Skip to content
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

Jump to bottom

[Bug][Zeta] Fix task notifyTaskStatusToMaster failed when job not running or failed before run #4847

Merged
merged 3 commits into from
May 30, 2023
Merged

[Bug][Zeta] Fix task notifyTaskStatusToMaster failed when job not running or failed before run #4847

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
b83574e
[Bug][Zeta] Fix task repeat notify failed when job not running
Hisoka-X May 26, 2023
9dd89ec
[Bug][Zeta] Fix notifyTaskStatusToMaster not release lock and NPE
Hisoka-X May 29, 2023
3765700
improve
Hisoka-X May 29, 2023
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
3 changes: 2 additions & 1 deletion ...el-engine-server/src/main/java/org/apache/seatunnel/engine/server/CoordinatorService.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,7 @@
import org.apache.seatunnel.engine.common.Constant;
import org.apache.seatunnel.engine.common.config.EngineConfig;
import org.apache.seatunnel.engine.common.exception.JobException;
import org.apache.seatunnel.engine.common.exception.JobNotFoundException;
import org.apache.seatunnel.engine.common.exception.SeaTunnelEngineException;
import org.apache.seatunnel.engine.common.utils.PassiveCompletableFuture;
import org.apache.seatunnel.engine.core.job.JobDAGInfo;
Expand Down Expand Up @@ -525,7 +526,7 @@ public void updateTaskExecutionState(TaskExecutionState taskExecutionState) {
TaskGroupLocation taskGroupLocation = taskExecutionState.getTaskGroupLocation();
JobMaster runningJobMaster = runningJobMasterMap.get(taskGroupLocation.getJobId());
if (runningJobMaster == null) {
throw new JobException(
throw new JobNotFoundException(
String.format("Job %s not running", taskGroupLocation.getJobId()));
}
runningJobMaster.updateTaskExecutionState(taskExecutionState);
Expand Down
49 changes: 36 additions & 13 deletions ...-engine-server/src/main/java/org/apache/seatunnel/engine/server/TaskExecutionService.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@
import org.apache.seatunnel.engine.common.config.ConfigProvider;
import org.apache.seatunnel.engine.common.config.SeaTunnelConfig;
import org.apache.seatunnel.engine.common.config.server.ThreadShareMode;
import org.apache.seatunnel.engine.common.exception.JobNotFoundException;
import org.apache.seatunnel.engine.common.loader.SeaTunnelChildFirstClassLoader;
import org.apache.seatunnel.engine.common.utils.PassiveCompletableFuture;
import org.apache.seatunnel.engine.server.exception.TaskGroupContextNotFoundException;
Expand Down Expand Up @@ -336,17 +337,31 @@ public PassiveCompletableFuture<TaskExecutionState> deployLocalTask(
logger.severe(ExceptionUtils.getMessage(t));
resultFuture.completeExceptionally(t);
}
resultFuture.whenComplete(
resultFuture.whenCompleteAsync(
withTryCatch(
logger,
(r, s) -> {
if (s != null) {
logger.severe(
String.format(
"Task %s complete with error %s",
taskGroup.getTaskGroupLocation(),
ExceptionUtils.getMessage(s)));
}
if (r == null) {
r =
new TaskExecutionState(
taskGroup.getTaskGroupLocation(),
ExecutionState.FAILED,
s);
}
logger.info(
String.format(
"Task %s complete with state %s",
r != null ? r.getTaskGroupLocation() : "null",
r != null ? r.getExecutionState() : "null"));
r.getTaskGroupLocation(), r.getExecutionState()));
notifyTaskStatusToMaster(taskGroup.getTaskGroupLocation(), r);
}));
}),
executorService);
return new PassiveCompletableFuture<>(resultFuture);
}

Expand All @@ -370,16 +385,24 @@ private void notifyTaskStatusToMaster(
notifyStateSuccess = true;
} catch (InterruptedException e) {
logger.severe("send notify task status failed", e);
} catch (JobNotFoundException e) {
logger.warning("send notify task status failed because can't find job", e);
notifyStateSuccess = true;
} catch (ExecutionException e) {
logger.warning(ExceptionUtils.getMessage(e));
logger.warning(
String.format(
"notify the job of the task(%s) status failed, retry in %s millis",
taskGroupLocation, sleepTime));
try {
Thread.sleep(sleepTime);
} catch (InterruptedException ex) {
logger.severe(e);
if (e.getCause() instanceof JobNotFoundException) {
logger.warning("send notify task status failed because can't find job", e);
notifyStateSuccess = true;
} else {
logger.warning(ExceptionUtils.getMessage(e));
logger.warning(
String.format(
"notify the job of the task(%s) status failed, retry in %s millis",
taskGroupLocation, sleepTime));
try {
Thread.sleep(sleepTime);
} catch (InterruptedException ex) {
logger.severe(e);
}
}
}
}
Expand Down