seaweedfs/weed/filer/stream.go

371 lines
10 KiB
Go
Raw Normal View History

2020-09-01 07:21:19 +00:00
package filer
import (
"bytes"
2020-10-13 20:53:34 +00:00
"fmt"
"golang.org/x/exp/slices"
"io"
"math"
"strings"
2021-08-26 00:28:50 +00:00
"sync"
2021-05-24 07:14:50 +00:00
"time"
"github.com/seaweedfs/seaweedfs/weed/glog"
"github.com/seaweedfs/seaweedfs/weed/pb/filer_pb"
"github.com/seaweedfs/seaweedfs/weed/stats"
"github.com/seaweedfs/seaweedfs/weed/util"
"github.com/seaweedfs/seaweedfs/weed/wdclient"
)
var getLookupFileIdBackoffSchedule = []time.Duration{
150 * time.Millisecond,
600 * time.Millisecond,
1800 * time.Millisecond,
}
2021-08-16 02:46:45 +00:00
func HasData(entry *filer_pb.Entry) bool {
if len(entry.Content) > 0 {
return true
}
return len(entry.GetChunks()) > 0
2021-08-16 02:46:45 +00:00
}
func IsSameData(a, b *filer_pb.Entry) bool {
if len(a.Content) > 0 || len(b.Content) > 0 {
return bytes.Equal(a.Content, b.Content)
}
return isSameChunks(a.Chunks, b.Chunks)
}
func isSameChunks(a, b []*filer_pb.FileChunk) bool {
if len(a) != len(b) {
return false
}
slices.SortFunc(a, func(i, j *filer_pb.FileChunk) bool {
return strings.Compare(i.ETag, j.ETag) < 0
})
slices.SortFunc(b, func(i, j *filer_pb.FileChunk) bool {
return strings.Compare(i.ETag, j.ETag) < 0
})
2021-08-16 02:46:45 +00:00
for i := 0; i < len(a); i++ {
if a[i].ETag != b[i].ETag {
2021-08-16 02:46:45 +00:00
return false
}
}
return true
}
func NewFileReader(filerClient filer_pb.FilerClient, entry *filer_pb.Entry) io.Reader {
if len(entry.Content) > 0 {
return bytes.NewReader(entry.Content)
}
return NewChunkStreamReader(filerClient, entry.GetChunks())
2021-08-16 02:46:45 +00:00
}
func StreamContent(masterClient wdclient.HasLookupFileIdFunction, writer io.Writer, chunks []*filer_pb.FileChunk, offset int64, size int64) error {
return StreamContentWithThrottler(masterClient, writer, chunks, offset, size, 0)
}
func StreamContentWithThrottler(masterClient wdclient.HasLookupFileIdFunction, writer io.Writer, chunks []*filer_pb.FileChunk, offset int64, size int64, downloadMaxBytesPs int64) error {
2022-08-15 17:03:52 +00:00
glog.V(4).Infof("start to stream content for chunks: %d", len(chunks))
chunkViews := ViewFromChunks(masterClient.GetLookupFileIdFunction(), chunks, offset, size)
fileId2Url := make(map[string][]string)
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
for x := chunkViews.Front(); x != nil; x = x.Next {
chunkView := x.Value
var urlStrings []string
var err error
for _, backoff := range getLookupFileIdBackoffSchedule {
urlStrings, err = masterClient.GetLookupFileIdFunction()(chunkView.FileId)
2022-07-26 06:56:45 +00:00
if err == nil && len(urlStrings) > 0 {
break
}
2022-08-15 17:03:52 +00:00
glog.V(4).Infof("waiting for chunk: %s", chunkView.FileId)
2022-07-26 06:56:45 +00:00
time.Sleep(backoff)
}
if err != nil {
glog.V(1).Infof("operation LookupFileId %s failed, err: %v", chunkView.FileId, err)
return err
2021-03-11 18:34:36 +00:00
} else if len(urlStrings) == 0 {
2022-07-15 10:29:15 +00:00
errUrlNotFound := fmt.Errorf("operation LookupFileId %s failed, err: urls not found", chunkView.FileId)
glog.Error(errUrlNotFound)
return errUrlNotFound
}
fileId2Url[chunkView.FileId] = urlStrings
}
downloadThrottler := util.NewWriteThrottler(downloadMaxBytesPs)
