Skip to content

executor: fix hang in hash agg when exceeding memory limit leads to panic #57641

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
Nov 26, 2024
Merged

executor: fix hang in hash agg when exceeding memory limit leads to panic #57641

Changes from 4 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
76ed9c0
fix
xzhangxian1008 Nov 22, 2024
476ed5d
move
xzhangxian1008 Nov 25, 2024
17cebd9
tweaking
xzhangxian1008 Nov 25, 2024
a2b8dd7
fix
xzhangxian1008 Nov 25, 2024
9256bbd
tweaking
xzhangxian1008 Nov 25, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
30 changes: 16 additions & 14 deletions pkg/executor/aggregate/agg_hash_partial_worker.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -71,25 +71,16 @@ type HashAggPartialWorker struct {
inflightChunkSync *sync.WaitGroup
}

func (w *HashAggPartialWorker) getChildInput() bool {
func (w *HashAggPartialWorker) getChildInput() (*chunk.Chunk, bool) {
select {
case <-w.finishCh:
return false
return nil, false
case chk, ok := <-w.inputCh:
if !ok {
return false
}

sizeBefore := w.chk.MemoryUsage()
w.chk.SwapColumns(chk)
w.memTracker.Consume(w.chk.MemoryUsage() - sizeBefore)

w.giveBackCh <- &HashAggInput{
chk: chk,
giveBackCh: w.inputCh,
return nil, false
}
return chk, true
}
return true
}

func (w *HashAggPartialWorker) fetchChunkAndProcess(ctx sessionctx.Context, hasError *bool, needShuffle *bool) bool {
Expand All @@ -99,7 +90,7 @@ func (w *HashAggPartialWorker) fetchChunkAndProcess(ctx sessionctx.Context, hasE
}

waitStart := time.Now()
ok := w.getChildInput()
chk, ok := w.getChildInput()
updateWaitTime(w.stats, waitStart)

if !ok {
Expand All @@ -108,6 +99,17 @@ func (w *HashAggPartialWorker) fetchChunkAndProcess(ctx sessionctx.Context, hasE

defer w.inflightChunkSync.Done()

w.intestDuringPartialWorkerRun()

sizeBefore := w.chk.MemoryUsage()
w.chk.SwapColumns(chk)
w.memTracker.Consume(w.chk.MemoryUsage() - sizeBefore)

w.giveBackCh <- &HashAggInput{
chk: chk,
giveBackCh: w.inputCh,
}

execStart := time.Now()
if err := w.updatePartialResult(ctx, w.chk, len(w.partialResultsMap)); err != nil {
*hasError = true
Expand Down