forked from mirror/redis
Update cluster node latency asynchronously
This commit is contained in:
parent
8b4fa6d443
commit
1a4d34448d
53
cluster.go
53
cluster.go
|
@ -2,6 +2,7 @@ package redis
|
|||
|
||||
import (
|
||||
"fmt"
|
||||
"math"
|
||||
"math/rand"
|
||||
"net"
|
||||
"sync"
|
||||
|
@ -118,11 +119,11 @@ func (opt *ClusterOptions) clientOptions() *Options {
|
|||
//------------------------------------------------------------------------------
|
||||
|
||||
type clusterNode struct {
|
||||
Client *Client
|
||||
Latency time.Duration
|
||||
Client *Client
|
||||
|
||||
loading time.Time
|
||||
latency uint32 // atomic
|
||||
generation uint32
|
||||
loading int64 // atomic
|
||||
}
|
||||
|
||||
func newClusterNode(clOpt *ClusterOptions, addr string) *clusterNode {
|
||||
|
@ -132,8 +133,9 @@ func newClusterNode(clOpt *ClusterOptions, addr string) *clusterNode {
|
|||
Client: NewClient(opt),
|
||||
}
|
||||
|
||||
node.latency = math.MaxUint32
|
||||
if clOpt.RouteByLatency {
|
||||
node.updateLatency()
|
||||
go node.updateLatency()
|
||||
}
|
||||
|
||||
return &node
|
||||
|
@ -141,16 +143,46 @@ func newClusterNode(clOpt *ClusterOptions, addr string) *clusterNode {
|
|||
|
||||
func (n *clusterNode) updateLatency() {
|
||||
const probes = 10
|
||||
|
||||
var latency uint32
|
||||
for i := 0; i < probes; i++ {
|
||||
start := time.Now()
|
||||
n.Client.Ping()
|
||||
n.Latency += time.Since(start)
|
||||
probe := uint32(time.Since(start) / time.Microsecond)
|
||||
latency = (latency + probe) / 2
|
||||
}
|
||||
n.Latency = n.Latency / probes
|
||||
atomic.StoreUint32(&n.latency, latency)
|
||||
}
|
||||
|
||||
func (n *clusterNode) Close() error {
|
||||
return n.Client.Close()
|
||||
}
|
||||
|
||||
func (n *clusterNode) Test() error {
|
||||
return n.Client.ClusterInfo().Err()
|
||||
}
|
||||
|
||||
func (n *clusterNode) Latency() time.Duration {
|
||||
latency := atomic.LoadUint32(&n.latency)
|
||||
return time.Duration(latency) * time.Microsecond
|
||||
}
|
||||
|
||||
func (n *clusterNode) MarkAsLoading() {
|
||||
atomic.StoreInt64(&n.loading, time.Now().Unix())
|
||||
}
|
||||
|
||||
func (n *clusterNode) Loading() bool {
|
||||
return !n.loading.IsZero() && time.Since(n.loading) < time.Minute
|
||||
const minute = int64(time.Minute / time.Second)
|
||||
|
||||
loading := atomic.LoadInt64(&n.loading)
|
||||
if loading == 0 {
|
||||
return false
|
||||
}
|
||||
if time.Now().Unix()-loading < minute {
|
||||
return true
|
||||
}
|
||||
atomic.StoreInt64(&n.loading, 0)
|
||||
return false
|
||||
}
|
||||
|
||||
func (n *clusterNode) Generation() uint32 {
|
||||
|
@ -310,7 +342,7 @@ func (c *clusterNodes) Random() (*clusterNode, error) {
|
|||
return nil, err
|
||||
}
|
||||
|
||||
nodeErr = node.Client.ClusterInfo().Err()
|
||||
nodeErr = node.Test()
|
||||
if nodeErr == nil {
|
||||
return node, nil
|
||||
}
|
||||
|
@ -416,7 +448,7 @@ func (c *clusterState) slotClosestNode(slot int) (*clusterNode, error) {
|
|||
if n.Loading() {
|
||||
continue
|
||||
}
|
||||
if node == nil || node.Latency-n.Latency > threshold {
|
||||
if node == nil || node.Latency()-n.Latency() > threshold {
|
||||
node = n
|
||||
}
|
||||
}
|
||||
|
@ -687,8 +719,7 @@ func (c *ClusterClient) defaultProcess(cmd Cmder) error {
|
|||
|
||||
// If slave is loading - read from master.
