Skip to content

Commit

Permalink
Merge pull request #7628 from bprashanth/rc_ttl
Browse files Browse the repository at this point in the history
Update expiration timeout based on observed latencies
  • Loading branch information
Quinton Hoole committed May 1, 2015
2 parents 204170e + a8fdf3d commit c68723e
Show file tree
Hide file tree
Showing 3 changed files with 32 additions and 12 deletions.
2 changes: 2 additions & 0 deletions pkg/client/cache/expiration_cache.go
Expand Up @@ -18,6 +18,7 @@ package cache

import (
"github.com/GoogleCloudPlatform/kubernetes/pkg/util"
"github.com/golang/glog"
"time"
)

Expand Down Expand Up @@ -81,6 +82,7 @@ func (c *ExpirationCache) getOrExpire(key string) (interface{}, bool) {
return nil, false
}
if c.expirationPolicy.IsExpired(timestampedItem) {
glog.V(4).Infof("Entry %v: %+v has expired", key, timestampedItem.obj)
// Since expiration happens lazily on read, don't hold up
// the reader trying to acquire a write lock for the delete.
// The next reader will retry the delete even if this one
Expand Down
9 changes: 7 additions & 2 deletions pkg/controller/controller_utils.go
Expand Up @@ -19,6 +19,8 @@ package controller
import (
"encoding/json"
"fmt"
"time"

"github.com/GoogleCloudPlatform/kubernetes/pkg/api"
"github.com/GoogleCloudPlatform/kubernetes/pkg/api/validation"
"github.com/GoogleCloudPlatform/kubernetes/pkg/client"
Expand All @@ -28,7 +30,6 @@ import (
"github.com/GoogleCloudPlatform/kubernetes/pkg/util/wait"
"github.com/golang/glog"
"sync/atomic"
"time"
)

const CreatedByAnnotation = "kubernetes.io/created-by"
Expand Down Expand Up @@ -106,7 +107,9 @@ func (r *RCExpectations) setExpectations(rc *api.ReplicationController, add, del
if err != nil {
return err
}
return r.Add(&PodExpectations{add: int64(add), del: int64(del), key: rcKey})
podExp := &PodExpectations{add: int64(add), del: int64(del), key: rcKey}
glog.V(4).Infof("Setting expectations %+v", podExp)
return r.Add(podExp)
}

func (r *RCExpectations) ExpectCreations(rc *api.ReplicationController, adds int) error {
Expand All @@ -124,6 +127,8 @@ func (r *RCExpectations) lowerExpectations(rc *api.ReplicationController, add, d
glog.V(2).Infof("Controller has both add and del expectations %+v", podExp)
}
podExp.Seen(int64(add), int64(del))
// The expectations might've been modified since the update on the previous line.
glog.V(4).Infof("Lowering expectations %+v", podExp)
}
}

Expand Down
33 changes: 23 additions & 10 deletions pkg/controller/replication_controller.go
Expand Up @@ -42,20 +42,28 @@ var (

const (
// We'll attempt to recompute the required replicas of all replication controllers
// the have fulfilled their expectations at least this often.
// the have fulfilled their expectations at least this often. This recomputation
// happens based on contents in local pod storage.
FullControllerResyncPeriod = 30 * time.Second

// If a watch misdelivers info about a pod, it'll take this long
// to rectify the number of replicas.
// If a watch misdelivers info about a pod, it'll take at least this long
// to rectify the number of replicas. Note that dropped deletes are only
// rectified after the expectation times out because we don't know the
// final resting state of the pod.
PodRelistPeriod = 5 * time.Minute

// If a watch drops an (add, delete) event for a pod, it'll take this long
// before a dormant rc waiting for those packets is woken up anyway. This
// should typically be somewhere between the PodRelistPeriod and the
// FullControllerResyncPeriod. It is specifically targeted at the case
// where some problem prevents an update of expectations, without it the
// RC could stay asleep forever.
ExpectationsTimeout = 2 * time.Minute
// If a watch drops a delete event for a pod, it'll take this long
// before a dormant rc waiting for those packets is woken up anyway. It is
// specifically targeted at the case where some problem prevents an update
// of expectations, without it the RC could stay asleep forever. This should
// be set based on the expected latency of watch events.

// TODO: Set this per expectation, based on its size.
// Currently an rc can service (create *and* observe the watch events for said
// creation) about 10-20 pods a second, so it takes about 3.5 min to service
// 3000 pods. Just creation is limited to 30qps, and watching happens with
// ~10-30s latency/pod at scale.
ExpectationsTimeout = 6 * time.Minute
)

// ReplicationManager is responsible for synchronizing ReplicationController objects stored
Expand Down Expand Up @@ -220,6 +228,11 @@ func (rm *ReplicationManager) deletePod(obj interface{}) {
}
return
}
// When a delete is dropped, the relist will notice a pod in the store not
// in the list, leading to the insertion of a tombstone key. Since we don't
// know which rc to wake up/update expectations, we rely on the ttl on the
// expectation expiring. The rc syncs via the 30s periodic resync and notices
// fewer pods than its replica count.
podKey, err := framework.DeletionHandlingMetaNamespaceKeyFunc(obj)
if err != nil {
glog.Errorf("Couldn't get key for object %+v: %v", obj, err)
Expand Down

0 comments on commit c68723e

Please sign in to comment.