lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

Removing rollbacks instants from timeline for restore operation (#4518)

Browse Source
This commit is contained in:
Sivabalan Narayanan
2022-01-09 21:14:28 -05:00
committed by GitHub
parent e9a7f49f55
commit 56f93f4ebd
2 changed files with 12 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
hudi-client/hudi-client-common/src/main/java/org/apache/hudi/table/action/restore/BaseRestoreActionExecutor.java
View File

@@ -99,6 +99,16 @@ public abstract class BaseRestoreActionExecutor<T extends HoodieRecordPayload, I
writeToMetadata(restoreMetadata);
table.getActiveTimeline().saveAsComplete(new HoodieInstant(true, HoodieTimeline.RESTORE_ACTION, instantTime),
TimelineMetadataUtils.serializeRestoreMetadata(restoreMetadata));
// get all rollbacks instants after restore instant time and delete them.
// if not, rollbacks will be considered not completed and might hinder metadata table compaction.
List<HoodieInstant> instantsToRollback = table.getActiveTimeline().getRollbackTimeline()
.getReverseOrderedInstants()
.filter(instant -> HoodieActiveTimeline.GREATER_THAN.test(instant.getTimestamp(), restoreInstantTime))
.collect(Collectors.toList());
instantsToRollback.forEach(entry -> {
table.getActiveTimeline().deletePending(new HoodieInstant(HoodieInstant.State.INFLIGHT, HoodieTimeline.ROLLBACK_ACTION, entry.getTimestamp()));
table.getActiveTimeline().deletePending(new HoodieInstant(HoodieInstant.State.REQUESTED, HoodieTimeline.ROLLBACK_ACTION, entry.getTimestamp()));
});
LOG.info("Commits " + instantsRolledBack + " rollback is complete. Restored table to " + restoreInstantTime);
return restoreMetadata;
}

2
hudi-client/hudi-spark-client/src/test/java/org/apache/hudi/client/functional/TestHoodieClientOnCopyOnWriteStorage.java
View File

@@ -577,6 +577,8 @@ public class TestHoodieClientOnCopyOnWriteStorage extends HoodieClientTestBase {
client = getHoodieWriteClient(newConfig);
client.restoreToInstant("004");
assertFalse(metaClient.reloadActiveTimeline().getRollbackTimeline().lastInstant().isPresent());
// Check the entire dataset has all records still
String[] fullPartitionPaths = new String[dataGen.getPartitionPaths().length];
for (int i = 0; i < fullPartitionPaths.length; i++) {