Shashank Thillainathan created HBASE-25827:
----------------------------------------------
Summary: Per Cell TTL tags get duplicated with increments causing
tags length overflow
Key: HBASE-25827
URL: https://issues.apache.org/jira/browse/HBASE-25827
Project: HBase
Issue Type: Bug
Components: regionserver
Affects Versions: 2.2.6, 2.1.9
Reporter: Shashank Thillainathan
Incrementing with per cell TTL and flushing corrupts the HFile.
Reproducing the issue:
Incrementing a row and a column with per cell TTL for about 3 thousand times
and flushing corrupts the HFile leaving the table unusable.
Cause:
On reading the HFile, it is found that duplicate TTL tags get appended for each
cell.
Though this case has already been addressed here at
[HBASE-18030|https://issues.apache.org/jira/browse/HBASE-18030], corruption
still occurs even with this patch.
{code:java}
// java.lang.IllegalStateException: Invalid currTagsLen -31260. Block offset:
250962, block length: 76568, position: 42207 (without header).
path=hdfs://hdfs/file/path
at
org.apache.hadoop.hbase.io.hfile.HFileReaderImpl$HFileScannerImpl.checkTagsLen(HFileReaderImpl.java:642)
at
org.apache.hadoop.hbase.io.hfile.HFileReaderImpl$HFileScannerImpl.readKeyValueLen(HFileReaderImpl.java:630)
at
org.apache.hadoop.hbase.io.hfile.HFileReaderImpl$HFileScannerImpl._next(HFileReaderImpl.java:1080)
at
org.apache.hadoop.hbase.io.hfile.HFileReaderImpl$HFileScannerImpl.next(HFileReaderImpl.java:1097)
at
org.apache.hadoop.hbase.regionserver.StoreFileScanner.next(StoreFileScanner.java:208)
at
org.apache.hadoop.hbase.regionserver.KeyValueHeap.next(KeyValueHeap.java:120)
at
org.apache.hadoop.hbase.regionserver.StoreScanner.next(StoreScanner.java:654)
at
org.apache.hadoop.hbase.regionserver.compactions.Compactor.performCompaction(Compactor.java:388)
at
org.apache.hadoop.hbase.regionserver.compactions.Compactor.compact(Compactor.java:327)
at
org.apache.hadoop.hbase.regionserver.compactions.DefaultCompactor.compact(DefaultCompactor.java:65)
at
org.apache.hadoop.hbase.regionserver.DefaultStoreEngine$DefaultCompactionContext.compact(DefaultStoreEngine.java:126)
at org.apache.hadoop.hbase.regionserver.HStore.compact(HStore.java:1432)
at
org.apache.hadoop.hbase.regionserver.HRegion.compact(HRegion.java:2192)
at
org.apache.hadoop.hbase.regionserver.CompactSplit$CompactionRunner.doCompaction(CompactSplit.java:577)
at
org.apache.hadoop.hbase.regionserver.CompactSplit$CompactionRunner.run(CompactSplit.java:619)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
