prometheus/tsdb/wal/wal_test.go

588 lines
15 KiB
Go
Raw Normal View History

// Copyright 2017 The Prometheus Authors
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package wal
import (
"bytes"
"fmt"
"io"
"io/ioutil"
"math/rand"
"os"
"path/filepath"
"testing"
client_testutil "github.com/prometheus/client_golang/prometheus/testutil"
"github.com/stretchr/testify/require"
"go.uber.org/goleak"
"github.com/prometheus/prometheus/tsdb/fileutil"
"github.com/prometheus/prometheus/util/testutil"
)
func TestMain(m *testing.M) {
goleak.VerifyTestMain(m)
}
// TestWALRepair_ReadingError ensures that a repair is run for an error
// when reading a record.
func TestWALRepair_ReadingError(t *testing.T) {
for name, test := range map[string]struct {
corrSgm int // Which segment to corrupt.
corrFunc func(f *os.File) // Func that applies the corruption.
intactRecs int // Total expected records left after the repair.
}{
"torn_last_record": {
2,
func(f *os.File) {
_, err := f.Seek(pageSize*2, 0)
require.NoError(t, err)
_, err = f.Write([]byte{byte(recFirst)})
require.NoError(t, err)
},
8,
},
// Ensures that the page buffer is big enough to fit
// an entire page size without panicking.
Make head Postings only return series in time range benchmark old ns/op new ns/op delta BenchmarkQuerierSelect/Head/1of1000000-8 405805161 120436132 -70.32% BenchmarkQuerierSelect/Head/10of1000000-8 403079620 120624292 -70.07% BenchmarkQuerierSelect/Head/100of1000000-8 404678647 120923522 -70.12% BenchmarkQuerierSelect/Head/1000of1000000-8 403145813 118636563 -70.57% BenchmarkQuerierSelect/Head/10000of1000000-8 405020046 125716206 -68.96% BenchmarkQuerierSelect/Head/100000of1000000-8 426305002 175808499 -58.76% BenchmarkQuerierSelect/Head/1000000of1000000-8 619002108 567013003 -8.40% BenchmarkQuerierSelect/SortedHead/1of1000000-8 1276316086 120281094 -90.58% BenchmarkQuerierSelect/SortedHead/10of1000000-8 1282631170 121836526 -90.50% BenchmarkQuerierSelect/SortedHead/100of1000000-8 1325824787 121174967 -90.86% BenchmarkQuerierSelect/SortedHead/1000of1000000-8 1271386268 121025117 -90.48% BenchmarkQuerierSelect/SortedHead/10000of1000000-8 1280223345 130838948 -89.78% BenchmarkQuerierSelect/SortedHead/100000of1000000-8 1271401620 243635515 -80.84% BenchmarkQuerierSelect/SortedHead/1000000of1000000-8 1360256090 1307744674 -3.86% BenchmarkQuerierSelect/Block/1of1000000-8 748183120 707888498 -5.39% BenchmarkQuerierSelect/Block/10of1000000-8 741084129 716317249 -3.34% BenchmarkQuerierSelect/Block/100of1000000-8 722157273 735624256 +1.86% BenchmarkQuerierSelect/Block/1000of1000000-8 727587744 731981838 +0.60% BenchmarkQuerierSelect/Block/10000of1000000-8 727518578 726860308 -0.09% BenchmarkQuerierSelect/Block/100000of1000000-8 765577046 757382386 -1.07% BenchmarkQuerierSelect/Block/1000000of1000000-8 1126722881 1084779083 -3.72% benchmark old allocs new allocs delta BenchmarkQuerierSelect/Head/1of1000000-8 4000018 24 -100.00% BenchmarkQuerierSelect/Head/10of1000000-8 4000036 82 -100.00% BenchmarkQuerierSelect/Head/100of1000000-8 4000216 625 -99.98% BenchmarkQuerierSelect/Head/1000of1000000-8 4002016 6028 -99.85% BenchmarkQuerierSelect/Head/10000of1000000-8 4020016 60037 -98.51% BenchmarkQuerierSelect/Head/100000of1000000-8 4200016 600047 -85.71% BenchmarkQuerierSelect/Head/1000000of1000000-8 6000016 6000016 +0.00% BenchmarkQuerierSelect/SortedHead/1of1000000-8 4000055 28 -100.00% BenchmarkQuerierSelect/SortedHead/10of1000000-8 4000073 87 -100.00% BenchmarkQuerierSelect/SortedHead/100of1000000-8 4000253 630 -99.98% BenchmarkQuerierSelect/SortedHead/1000of1000000-8 4002053 6036 -99.85% BenchmarkQuerierSelect/SortedHead/10000of1000000-8 4020053 60054 -98.51% BenchmarkQuerierSelect/SortedHead/100000of1000000-8 4200053 600074 -85.71% BenchmarkQuerierSelect/SortedHead/1000000of1000000-8 6000053 6000053 +0.00% BenchmarkQuerierSelect/Block/1of1000000-8 6000021 6000021 +0.00% BenchmarkQuerierSelect/Block/10of1000000-8 6000057 6000057 +0.00% BenchmarkQuerierSelect/Block/100of1000000-8 6000417 6000417 +0.00% BenchmarkQuerierSelect/Block/1000of1000000-8 6004017 6004017 +0.00% BenchmarkQuerierSelect/Block/10000of1000000-8 6040017 6040017 +0.00% BenchmarkQuerierSelect/Block/100000of1000000-8 6400017 6400017 +0.00% BenchmarkQuerierSelect/Block/1000000of1000000-8 10000018 10000018 +0.00% benchmark old bytes new bytes delta BenchmarkQuerierSelect/Head/1of1000000-8 176001177 1392 -100.00% BenchmarkQuerierSelect/Head/10of1000000-8 176002329 4368 -100.00% BenchmarkQuerierSelect/Head/100of1000000-8 176013849 33520 -99.98% BenchmarkQuerierSelect/Head/1000of1000000-8 176129056 321456 -99.82% BenchmarkQuerierSelect/Head/10000of1000000-8 177281049 3427376 -98.07% BenchmarkQuerierSelect/Head/100000of1000000-8 188801049 35055408 -81.43% BenchmarkQuerierSelect/Head/1000000of1000000-8 304001059 304001049 -0.00% BenchmarkQuerierSelect/SortedHead/1of1000000-8 229192188 2488 -100.00% BenchmarkQuerierSelect/SortedHead/10of1000000-8 229193340 5568 -100.00% BenchmarkQuerierSelect/SortedHead/100of1000000-8 229204860 35536 -99.98% BenchmarkQuerierSelect/SortedHead/1000of1000000-8 229320060 345104 -99.85% BenchmarkQuerierSelect/SortedHead/10000of1000000-8 230472060 3894672 -98.31% BenchmarkQuerierSelect/SortedHead/100000of1000000-8 241992060 40511632 -83.26% BenchmarkQuerierSelect/SortedHead/1000000of1000000-8 357192060 357192060 +0.00% BenchmarkQuerierSelect/Block/1of1000000-8 227201516 227201506 -0.00% BenchmarkQuerierSelect/Block/10of1000000-8 227203057 227203041 -0.00% BenchmarkQuerierSelect/Block/100of1000000-8 227217161 227217165 +0.00% BenchmarkQuerierSelect/Block/1000of1000000-8 227358279 227358289 +0.00% BenchmarkQuerierSelect/Block/10000of1000000-8 228769485 228769475 -0.00% BenchmarkQuerierSelect/Block/100000of1000000-8 242881487 242881477 -0.00% BenchmarkQuerierSelect/Block/1000000of1000000-8 384001705 384001705 +0.00% Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2020-01-21 11:30:20 -08:00
// https://github.com/prometheus/tsdb/pull/414
"bad_header": {
1,
func(f *os.File) {
_, err := f.Seek(pageSize, 0)
require.NoError(t, err)
_, err = f.Write([]byte{byte(recPageTerm)})
require.NoError(t, err)
},
4,
},
"bad_fragment_sequence": {
1,
func(f *os.File) {
_, err := f.Seek(pageSize, 0)
require.NoError(t, err)
_, err = f.Write([]byte{byte(recLast)})
require.NoError(t, err)
},
4,
},
"bad_fragment_flag": {
1,
func(f *os.File) {
_, err := f.Seek(pageSize, 0)
require.NoError(t, err)
_, err = f.Write([]byte{123})
require.NoError(t, err)
},
4,
},
"bad_checksum": {
1,
func(f *os.File) {
_, err := f.Seek(pageSize+4, 0)
require.NoError(t, err)
_, err = f.Write([]byte{0})
require.NoError(t, err)
},
4,
},
"bad_length": {
1,
func(f *os.File) {
_, err := f.Seek(pageSize+2, 0)
require.NoError(t, err)
_, err = f.Write([]byte{0})
require.NoError(t, err)
},
4,
},
"bad_content": {
1,
func(f *os.File) {
_, err := f.Seek(pageSize+100, 0)
require.NoError(t, err)
_, err = f.Write([]byte("beef"))
require.NoError(t, err)
},
4,
},
} {
t.Run(name, func(t *testing.T) {
dir, err := ioutil.TempDir("", "wal_repair")
require.NoError(t, err)
defer func() {
require.NoError(t, os.RemoveAll(dir))
}()
// We create 3 segments with 3 records each and
// then corrupt a given record in a given segment.
