VictoriaMetrics/lib/storage/merge.go
Aliaksandr Valialkin af2ceaaa0b lib/storage: mention source parts on merge error
This should improve determining broken source part.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/76
2019-06-24 14:08:43 +03:00

206 lines
5.9 KiB
Go

package storage
import (
"fmt"
"sync"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/decimal"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
)
// mergeBlockStreams merges bsrs into bsw and updates ph.
//
// mergeBlockStreams returns immediately if stopCh is closed.
//
// rowsMerged is atomically updated with the number of merged rows during the merge.
func mergeBlockStreams(ph *partHeader, bsw *blockStreamWriter, bsrs []*blockStreamReader, stopCh <-chan struct{}, rowsMerged *uint64,
deletedMetricIDs map[uint64]struct{}, rowsDeleted *uint64) error {
ph.Reset()
bsm := bsmPool.Get().(*blockStreamMerger)
bsm.Init(bsrs)
err := mergeBlockStreamsInternal(ph, bsw, bsm, stopCh, rowsMerged, deletedMetricIDs, rowsDeleted)
bsm.reset()
bsmPool.Put(bsm)
bsw.MustClose()
if err == nil {
return nil
}
if err == errForciblyStopped {
return err
}
return fmt.Errorf("cannot merge %d streams: %s: %s", len(bsrs), bsrs, err)
}
var bsmPool = &sync.Pool{
New: func() interface{} {
return &blockStreamMerger{}
},
}
var errForciblyStopped = fmt.Errorf("forcibly stopped")
func mergeBlockStreamsInternal(ph *partHeader, bsw *blockStreamWriter, bsm *blockStreamMerger, stopCh <-chan struct{}, rowsMerged *uint64,
deletedMetricIDs map[uint64]struct{}, rowsDeleted *uint64) error {
// Search for the first block to merge
var pendingBlock *Block
for bsm.NextBlock() {
select {
case <-stopCh:
return errForciblyStopped
default:
}
if _, deleted := deletedMetricIDs[bsm.Block.bh.TSID.MetricID]; deleted {
// Skip blocks for deleted metrics.
*rowsDeleted += uint64(bsm.Block.bh.RowsCount)
continue
}
pendingBlock = getBlock()
pendingBlock.CopyFrom(bsm.Block)
break
}
if pendingBlock != nil {
defer putBlock(pendingBlock)
}
// Merge blocks.
tmpBlock := getBlock()
defer putBlock(tmpBlock)
for bsm.NextBlock() {
select {
case <-stopCh:
return errForciblyStopped
default:
}
if _, deleted := deletedMetricIDs[bsm.Block.bh.TSID.MetricID]; deleted {
// Skip blocks for deleted metrics.
*rowsDeleted += uint64(bsm.Block.bh.RowsCount)
continue
}
// Verify whether pendingBlock may be merged with bsm.Block (the current block).
if pendingBlock.bh.TSID.MetricID != bsm.Block.bh.TSID.MetricID {
// Fast path - blocks belong to distinct time series.
// Write the pendingBlock and then deal with bsm.Block.
if bsm.Block.bh.TSID.Less(&pendingBlock.bh.TSID) {
logger.Panicf("BUG: the next TSID=%+v is smaller than the current TSID=%+v", &bsm.Block.bh.TSID, &pendingBlock.bh.TSID)
}
bsw.WriteExternalBlock(pendingBlock, ph, rowsMerged)
pendingBlock.CopyFrom(bsm.Block)
continue
}
if pendingBlock.tooBig() && pendingBlock.bh.MaxTimestamp <= bsm.Block.bh.MinTimestamp {
// Fast path - pendingBlock is too big and it doesn't overlap with bsm.Block.
