Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

etcdserver: clarify read index warnings, add metrics #9897

Merged
merged 2 commits into from
Jul 3, 2018
Merged

etcdserver: clarify read index warnings, add metrics #9897

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
4733a1d
etcdserver: clarify read index warnings
gyuho Jul 2, 2018
37000cc
etcdserver: add "etcd_server_slow_read_indexes_total"
gyuho Jul 2, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions etcdserver/metrics.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,6 +78,12 @@ var (
Name: "proposals_failed_total",
Help: "The total number of failed proposals seen.",
})
slowReadIndex = prometheus.NewCounter(prometheus.CounterOpts{
Namespace: "etcd",
Subsystem: "server",
Name: "slow_read_indexes_total",
Help: "The total number of pending read indexes not in sync with leader's or timed out read index requests.",
})
leaseExpired = prometheus.NewCounter(prometheus.CounterOpts{
Namespace: "etcd_debugging",
Subsystem: "server",
Expand Down Expand Up @@ -109,6 +115,7 @@ func init() {
prometheus.MustRegister(proposalsApplied)
prometheus.MustRegister(proposalsPending)
prometheus.MustRegister(proposalsFailed)
prometheus.MustRegister(slowReadIndex)
prometheus.MustRegister(leaseExpired)
prometheus.MustRegister(quotaBackendBytes)
prometheus.MustRegister(currentVersion)
Expand Down
24 changes: 15 additions & 9 deletions etcdserver/v3_server.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,6 @@ import (
"bytes"
"context"
"encoding/binary"
"fmt"
"time"

"github.com/coreos/etcd/auth"
Expand Down Expand Up @@ -632,8 +631,9 @@ func (s *EtcdServer) linearizableReadLoop() {
var rs raft.ReadState

for {
ctx := make([]byte, 8)
binary.BigEndian.PutUint64(ctx, s.reqIDGen.Next())
ctxToSend := make([]byte, 8)
id1 := s.reqIDGen.Next()
binary.BigEndian.PutUint64(ctxToSend, id1)

select {
case <-s.readwaitc:
Expand All @@ -650,7 +650,7 @@ func (s *EtcdServer) linearizableReadLoop() {

lg := s.getLogger()
cctx, cancel := context.WithTimeout(context.Background(), s.Cfg.ReqTimeout())
if err := s.r.ReadIndex(cctx, ctx); err != nil {
if err := s.r.ReadIndex(cctx, ctxToSend); err != nil {
cancel()
if err == raft.ErrStopped {
return
Expand All @@ -672,19 +672,24 @@ func (s *EtcdServer) linearizableReadLoop() {
for !timeout && !done {
select {
case rs = <-s.r.readStateC:
done = bytes.Equal(rs.RequestCtx, ctx)
done = bytes.Equal(rs.RequestCtx, ctxToSend)
if !done {
// a previous request might time out. now we should ignore the response of it and
// continue waiting for the response of the current requests.
id2 := uint64(0)
if len(rs.RequestCtx) == 8 {
id2 = binary.BigEndian.Uint64(rs.RequestCtx)
}
if lg != nil {
lg.Warn(
"ignored out-of-date read index response",
zap.String("ctx-expected", fmt.Sprintf("%+v", string(rs.RequestCtx))),
zap.String("ctx-got", fmt.Sprintf("%+v", string(ctx))),
"ignored out-of-date read index response; local node read indexes queueing up and waiting to be in sync with leader",
zap.Uint64("sent-request-id", id1),
zap.Uint64("received-request-id", id2),
)
} else {
plog.Warningf("ignored out-of-date read index response (want %v, got %v)", rs.RequestCtx, ctx)
plog.Warningf("ignored out-of-date read index response; local node read indexes queueing up and waiting to be in sync with leader (request ID want %d, got %d)", id1, id2)
}
slowReadIndex.Inc()
}
case <-time.After(s.Cfg.ReqTimeout()):
if lg != nil {
Expand All @@ -694,6 +699,7 @@ func (s *EtcdServer) linearizableReadLoop() {
}
nr.notify(ErrTimeout)
timeout = true
slowReadIndex.Inc()
case <-s.stopping:
return
}
Expand Down