[
https://issues.apache.org/jira/browse/HBASE-12751?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14901693#comment-14901693
]
stack commented on HBASE-12751:
-------------------------------
The TestIOFencing failure is because test is parameterized and the DLR run was
hanging. Fixed.
I removed TestVisibilityLabelsWithDistributedLogReplay because this patch
breaks DLR and DLR is deprecated as-is.
The TestDistributedLogSplitting and TestServerCrashProcedure are both hung up
in MVCC waiting on write number to progress. The accounting must be wrong. Let
me see.... If I do the below hack, tests pass...
{code}
diff --git
a/hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/HRegion.java
b/hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/HRegion.java
index a8ffa8d..e70e7bd 100644
---
a/hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/HRegion.java
+++
b/hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/HRegion.java
@@ -2099,7 +2099,7 @@ public class HRegion implements HeapSize,
PropagatingConfigurationObserver, Regi
writeFlushRequestMarkerToWAL(wal, writeFlushWalMarker));
// TODO: Lets see if we hang here, if there is a scenario where an
outstanding reader
// with a read point is in advance of this write point.
- mvcc.completeAndWait(writeEntry);
+ mvcc.complete(writeEntry);
writeEntry = null;
return new PrepareFlushResult(flushResult, myseqid);
} else {
@@ -2254,7 +2254,7 @@ public class HRegion implements HeapSize,
PropagatingConfigurationObserver, Regi
// uncommitted transactions from being written into HFiles.
// We have to block before we start the flush, otherwise keys that
// were removed via a rollbackMemstore could be written to Hfiles.
- mvcc.completeAndWait(writeEntry);
+ mvcc.complete(writeEntry);
// set writeEntry to null to prevent mvcc.complete from being called
again inside finally
// block
writeEntry = null;
@@ -3158,7 +3158,7 @@ public class HRegion implements HeapSize,
PropagatingConfigurationObserver, Regi
// STEP 8. Advance mvcc. This will make this put visible to scanners and
getters.
// ------------------------------------------------------------------
if (writeEntry != null) {
- mvcc.completeAndWait(writeEntry);
+ mvcc.complete(writeEntry);
writeEntry = null;
} else if (isInReplay) {
// ensure that the sequence id of the region is at least as big as
orig log seq id
@@ -3203,7 +3203,7 @@ public class HRegion implements HeapSize,
PropagatingConfigurationObserver, Regi
}
if (writeEntry != null) mvcc.complete(writeEntry);
} else if (writeEntry != null) {
- mvcc.completeAndWait(writeEntry);
+ mvcc.complete(writeEntry);
}
if (locked) {
@@ -6856,7 +6856,7 @@ public class HRegion implements HeapSize,
PropagatingConfigurationObserver, Regi
}
// 13. Roll mvcc forward
if (writeEntry != null) {
- mvcc.completeAndWait(writeEntry);
+ mvcc.complete(writeEntry);
}
{code}
TestReplicasClient seems broke in a different way altogether. Digging.
> Allow RowLock to be reader writer
> ---------------------------------
>
> Key: HBASE-12751
> URL: https://issues.apache.org/jira/browse/HBASE-12751
> Project: HBase
> Issue Type: Bug
> Components: regionserver
> Affects Versions: 2.0.0, 1.3.0
> Reporter: Elliott Clark
> Assignee: Elliott Clark
> Fix For: 2.0.0, 1.3.0
>
> Attachments: 12751.rebased.v25.txt, 12751.rebased.v26.txt,
> 12751.rebased.v26.txt, 12751.rebased.v27.txt, 12751.rebased.v29.txt,
> 12751.rebased.v31.txt, 12751.rebased.v32.txt, 12751.rebased.v32.txt,
> 12751.rebased.v33.txt, 12751.rebased.v34.txt, 12751.rebased.v35.txt,
> 12751.rebased.v35.txt, 12751.rebased.v35.txt, 12751.v37.txt, 12751.v38.txt,
> 12751v22.txt, 12751v23.txt, 12751v23.txt, 12751v23.txt, 12751v23.txt,
> 12751v36.txt, HBASE-12751-v1.patch, HBASE-12751-v10.patch,
> HBASE-12751-v10.patch, HBASE-12751-v11.patch, HBASE-12751-v12.patch,
> HBASE-12751-v13.patch, HBASE-12751-v14.patch, HBASE-12751-v15.patch,
> HBASE-12751-v16.patch, HBASE-12751-v17.patch, HBASE-12751-v18.patch,
> HBASE-12751-v19 (1).patch, HBASE-12751-v19.patch, HBASE-12751-v2.patch,
> HBASE-12751-v20.patch, HBASE-12751-v20.patch, HBASE-12751-v21.patch,
> HBASE-12751-v3.patch, HBASE-12751-v4.patch, HBASE-12751-v5.patch,
> HBASE-12751-v6.patch, HBASE-12751-v7.patch, HBASE-12751-v8.patch,
> HBASE-12751-v9.patch, HBASE-12751.patch
>
>
> Right now every write operation grabs a row lock. This is to prevent values
> from changing during a read modify write operation (increment or check and
> put). However it limits parallelism in several different scenarios.
> If there are several puts to the same row but different columns or stores
> then this is very limiting.
> If there are puts to the same column then mvcc number should ensure a
> consistent ordering. So locking is not needed.
> However locking for check and put or increment is still needed.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