// Write the pendingBlock and then deal with bsm.Block.
bsw.WriteExternalBlock(pendingBlock, ph, rowsMerged)
pendingBlock.CopyFrom(bsm.Block)
continue
}
// Slow path - pendingBlock and bsm.Block belong to the same time series,
// so they must be merged.
if err := unmarshalAndCalibrateScale(pendingBlock, bsm.Block); err != nil {
return fmt.Errorf("cannot unmarshal and calibrate scale for blocks to be merged: %s", err)
}
tmpBlock.Reset()
tmpBlock.bh.TSID = bsm.Block.bh.TSID
tmpBlock.bh.Scale = bsm.Block.bh.Scale
tmpBlock.bh.PrecisionBits = minUint8(pendingBlock.bh.PrecisionBits, bsm.Block.bh.PrecisionBits)
mergeBlocks(tmpBlock, pendingBlock, bsm.Block)
if len(tmpBlock.timestamps) <= maxRowsPerBlock {
// More entries may be added to tmpBlock. Swap it with pendingBlock,
// so more entries may be added to pendingBlock on the next iteration.
tmpBlock.fixupTimestamps()
pendingBlock, tmpBlock = tmpBlock, pendingBlock
continue
}
// Write the first len(maxRowsPerBlock) of tmpBlock.timestamps to bsw,
// leave the rest in pendingBlock.
tmpBlock.nextIdx = maxRowsPerBlock
pendingBlock.CopyFrom(tmpBlock)
pendingBlock.fixupTimestamps()
tmpBlock.nextIdx = 0
tmpBlock.timestamps = tmpBlock.timestamps[:maxRowsPerBlock]
tmpBlock.values = tmpBlock.values[:maxRowsPerBlock]
tmpBlock.fixupTimestamps()
bsw.WriteExternalBlock(tmpBlock, ph, rowsMerged)
}
if err := bsm.Error(); err != nil {
return fmt.Errorf("cannot read block to be merged: %s", err)
}
if pendingBlock != nil {
bsw.WriteExternalBlock(pendingBlock, ph, rowsMerged)
}
return nil
}
// mergeBlocks merges ib1 and ib2 to ob.
func mergeBlocks(ob, ib1, ib2 *Block) {
ib1.assertMergeable(ib2)
ib1.assertUnmarshaled()
ib2.assertUnmarshaled()
if ib1.bh.MaxTimestamp < ib2.bh.MinTimestamp {
// Fast path - ib1 values have smaller timestamps than ib2 values.
appendRows(ob, ib1)
appendRows(ob, ib2)
return
}
if ib2.bh.MaxTimestamp < ib1.bh.MinTimestamp {
// Fast path - ib2 values have smaller timestamps than ib1 values.
appendRows(ob, ib2)
appendRows(ob, ib1)
return
}
if ib1.nextIdx >= len(ib1.timestamps) {
appendRows(ob, ib2)
return
}
if ib2.nextIdx >= len(ib2.timestamps) {
appendRows(ob, ib1)
return
}
for {
i := ib1.nextIdx
ts2 := ib2.timestamps[ib2.nextIdx]
for i < len(ib1.timestamps) && ib1.timestamps[i] <= ts2 {
i++
}
ob.timestamps = append(ob.timestamps, ib1.timestamps[ib1.nextIdx:i]...)
ob.values = append(ob.values, ib1.values[ib1.nextIdx:i]...)
ib1.nextIdx = i
if ib1.nextIdx >= len(ib1.timestamps) {
appendRows(ob, ib2)
return
}
ib1, ib2 = ib2, ib1
}
}
func appendRows(ob, ib *Block) {
ob.timestamps = append(ob.timestamps, ib.timestamps[ib.nextIdx:]...)
ob.values = append(ob.values, ib.values[ib.nextIdx:]...)
}
func unmarshalAndCalibrateScale(b1, b2 *Block) error {
if err := b1.UnmarshalData(); err != nil {
return err
}
if err := b2.UnmarshalData(); err != nil {
return err
}
scale := decimal.CalibrateScale(b1.values, b1.bh.Scale, b2.values, b2.bh.Scale)
b1.bh.Scale = scale
b2.bh.Scale = scale
return nil
}
func minUint8(a, b uint8) uint8 {
if a < b {
return a
}
return b
}