Like Prometheus, but for logs.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 
loki/pkg/logql/metrics.go

434 lines
15 KiB

package logql
import (
"context"
"strconv"
"strings"
"time"
"github.com/dustin/go-humanize"
"github.com/go-kit/log"
"github.com/go-kit/log/level"
"github.com/prometheus/client_golang/prometheus"
"github.com/prometheus/client_golang/prometheus/promauto"
promql_parser "github.com/prometheus/prometheus/promql/parser"
"github.com/grafana/loki/pkg/analytics"
"github.com/grafana/loki/pkg/logql/syntax"
"github.com/grafana/loki/pkg/logqlmodel"
logql_stats "github.com/grafana/loki/pkg/logqlmodel/stats"
"github.com/grafana/loki/pkg/querier/astmapper"
"github.com/grafana/loki/pkg/util"
"github.com/grafana/loki/pkg/util/constants"
"github.com/grafana/loki/pkg/util/httpreq"
util_log "github.com/grafana/loki/pkg/util/log"
"github.com/grafana/loki/pkg/util/spanlogger"
)
const (
QueryTypeMetric = "metric"
QueryTypeFilter = "filter"
QueryTypeLimited = "limited"
QueryTypeLabels = "labels"
QueryTypeSeries = "series"
QueryTypeStats = "stats"
QueryTypeVolume = "volume"
latencyTypeSlow = "slow"
latencyTypeFast = "fast"
slowQueryThresholdSecond = float64(10)
)
var (
bytesPerSecond = promauto.NewHistogramVec(prometheus.HistogramOpts{
Namespace: constants.Loki,
Name: "logql_querystats_bytes_processed_per_seconds",
Help: "Distribution of bytes processed per second for LogQL queries.",
// 50MB 100MB 200MB 400MB 600MB 800MB 1GB 2GB 3GB 4GB 5GB 6GB 7GB 8GB 9GB 10GB 15GB 20GB 30GB, 40GB 50GB 60GB
Buckets: []float64{50 * 1e6, 100 * 1e6, 400 * 1e6, 600 * 1e6, 800 * 1e6, 1 * 1e9, 2 * 1e9, 3 * 1e9, 4 * 1e9, 5 * 1e9, 6 * 1e9, 7 * 1e9, 8 * 1e9, 9 * 1e9, 10 * 1e9, 15 * 1e9, 20 * 1e9, 30 * 1e9, 40 * 1e9, 50 * 1e9, 60 * 1e9},
}, []string{"status_code", "type", "range", "latency_type", "sharded"})
execLatency = promauto.NewHistogramVec(prometheus.HistogramOpts{
Namespace: constants.Loki,
Name: "logql_querystats_latency_seconds",
Help: "Distribution of latency for LogQL queries.",
// 0.25 0.5 1 2 4 8 16 32 64 128
Buckets: prometheus.ExponentialBuckets(0.250, 2, 10),
}, []string{"status_code", "type", "range"})
chunkDownloadLatency = promauto.NewHistogramVec(prometheus.HistogramOpts{
Namespace: constants.Loki,
Name: "logql_querystats_chunk_download_latency_seconds",
Help: "Distribution of chunk downloads latency for LogQL queries.",
// 0.25 0.5 1 2 4 8 16 32 64 128
Buckets: prometheus.ExponentialBuckets(0.250, 2, 10),
}, []string{"status_code", "type", "range"})
duplicatesTotal = promauto.NewCounter(prometheus.CounterOpts{
Namespace: constants.Loki,
Name: "logql_querystats_duplicates_total",
Help: "Total count of duplicates found while executing LogQL queries.",
})
chunkDownloadedTotal = promauto.NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Name: "logql_querystats_downloaded_chunk_total",
Help: "Total count of chunks downloaded found while executing LogQL queries.",
}, []string{"status_code", "type", "range"})
ingesterLineTotal = promauto.NewCounter(prometheus.CounterOpts{
Namespace: constants.Loki,
Name: "logql_querystats_ingester_sent_lines_total",
Help: "Total count of lines sent from ingesters while executing LogQL queries.",
})
bytePerSecondMetricUsage = analytics.NewStatistics("query_metric_bytes_per_second")
bytePerSecondLogUsage = analytics.NewStatistics("query_log_bytes_per_second")
linePerSecondMetricUsage = analytics.NewStatistics("query_metric_lines_per_second")
linePerSecondLogUsage = analytics.NewStatistics("query_log_lines_per_second")
)
func RecordRangeAndInstantQueryMetrics(
ctx context.Context,
log log.Logger,
p Params,
status string,
stats logql_stats.Result,
result promql_parser.Value,
) {
var (
logger = fixLogger(ctx, log)
rt = string(GetRangeType(p))
latencyType = latencyTypeFast
returnedLines = 0
)
queryType, err := QueryType(p.GetExpression())
if err != nil {
level.Warn(logger).Log("msg", "error parsing query type", "err", err)
}
// Tag throughput metric by latency type based on a threshold.
