[CASSANDRA-11616] cassandra very high cpu rate

• Type: Bug
• Status: Resolved
• Priority: Normal
• Resolution: Not A Problem
• Component/s: Legacy/Local Write-Read Paths

Environment

CentOS release 6.4
4 nodes cluster
cassandra 3.0.5
nodetool cfstats mykeyspace show the table data volume: Number of keys (estimate): 77570676

• Severity: Normal
• Since Version: 3.0.5
• Assignee: Sam Tunnicliffe
• Reporter: PengtaoGeng

Authors

Sam Tunnicliffe

• Created: 20/Apr/16 08:07
• Updated: 16/Apr/19 09:30
• Resolved: 21/Apr/16 03:33

Description

Under the very low speed of query CPU utilization of 100%.
Query cql is only by partition key or by second index.

Blow is the desc table info:
CREATE TABLE mykeyspace.userlabel (
id text PRIMARY KEY,
cardno text,
phone text,
ccount text,
username text
) ;
CREATE INDEX userlabel_phone ON mykeyspace.userlabel (phone)

top -H and jstack find the utilization cpu threads, they are all come from "SharedPool-Worker".
Show one thread jstack info:

"SharedPool-Worker-28" #205 daemon prio=5 os_prio=0 tid=0x00007f1610cc8780 nid=0xe7c0 runnable [0x00007f0ed566f000]
java.lang.Thread.State: RUNNABLE
at org.apache.cassandra.utils.MurmurHash.hash3_x64_128(MurmurHash.java:191)
at org.apache.cassandra.dht.Murmur3Partitioner.getHash(Murmur3Partitioner.java:181)
at org.apache.cassandra.dht.Murmur3Partitioner.decorateKey(Murmur3Partitioner.java:53)
at org.apache.cassandra.db.PartitionPosition$ForKey.get(PartitionPosition.java:49)
at org.apache.cassandra.db.marshal.PartitionerDefinedOrder.compareCustom(PartitionerDefinedOrder.java:93)
at org.apache.cassandra.db.marshal.AbstractType.compare(AbstractType.java:158)
at org.apache.cassandra.db.ClusteringComparator.compareComponent(ClusteringComparator.java:166)
at org.apache.cassandra.db.ClusteringComparator.compare(ClusteringComparator.java:137)
at org.apache.cassandra.db.ClusteringComparator.compare(ClusteringComparator.java:126)
at org.apache.cassandra.db.ClusteringComparator.compare(ClusteringComparator.java:44)
at org.apache.cassandra.utils.MergeIterator$Candidate.compareTo(MergeIterator.java:378)
at org.apache.cassandra.utils.MergeIterator$ManyToOne.replaceAndSink(MergeIterator.java:266)
at org.apache.cassandra.utils.MergeIterator$ManyToOne.advance(MergeIterator.java:189)
at org.apache.cassandra.utils.MergeIterator$ManyToOne.computeNext(MergeIterator.java:158)
at org.apache.cassandra.utils.AbstractIterator.hasNext(AbstractIterator.java:47)
at org.apache.cassandra.db.rows.UnfilteredRowIterators$UnfilteredRowMergeIterator.computeNext(UnfilteredRowIterators.java:428)
at org.apache.cassandra.db.rows.UnfilteredRowIterators$UnfilteredRowMergeIterator.computeNext(UnfilteredRowIterators.java:288)
at org.apache.cassandra.utils.AbstractIterator.hasNext(AbstractIterator.java:47)
at org.apache.cassandra.db.transform.BaseRows.hasNext(BaseRows.java:108)
at org.apache.cassandra.index.internal.composites.CompositesSearcher$1.prepareNext(CompositesSearcher.java:130)
at org.apache.cassandra.index.internal.composites.CompositesSearcher$1.hasNext(CompositesSearcher.java:83)
at org.apache.cassandra.db.transform.BasePartitions.hasNext(BasePartitions.java:72)
at org.apache.cassandra.db.partitions.UnfilteredPartitionIterators$Serializer.serialize(UnfilteredPartitionIterators.java:295)
at org.apache.cassandra.db.ReadResponse$LocalDataResponse.build(ReadResponse.java:134)
at org.apache.cassandra.db.ReadResponse$LocalDataResponse.<init>(ReadResponse.java:127)
at org.apache.cassandra.db.ReadResponse$LocalDataResponse.<init>(ReadResponse.java:123)
at org.apache.cassandra.db.ReadResponse.createDataResponse(ReadResponse.java:65)
at org.apache.cassandra.db.ReadCommand.createResponse(ReadCommand.java:289)
at org.apache.cassandra.db.ReadCommandVerbHandler.doVerb(ReadCommandVerbHandler.java:47)
at org.apache.cassandra.net.MessageDeliveryTask.run(MessageDeliveryTask.java:67)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at org.apache.cassandra.concurrent.AbstractLocalAwareExecutorService$FutureTask.run(AbstractLocalAwareExecutorService.java:164)
at org.apache.cassandra.concurrent.AbstractLocalAwareExecutorService$LocalSessionFutureTask.run(AbstractLocalAwareExecutorService.java:136)
at org.apache.cassandra.concurrent.SEPWorker.run(SEPWorker.java:105)
at java.lang.Thread.run(Thread.java:745)

I'm using Cassandra version 3.0.5, just upgraded from 3.0.4 recently for bug of CASSANDRA-11304.

Does it cause by "org.apache.cassandra.db.partitions.UnfilteredPartitionIterator#prepareNext" ?