Skip to content

task runner actor local: resume inputs after all checkpoint… (backport #24796) #24829

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 2 commits into from
Sep 13, 2025
Merged

task runner actor local: resume inputs after all checkpoint… (backport #24796) #24829

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
3f2a3d4
task runner actor local/checkpoint: resume inputs after all checkoint…
yumkam Sep 12, 2025
73f80c5
review fixes: refactor/improve
yumkam Sep 12, 2025
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
24 changes: 21 additions & 3 deletions ydb/library/yql/dq/actors/task_runner/task_runner_actor_local.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -284,6 +284,7 @@ class TLocalTaskRunnerActor
LOG_E("Failed to save state: " << e.what());
mkqlProgramState = nullptr;
}
StartResumeByCheckpoint();
HasActiveCheckpoint = false;
if (!WatermarkRequests.empty()) {
auto nextWatermarkRequest = WatermarkRequests.front();
Expand Down Expand Up @@ -480,7 +481,7 @@ class TLocalTaskRunnerActor
checkpoint = hasCheckpoint ? std::move(poppedCheckpoint) : TMaybe<NDqProto::TCheckpoint>();

if (hasCheckpoint) {
ResumeByCheckpoint();
AdvanceResumeByCheckpoint();
break;
}

Expand All @@ -504,12 +505,28 @@ class TLocalTaskRunnerActor
ev->Cookie);
}

void ResumeByCheckpoint() {
void MaybeResumeByCheckpoint() {
if (UnsentCheckpoints > 0) {
LOG_T("Pending " << UnsentCheckpoints << " checkpoints to be sent");
return;
}
for (const auto& inputId : InputsWithCheckpoints) {
TaskRunner->GetInputChannel(inputId)->ResumeByCheckpoint();
}
}

void StartResumeByCheckpoint() {
Y_ENSURE(UnsentCheckpoints == 0);
UnsentCheckpoints = Sinks.size() + Outputs.size();
MaybeResumeByCheckpoint();
}

void AdvanceResumeByCheckpoint() {
Y_ENSURE(UnsentCheckpoints > 0);
--UnsentCheckpoints;
MaybeResumeByCheckpoint();
}

void ResumeByWatermark(TInstant watermark) {
for (const auto& inputId : InputsWithWatermarks) {
TaskRunner->GetInputChannel(inputId)->ResumeByWatermark(watermark);
Expand Down Expand Up @@ -539,7 +556,7 @@ class TLocalTaskRunnerActor
if (hasCheckpoint) {
checkpointSize = checkpoint.ByteSize();
maybeCheckpoint.ConstructInPlace(std::move(checkpoint));
ResumeByCheckpoint();
AdvanceResumeByCheckpoint();
}
const bool finished = sink->IsFinished();
const bool changed = finished || size > 0 || hasCheckpoint;
Expand Down Expand Up @@ -702,6 +719,7 @@ class TLocalTaskRunnerActor
TMaybe<TInstant> LastWatermark;
std::deque<TInstant> WatermarkRequests;
bool HasActiveCheckpoint = false;
ui32 UnsentCheckpoints = 0;
};

struct TLocalTaskRunnerActorFactory: public ITaskRunnerActorFactory {
Expand Down
Loading