crowdsec/pkg/leakybucket/bucket.go
2020-08-05 11:39:54 +02:00

292 lines
8.4 KiB
Go

package leakybucket
import (
"sync/atomic"
"time"
//"log"
"github.com/crowdsecurity/crowdsec/pkg/time/rate"
"github.com/crowdsecurity/crowdsec/pkg/types"
"github.com/goombaio/namegenerator"
//rate "time/rate"
"github.com/davecgh/go-spew/spew"
"github.com/prometheus/client_golang/prometheus"
log "github.com/sirupsen/logrus"
//"golang.org/x/time/rate"
)
const (
LIVE = iota
TIMEMACHINE
)
//Leaky represents one instance of a bucket
type Leaky struct {
Name string
Mode int //LIVE or TIMEMACHINE
//the limiter is what holds the proper "leaky aspect", it determines when/if we can pour objects
Limiter rate.RateLimiter `json:"-"`
SerializedState rate.Lstate
//Queue is used to held the cache of objects in the bucket, it is used to know 'how many' objects we have in buffer.
Queue *Queue
//Leaky buckets are receiving message through a chan
In chan types.Event `json:"-"`
//Leaky buckets are pushing their overflows through a chan
Out chan *Queue `json:"-"`
// shared for all buckets (the idea is to kill this afterwards)
AllOut chan types.Event `json:"-"`
KillSwitch chan bool `json:"-"`
//max capacity (for burst)
Capacity int
//CacheRatio is the number of elements that should be kept in memory (compared to capacity)
CacheSize int
//the unique identifier of the bucket (a hash)
Mapkey string
// chan for signaling
Signal chan bool `json:"-"`
Reprocess bool
Uuid string
First_ts time.Time
Last_ts time.Time
Ovflw_ts time.Time
Total_count int
Leakspeed time.Duration
BucketConfig *BucketFactory
Duration time.Duration
Pour func(*Leaky, types.Event) `json:"-"`
//Profiling when set to true enables profiling of bucket
Profiling bool
timedOverflow bool
logger *log.Entry
}
var BucketsPour = prometheus.NewCounterVec(
prometheus.CounterOpts{
Name: "cs_bucket_poured_total",
Help: "Total events were poured in bucket.",
},
[]string{"source", "name"},
)
var BucketsOverflow = prometheus.NewCounterVec(
prometheus.CounterOpts{
Name: "cs_bucket_overflowed_total",
Help: "Total buckets overflowed.",
},
[]string{"name"},
)
var BucketsUnderflow = prometheus.NewCounterVec(
prometheus.CounterOpts{
Name: "cs_bucket_underflowed_total",
Help: "Total buckets underflowed.",
},
[]string{"name"},
)
var BucketsInstanciation = prometheus.NewCounterVec(
prometheus.CounterOpts{
Name: "cs_bucket_created_total",
Help: "Total buckets were instanciated.",
},
[]string{"name"},
)
var BucketsCurrentCount = prometheus.NewGaugeVec(
prometheus.GaugeOpts{
Name: "cs_buckets",
Help: "Number of buckets that currently exist.",
},
[]string{"name"},
)
var LeakyRoutineCount int64
// Newleaky creates a new leaky bucket from a BucketFactory
// Events created by the bucket (overflow, bucket empty) are sent to a chan defined by BucketFactory
// The leaky bucket implementation is based on rate limiter (see https://godoc.org/golang.org/x/time/rate)
// There's a trick to have an event said when the bucket gets empty to allow its destruction
func NewLeaky(g BucketFactory) *Leaky {
g.logger.Tracef("Instantiating live bucket %s", g.Name)
return FromFactory(g)
}
func FromFactory(g BucketFactory) *Leaky {
var limiter rate.RateLimiter
//golang rate limiter. It's mainly intended for http rate limiter
Qsize := g.Capacity
if g.CacheSize > 0 {
//cache is smaller than actual capacity
if g.CacheSize <= g.Capacity {
Qsize = g.CacheSize
//bucket might be counter (infinite size), allow cache limitation
} else if g.Capacity == -1 {
Qsize = g.CacheSize
}
}
if g.Capacity == -1 {
//In this case we allow all events to pass.
