You are viewing a plain text version of this content. The canonical link for it is here.
Posted to oak-commits@jackrabbit.apache.org by re...@apache.org on 2017/03/07 14:55:32 UTC
svn commit: r1785844 - in /jackrabbit/oak/branches/1.6: ./
oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/SplitDocumentCleanUp.java
oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/mongo/MongoVersionGCSupport.java
Author: reschke
Date: Tue Mar 7 14:55:32 2017
New Revision: 1785844
URL: http://svn.apache.org/viewvc?rev=1785844&view=rev
Log:
OAK-5878: SplitDocumentCleanup iterates twice over splitDocGarbage (ported to 1.6)
Delete split documents right after disconnecting them.
Modified:
jackrabbit/oak/branches/1.6/ (props changed)
jackrabbit/oak/branches/1.6/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/SplitDocumentCleanUp.java
jackrabbit/oak/branches/1.6/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/mongo/MongoVersionGCSupport.java
Propchange: jackrabbit/oak/branches/1.6/
------------------------------------------------------------------------------
--- svn:mergeinfo (original)
+++ svn:mergeinfo Tue Mar 7 14:55:32 2017
@@ -1,3 +1,3 @@
/jackrabbit/oak/branches/1.0:1665962
-/jackrabbit/oak/trunk:1781068,1781075,1781248,1781386,1781846,1781907,1782000,1782029,1782196,1782447,1782770,1782945,1782973,1782990,1783061,1783066,1783089,1783104-1783105,1783619,1783731,1783733,1783742,1783855,1783891,1784023,1784130,1784162,1784251,1784401,1784551,1785095,1785108,1785283
+/jackrabbit/oak/trunk:1781068,1781075,1781248,1781386,1781846,1781907,1782000,1782029,1782196,1782447,1782770,1782945,1782973,1782990,1783061,1783066,1783089,1783104-1783105,1783619,1783731,1783733,1783742,1783855,1783891,1784023,1784130,1784162,1784251,1784401,1784551,1785095,1785108,1785283,1785838
/jackrabbit/trunk:1345480
Modified: jackrabbit/oak/branches/1.6/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/SplitDocumentCleanUp.java
URL: http://svn.apache.org/viewvc/jackrabbit/oak/branches/1.6/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/SplitDocumentCleanUp.java?rev=1785844&r1=1785843&r2=1785844&view=diff
==============================================================================
--- jackrabbit/oak/branches/1.6/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/SplitDocumentCleanUp.java (original)
+++ jackrabbit/oak/branches/1.6/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/SplitDocumentCleanUp.java Tue Mar 7 14:55:32 2017
@@ -20,7 +20,6 @@ import java.io.Closeable;
import java.io.IOException;
import java.util.List;
-import com.google.common.base.Function;
import com.google.common.collect.Lists;
import org.apache.jackrabbit.oak.plugins.document.VersionGarbageCollector.VersionGCStats;
@@ -29,21 +28,25 @@ import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import static com.google.common.base.Preconditions.checkArgument;
-import static com.google.common.collect.Iterables.transform;
import static org.apache.jackrabbit.oak.plugins.document.Collection.NODES;
import static org.apache.jackrabbit.oak.plugins.document.NodeDocument.SplitDocType.INTERMEDIATE;
import static org.apache.jackrabbit.oak.plugins.document.NodeDocument.SplitDocType.NONE;
/**
-* Implements a split document cleanup.
-*/
+ * Implements a split document cleanup.
+ */
public class SplitDocumentCleanUp implements Closeable {
private static final Logger LOG = LoggerFactory.getLogger(SplitDocumentCleanUp.class);
+ // number of document IDs to collect before removing them in a single call
+ private static final int DELETE_BATCH_SIZE = 100;
+
protected final DocumentStore store;
protected final Iterable<NodeDocument> splitDocGarbage;
protected final VersionGCStats stats;
+ protected final List<String> idsToBeDeleted = Lists.newArrayList();
+ protected int deleteCount;
protected SplitDocumentCleanUp(DocumentStore store,
VersionGCStats stats,
@@ -56,20 +59,31 @@ public class SplitDocumentCleanUp implem
protected SplitDocumentCleanUp disconnect() {
for (NodeDocument splitDoc : splitDocGarbage) {
disconnect(splitDoc);
+ collectIdToBeDeleted(splitDoc.getId());
}
return this;
}
+ /**
+ * Collects document IDs for subsequent deletion.
+ * <p>
+ * Implementations that override
+ * {@link SplitDocumentCleanUp#deleteSplitDocuments()} should override this
+ * method as well.
+ */
+ protected void collectIdToBeDeleted(String id) {
+ idsToBeDeleted.add(id);
+ // proceed to delete early if we reach DELETE_BATCH_SIZE
+ if (idsToBeDeleted.size() >= DELETE_BATCH_SIZE) {
+ store.remove(NODES, idsToBeDeleted);
+ deleteCount += idsToBeDeleted.size();
+ idsToBeDeleted.clear();
+ }
+ }
+
protected int deleteSplitDocuments() {
- List<String> docsToDelete = Lists.newArrayList(transform(splitDocGarbage,
- new Function<NodeDocument, String>() {
- @Override
- public String apply(NodeDocument input) {
- return input.getId();
- }
- }));
- store.remove(NODES, docsToDelete);
- return docsToDelete.size();
+ store.remove(NODES, idsToBeDeleted);
+ return idsToBeDeleted.size() + deleteCount;
}
private void disconnect(NodeDocument splitDoc) {
Modified: jackrabbit/oak/branches/1.6/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/mongo/MongoVersionGCSupport.java
URL: http://svn.apache.org/viewvc/jackrabbit/oak/branches/1.6/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/mongo/MongoVersionGCSupport.java?rev=1785844&r1=1785843&r2=1785844&view=diff
==============================================================================
--- jackrabbit/oak/branches/1.6/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/mongo/MongoVersionGCSupport.java (original)
+++ jackrabbit/oak/branches/1.6/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/document/mongo/MongoVersionGCSupport.java Tue Mar 7 14:55:32 2017
@@ -171,6 +171,11 @@ public class MongoVersionGCSupport exten
}
@Override
+ protected void collectIdToBeDeleted(String id) {
+ // nothing to do here, as we're overwriting deleteSplitDocuments()
+ }
+
+ @Override
protected int deleteSplitDocuments() {
DBObject query = createQuery(gcTypes, oldestRevTimeStamp);