// As a result we want a repaired WAL with given intact records.
segSize := 3 * pageSize
w, err := NewSize(nil, nil, dir, segSize, false)
require.NoError(t, err)
var records [][]byte
for i := 1; i <= 9; i++ {
b := make([]byte, pageSize-recordHeaderSize)
b[0] = byte(i)
records = append(records, b)
require.NoError(t, w.Log(b))
}
first, last, err := Segments(w.Dir())
require.NoError(t, err)
require.Equal(t, 3, 1+last-first, "wal creation didn't result in expected number of segments")
require.NoError(t, w.Close())
f, err := os.OpenFile(SegmentName(dir, test.corrSgm), os.O_RDWR, 0666)
require.NoError(t, err)
// Apply corruption function.
test.corrFunc(f)
require.NoError(t, f.Close())
w, err = NewSize(nil, nil, dir, segSize, false)
require.NoError(t, err)
defer w.Close()
first, last, err = Segments(w.Dir())
require.NoError(t, err)
// Backfill segments from the most recent checkpoint onwards.
for i := first; i <= last; i++ {
s, err := OpenReadSegment(SegmentName(w.Dir(), i))
require.NoError(t, err)
sr := NewSegmentBufReader(s)
require.NoError(t, err)
r := NewReader(sr)
for r.Next() {
}
//Close the segment so we don't break things on Windows.
s.Close()
// No corruption in this segment.
if r.Err() == nil {
continue
}
require.NoError(t, w.Repair(r.Err()))
break
}
sr, err := NewSegmentsReader(dir)
require.NoError(t, err)
defer sr.Close()
r := NewReader(sr)
var result [][]byte
for r.Next() {
var b []byte
result = append(result, append(b, r.Record()...))
}
require.NoError(t, r.Err())
require.Equal(t, test.intactRecs, len(result), "Wrong number of intact records")
for i, r := range result {
if !bytes.Equal(records[i], r) {
t.Fatalf("record %d diverges: want %x, got %x", i, records[i][:10], r[:10])
}
}
// Make sure there is a new 0 size Segment after the corrupted Segment.
_, last, err = Segments(w.Dir())
require.NoError(t, err)
require.Equal(t, test.corrSgm+1, last)
fi, err := os.Stat(SegmentName(dir, last))
require.NoError(t, err)
require.Equal(t, int64(0), fi.Size())
})
}
}
// TestCorruptAndCarryOn writes a multi-segment WAL; corrupts the first segment and
// ensures that an error during reading that segment are correctly repaired before
// moving to write more records to the WAL.
func TestCorruptAndCarryOn(t *testing.T) {
dir, err := ioutil.TempDir("", "wal_repair")
require.NoError(t, err)
defer func() {
require.NoError(t, os.RemoveAll(dir))
}()
var (
logger = testutil.NewLogger(t)
segmentSize = pageSize * 3
recordSize = (pageSize / 3) - recordHeaderSize
)
// Produce a WAL with a two segments of 3 pages with 3 records each,
// so when we truncate the file we're guaranteed to split a record.