//maybe in the future we could avoid using a limiter
limiter = &rate.AlwaysFull{}
} else {
limiter = rate.NewLimiter(rate.Every(g.leakspeed), g.Capacity)
}
BucketsInstanciation.With(prometheus.Labels{"name": g.Name}).Inc()
//create the leaky bucket per se
l := &Leaky{
Name: g.Name,
Limiter: limiter,
Uuid: namegenerator.NewNameGenerator(time.Now().UTC().UnixNano()).Generate(),
Queue: NewQueue(Qsize),
CacheSize: g.CacheSize,
Out: make(chan *Queue, 1),
AllOut: g.ret,
Capacity: g.Capacity,
Leakspeed: g.leakspeed,
BucketConfig: &g,
Pour: Pour,
Reprocess: g.Reprocess,
Profiling: g.Profiling,
Mode: LIVE,
}
if l.BucketConfig.Capacity > 0 && l.BucketConfig.leakspeed != time.Duration(0) {
l.Duration = time.Duration(l.BucketConfig.Capacity+1) * l.BucketConfig.leakspeed
}
if l.BucketConfig.duration != time.Duration(0) {
l.Duration = l.BucketConfig.duration
l.timedOverflow = true
}
return l
}
/* for now mimic a leak routine */
//LeakRoutine us the life of a bucket. It dies when the bucket underflows or overflows
func LeakRoutine(l *Leaky) {
var (
durationTicker <-chan time.Time = make(<-chan time.Time)
)
BucketsCurrentCount.With(prometheus.Labels{"name": l.Name}).Inc()
defer BucketsCurrentCount.With(prometheus.Labels{"name": l.Name}).Dec()
/*todo : we create a logger at runtime while we want leakroutine to be up asap, might not be a good idea*/
l.logger = l.BucketConfig.logger.WithFields(log.Fields{"capacity": l.Capacity, "partition": l.Mapkey, "bucket_id": l.Uuid})
l.Signal <- true
atomic.AddInt64(&LeakyRoutineCount, 1)
defer atomic.AddInt64(&LeakyRoutineCount, -1)
for _, f := range l.BucketConfig.processors {
err := f.OnBucketInit(l.BucketConfig)
if err != nil {
l.logger.Errorf("Problem at bucket initializiation. Bail out %T : %v", f, err)
close(l.Signal)
return
}
}
l.logger.Debugf("Leaky routine starting, lifetime : %s", l.Duration)
for {
select {
/*receiving an event*/
case msg := <-l.In:
/*the msg var use is confusing and is redeclared in a different type :/*/
for _, f := range l.BucketConfig.processors {
msg := f.OnBucketPour(l.BucketConfig)(msg, l)
// if &msg == nil we stop processing
if msg == nil {
goto End
}
}
l.logger.Tracef("Pour event: %s", spew.Sdump(msg))
l.logger.Debugf("Pouring event.")
BucketsPour.With(prometheus.Labels{"name": l.Name, "source": msg.Line.Src}).Inc()
l.Pour(l, msg) // glue for now
//Clear cache on behalf of pour
tmp := time.NewTicker(l.Duration)
durationTicker = tmp.C
l.Signal <- true
defer tmp.Stop()
/*a kill chan to allow externally killing the leaky routines*/
case <-l.KillSwitch:
close(l.Signal)
l.logger.Debugf("Bucket externally killed, return")
l.AllOut <- types.Event{Overflow: types.SignalOccurence{MapKey: l.Mapkey}, Type: types.OVFLW}
return
/*we overflowed*/
case ofw := <-l.Out:
close(l.Signal)
sig := FormatOverflow(l, ofw)
l.logger.Tracef("Overflow hooks time : %v", l.BucketConfig.processors)
for _, f := range l.BucketConfig.processors {
sig, ofw = f.OnBucketOverflow(l.BucketConfig)(l, sig, ofw)
if ofw == nil {
l.logger.Debugf("Overflow has been discard (%T)", f)
break
}
}
l.logger.Tracef("Overflow event: %s", spew.Sdump(types.Event{Overflow: sig}))
mt, _ := l.Ovflw_ts.MarshalText()
l.logger.Tracef("overflow time : %s", mt)
BucketsOverflow.With(prometheus.Labels{"name": l.Name}).Inc()
l.AllOut <- types.Event{Overflow: sig, Type: types.OVFLW, MarshaledTime: string(mt)}
return
/*we underflow or reach bucket deadline (timers)*/
case <-durationTicker:
l.Ovflw_ts = time.Now()
close(l.Signal)
ofw := l.Queue
sig := types.SignalOccurence{MapKey: l.Mapkey}
if l.timedOverflow {
BucketsOverflow.With(prometheus.Labels{"name": l.Name}).Inc()
sig = FormatOverflow(l, ofw)
for _, f := range l.BucketConfig.processors {
sig, ofw = f.OnBucketOverflow(l.BucketConfig)(l, sig, ofw)
if ofw == nil {
l.logger.Debugf("Overflow has been discard (%T)", f)
break
}
}
l.logger.Infof("Timed Overflow")
} else {
l.logger.Debugf("bucket underflow, destroy")
BucketsUnderflow.With(prometheus.Labels{"name": l.Name}).Inc()
}
l.logger.Tracef("Overflow event: %s", spew.Sdump(types.Event{Overflow: sig}))
l.AllOut <- types.Event{Overflow: sig, Type: types.OVFLW}
l.logger.Tracef("Returning from leaky routine.")
return
}
End:
}
}
func Pour(l *Leaky, msg types.Event) {
l.Total_count += 1
if l.First_ts.IsZero() {
l.First_ts = time.Now()
}
l.Last_ts = time.Now()
if l.Limiter.Allow() {
l.Queue.Add(msg)
} else {
l.Ovflw_ts = time.Now()
l.logger.Debugf("Last event to be poured, bucket overflow.")
l.Queue.Add(msg)
l.Out <- l.Queue
}
}