|
||||
if c.opt.ReadOnly && internal.IsLoadingError(err) {
|
||||
// TODO: race
|
||||
node.loading = time.Now()
|
||||
node.MarkAsLoading()
|
||||
continue
|
||||
}
|
||||
|
||||
|
|
|
@ -320,6 +320,14 @@ var _ = Describe("ClusterClient", func() {
|
|||
Expect(err).NotTo(HaveOccurred())
|
||||
Expect(cmds).To(HaveLen(14))
|
||||
|
||||
_ = client.ForEachNode(func(node *redis.Client) error {
|
||||
defer GinkgoRecover()
|
||||
Eventually(func() int64 {
|
||||
return node.DBSize().Val()
|
||||
}, 30*time.Second).ShouldNot(BeZero())
|
||||
return nil
|
||||
})
|
||||
|
||||
for _, key := range keys {
|
||||
slot := hashtag.Slot(key)
|
||||
client.SwapSlotNodes(slot)
|
||||
|
@ -576,7 +584,7 @@ var _ = Describe("ClusterClient", func() {
|
|||
|
||||
_ = client.ForEachSlave(func(slave *redis.Client) error {
|
||||
Eventually(func() int64 {
|
||||
return client.DBSize().Val()
|
||||
return slave.DBSize().Val()
|
||||
}, 30*time.Second).Should(Equal(int64(0)))
|
||||
return slave.ClusterFailover().Err()
|
||||
})
|
||||
|
@ -717,7 +725,7 @@ var _ = Describe("ClusterClient timeout", func() {
|
|||
})
|
||||
}
|
||||
|
||||
const pause = time.Second
|
||||
const pause = 2 * time.Second
|
||||
|
||||
Context("read/write timeout", func() {
|
||||
BeforeEach(func() {
|
||||
|
|
|
@ -447,7 +447,7 @@ var _ = Describe("Commands", func() {
|
|||
|
||||
pttl := client.PTTL("key")
|
||||
Expect(pttl.Err()).NotTo(HaveOccurred())
|
||||
Expect(pttl.Val()).To(BeNumerically("~", expiration, 10*time.Millisecond))
|
||||
Expect(pttl.Val()).To(BeNumerically("~", expiration, 100*time.Millisecond))
|
||||
})
|
||||
|
||||
It("should PExpireAt", func() {
|
||||
|
@ -466,7 +466,7 @@ var _ = Describe("Commands", func() {
|
|||
|
||||
pttl := client.PTTL("key")
|
||||
Expect(pttl.Err()).NotTo(HaveOccurred())
|
||||
Expect(pttl.Val()).To(BeNumerically("~", expiration, 10*time.Millisecond))
|
||||
Expect(pttl.Val()).To(BeNumerically("~", expiration, 100*time.Millisecond))
|
||||
})
|
||||
|
||||
It("should PTTL", func() {
|
||||
|
@ -481,7 +481,7 @@ var _ = Describe("Commands", func() {
|
|||
|
||||
pttl := client.PTTL("key")
|
||||
Expect(pttl.Err()).NotTo(HaveOccurred())
|
||||
Expect(pttl.Val()).To(BeNumerically("~", expiration, 10*time.Millisecond))
|
||||
Expect(pttl.Val()).To(BeNumerically("~", expiration, 100*time.Millisecond))
|
||||
})
|
||||
|
||||
It("should RandomKey", func() {
|
||||
|
|
Loading…
Reference in New Issue