2022-01-13 14:20:33 +00:00
remaining := size
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
for x := chunkViews.Front(); x != nil; x = x.Next {
chunkView := x.Value
if offset < chunkView.ViewOffset {
gap := chunkView.ViewOffset - offset
2022-01-13 14:20:33 +00:00
remaining -= gap
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
glog.V(4).Infof("zero [%d,%d)", offset, chunkView.ViewOffset)
2022-01-13 14:20:33 +00:00
err := writeZero(writer, gap)
if err != nil {
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
return fmt.Errorf("write zero [%d,%d)", offset, chunkView.ViewOffset)
2022-01-13 14:20:33 +00:00
}
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
offset = chunkView.ViewOffset
2022-01-13 14:20:33 +00:00
}
urlStrings := fileId2Url[chunkView.FileId]
2021-05-24 07:14:50 +00:00
start := time.Now()
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
err := retriedStreamFetchChunkData(writer, urlStrings, chunkView.CipherKey, chunkView.IsGzipped, chunkView.IsFullChunk(), chunkView.OffsetInChunk, int(chunkView.ViewSize))
offset += int64(chunkView.ViewSize)
remaining -= int64(chunkView.ViewSize)
2021-05-24 07:14:50 +00:00
stats.FilerRequestHistogram.WithLabelValues("chunkDownload").Observe(time.Since(start).Seconds())
2020-10-10 23:02:10 +00:00
if err != nil {
2021-05-24 07:14:50 +00:00
stats.FilerRequestCounter.WithLabelValues("chunkDownloadError").Inc()
2020-10-13 20:53:34 +00:00
return fmt.Errorf("read chunk: %v", err)
}
2021-05-24 07:14:50 +00:00
stats.FilerRequestCounter.WithLabelValues("chunkDownload").Inc()
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
downloadThrottler.MaybeSlowdown(int64(chunkView.ViewSize))
}
2022-03-14 10:19:16 +00:00
if remaining > 0 {
glog.V(4).Infof("zero [%d,%d)", offset, offset+remaining)
err := writeZero(writer, remaining)
if err != nil {
return fmt.Errorf("write zero [%d,%d)", offset, offset+remaining)
}
2022-01-13 14:20:33 +00:00
}
return nil
}
2020-04-28 07:05:47 +00:00
// ---------------- ReadAllReader ----------------------------------
2022-01-13 14:20:33 +00:00
func writeZero(w io.Writer, size int64) (err error) {
zeroPadding := make([]byte, 1024)
var written int
for size > 0 {
if size > 1024 {
written, err = w.Write(zeroPadding)
} else {
written, err = w.Write(zeroPadding[:size])
}
size -= int64(written)
if err != nil {
return
}
}
return
}
2022-02-26 10:16:47 +00:00
func ReadAll(buffer []byte, masterClient *wdclient.MasterClient, chunks []*filer_pb.FileChunk) error {
2020-04-28 07:05:47 +00:00
lookupFileIdFn := func(fileId string) (targetUrls []string, err error) {
2020-04-28 07:05:47 +00:00
return masterClient.LookupFileId(fileId)
}
2022-02-26 10:16:47 +00:00
chunkViews := ViewFromChunks(lookupFileIdFn, chunks, 0, int64(len(buffer)))
idx := 0
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
for x := chunkViews.Front(); x != nil; x = x.Next {
chunkView := x.Value
urlStrings, err := lookupFileIdFn(chunkView.FileId)
2020-04-28 07:05:47 +00:00
if err != nil {
glog.V(1).Infof("operation LookupFileId %s failed, err: %v", chunkView.FileId, err)
2022-02-26 10:16:47 +00:00
return err
2020-04-28 07:05:47 +00:00
}
2020-10-09 06:19:42 +00:00
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
n, err := retriedFetchChunkData(buffer[idx:idx+int(chunkView.ViewSize)], urlStrings, chunkView.CipherKey, chunkView.IsGzipped, chunkView.IsFullChunk(), chunkView.OffsetInChunk)