// Latency below the threshold is fast, above is slow.
if stats.Summary.ExecTime > slowQueryThresholdSecond {
latencyType = latencyTypeSlow
}
if result != nil && result.Type() == logqlmodel.ValueTypeStreams {
returnedLines = int(result.(logqlmodel.Streams).Lines())
}
queryTags, _ := ctx.Value(httpreq.QueryTagsHTTPHeader).(string) // it's ok to be empty.
logValues := make([]interface{}, 0, 50)
logValues = append(logValues, []interface{}{
"latency", latencyType, // this can be used to filter log lines.
"query", p.QueryString(),
"query_hash", util.HashedQuery(p.QueryString()),
"query_type", queryType,
"range_type", rt,
"length", p.End().Sub(p.Start()),
"start_delta", time.Since(p.Start()),
"end_delta", time.Since(p.End()),
"step", p.Step(),
"duration", logql_stats.ConvertSecondsToNanoseconds(stats.Summary.ExecTime),
"status", status,
"limit", p.Limit(),
"returned_lines", returnedLines,
"throughput", strings.Replace(humanize.Bytes(uint64(stats.Summary.BytesProcessedPerSecond)), " ", "", 1),
"total_bytes", strings.Replace(humanize.Bytes(uint64(stats.Summary.TotalBytesProcessed)), " ", "", 1),
"total_bytes_structured_metadata", strings.Replace(humanize.Bytes(uint64(stats.Summary.TotalStructuredMetadataBytesProcessed)), " ", "", 1),
"lines_per_second", stats.Summary.LinesProcessedPerSecond,
"total_lines", stats.Summary.TotalLinesProcessed,
"post_filter_lines", stats.Summary.TotalPostFilterLines,
"total_entries", stats.Summary.TotalEntriesReturned,
"store_chunks_download_time", stats.ChunksDownloadTime(),
"queue_time", logql_stats.ConvertSecondsToNanoseconds(stats.Summary.QueueTime),
"splits", stats.Summary.Splits,
"shards", stats.Summary.Shards,
"query_referenced_structured_metadata", stats.QueryReferencedStructuredMetadata(),
"chunk_refs_fetch_time", stats.ChunkRefsFetchTime(),
"cache_chunk_req", stats.Caches.Chunk.EntriesRequested,
"cache_chunk_hit", stats.Caches.Chunk.EntriesFound,
"cache_chunk_bytes_stored", stats.Caches.Chunk.BytesSent,
"cache_chunk_bytes_fetched", stats.Caches.Chunk.BytesReceived,
"cache_chunk_download_time", stats.Caches.Chunk.CacheDownloadTime(),
"cache_index_req", stats.Caches.Index.EntriesRequested,
"cache_index_hit", stats.Caches.Index.EntriesFound,
"cache_index_download_time", stats.Caches.Index.CacheDownloadTime(),
"cache_stats_results_req", stats.Caches.StatsResult.EntriesRequested,
"cache_stats_results_hit", stats.Caches.StatsResult.EntriesFound,
"cache_stats_results_download_time", stats.Caches.StatsResult.CacheDownloadTime(),
"cache_volume_results_req", stats.Caches.VolumeResult.EntriesRequested,
"cache_volume_results_hit", stats.Caches.VolumeResult.EntriesFound,
"cache_volume_results_download_time", stats.Caches.VolumeResult.CacheDownloadTime(),
"cache_result_req", stats.Caches.Result.EntriesRequested,
"cache_result_hit", stats.Caches.Result.EntriesFound,
"cache_result_download_time", stats.Caches.Result.CacheDownloadTime(),
}...)
logValues = append(logValues, tagsToKeyValues(queryTags)...)
level.Info(logger).Log(
logValues...,
)
sharded := strconv.FormatBool(false)
if stats.Summary.Shards > 1 {
sharded = strconv.FormatBool(true)
}
bytesPerSecond.WithLabelValues(status, queryType, rt, latencyType, sharded).
