vitalif
/
etcd
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Merge pull request #3071 from yichengq/rename-rafthttp-metrics 

rafthttp: message_sent_latency metrics: channel -> sendingType
release-2.1
Yicheng Qin 2015-06-29 10:58:36 -07:00
parent e01d53b853 606876154d
commit f421eaeff7
2 changed files with 11 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
Documentation/metrics.md
View File

@ -49,17 +49,17 @@ Abnormally high snapshot duration (`snapshot_save_total_durations_microseconds`)
### rafthttp
| Name | Description | Type | Labels |
|-----------------------------------|--------------------------------------------|---------|----------------------------|
| message_sent_latency_microseconds | The latency distributions of messages sent | Summary | channel, msgType, remoteID |
| message_sent_failed_total | The total number of failed messages sent | Summary | channel, msgType, remoteID |
| Name | Description | Type | Labels |
|-----------------------------------|--------------------------------------------|---------|--------------------------------|
| message_sent_latency_microseconds | The latency distributions of messages sent | Summary | sendingType, msgType, remoteID |
| message_sent_failed_total | The total number of failed messages sent | Summary | sendingType, msgType, remoteID |
Abnormally high message duration (`message_sent_latency_microseconds`) indicates network issues and might cause the cluster to be unstable.
An increase in message failures (`message_sent_failed_total`) indicates more severe network issues and might cause the cluster to be unstable.
Label `channel` is the channel to send message. `message`, `msgapp` and `msgappv2` channels use HTTP streaming, while `pipeline` does HTTP request for each message.
Label `sendingType` is the connection type to send messages. `message`, `msgapp` and `msgappv2` use HTTP streaming, while `pipeline` does HTTP request for each message.
Label `msgType` is the type of raft message. `MsgApp` is log replication message; `MsgSnap` is snapshot install message; `MsgProp` is proposal forward message; the others are used to maintain raft internal status. If you have a large snapshot, you would expect a long msgSnap sending latency. For other types of messages, you would expect low latency, which is comparable to your ping latency if you have enough network bandwidth.

12
rafthttp/metrics.go
View File

@ -30,7 +30,7 @@ var (
Name: "message_sent_latency_microseconds",
Help: "message sent latency distributions.",
},
[]string{"channel", "remoteID", "msgType"},
[]string{"sendingType", "remoteID", "msgType"},
)
msgSentFailed = prometheus.NewCounterVec(prometheus.CounterOpts{
@ -39,7 +39,7 @@ var (
Name: "message_sent_failed_total",
Help: "The total number of failed messages sent.",
},
[]string{"channel", "remoteID", "msgType"},
[]string{"sendingType", "remoteID", "msgType"},
)
)
@ -48,18 +48,18 @@ func init() {
prometheus.MustRegister(msgSentFailed)
}
func reportSentDuration(channel string, m raftpb.Message, duration time.Duration) {
func reportSentDuration(sendingType string, m raftpb.Message, duration time.Duration) {
typ := m.Type.String()
if isLinkHeartbeatMessage(m) {
typ = "MsgLinkHeartbeat"
}
msgSentDuration.WithLabelValues(channel, types.ID(m.To).String(), typ).Observe(float64(duration.Nanoseconds() / int64(time.Microsecond)))
msgSentDuration.WithLabelValues(sendingType, types.ID(m.To).String(), typ).Observe(float64(duration.Nanoseconds() / int64(time.Microsecond)))
}
func reportSentFailure(channel string, m raftpb.Message) {
func reportSentFailure(sendingType string, m raftpb.Message) {
typ := m.Type.String()
if isLinkHeartbeatMessage(m) {
typ = "MsgLinkHeartbeat"
}
msgSentFailed.WithLabelValues(channel, types.ID(m.To).String(), typ).Inc()
msgSentFailed.WithLabelValues(sendingType, types.ID(m.To).String(), typ).Inc()
}