mirror of
https://github.com/etcd-io/etcd.git
synced 2024-09-27 06:25:44 +00:00
Merge pull request #3323 from xiang90/cl_health
etcdctl: use health endpoint to greatly simplify health checking
This commit is contained in:
commit
b0303e948c
@ -2,12 +2,10 @@ package command
|
||||
|
||||
import (
|
||||
"encoding/json"
|
||||
"errors"
|
||||
"fmt"
|
||||
"net/http"
|
||||
"os"
|
||||
"os/signal"
|
||||
"sort"
|
||||
"time"
|
||||
|
||||
"github.com/coreos/etcd/Godeps/_workspace/src/github.com/codegangsta/cli"
|
||||
@ -42,124 +40,60 @@ func handleClusterHealth(c *cli.Context) {
|
||||
handleError(ExitServerError, err)
|
||||
}
|
||||
|
||||
// TODO: update members when forever is set.
|
||||
hc := http.Client{
|
||||
Transport: tr,
|
||||
}
|
||||
|
||||
mi := mustNewMembersAPI(c)
|
||||
ms, err := mi.List(context.TODO())
|
||||
if err != nil {
|
||||
fmt.Println("cluster may be unhealthy: failed to list members")
|
||||
handleError(ExitServerError, err)
|
||||
}
|
||||
cl := make([]string, 0)
|
||||
for _, m := range ms {
|
||||
cl = append(cl, m.ClientURLs...)
|
||||
}
|
||||
|
||||
for {
|
||||
// check the /health endpoint of all members first
|
||||
health := false
|
||||
for _, m := range ms {
|
||||
checked := false
|
||||
for _, url := range m.ClientURLs {
|
||||
resp, err := hc.Get(url + "/health")
|
||||
if err != nil {
|
||||
fmt.Printf("failed to check the health of member %s on %s: %v\n", m.ID, url, err)
|
||||
continue
|
||||
}
|
||||
|
||||
ep, rs0, err := getLeaderStatus(tr, cl)
|
||||
if err != nil {
|
||||
fmt.Println("cluster may be unhealthy: failed to connect", cl)
|
||||
if forever {
|
||||
time.Sleep(10 * time.Second)
|
||||
continue
|
||||
result := struct{ Health string }{}
|
||||
d := json.NewDecoder(resp.Body)
|
||||
err = d.Decode(&result)
|
||||
resp.Body.Close()
|
||||
if err != nil {
|
||||
fmt.Printf("failed to check the health of member %s on %s: %v\n", m.ID, url, err)
|
||||
continue
|
||||
}
|
||||
|
||||
checked = true
|
||||
if result.Health == "true" {
|
||||
checked = true
|
||||
fmt.Printf("member %s is healthy: got healthy result from %s\n", m.ID, url)
|
||||
} else {
|
||||
fmt.Printf("member %s is unhealthy: got unhealthy result from %s\n", m.ID, url)
|
||||
}
|
||||
break
|
||||
}
|
||||
os.Exit(1)
|
||||
}
|
||||
|
||||
time.Sleep(time.Second)
|
||||
|
||||
// are all the members makeing progress?
|
||||
_, rs1, err := getLeaderStatus(tr, []string{ep})
|
||||
if err != nil {
|
||||
fmt.Println("cluster is unhealthy")
|
||||
if forever {
|
||||
time.Sleep(10 * time.Second)
|
||||
continue
|
||||
if !checked {
|
||||
fmt.Printf("member %s is unreachable: %v are all unreachable\n", m.ID, m.ClientURLs)
|
||||
}
|
||||
os.Exit(1)
|
||||
}
|
||||
|
||||
if rs1.Commit > rs0.Commit {
|
||||
fmt.Printf("cluster is healthy: raft is making progress [commit index: %v->%v]\n", rs0.Commit, rs1.Commit)
|
||||
if health {
|
||||
fmt.Println("cluster is healthy")
|
||||
} else {
|
||||
fmt.Printf("cluster is unhealthy: raft is not making progress [commit index: %v]\n", rs0.Commit)
|
||||
}
|
||||
fmt.Printf("leader is %v\n", rs0.Lead)
|
||||
|
||||
var prints []string
|
||||
|
||||
for id, pr0 := range rs0.Progress {
|
||||
pr1, ok := rs1.Progress[id]
|
||||
if !ok {
|
||||
// TODO: forever should handle configuration change.
|
||||
fmt.Println("Cluster configuration changed during health checking. Please retry.")
|
||||
os.Exit(1)
|
||||
}
|
||||
if pr1.Match <= pr0.Match {
|
||||
prints = append(prints, fmt.Sprintf("member %s is unhealthy: raft is not making progress [match: %v->%v]\n", id, pr0.Match, pr1.Match))
|
||||
} else {
|
||||
prints = append(prints, fmt.Sprintf("member %s is healthy: raft is making progress [match: %v->%v]\n", id, pr0.Match, pr1.Match))
|
||||
}
|
||||
}
|
||||
|
||||
sort.Strings(prints)
|
||||
for _, p := range prints {
|
||||
fmt.Print(p)
|
||||
fmt.Println("cluster is unhealthy")
|
||||
}
|
||||
|
||||
if !forever {
|
||||
return
|
||||
break
|
||||
}
|
||||
|
||||
fmt.Printf("\nnext check after 10 second...\n\n")
|
||||
time.Sleep(10 * time.Second)
|
||||
}
|
||||
}
|
||||
|
||||
type raftStatus struct {
|
||||
ID string `json:"id"`
|
||||
Term uint64 `json:"term"`
|
||||
Vote string `json:"vote"`
|
||||
Commit uint64 `json:"commit"`
|
||||
Lead string `json:"lead"`
|
||||
RaftState string `json:"raftState"`
|
||||
Progress map[string]struct {
|
||||
Match uint64 `json:"match"`
|
||||
Next uint64 `json:"next"`
|
||||
State string `json:"state"`
|
||||
} `json:"progress"`
|
||||
}
|
||||
|
||||
type vars struct {
|
||||
RaftStatus raftStatus `json:"raft.status"`
|
||||
}
|
||||
|
||||
func getLeaderStatus(tr *http.Transport, endpoints []string) (string, raftStatus, error) {
|
||||
// TODO: use new etcd client
|
||||
httpclient := http.Client{
|
||||
Transport: tr,
|
||||
}
|
||||
|
||||
for _, ep := range endpoints {
|
||||
resp, err := httpclient.Get(ep + "/debug/vars")
|
||||
if err != nil {
|
||||
continue
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
if resp.StatusCode != http.StatusOK {
|
||||
continue
|
||||
}
|
||||
|
||||
vs := &vars{}
|
||||
d := json.NewDecoder(resp.Body)
|
||||
err = d.Decode(vs)
|
||||
if err != nil {
|
||||
continue
|
||||
}
|
||||
if vs.RaftStatus.Lead != vs.RaftStatus.ID {
|
||||
continue
|
||||
}
|
||||
return ep, vs.RaftStatus, nil
|
||||
}
|
||||
return "", raftStatus{}, errors.New("no leader")
|
||||
}
|
||||
|
Loading…
x
Reference in New Issue
Block a user