Observe(float64(stats.Summary.BytesProcessedPerSecond))
execLatency.WithLabelValues(status, queryType, rt).
Observe(stats.Summary.ExecTime)
chunkDownloadLatency.WithLabelValues(status, queryType, rt).
Observe(stats.ChunksDownloadTime().Seconds())
duplicatesTotal.Add(float64(stats.TotalDuplicates()))
chunkDownloadedTotal.WithLabelValues(status, queryType, rt).
Add(float64(stats.TotalChunksDownloaded()))
ingesterLineTotal.Add(float64(stats.Ingester.TotalLinesSent))
recordUsageStats(queryType, stats)
}
func RecordLabelQueryMetrics(
ctx context.Context,
log log.Logger,
start, end time.Time,
label, query, status string,
stats logql_stats.Result,
) {
var (
logger = fixLogger(ctx, log)
latencyType = latencyTypeFast
queryType = QueryTypeLabels
)
// Tag throughput metric by latency type based on a threshold.
// Latency below the threshold is fast, above is slow.
if stats.Summary.ExecTime > slowQueryThresholdSecond {
latencyType = latencyTypeSlow
}
level.Info(logger).Log(
"latency", latencyType,
"query_type", queryType,
"splits", stats.Summary.Splits,
"start", start.Format(time.RFC3339Nano),
"end", end.Format(time.RFC3339Nano),
"start_delta", time.Since(start),
"end_delta", time.Since(end),
"length", end.Sub(start),
"duration", time.Duration(int64(stats.Summary.ExecTime*float64(time.Second))),
"status", status,
"label", label,
"query", query,
"query_hash", util.HashedQuery(query),
"total_entries", stats.Summary.TotalEntriesReturned,
)
execLatency.WithLabelValues(status, queryType, "").Observe(stats.Summary.ExecTime)
}
// fixLogger forces the given logger to include a caller=metrics.go kv pair.
// The given logger might be a spanlogger instance, in which case it only logs caller=spanlogger.go:<line>.
// We use `caller=metrics.go` when querying our logs for performance issues, and some logs were missing.
func fixLogger(ctx context.Context, logger log.Logger) log.Logger {
nl := util_log.WithContext(ctx, logger)
if _, ok := logger.(*spanlogger.SpanLogger); ok {
return log.With(nl, "caller", "metrics.go")
}
return nl
}
func PrintMatches(matches []string) string {
// not using comma (,) as separator as matcher may already have comma (e.g: `{a="b", c="d"}`)
return strings.Join(matches, ":")
}
func RecordSeriesQueryMetrics(ctx context.Context, log log.Logger, start, end time.Time, match []string, status string, shards []string, stats logql_stats.Result) {
var (
logger = fixLogger(ctx, log)
latencyType = latencyTypeFast
queryType = QueryTypeSeries
)
// Tag throughput metric by latency type based on a threshold.
// Latency below the threshold is fast, above is slow.
if stats.Summary.ExecTime > slowQueryThresholdSecond {
latencyType = latencyTypeSlow
}
shard := extractShard(shards)
logValues := make([]interface{}, 0, 15)
logValues = append(logValues,
"latency", latencyType,
"query_type", queryType,
"splits", stats.Summary.Splits,
"start", start.Format(time.RFC3339Nano),
"end", end.Format(time.RFC3339Nano),
"start_delta", time.Since(start),
"end_delta", time.Since(end),
"length", end.Sub(start),
"duration", time.Duration(int64(stats.Summary.ExecTime*float64(time.Second))),
"status", status,
"match", PrintMatches(match),
"query_hash", util.HashedQuery(PrintMatches(match)),
"total_entries", stats.Summary.TotalEntriesReturned)
if shard != nil {
logValues = append(logValues,
"shard_num", shard.Shard,
"shard_count", shard.Of,
)
}
level.Info(logger).Log(logValues...)
execLatency.WithLabelValues(status, queryType, "").Observe(stats.Summary.ExecTime)
}
func RecordStatsQueryMetrics(ctx context.Context, log log.Logger, start, end time.Time, query string, status string, stats logql_stats.Result) {
var (
logger = fixLogger(ctx, log)
latencyType = latencyTypeFast
queryType = QueryTypeStats
)
// Tag throughput metric by latency type based on a threshold.
