anishek commented on a change in pull request #549: HIVE-21314 : Hive
Replication not retaining the owner in the replicated table
URL: https://github.com/apache/hive/pull/549#discussion_r259809837
##########
File path:
ql/src/java/org/apache/hadoop/hive/ql/exec/repl/ExternalTableCopyTaskBuilder.java
##########
@@ -54,14 +59,53 @@
List<Task<? extends Serializable>> tasks(TaskTracker tracker) {
List<Task<? extends Serializable>> tasks = new ArrayList<>();
Iterator<DirCopyWork> itr = work.getPathsToCopyIterator();
- while (tracker.canAddMoreTasks() && itr.hasNext()) {
+ int numTaskCanBeAdded = tracker.numTaskCanBeAdded();
+ Task<? extends Serializable> barrierTask = TaskFactory.get(new
DependencyCollectionWork(), conf);
+ while (numTaskCanBeAdded-- > 0 && itr.hasNext()) {
DirCopyWork dirCopyWork = itr.next();
Task<DirCopyWork> task = TaskFactory.get(dirCopyWork, conf);
tasks.add(task);
- tracker.addTask(task);
+ barrierTask.addDependentTask(task);
LOG.debug("added task for {}", dirCopyWork);
}
- return tasks;
+
+ if (!tasks.isEmpty()) {
+ tracker.addDependentTask(barrierTask);
+ tracker.addTaskList(tasks);
+ return Collections.singletonList(barrierTask);
+ } else {
+ return tasks;
+ }
+ }
+
+ private static Integer setTargetPathOwnerInt(Path targetPath, Path
sourcePath, HiveConf conf) throws IOException {
+ FileSystem targetFs = targetPath.getFileSystem(conf);
+ if (!targetFs.exists(targetPath)) {
+ targetFs.create(targetPath);
+ }
+ FileStatus status =
sourcePath.getFileSystem(conf).getFileStatus(sourcePath);
+ if (status == null) {
+ throw new IOException("source path missing " + sourcePath);
Review comment:
this would throw exception if the source path will not be available during
load if the table was dropped in source, should handle that case, the owner
should always be the one at the directory level for table / partition
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