{
w, err := NewSize(logger, nil, dir, segmentSize, false)
require.NoError(t, err)
for i := 0; i < 18; i++ {
buf := make([]byte, recordSize)
_, err := rand.Read(buf)
require.NoError(t, err)
err = w.Log(buf)
require.NoError(t, err)
}
err = w.Close()
require.NoError(t, err)
}
// Check all the segments are the correct size.
{
segments, err := listSegments(dir)
require.NoError(t, err)
for _, segment := range segments {
f, err := os.OpenFile(filepath.Join(dir, fmt.Sprintf("%08d", segment.index)), os.O_RDONLY, 0666)
require.NoError(t, err)
fi, err := f.Stat()
require.NoError(t, err)
t.Log("segment", segment.index, "size", fi.Size())
require.Equal(t, int64(segmentSize), fi.Size())
err = f.Close()
require.NoError(t, err)
}
}
// Truncate the first file, splitting the middle record in the second
// page in half, leaving 4 valid records.
{
f, err := os.OpenFile(filepath.Join(dir, fmt.Sprintf("%08d", 0)), os.O_RDWR, 0666)
require.NoError(t, err)
fi, err := f.Stat()
require.NoError(t, err)
require.Equal(t, int64(segmentSize), fi.Size())
err = f.Truncate(int64(segmentSize / 2))
require.NoError(t, err)
err = f.Close()
require.NoError(t, err)
}
// Now try and repair this WAL, and write 5 more records to it.
{
sr, err := NewSegmentsReader(dir)
require.NoError(t, err)
reader := NewReader(sr)
i := 0
for ; i < 4 && reader.Next(); i++ {
require.Equal(t, recordSize, len(reader.Record()))
}
require.Equal(t, 4, i, "not enough records")
require.False(t, reader.Next(), "unexpected record")
corruptionErr := reader.Err()
require.Error(t, corruptionErr)
err = sr.Close()
require.NoError(t, err)
w, err := NewSize(logger, nil, dir, segmentSize, false)
require.NoError(t, err)
err = w.Repair(corruptionErr)
require.NoError(t, err)
Spelling (#6517) * spelling: alertmanager Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: attributes Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: autocomplete Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: bootstrap Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: caught Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: chunkenc Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: compaction Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: corrupted Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: deletable Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: expected Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: fine-grained Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: initialized Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: iteration Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: javascript Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: multiple Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: number Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: overlapping Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: possible Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: postings Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: procedure Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: programmatic Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: queuing Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: querier Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: repairing Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: received Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: reproducible Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: retention Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: sample Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: segements Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: semantic Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: software [LICENSE] Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: staging Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: timestamp Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: unfortunately Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: uvarint Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: subsequently Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: ressamples Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>
2020-01-02 06:54:09 -08:00
// Ensure that we have a completely clean slate after repairing.
require.Equal(t, w.segment.Index(), 1) // We corrupted segment 0.
require.Equal(t, w.donePages, 0)
for i := 0; i < 5; i++ {
buf := make([]byte, recordSize)
_, err := rand.Read(buf)
require.NoError(t, err)
err = w.Log(buf)
require.NoError(t, err)
}
err = w.Close()
require.NoError(t, err)
}
// Replay the WAL. Should get 9 records.
{
sr, err := NewSegmentsReader(dir)
require.NoError(t, err)
reader := NewReader(sr)
i := 0
for ; i < 9 && reader.Next(); i++ {
require.Equal(t, recordSize, len(reader.Record()))
}
require.Equal(t, 9, i, "wrong number of records")
require.False(t, reader.Next(), "unexpected record")
require.Equal(t, nil, reader.Err())
sr.Close()
}
}
// TestClose ensures that calling Close more than once doesn't panic and doesn't block.
func TestClose(t *testing.T) {
dir, err := ioutil.TempDir("", "wal_repair")
require.NoError(t, err)
defer func() {
require.NoError(t, os.RemoveAll(dir))
}()
w, err := NewSize(nil, nil, dir, pageSize, false)
require.NoError(t, err)
require.NoError(t, w.Close())
require.Error(t, w.Close())
}
func TestSegmentMetric(t *testing.T) {
var (
segmentSize = pageSize
recordSize = (pageSize / 2) - recordHeaderSize
)
dir, err := ioutil.TempDir("", "segment_metric")
require.NoError(t, err)
defer func() {
require.NoError(t, os.RemoveAll(dir))
}()
w, err := NewSize(nil, nil, dir, segmentSize, false)
require.NoError(t, err)
initialSegment := client_testutil.ToFloat64(w.metrics.currentSegment)
// Write 3 records, each of which is half the segment size, meaning we should rotate to the next segment.