// Latency below the threshold is fast, above is slow.
if stats.Summary.ExecTime > slowQueryThresholdSecond {
latencyType = latencyTypeSlow
}
logValues := make([]interface{}, 0, 15)
logValues = append(logValues,
"latency", latencyType,
"query_type", queryType,
"start", start.Format(time.RFC3339Nano),
"end", end.Format(time.RFC3339Nano),
"start_delta", time.Since(start),
"end_delta", time.Since(end),
"length", end.Sub(start),
"duration", time.Duration(int64(stats.Summary.ExecTime*float64(time.Second))),
"status", status,
"query", query,
"query_hash", util.HashedQuery(query),
"total_entries", stats.Summary.TotalEntriesReturned)
level.Info(logger).Log(logValues...)
execLatency.WithLabelValues(status, queryType, "").Observe(stats.Summary.ExecTime)
}
func RecordVolumeQueryMetrics(ctx context.Context, log log.Logger, start, end time.Time, query string, limit uint32, step time.Duration, status string, stats logql_stats.Result) {
var (
logger = fixLogger(ctx, log)
latencyType = latencyTypeFast
queryType = QueryTypeVolume
)
// Tag throughput metric by latency type based on a threshold.
// Latency below the threshold is fast, above is slow.
if stats.Summary.ExecTime > slowQueryThresholdSecond {
latencyType = latencyTypeSlow
}
rangeType := "range"
if step == 0 {
rangeType = "instant"
}
level.Info(logger).Log(
"latency", latencyType,
"query_type", queryType,
"query", query,
"query_hash", util.HashedQuery(query),
"start", start.Format(time.RFC3339Nano),
"end", end.Format(time.RFC3339Nano),
"start_delta", time.Since(start),
"end_delta", time.Since(end),
"range_type", rangeType,
"step", step,
"limit", limit,
"length", end.Sub(start),
"duration", time.Duration(int64(stats.Summary.ExecTime*float64(time.Second))),
"status", status,
"splits", stats.Summary.Splits,
"total_entries", stats.Summary.TotalEntriesReturned,
// cache is accumulated by middleware used by the frontend only; logs from the queriers will not show cache stats
"cache_volume_results_req", stats.Caches.VolumeResult.EntriesRequested,
"cache_volume_results_hit", stats.Caches.VolumeResult.EntriesFound,
"cache_volume_results_stored", stats.Caches.VolumeResult.EntriesStored,
"cache_volume_results_download_time", stats.Caches.VolumeResult.CacheDownloadTime(),
)
execLatency.WithLabelValues(status, queryType, "").Observe(stats.Summary.ExecTime)
}
func recordUsageStats(queryType string, stats logql_stats.Result) {
if queryType == QueryTypeMetric {
bytePerSecondMetricUsage.Record(float64(stats.Summary.BytesProcessedPerSecond))
linePerSecondMetricUsage.Record(float64(stats.Summary.LinesProcessedPerSecond))
} else {
bytePerSecondLogUsage.Record(float64(stats.Summary.BytesProcessedPerSecond))
linePerSecondLogUsage.Record(float64(stats.Summary.LinesProcessedPerSecond))
}
}
func QueryType(expr syntax.Expr) (string, error) {
switch e := expr.(type) {
case syntax.SampleExpr:
return QueryTypeMetric, nil
case syntax.LogSelectorExpr:
if e.HasFilter() {
return QueryTypeFilter, nil
}
return QueryTypeLimited, nil
default:
return "", nil
}
}
// tagsToKeyValues converts QueryTags to form that is easy to log.
// e.g: `Source=foo,Feature=beta` -> []interface{}{"source", "foo", "feature", "beta"}
// so that we could log nicely!
// If queryTags is not in canonical form then its completely ignored (e.g: `key1=value1,key2=value`)
func tagsToKeyValues(queryTags string) []interface{} {
toks := strings.FieldsFunc(queryTags, func(r rune) bool {
return r == ','
})
vals := make([]string, 0)
for _, tok := range toks {
val := strings.FieldsFunc(tok, func(r rune) bool {
return r == '='
})
if len(val) != 2 {
continue
}
vals = append(vals, strings.ToLower(val[0]), val[1])
}
res := make([]interface{}, 0, len(vals))
for _, val := range vals {
res = append(res, val)
}
return res
}
func extractShard(shards []string) *astmapper.ShardAnnotation {
if len(shards) == 0 {
return nil
}
var shard astmapper.ShardAnnotation
shard, err := astmapper.ParseShard(shards[0])
if err != nil {
return nil
}
return &shard
}