2020-10-09 06:19:42 +00:00
if err != nil {
2022-02-26 10:16:47 +00:00
return err
2020-04-28 07:05:47 +00:00
}
2022-02-26 10:16:47 +00:00
idx += n
2020-04-28 07:05:47 +00:00
}
2022-02-26 10:16:47 +00:00
return nil
2020-04-28 07:05:47 +00:00
}
// ---------------- ChunkStreamReader ----------------------------------
type ChunkStreamReader struct {
head *Interval[*ChunkView]
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
chunkView *Interval[*ChunkView]
2021-09-01 09:45:42 +00:00
totalSize int64
logicOffset int64
buffer []byte
bufferOffset int64
bufferLock sync.Mutex
chunk string
lookupFileId wdclient.LookupFileIdFunctionType
}
var _ = io.ReadSeeker(&ChunkStreamReader{})
var _ = io.ReaderAt(&ChunkStreamReader{})
2021-08-07 22:35:27 +00:00
func doNewChunkStreamReader(lookupFileIdFn wdclient.LookupFileIdFunctionType, chunks []*filer_pb.FileChunk) *ChunkStreamReader {
chunkViews := ViewFromChunks(lookupFileIdFn, chunks, 0, math.MaxInt64)
2021-08-07 21:46:23 +00:00
var totalSize int64
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
for x := chunkViews.Front(); x != nil; x = x.Next {
chunk := x.Value
totalSize += int64(chunk.ViewSize)
2021-08-07 21:46:23 +00:00
}
return &ChunkStreamReader{
head: chunkViews.Front(),
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
chunkView: chunkViews.Front(),
lookupFileId: lookupFileIdFn,
2021-08-07 21:46:23 +00:00
totalSize: totalSize,
}
}
2021-08-07 22:35:27 +00:00
func NewChunkStreamReaderFromFiler(masterClient *wdclient.MasterClient, chunks []*filer_pb.FileChunk) *ChunkStreamReader {
2021-08-07 22:35:27 +00:00
lookupFileIdFn := func(fileId string) (targetUrl []string, err error) {
return masterClient.LookupFileId(fileId)
}
2021-08-07 22:35:27 +00:00
return doNewChunkStreamReader(lookupFileIdFn, chunks)
}
2021-08-07 22:35:27 +00:00
func NewChunkStreamReader(filerClient filer_pb.FilerClient, chunks []*filer_pb.FileChunk) *ChunkStreamReader {
2021-08-07 21:46:23 +00:00
2021-08-07 22:35:27 +00:00
lookupFileIdFn := LookupFn(filerClient)
return doNewChunkStreamReader(lookupFileIdFn, chunks)
}
func (c *ChunkStreamReader) ReadAt(p []byte, off int64) (n int, err error) {
2021-08-26 00:28:50 +00:00
c.bufferLock.Lock()
defer c.bufferLock.Unlock()
if err = c.prepareBufferFor(off); err != nil {
return
}
c.logicOffset = off
2021-08-26 00:28:50 +00:00
return c.doRead(p)
}
func (c *ChunkStreamReader) Read(p []byte) (n int, err error) {
2021-08-26 00:28:50 +00:00
c.bufferLock.Lock()
defer c.bufferLock.Unlock()
return c.doRead(p)
}
func (c *ChunkStreamReader) doRead(p []byte) (n int, err error) {
// fmt.Printf("do read [%d,%d) at %s[%d,%d)\n", c.logicOffset, c.logicOffset+int64(len(p)), c.chunk, c.bufferOffset, c.bufferOffset+int64(len(c.buffer)))
2020-04-29 09:42:58 +00:00
for n < len(p) {
2021-08-26 00:28:50 +00:00
// println("read", c.logicOffset)
if err = c.prepareBufferFor(c.logicOffset); err != nil {
return
}
2021-08-26 00:28:50 +00:00
t := copy(p[n:], c.buffer[c.logicOffset-c.bufferOffset:])
2020-04-29 09:42:58 +00:00
n += t
2021-08-08 08:21:42 +00:00
c.logicOffset += int64(t)
}
return
}
func (c *ChunkStreamReader) isBufferEmpty() bool {
2021-09-01 09:45:42 +00:00
return len(c.buffer) <= int(c.logicOffset-c.bufferOffset)
}