for i := 0; i < 3; i++ {
buf := make([]byte, recordSize)
_, err := rand.Read(buf)
require.NoError(t, err)
err = w.Log(buf)
require.NoError(t, err)
}
require.Equal(t, initialSegment+1, client_testutil.ToFloat64(w.metrics.currentSegment), "segment metric did not increment after segment rotation")
require.NoError(t, w.Close())
}
func TestCompression(t *testing.T) {
Spelling (#6517) * spelling: alertmanager Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: attributes Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: autocomplete Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: bootstrap Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: caught Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: chunkenc Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: compaction Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: corrupted Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: deletable Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: expected Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: fine-grained Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: initialized Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: iteration Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: javascript Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: multiple Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: number Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: overlapping Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: possible Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: postings Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: procedure Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: programmatic Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: queuing Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: querier Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: repairing Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: received Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: reproducible Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: retention Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: sample Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: segements Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: semantic Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: software [LICENSE] Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: staging Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: timestamp Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: unfortunately Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: uvarint Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: subsequently Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: ressamples Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>
2020-01-02 06:54:09 -08:00
bootstrap := func(compressed bool) string {
const (
segmentSize = pageSize
recordSize = (pageSize / 2) - recordHeaderSize
records = 100
)
dirPath, err := ioutil.TempDir("", fmt.Sprintf("TestCompression_%t", compressed))
require.NoError(t, err)
w, err := NewSize(nil, nil, dirPath, segmentSize, compressed)
require.NoError(t, err)
buf := make([]byte, recordSize)
for i := 0; i < records; i++ {
require.NoError(t, w.Log(buf))
}
require.NoError(t, w.Close())
return dirPath
}
Spelling (#6517) * spelling: alertmanager Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: attributes Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: autocomplete Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: bootstrap Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: caught Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: chunkenc Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: compaction Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: corrupted Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: deletable Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: expected Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: fine-grained Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: initialized Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: iteration Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: javascript Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: multiple Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: number Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: overlapping Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: possible Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: postings Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: procedure Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: programmatic Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: queuing Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: querier Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: repairing Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: received Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: reproducible Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: retention Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: sample Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: segements Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: semantic Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: software [LICENSE] Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: staging Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: timestamp Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: unfortunately Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: uvarint Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: subsequently Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: ressamples Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>
2020-01-02 06:54:09 -08:00
dirCompressed := bootstrap(true)
defer func() {
require.NoError(t, os.RemoveAll(dirCompressed))
}()
Spelling (#6517) * spelling: alertmanager Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: attributes Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: autocomplete Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: bootstrap Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: caught Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: chunkenc Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: compaction Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: corrupted Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: deletable Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: expected Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: fine-grained Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: initialized Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: iteration Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: javascript Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: multiple Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: number Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: overlapping Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: possible Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: postings Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: procedure Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: programmatic Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: queuing Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: querier Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: repairing Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: received Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: reproducible Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: retention Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: sample Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: segements Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: semantic Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: software [LICENSE] Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: staging Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: timestamp Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: unfortunately Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: uvarint Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: subsequently Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> * spelling: ressamples Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>
2020-01-02 06:54:09 -08:00
dirUnCompressed := bootstrap(false)
defer func() {
require.NoError(t, os.RemoveAll(dirUnCompressed))
}()
uncompressedSize, err := fileutil.DirSize(dirUnCompressed)
require.NoError(t, err)
compressedSize, err := fileutil.DirSize(dirCompressed)
require.NoError(t, err)
require.Greater(t, float64(uncompressedSize)*0.75, float64(compressedSize), "Compressing zeroes should save at least 25%% space - uncompressedSize: %d, compressedSize: %d", uncompressedSize, compressedSize)
}
func TestLogPartialWrite(t *testing.T) {
const segmentSize = pageSize * 2
record := []byte{1, 2, 3, 4, 5}
tests := map[string]struct {
numRecords int
faultyRecord int
}{
"partial write when logging first record in a page": {
numRecords: 10,
faultyRecord: 1,
},
"partial write when logging record in the middle of a page": {
numRecords: 10,
faultyRecord: 3,
},
"partial write when logging last record of a page": {
numRecords: (pageSize / (recordHeaderSize + len(record))) + 10,
faultyRecord: pageSize / (recordHeaderSize + len(record)),
},
// TODO the current implementation suffers this:
//"partial write when logging a record overlapping two pages": {
// numRecords: (pageSize / (recordHeaderSize + len(record))) + 10,
// faultyRecord: pageSize/(recordHeaderSize+len(record)) + 1,
//},
}
for testName, testData := range tests {
t.Run(testName, func(t *testing.T) {
dirPath, err := ioutil.TempDir("", "logpartialwrite")
require.NoError(t, err)
defer func() { require.NoError(t, os.RemoveAll(dirPath)) }()
w, err := NewSize(nil, nil, dirPath, segmentSize, false)
require.NoError(t, err)
// Replace the underlying segment file with a mocked one that injects a failure.
