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

Add better observability to queryReadiness #5946

Merged
merged 9 commits into from
Apr 28, 2022
Merged

Add better observability to queryReadiness #5946

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
0f90a99
Add two metrics to the IndexGateway.
DylanGuedes Apr 18, 2022
63321f9
Remove `usersToBeQueryReadyForTotal`.
DylanGuedes Apr 20, 2022
cbbd4b6
Rename metric help text to not mislead people.
DylanGuedes Apr 20, 2022
f9c520a
Log queryReadiness duration.
DylanGuedes Apr 20, 2022
b4737dc
Fix where log message and duration and triggered.
DylanGuedes Apr 25, 2022
ddd4380
Join users list in a single string.
DylanGuedes Apr 26, 2022
e90a816
Tweak queryReadiness log messages.
DylanGuedes Apr 27, 2022
bb1ed79
Ensure queryReadinessDuration metric.
DylanGuedes Apr 28, 2022
b1da770
noop
DylanGuedes Apr 28, 2022
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
14 changes: 14 additions & 0 deletions pkg/storage/stores/shipper/downloads/metrics.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@ package downloads
import (
"github.com/prometheus/client_golang/prometheus"
"github.com/prometheus/client_golang/prometheus/promauto"
"github.com/weaveworks/common/instrument"
)

const (
Expand All @@ -14,6 +15,8 @@ type metrics struct {
queryTimeTableDownloadDurationSeconds *prometheus.CounterVec
tablesSyncOperationTotal *prometheus.CounterVec
tablesDownloadOperationDurationSeconds prometheus.Gauge
ensureQueryReadinessDurationSeconds prometheus.Histogram
usersToBeQueryReadyForTotal prometheus.Gauge
}

func newMetrics(r prometheus.Registerer) *metrics {
Expand All @@ -23,6 +26,17 @@ func newMetrics(r prometheus.Registerer) *metrics {
Name: "query_time_table_download_duration_seconds",
Help: "Time (in seconds) spent in downloading of files per table at query time",
}, []string{"table"}),
ensureQueryReadinessDurationSeconds: promauto.With(r).NewHistogram(prometheus.HistogramOpts{
Namespace: "loki_boltdb_shipper",
Name: "query_readiness_duration_seconds",
Help: "Time (in seconds) spent making an index gateway ready to be queried",
Buckets: instrument.DefBuckets,
}),
usersToBeQueryReadyForTotal: promauto.With(r).NewGauge(prometheus.GaugeOpts{
Namespace: "loki_boltdb_shipper",
Name: "users_to_be_query_ready_for_total",
Help: "Total number of users an index gateway instance has to be query ready for.",
}),
DylanGuedes marked this conversation as resolved.
Show resolved Hide resolved
tablesSyncOperationTotal: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: "loki_boltdb_shipper",
Name: "tables_sync_operation_total",
Expand Down
11 changes: 11 additions & 0 deletions pkg/storage/stores/shipper/downloads/table_manager.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -241,6 +241,14 @@ func (tm *TableManager) cleanupCache() error {

// ensureQueryReadiness compares tables required for being query ready with the tables we already have and downloads the missing ones.
func (tm *TableManager) ensureQueryReadiness(ctx context.Context) error {
start := time.Now()
usersToBeQueryReadyLen := 0
defer func() {
duration := time.Since(start)
tm.metrics.ensureQueryReadinessDurationSeconds.Observe(duration.Seconds())
DylanGuedes marked this conversation as resolved.
Show resolved Hide resolved
tm.metrics.usersToBeQueryReadyForTotal.Set(float64(usersToBeQueryReadyLen))
}()

activeTableNumber := getActiveTableNumber()

// find the largest query readiness number
Expand Down Expand Up @@ -309,6 +317,9 @@ func (tm *TableManager) ensureQueryReadiness(ctx context.Context) error {
return err
}

usersToBeQueryReadyLen += len(usersToBeQueryReadyFor)
DylanGuedes marked this conversation as resolved.
Show resolved Hide resolved

level.Debug(util_log.Logger).Log("msg", "instance should be query ready for users", "users", usersToBeQueryReadyFor)
DylanGuedes marked this conversation as resolved.
Show resolved Hide resolved
if err := table.EnsureQueryReadiness(ctx, usersToBeQueryReadyFor); err != nil {
return err
}
Expand Down