Spring Integration - Oracle lock contention - spring-integration

Question

The jdbc inbound channel adapter and message stores in our application consistenly stops querying for data and I've been able to track it down to a blocked Oracle session. In Oracle, we are getting enq: TX - row lock contention. The DBA tracked it down to the following:
SELECT COMPLETE, LAST_RELEASED_SEQUENCE, CREATED_DATE, UPDATED_DATE
from INT_MESSAGE_GROUP
where GROUP_KEY = :1 and REGION=:2
Any suggestions on how to resolve this would be greatly appreciated.
SI configuration (extract):
<int-jdbc:message-store id="jdbc-messageStore" data-source="dataSource" />
<int-jdbc:inbound-channel-adapter id="JDBCInboundChannel" query="${cache.integration.jdbc.selectQuery}"
channel="inboundMessagesChannel" data-source="dataSource" update="update CACHE_REPOSITORY set STATUS='P' WHERE GUID IN (:guid)" row-mapper="rowMapper"
max-rows-per-poll="${cache.integration.jdbc.maxRowsPerPoll}" auto-startup="false" >
<int:poller id="jdbcPoller" fixed-delay="${cache.integration.jdbc.fixedDelay}" >
<int:transactional />
</int:poller>
</int-jdbc:inbound-channel-adapter>
<int:chain...>
<int:aggregator id="reportTypeAggregator" ref="aggregatorBean" method="collect"
message-store="jdbc-messageStore" release-strategy="releaseStrategry"
release-strategy-method="canRelease" correlation-strategy="reportTypeCorrelationStrategry"
correlation-strategy-method="correlate" send-partial-result-on-expiry="true"
expire-groups-upon-completion="true" empty-group-min-timeout="30000" />
</int:chain>
<bean id="queryProvider" class="org.springframework.integration.jdbc.store.channel.OracleChannelMessageStoreQueryProvider"/>
<bean id="jdbc-channel-messageStore" class="org.springframework.integration.jdbc.store.JdbcChannelMessageStore">
<property name="dataSource" ref="dataSource"/>
<property name="channelMessageStoreQueryProvider" ref="queryProvider"/>
<property name="region" value="${cache.integration.channelMessageStore.region}"/>
</bean>
<int:channel id="archiveCreationChannel" >
<int:queue message-store="jdbc-channel-messageStore" />
<int:interceptors>
<int:wire-tap channel="logger" timeout="-1"/>
</int:interceptors>
</int:channel>
<bean id="documentMessageStoreReaper" class="org.springframework.integration.store.MessageGroupStoreReaper">
<property name="messageGroupStore" ref="jdbc-messageStore" />
<property name="timeout" value="${cache.integration.reaper.timeout}" />
<property name="autoStartup" value="false" />
</bean>
<task:scheduled-tasks>
<task:scheduled ref="documentMessageStoreReaper" method="run" fixed-rate="10000" />
</task:scheduled-tasks>

Answer 1

I suppose the main issue is around fixed-rate for documentMessageStoreReaper, which initiate the DELETE from INT_MESSAGE_GROUP. You should change it to the fixed-dealy to avoid concurrent tasks.
Since you say that the DELETE from INT_MESSAGE_GROUP is very long, hence 10 sec between tasks might be not enough.