func (c *ChunkStreamReader) Seek(offset int64, whence int) (int64, error) {
2021-08-26 00:28:50 +00:00
c.bufferLock.Lock()
defer c.bufferLock.Unlock()
var err error
switch whence {
case io.SeekStart:
case io.SeekCurrent:
2021-08-08 08:21:42 +00:00
offset += c.logicOffset
case io.SeekEnd:
2021-08-07 21:46:23 +00:00
offset = c.totalSize + offset
}
2021-08-07 21:46:23 +00:00
if offset > c.totalSize {
err = io.ErrUnexpectedEOF
2021-08-08 08:21:42 +00:00
} else {
c.logicOffset = offset
}
2021-08-08 08:21:42 +00:00
return offset, err
}
2021-08-26 00:28:50 +00:00
func insideChunk(offset int64, chunk *ChunkView) bool {
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
return chunk.ViewOffset <= offset && offset < chunk.ViewOffset+int64(chunk.ViewSize)
2021-08-26 00:28:50 +00:00
}
2021-08-08 08:21:42 +00:00
func (c *ChunkStreamReader) prepareBufferFor(offset int64) (err error) {
2021-08-07 22:35:27 +00:00
// stay in the same chunk
2021-08-26 00:28:50 +00:00
if c.bufferOffset <= offset && offset < c.bufferOffset+int64(len(c.buffer)) {
return nil
2021-08-07 22:35:27 +00:00
}
2023-02-17 09:31:44 +00:00
// glog.V(2).Infof("c.chunkView: %v buffer:[%d,%d) offset:%d totalSize:%d", c.chunkView, c.bufferOffset, c.bufferOffset+int64(len(c.buffer)), offset, c.totalSize)
// find a possible chunk view
p := c.chunkView
for p != nil {
chunk := p.Value
2023-02-17 09:31:44 +00:00
// glog.V(2).Infof("prepareBufferFor check chunk:[%d,%d)", chunk.ViewOffset, chunk.ViewOffset+int64(chunk.ViewSize))
if insideChunk(offset, chunk) {
if c.isBufferEmpty() || c.bufferOffset != chunk.ViewOffset {
c.chunkView = p
return c.fetchChunkToBuffer(chunk)
}
}
if offset < c.bufferOffset {
p = p.Prev
} else {
p = p.Next
}
}
return io.EOF
}
func (c *ChunkStreamReader) fetchChunkToBuffer(chunkView *ChunkView) error {
urlStrings, err := c.lookupFileId(chunkView.FileId)
if err != nil {
glog.V(1).Infof("operation LookupFileId %s failed, err: %v", chunkView.FileId, err)
return err
}
var buffer bytes.Buffer
2020-10-13 07:29:46 +00:00
var shouldRetry bool
for _, urlString := range urlStrings {
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
shouldRetry, err = util.ReadUrlAsStream(urlString+"?readDeleted=true", chunkView.CipherKey, chunkView.IsGzipped, chunkView.IsFullChunk(), chunkView.OffsetInChunk, int(chunkView.ViewSize), func(data []byte) {
buffer.Write(data)
})
2020-10-13 07:29:46 +00:00
if !shouldRetry {
break
}
if err != nil {
glog.V(1).Infof("read %s failed, err: %v", chunkView.FileId, err)
buffer.Reset()
} else {
break
}
}
if err != nil {
return err
}
c.buffer = buffer.Bytes()
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
c.bufferOffset = chunkView.ViewOffset
2021-08-26 00:28:50 +00:00
c.chunk = chunkView.FileId
more solid weed mount (#4089) * compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-03 07:20:45 +00:00
// glog.V(0).Infof("fetched %s [%d,%d)", chunkView.FileId, chunkView.ViewOffset, chunkView.ViewOffset+int64(chunkView.ViewSize))
return nil
}
2020-04-28 07:05:47 +00:00
func (c *ChunkStreamReader) Close() {
// TODO try to release and reuse buffer
}
2020-03-23 07:01:34 +00:00
func VolumeId(fileId string) string {
lastCommaIndex := strings.LastIndex(fileId, ",")
if lastCommaIndex > 0 {
return fileId[:lastCommaIndex]
}
2020-03-23 07:01:34 +00:00
return fileId
}