[
https://issues.apache.org/jira/browse/CASSANDRA-12215?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15386470#comment-15386470
]
Hau Phan commented on CASSANDRA-12215:
--------------------------------------
One thing I noticed, when setting the gc_grace_seconds to 8640000 (100 days),
the tables are readable and compaction runs without error.
Enabling tracing, doing a select * from the table still shows tombstones:
{code}
Read 124 live and 4 tombstone cells [SharedPool-Worker-2] | 2016-07-20
14:09:17.236000 | 127.0.0.1 | 1849
{code}
I have run a COPY TO, truncated the table, then ran a COPY FROM, ran the select
* again, and the tombstones still exist.
Reviewing the 'dump' file, I'm not seeing anything marked tombstones and the
data looks fine. My understanding is, compaction should remove the tombstone
records if it's past the gc_grace_seconds limit. Also, if the 'dump' file
doesn't show tombstones, where is cassandra retrieving that info?
> NullPointerException during Compaction
> --------------------------------------
>
> Key: CASSANDRA-12215
> URL: https://issues.apache.org/jira/browse/CASSANDRA-12215
> Project: Cassandra
> Issue Type: Bug
> Components: Compaction
> Environment: Cassandra 3.0.8, cqlsh 5.0.1
> Reporter: Hau Phan
> Fix For: 3.0.x
>
>
> Running 3.0.8 on a single standalone node with cqlsh 5.0.1, the keyspace RF =
> 1 and class SimpleStrategy.
> Attempting to run a 'select * from <table>' and receiving this error:
> ReadFailure: code=1300 [Replica(s) failed to execute read] message="Operation
> failed - received 0 responses and 1 failures" info={'failures': 1,
> 'received_responses': 0, 'required_responses': 1, 'consistency': 'ONE'}
> Cassandra system.log prints this:
> {code}
> ERROR [CompactionExecutor:5] 2016-07-15 13:42:13,219 CassandraDaemon.java:201
> - Exception in thread Thread[CompactionExecutor:5,1,main]
> java.lang.NullPointerException: null
> at
> org.apache.cassandra.db.transform.UnfilteredRows.isEmpty(UnfilteredRows.java:58)
> ~[apache-cassandra-3.0.8.jar:3.0.8]
> at
> org.apache.cassandra.db.partitions.PurgeFunction.applyToPartition(PurgeFunction.java:64)
> ~[apache-cassandra-3.0.8.jar:3.0.8]
> at
> org.apache.cassandra.db.partitions.PurgeFunction.applyToPartition(PurgeFunction.java:24)
> ~[apache-cassandra-3.0.8.jar:3.0.8]
> at
> org.apache.cassandra.db.transform.BasePartitions.hasNext(BasePartitions.java:96)
> ~[apache-cassandra-3.0.8.jar:3.0.8]
> at
> org.apache.cassandra.db.compaction.CompactionIterator.hasNext(CompactionIterator.java:226)
> ~[apache-cassandra-3.0.8.jar:3.0.8]
> at
> org.apache.cassandra.db.compaction.CompactionTask.runMayThrow(CompactionTask.java:177)
> ~[apache-cassandra-3.0.8.jar:3.0.8]
> at
> org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28)
> ~[apache-cassandra-3.0.8.jar:3.0.8]
> at
> org.apache.cassandra.db.compaction.CompactionTask.executeInternal(CompactionTask.java:78)
> ~[apache-cassandra-3.0.8.jar:3.0.8]
> at
> org.apache.cassandra.db.compaction.AbstractCompactionTask.execute(AbstractCompactionTask.java:60)
> ~[apache-cassandra-3.0.8.jar:3.0.8]
> at
> org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionCandidate.run(CompactionManager.java:263)
> ~[apache-cassandra-3.0.8.jar:3.0.8]
> at
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> ~[na:1.8.0_65]
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> ~[na:1.8.0_65]
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> ~[na:1.8.0_65]
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> [na:1.8.0_65]
> at java.lang.Thread.run(Thread.java:745) [na:1.8.0_65]
> {code}
> Doing a sstabledump -d shows a few rows with the column value of
> "<tombstone>", telling me compaction doesn't seem to be working correctly.
> # nodetool compactionstats
> pending tasks: 1
> attempting to run a compaction gets:
> # nodetool compact <table> <cf>
> error: null
> -- StackTrace --
> java.lang.NullPointerException
> at
> org.apache.cassandra.db.transform.UnfilteredRows.isEmpty(UnfilteredRows.java:58)
> at
> org.apache.cassandra.db.partitions.PurgeFunction.applyToPartition(PurgeFunction.java:64)
> at
> org.apache.cassandra.db.partitions.PurgeFunction.applyToPartition(PurgeFunction.java:24)
> at
> org.apache.cassandra.db.transform.BasePartitions.hasNext(BasePartitions.java:96)
> at
> org.apache.cassandra.db.compaction.CompactionIterator.hasNext(CompactionIterator.java:226)
> at
> org.apache.cassandra.db.compaction.CompactionTask.runMayThrow(CompactionTask.java:177)
> at
> org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28)
> at
> org.apache.cassandra.db.compaction.CompactionTask.executeInternal(CompactionTask.java:78)
> at
> org.apache.cassandra.db.compaction.AbstractCompactionTask.execute(AbstractCompactionTask.java:60)
> at
> org.apache.cassandra.db.compaction.CompactionManager$8.runMayThrow(CompactionManager.java:606)
> at
> org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28)
> at
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> at java.lang.Thread.run(Thread.java:745)
> Since the table is pretty small, I can do a copy to, truncate the table, and
> copy from, and the table is fine. But my concern is if compaction fails to
> remove those rows, and the table will eventually be very large in a
> production environment, the copy, truncate, and copy will no longer be an
> option.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
