.T) {
lim := NewLimiter(10, 2)
runReserve(t, lim, request{t0, 2, t0, true})
runReserve(t, lim, request{t0, 2, t2, true})
runReserve(t, lim, request{t3, 2, t4, true})
}
func TestMix(t *testing.T) {
lim := NewLimiter(10, 2)
runReserve(t, lim, request{t0, 3, t1, false}) // should return false because n > Burst
runReserve(t, lim, request{t0, 2, t0, true})
run(t, lim, []allow{{t1, 2, false}}) // not enough tokens - don't allow
runReserve(t, lim, request{t1, 2, t2, true})
run(t, lim, []allow{{t1, 1, false}}) // negative tokens - don't allow
run(t, lim, []allow{{t3, 1, true}})
}
func TestCancelInvalid(t *testing.T) {
lim := NewLimiter(10, 2)
runReserve(t, lim, request{t0, 2, t0, true})
r := runReserve(t, lim, request{t0, 3, t3, false})
r.CancelAt(t0) // should have no effect
runReserve(t, lim, request{t0, 2, t2, true}) // did not get extra tokens
}
func TestCancelLast(t *testing.T) {
lim := NewLimiter(10, 2)
runReserve(t, lim, request{t0, 2, t0, true})
r := runReserve(t, lim, request{t0, 2, t2, true})
r.CancelAt(t1) // got 2 tokens back
runReserve(t, lim, request{t1, 2, t2, true})
}
func TestCancelTooLate(t *testing.T) {
lim := NewLimiter(10, 2)
runReserve(t, lim, request{t0, 2, t0, true})
r := runReserve(t, lim, request{t0, 2, t2, true})
r.CancelAt(t3) // too late to cancel - should have no effect
runReserve(t, lim, request{t3, 2, t4, true})
}
func TestCancel0Tokens(t *testing.T) {
lim := NewLimiter(10, 2)
runReserve(t, lim, request{t0, 2, t0, true})
r := runReserve(t, lim, request{t0, 1, t1, true})
runReserve(t, lim, request{t0, 1, t2, true})
r.CancelAt(t0) // got 0 tokens back
runReserve(t, lim, request{t0, 1, t3, true})
}
func TestCancel1Token(t *testing.T) {
lim := NewLimiter(10, 2)
runReserve(t, lim, request{t0, 2, t0, true})
r := runReserve(t, lim, request{t0, 2, t2, true})
runReserve(t, lim, request{t0, 1, t3, true})
r.CancelAt(t2) // got 1 token back
runReserve(t, lim, request{t2, 2, t4, true})
}
func TestCancelMulti(t *testing.T) {
lim := NewLimiter(10, 4)
runReserve(t, lim, request{t0, 4, t0, true})
rA := runReserve(t, lim, request{t0, 3, t3, true})
runReserve(t, lim, request{t0, 1, t4, true})
rC := runReserve(t, lim, request{t0, 1, t5, true})
rC.CancelAt(t1) // get 1 token back
rA.CancelAt(t1) // get 2 tokens back, as if C was never reserved
runReserve(t, lim, request{t1, 3, t5, true})
}
func TestReserveJumpBack(t *testing.T) {
lim := NewLimiter(10, 2)
runReserve(t, lim, request{t1, 2, t1, true}) // start at t1
runReserve(t, lim, request{t0, 1, t1, true}) // should violate Limit,Burst
runReserve(t, lim, request{t2, 2, t3, true})
}
func TestReserveJumpBackCancel(t *testing.T) {
lim := NewLimiter(10, 2)
runReserve(t, lim, request{t1, 2, t1, true}) // start at t1
r := runReserve(t, lim, request{t1, 2, t3, true})
runReserve(t, lim, request{t1, 1, t4, true})
r.CancelAt(t0) // cancel at t0, get 1 token back
runReserve(t, lim, request{t1, 2, t4, true}) // should violate Limit,Burst
}
func TestReserveSetLimit(t *testing.T) {
lim := NewLimiter(5, 2)
runReserve(t, lim, request{t0, 2, t0, true})
runReserve(t, lim, request{t0, 2, t4, true})
lim.SetLimitAt(t2, 10)
