Skip to content

TEZ-4445: Tez task can get stuck when waiting for all initializers on LogicalIOProcessorRuntimeTask:initialize #238

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

Open
zhangbutao wants to merge 1 commit into
base: master
Choose a base branch
from
+24 −3
Open

TEZ-4445: Tez task can get stuck when waiting for all initializers on LogicalIOProcessorRuntimeTask:initialize #238

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
9 changes: 9 additions & 0 deletions tez-api/src/main/java/org/apache/tez/dag/api/TezConfiguration.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2228,4 +2228,13 @@ static Set<String> getPropertySet() {
@ConfigurationProperty
public static final String TEZ_MRREADER_CONFIG_UPDATE_PROPERTIES = "tez.mrreader.config.update.properties";

/**
* Long value. Specifies the timeout when waiting all initializers to be finished.
* Default time unit is: ms. Set to a negative number to disable .
*/
@ConfigurationScope(Scope.AM)
@ConfigurationProperty(type="long")
public static final String TEZ_AM_INITIALIZE_WAIT_INITIALIZERS_FINISH_TIMEOUT =
TEZ_AM_PREFIX + "initialize.wait.initializers.timeout";
public static final long TEZ_AM_INITIALIZE_WAIT_INITIALIZERS_FINISH_TIMEOUT_DEFAULT = -1;
}
18 changes: 15 additions & 3 deletions ...runtime-internals/src/main/java/org/apache/tez/runtime/LogicalIOProcessorRuntimeTask.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -164,6 +164,7 @@ public class LogicalIOProcessorRuntimeTask extends RuntimeTask {
private final TezExecutors sharedExecutor;
/** nanoTime of the task initialization start. */
private Long initStartTimeNs = null;
private Long waitInitializersFinishTimeout;

public LogicalIOProcessorRuntimeTask(TaskSpec taskSpec, int appAttemptNumber,
Configuration tezConf, String[] localDirs, TezUmbilical tezUmbilical,
Expand Down Expand Up @@ -225,6 +226,8 @@ public LogicalIOProcessorRuntimeTask(TaskSpec taskSpec, int appAttemptNumber,
this.maxEventBacklog = tezConf.getInt(TezConfiguration.TEZ_TASK_MAX_EVENT_BACKLOG,
TezConfiguration.TEZ_TASK_MAX_EVENT_BACKLOG_DEFAULT);
this.sharedExecutor = sharedExecutor;
this.waitInitializersFinishTimeout = tezConf.getLong(TezConfiguration.TEZ_AM_INITIALIZE_WAIT_INITIALIZERS_FINISH_TIMEOUT,
TezConfiguration.TEZ_AM_INITIALIZE_WAIT_INITIALIZERS_FINISH_TIMEOUT_DEFAULT);
}

/**
Expand Down Expand Up @@ -267,10 +270,19 @@ public void initialize() throws Exception {
int completedTasks = 0;
while (completedTasks < numTasks) {
LOG.info("Waiting for " + (numTasks-completedTasks) + " initializers to finish");
Future<Void> future = initializerCompletionService.take();
Future<Void> future;
if (waitInitializersFinishTimeout >= 0) {
future = initializerCompletionService.poll(waitInitializersFinishTimeout, TimeUnit.MILLISECONDS);
} else {
future = initializerCompletionService.take();
}
try {
future.get();
completedTasks++;
if (future != null) {
future.get();
completedTasks++;
} else {
throw new RuntimeException("Timed out while waiting for all initializers to finish");
}
} catch (ExecutionException e) {
if (e.getCause() instanceof Exception) {
throw (Exception) e.getCause();
Expand Down