w.segment.SegmentFile = &faultySegmentFile{
SegmentFile: w.segment.SegmentFile,
writeFailureAfter: ((recordHeaderSize + len(record)) * (testData.faultyRecord - 1)) + 2,
writeFailureErr: io.ErrShortWrite,
}
for i := 1; i <= testData.numRecords; i++ {
if err := w.Log(record); i == testData.faultyRecord {
require.Error(t, io.ErrShortWrite, err)
} else {
require.NoError(t, err)
}
}
require.NoError(t, w.Close())
// Read it back. We expect no corruption.
s, err := OpenReadSegment(SegmentName(dirPath, 0))
require.NoError(t, err)
defer func() { require.NoError(t, s.Close()) }()
r := NewReader(NewSegmentBufReader(s))
for i := 0; i < testData.numRecords; i++ {
require.True(t, r.Next())
require.NoError(t, r.Err())
require.Equal(t, record, r.Record())
}
require.False(t, r.Next())
require.NoError(t, r.Err())
})
}
}
type faultySegmentFile struct {
SegmentFile
written int
writeFailureAfter int
writeFailureErr error
}
func (f *faultySegmentFile) Write(p []byte) (int, error) {
if f.writeFailureAfter >= 0 && f.writeFailureAfter < f.written+len(p) {
partialLen := f.writeFailureAfter - f.written
if partialLen <= 0 || partialLen >= len(p) {
partialLen = 1
}
// Inject failure.
n, _ := f.SegmentFile.Write(p[:partialLen])
f.written += n
f.writeFailureAfter = -1
return n, f.writeFailureErr
}
// Proxy the write to the underlying file.
n, err := f.SegmentFile.Write(p)
f.written += n
return n, err
}
func BenchmarkWAL_LogBatched(b *testing.B) {
for _, compress := range []bool{true, false} {
b.Run(fmt.Sprintf("compress=%t", compress), func(b *testing.B) {
dir, err := ioutil.TempDir("", "bench_logbatch")
require.NoError(b, err)
defer func() {
require.NoError(b, os.RemoveAll(dir))
}()
w, err := New(nil, nil, dir, compress)
require.NoError(b, err)
defer w.Close()
var buf [2048]byte
var recs [][]byte
b.SetBytes(2048)
for i := 0; i < b.N; i++ {
recs = append(recs, buf[:])
if len(recs) < 1000 {
continue
}
err := w.Log(recs...)
require.NoError(b, err)
recs = recs[:0]
}
// Stop timer to not count fsync time on close.
// If it's counted batched vs. single benchmarks are very similar but
// do not show burst throughput well.
b.StopTimer()
})
}
}
func BenchmarkWAL_Log(b *testing.B) {
for _, compress := range []bool{true, false} {
b.Run(fmt.Sprintf("compress=%t", compress), func(b *testing.B) {
dir, err := ioutil.TempDir("", "bench_logsingle")
require.NoError(b, err)
defer func() {
require.NoError(b, os.RemoveAll(dir))
}()
w, err := New(nil, nil, dir, compress)
require.NoError(b, err)
defer w.Close()
var buf [2048]byte
b.SetBytes(2048)
for i := 0; i < b.N; i++ {
err := w.Log(buf[:])
require.NoError(b, err)
}
// Stop timer to not count fsync time on close.
// If it's counted batched vs. single benchmarks are very similar but
// do not show burst throughput well.
b.StopTimer()
})
}
}