https://jira.opendaylight.org This file is an XML representation of an issue en-us 8.20.10 820010 22-06-2022 https://jira.opendaylight.org/browse/CONTROLLER-1695 controller <p>When transitioning from a connecting connection to a local connection, if we have outstanding read replies and immediate aborts, the replay will fail with a ConcurrentModificationException:</p> <p>java.util.ConcurrentModificationException<br/> at java.util.LinkedHashMap$LinkedHashIterator.nextNode(LinkedHashMap.java:719)<span class="error">&#91;:1.8.0_121&#93;</span><br/> at java.util.LinkedHashMap$LinkedValueIterator.next(LinkedHashMap.java:747)<span class="error">&#91;:1.8.0_121&#93;</span><br/> at org.opendaylight.controller.cluster.databroker.actors.dds.ProxyHistory$ReconnectCohort.replayRequests(ProxyHistory.java:232)<span class="error">&#91;199:org.opendaylight.controller.sal-distributed-datastore:1.5.0.SNAPSHOT&#93;</span><br/> at org.opendaylight.controller.cluster.databroker.actors.dds.AbstractClientHistory$1.replayRequests(AbstractClientHistory.java:311)<span class="error">&#91;199:org.opendaylight.controller.sal-distributed-datastore:1.5.0.SNAPSHOT&#93;</span><br/> at org.opendaylight.controller.cluster.databroker.actors.dds.AbstractDataStoreClientBehavior.lambda$connectionUp$0(AbstractDataStoreClientBehavior.java:147)<span class="error">&#91;199:org.opendaylight.controller.sal-distributed-datastore:1.5.0.SNAPSHOT&#93;</span><br/> at org.opendaylight.controller.cluster.access.client.ClientActorBehavior.backendConnectFinished(ClientActorBehavior.java:293)<span class="error">&#91;197:org.opendaylight.controller.cds-access-client:1.1.0.SNAPSHOT&#93;</span></p> <p>Operating System: All<br/> Platform: All</p> CONTROLLER-1695

tell-based-protocol: ConcurrentModificationException during replay

Bug Resolved Done Robert Varga Robert Varga Tue, 23 May 2017 16:57:12 +0000 Tue, 25 Jul 2023 08:24:40 +0000 Tue, 30 May 2017 12:31:09 +0000 clustering 0 1 <p>Attachment cme.txt.gz has been added with description: Cleaned-up karaf.log with relevant data</p> <p>The reason for this failure is apparent here:</p> <p>2017-05-23 12:25:16,678 | DEBUG | ult-dispatcher-5 | ProxyHistory | 199 - org.opendaylight.controller.sal-distributed-datastore - 1.5.0.SNAPSHOT | LocalHistoryIdentifier{client=ClientIdentifier</p> {frontend=member-2-frontend-datastore-config, generation=0} <p>, history=0, cookie=0} created successor transaction proxy LocalReadOnlyProxyTransaction</p> {identifier=member-2-datastore-config-fe-0-txn-0-0, state=open}<br/> 2017-05-23 12:25:16,678 | DEBUG | ult-dispatcher-5 | AbstractProxyTransaction | 199 - org.opendaylight.controller.sal-distributed-datastore - 1.5.0.SNAPSHOT | Replaying queued request ReadTransactionRequest{target=member-2-datastore-config-fe-0-txn-0-0, sequence=0, replyTo=Actor[akka://opendaylight-cluster-data/user/$a#-555098784], path=/(urn:opendaylight:yang:aaa:cert?revision=2015-11-26)aaa-cert-service-config} to successor LocalReadOnlyProxyTransaction{identifier=member-2-datastore-config-fe-0-txn-0-0, state=open} <p>2017-05-23 12:25:16,679 | DEBUG | ult-dispatcher-5 | RemoteProxyTransaction | 199 - org.opendaylight.controller.sal-distributed-datastore - 1.5.0.SNAPSHOT | Read request completed with ReadTransactionSuccess</p> {target=member-2-datastore-config-fe-0-txn-0-0, sequence=0} <p>2017-05-23 12:25:16,707 | DEBUG | ult-dispatcher-5 | AbstractProxyTransaction | 199 - org.opendaylight.controller.sal-distributed-datastore - 1.5.0.SNAPSHOT | Transaction RemoteProxyTransaction</p> {identifier=member-2-datastore-config-fe-0-txn-0-0, state=successor} allocated sequence 1<br/> 2017-05-23 12:25:16,707 | DEBUG | ult-dispatcher-5 | AbstractProxyTransaction | 199 - org.opendaylight.controller.sal-distributed-datastore - 1.5.0.SNAPSHOT | Transaction proxy RemoteProxyTransaction{identifier=member-2-datastore-config-fe-0-txn-0-0, state=successor} <p> sending request ModifyTransactionRequest</p> {target=member-2-datastore-config-fe-0-txn-0-0, sequence=1, replyTo=Actor[akka://opendaylight-cluster-data/user/$a#-555098784], operations=[], protocol=ABORT} <p> callback org.opendaylight.controller.cluster.databroker.actors.dds.RemoteProxyTransaction$$Lambda$119/1241459742@11834a0c<br/> 2017-05-23 12:25:16,708 | DEBUG | ult-dispatcher-5 | AbstractClientConnection | 197 - org.opendaylight.controller.cds-access-client - 1.1.0.SNAPSHOT | member-2-frontend-datastore-config: timer already scheduled<br/> 2017-05-23 12:25:16,708 | DEBUG | ult-dispatcher-5 | ProxyHistory | 199 - org.opendaylight.controller.sal-distributed-datastore - 1.5.0.SNAPSHOT | Proxy org.opendaylight.controller.cluster.databroker.actors.dds.ProxyHistory$RemoteSingle@434df869 aborting transaction RemoteProxyTransaction</p> {identifier=member-2-datastore-config-fe-0-txn-0-0, state=successor} <p>I.e. transaction abort is completing immediately on the replay thread and touches the cohorts map, through which replay is currently iterating &#8211; hence causing CME as soon as we try to move to the next proxy.</p> <p><a href="https://git.opendaylight.org/gerrit/57759" class="external-link" target="_blank" rel="nofollow noopener">https://git.opendaylight.org/gerrit/57759</a><br/> <a href="https://git.opendaylight.org/gerrit/57770" class="external-link" target="_blank" rel="nofollow noopener">https://git.opendaylight.org/gerrit/57770</a></p> Blocks CONTROLLER-1664 Development External issue ID 8538 External issue URL Issue Type Rank 0|i02s5r: