VictoriaMetrics/lib/storage/search_test.go
Aliaksandr Valialkin 86a1cd700b lib/storage: remove inmemory index for recent hour, since it uses too much memory
Production workload shows that the index requires ~4Kb of RAM per active time series.
This is too much for high number of active time series, so let's delete this index.

Now the queries should fall back to the index for the current day instead of the index
for the recent hour. The query performance for the current day index should be good enough
given the 100M rows/sec scan speed per CPU core.
2019-11-13 17:58:07 +02:00

284 lines
7.9 KiB
Go

package storage
import (
"bytes"
"fmt"
"math/rand"
"os"
"reflect"
"regexp"
"sort"
"testing"
"testing/quick"
"time"
)
func TestSearchQueryMarshalUnmarshal(t *testing.T) {
rnd := rand.New(rand.NewSource(0))
typ := reflect.TypeOf(&SearchQuery{})
var buf []byte
var sq2 SearchQuery
for i := 0; i < 1000; i++ {
v, ok := quick.Value(typ, rnd)
if !ok {
t.Fatalf("cannot create random SearchQuery via testing/quick.Value")
}
sq1 := v.Interface().(*SearchQuery)
if sq1 == nil {
// Skip nil sq1.
continue
}
buf = sq1.Marshal(buf[:0])
tail, err := sq2.Unmarshal(buf)
if err != nil {
t.Fatalf("cannot unmarshal SearchQuery: %s", err)
}
if len(tail) > 0 {
t.Fatalf("unexpected tail left after SearchQuery unmarshaling; tail (len=%d): %q", len(tail), tail)
}
if sq1.MinTimestamp != sq2.MinTimestamp {
t.Fatalf("unexpected MinTimestamp; got %d; want %d", sq2.MinTimestamp, sq1.MinTimestamp)
}
if sq1.MaxTimestamp != sq2.MaxTimestamp {
t.Fatalf("unexpected MaxTimestamp; got %d; want %d", sq2.MaxTimestamp, sq1.MaxTimestamp)
}
if len(sq1.TagFilterss) != len(sq2.TagFilterss) {
t.Fatalf("unexpected TagFilterss len; got %d; want %d", len(sq2.TagFilterss), len(sq1.TagFilterss))
}
for ii := range sq1.TagFilterss {
tagFilters1 := sq1.TagFilterss[ii]
tagFilters2 := sq2.TagFilterss[ii]
for j := range tagFilters1 {
tf1 := &tagFilters1[j]
tf2 := &tagFilters2[j]
if string(tf1.Key) != string(tf2.Key) {
t.Fatalf("unexpected Key on iteration %d,%d; got %X; want %X", i, j, tf2.Key, tf1.Key)
}
if string(tf1.Value) != string(tf2.Value) {
t.Fatalf("unexpected Value on iteration %d,%d; got %X; want %X", i, j, tf2.Value, tf1.Value)
}
if tf1.IsNegative != tf2.IsNegative {
t.Fatalf("unexpected IsNegative on iteration %d,%d; got %v; want %v", i, j, tf2.IsNegative, tf1.IsNegative)
}
if tf1.IsRegexp != tf2.IsRegexp {
t.Fatalf("unexpected IsRegexp on iteration %d,%d; got %v; want %v", i, j, tf2.IsRegexp, tf1.IsRegexp)
}
}
}
}
}
func TestSearch(t *testing.T) {
t.Run("global_inverted_index", func(t *testing.T) {
testSearchGeneric(t, false)
})
t.Run("perday_inverted_index", func(t *testing.T) {
testSearchGeneric(t, true)
})
}
func testSearchGeneric(t *testing.T, forcePerDayInvertedIndex bool) {
path := fmt.Sprintf("TestSearch_%v", forcePerDayInvertedIndex)
st, err := OpenStorage(path, 0)
if err != nil {
t.Fatalf("cannot open storage %q: %s", path, err)
}
defer func() {
st.MustClose()
if err := os.RemoveAll(path); err != nil {
t.Fatalf("cannot remove storage %q: %s", path, err)
}
}()
// Add rows to storage.
const rowsCount = 2e4
const rowsPerBlock = 1e3
const metricGroupsCount = rowsCount / 5
const accountsCount = 2
mrs := make([]MetricRow, rowsCount)
var mn MetricName
mn.Tags = []Tag{
{[]byte("job"), []byte("super-service")},
{[]byte("instance"), []byte("8.8.8.8:1234")},
}
startTimestamp := timestampFromTime(time.Now())
startTimestamp -= startTimestamp % (1e3 * 60 * 30)
blockRowsCount := 0
for i := 0; i < rowsCount; i++ {
mn.MetricGroup = []byte(fmt.Sprintf("metric_%d", i%metricGroupsCount))
mr := &mrs[i]
mr.MetricNameRaw = mn.marshalRaw(nil)
mr.Timestamp = startTimestamp + int64(i)
mr.Value = float64(i)
blockRowsCount++
if blockRowsCount == rowsPerBlock {
if err := st.AddRows(mrs[i-blockRowsCount+1:i+1], defaultPrecisionBits); err != nil {
t.Fatalf("cannot add rows %d-%d: %s", i-blockRowsCount+1, i+1, err)
}
blockRowsCount = 0
}
}
if err := st.AddRows(mrs[rowsCount-blockRowsCount:], defaultPrecisionBits); err != nil {
t.Fatalf("cannot add rows %v-%v: %s", rowsCount-blockRowsCount, rowsCount, err)
}
endTimestamp := mrs[len(mrs)-1].Timestamp
// Re-open the storage in order to flush all the pending cached data.
st.MustClose()
st, err = OpenStorage(path, 0)
if err != nil {
t.Fatalf("cannot re-open storage %q: %s", path, err)
}
if forcePerDayInvertedIndex {
idb := st.idb()
idb.startDateForPerDayInvertedIndex = 0
idb.doExtDB(func(extDB *indexDB) {
extDB.startDateForPerDayInvertedIndex = 0
})
}
// Run search.
tr := TimeRange{
MinTimestamp: startTimestamp + int64(rowsCount)/3,
MaxTimestamp: endTimestamp - int64(rowsCount)/3,
}
t.Run("serial", func(t *testing.T) {
if err := testSearchInternal(st, tr, mrs, accountsCount); err != nil {
t.Fatalf("unexpected error: %s", err)
}
})
t.Run("concurrent", func(t *testing.T) {
ch := make(chan error, 3)
for i := 0; i < cap(ch); i++ {
go func() {
ch <- testSearchInternal(st, tr, mrs, accountsCount)
}()
}
for i := 0; i < cap(ch); i++ {
select {
case err := <-ch:
if err != nil {
t.Fatalf("unexpected error: %s", err)
}
case <-time.After(10 * time.Second):
t.Fatalf("timeout")
}
}
})
}
func testSearchInternal(st *Storage, tr TimeRange, mrs []MetricRow, accountsCount int) error {
var s Search
for i := 0; i < 10; i++ {
// Prepare TagFilters for search.
tfs := NewTagFilters()
metricGroupRe := fmt.Sprintf(`metric_\d*%d%d`, i, i)
if err := tfs.Add(nil, []byte(metricGroupRe), false, true); err != nil {
return fmt.Errorf("cannot add metricGroupRe=%q: %s", metricGroupRe, err)
}
if err := tfs.Add([]byte("job"), []byte("nonexisting-service"), true, false); err != nil {
return fmt.Errorf("cannot add tag filter %q=%q: %s", "job", "nonexsitsing-service", err)
}
if err := tfs.Add([]byte("instance"), []byte(".*"), false, true); err != nil {
return fmt.Errorf("cannot add tag filter %q=%q: %s", "instance", ".*", err)
}
// Build extectedMrs.
var expectedMrs []MetricRow
metricGroupRegexp := regexp.MustCompile(fmt.Sprintf("^%s$", metricGroupRe))
var mn MetricName
for j := range mrs {
mr := &mrs[j]
if mr.Timestamp < tr.MinTimestamp || mr.Timestamp > tr.MaxTimestamp {
continue
}
if err := mn.unmarshalRaw(mr.MetricNameRaw); err != nil {
return fmt.Errorf("cannot unmarshal MetricName: %s", err)
}
if !metricGroupRegexp.Match(mn.MetricGroup) {
continue
}
expectedMrs = append(expectedMrs, *mr)
}
// Search
s.Init(st, []*TagFilters{tfs}, tr, true, 1e5)
var mbs []MetricBlock
for s.NextMetricBlock() {
var b Block
b.CopyFrom(s.MetricBlock.Block)
var mb MetricBlock
mb.MetricName = append(mb.MetricName, s.MetricBlock.MetricName...)
mb.Block = &b
mbs = append(mbs, mb)
}
if err := s.Error(); err != nil {
return fmt.Errorf("search error: %s", err)
}
s.MustClose()
// Build foundMrs.
var foundMrs []MetricRow
for _, mb := range mbs {
rb := newTestRawBlock(mb.Block, tr)
if err := mn.Unmarshal(mb.MetricName); err != nil {
return fmt.Errorf("cannot unmarshal MetricName: %s", err)
}
metricNameRaw := mn.marshalRaw(nil)
for i, timestamp := range rb.Timestamps {
mr := MetricRow{
MetricNameRaw: metricNameRaw,
Timestamp: timestamp,
Value: rb.Values[i],
}
foundMrs = append(foundMrs, mr)
}
}
// Compare expectedMrs to foundMrs.
sort.Slice(expectedMrs, func(i, j int) bool {
a, b := &expectedMrs[i], &expectedMrs[j]
cmp := bytes.Compare(a.MetricNameRaw, b.MetricNameRaw)
if cmp < 0 {
return true
}
if cmp > 0 {
return false
}
return a.Timestamp < b.Timestamp
})
sort.Slice(foundMrs, func(i, j int) bool {
a, b := &foundMrs[i], &foundMrs[j]
cmp := bytes.Compare(a.MetricNameRaw, b.MetricNameRaw)
if cmp < 0 {
return true
}
if cmp > 0 {
return false
}
return a.Timestamp < b.Timestamp
})
if !reflect.DeepEqual(expectedMrs, foundMrs) {
return fmt.Errorf("unexpected rows found;\ngot\n%s\nwant\n%s", mrsToString(foundMrs), mrsToString(expectedMrs))
}
}
return nil
}
func mrsToString(mrs []MetricRow) string {
var bb bytes.Buffer
fmt.Fprintf(&bb, "len=%d\n", len(mrs))
for i := range mrs {
mr := &mrs[i]
fmt.Fprintf(&bb, "[%q, Timestamp=%d, Value=%f]\n", mr.MetricNameRaw, mr.Timestamp, mr.Value)
}
return bb.String()
}