runReserve(t, lim, request{t2, 1, t4, true}) // violates Limit and Burst
}
func TestReserveSetBurst(t *testing.T) {
lim := NewLimiter(5, 2)
runReserve(t, lim, request{t0, 2, t0, true})
runReserve(t, lim, request{t0, 2, t4, true})
lim.SetBurstAt(t3, 4)
runReserve(t, lim, request{t0, 4, t9, true}) // violates Limit and Burst
}
func TestReserveSetLimitCancel(t *testing.T) {
lim := NewLimiter(5, 2)
runReserve(t, lim, request{t0, 2, t0, true})
r := runReserve(t, lim, request{t0, 2, t4, true})
lim.SetLimitAt(t2, 10)
r.CancelAt(t2) // 2 tokens back
runReserve(t, lim, request{t2, 2, t3, true})
}
func TestReserveMax(t *testing.T) {
lim := NewLimiter(10, 2)
maxT := d
runReserveMax(t, lim, request{t0, 2, t0, true}, maxT)
runReserveMax(t, lim, request{t0, 1, t1, true}, maxT) // reserve for close future
runReserveMax(t, lim, request{t0, 1, t2, false}, maxT) // time to act too far in the future
}
type wait struct {
name string
ctx context.Context
n int
delay int // in multiples of d
nilErr bool
}
func runWait(t *testing.T, lim *Limiter, w wait) {
start := time.Now()
err := lim.WaitN(w.ctx, w.n)
delay := time.Since(start)
if (w.nilErr && err != nil) || (!w.nilErr && err == nil) || w.delay != dFromDuration(delay) {
errString := "<nil>"
if !w.nilErr {
errString = "<non-nil error>"
}
t.Errorf("lim.WaitN(%v, lim, %v) = %v with delay %v ; want %v with delay %v",
w.name, w.n, err, delay, errString, d*time.Duration(w.delay))
}
}
func TestWaitSimple(t *testing.T) {
lim := NewLimiter(10, 3)
ctx, cancel := context.WithCancel(context.Background())
cancel()
runWait(t, lim, wait{"already-cancelled", ctx, 1, 0, false})
runWait(t, lim, wait{"exceed-burst-error", context.Background(), 4, 0, false})
runWait(t, lim, wait{"act-now", context.Background(), 2, 0, true})
runWait(t, lim, wait{"act-later", context.Background(), 3, 2, true})
}
func TestWaitCancel(t *testing.T) {
lim := NewLimiter(10, 3)
ctx, cancel := context.WithCancel(context.Background())
runWait(t, lim, wait{"act-now", ctx, 2, 0, true}) // after this lim.tokens = 1
go func() {
time.Sleep(d)
cancel()
}()
runWait(t, lim, wait{"will-cancel", ctx, 3, 1, false})
// should get 3 tokens back, and have lim.tokens = 2
t.Logf("tokens:%v last:%v lastEvent:%v", lim.tokens, lim.last, lim.lastEvent)
runWait(t, lim, wait{"act-now-after-cancel", context.Background(), 2, 0, true})
}
func TestWaitTimeout(t *testing.T) {
lim := NewLimiter(10, 3)
ctx, cancel := context.WithTimeout(context.Background(), d)
defer cancel()
runWait(t, lim, wait{"act-now", ctx, 2, 0, true})
runWait(t, lim, wait{"w-timeout-err", ctx, 3, 0, false})
}
func TestWaitInf(t *testing.T) {
lim := NewLimiter(Inf, 0)
runWait(t, lim, wait{"exceed-burst-no-error", context.Background(), 3, 0, true})
}
func BenchmarkAllowN(b *testing.B) {
lim := NewLimiter(Every(1*time.Second), 1)
now := time.Now()
b.ReportAllocs()
b.ResetTimer()
b.RunParallel(func(pb *testing.PB) {
for pb.Next() {
lim.AllowN(now, 1)
}
})
}
func BenchmarkWaitNNoDelay(b *testing.B) {
lim := NewLimiter(Limit(b.N), b.N)
ctx := context.Background()
b.ReportAllocs()
b.ResetTimer()
for i := 0; i < b.N; i++ {
if err := lim.WaitN(ctx, 1); err != nil {
b.Errorf("failed limiter : %s", err)
}
}
}