[infinispan-issues] [JBoss JIRA] (ISPN-3330) Hotrod clients getWithMetadata doesn't work when locking isolation != NONE

[Adrian Nistor (JIRA)]

    [ https://issues.jboss.org/browse/ISPN-3330?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12804597#comment-12804597 ] 

Adrian Nistor commented on ISPN-3330:
-------------------------------------

Besides isolation level validation (which is ok) this PR also tries to validate key and value equivalence by constraining them to be able to compare byte[]. This is wrong for all caches that are not supposed to be accessed remotely, for example metadata caches spawned by query. These (especially if infinispan directory is used) do not have byte[] keys and values so the validation is not meaningful to them. Since we cannot know at this moment which caches are going to be accessed remotely I will remove this validation.

                
> Hotrod clients getWithMetadata doesn't work when locking isolation != NONE
> --------------------------------------------------------------------------
>
>                 Key: ISPN-3330
>                 URL: https://issues.jboss.org/browse/ISPN-3330
>             Project: Infinispan
>          Issue Type: Bug
>    Affects Versions: 5.3.0.Final, 6.0.0.Alpha4
>            Reporter: Jakub Markos
>            Assignee: Galder Zamarreño
>            Priority: Minor
>             Fix For: 6.0.0.Beta1, 6.0.0.Final
>
>
> I have a cluster of 2 infinispan servers with this configuration:
> {code:xml}<subsystem xmlns="urn:infinispan:server:core:5.3">
>     <cache-container name="default" default-cache="default" listener-executor="infinispan-listener">
>         <transport stack="udp" executor="infinispan-transport" lock-timeout="240000"/>
>         <distributed-cache name="default" start="EAGER" mode="SYNC" segments="1" owners="2" batching="false" l1-lifespan="0" remote-timeout="60000">
>             <locking isolation="REPEATABLE_READ"/>
>         </distributed-cache>
>     </cache-container>
> </subsystem>{code}
> Running this code:
> {code}remoteCache.put("k1", "v1", 10000000, TimeUnit.MICROSECONDS); // setting only lifespan
> MetadataValue<String> k1 = remoteCache.getWithMetadata("k1");
> assertTrue(k1.getValue().equals("v1"));
> // microseconds converted to seconds
> assertTrue(k1.getLifespan() == 10);
> assertTrue(k1.getMaxIdle() == -1);{code}
> fails with:
> {code}org.infinispan.client.hotrod.exceptions.HotRodClientException:Request for message id[5] returned server error (status=0x85): java.lang.ClassCastException: org.infinispan.container.entries.RepeatableReadEntry can
> not be cast to org.infinispan.container.entries.InternalCacheEntry
>         at org.infinispan.client.hotrod.impl.protocol.Codec10.checkForErrorsInResponseStatus(Codec10.java:143)
>         at org.infinispan.client.hotrod.impl.protocol.Codec10.readHeader(Codec10.java:99)
>         at org.infinispan.client.hotrod.impl.operations.HotRodOperation.readHeaderAndValidate(HotRodOperation.java:56)
>         at org.infinispan.client.hotrod.impl.operations.AbstractKeyOperation.sendKeyOperation(AbstractKeyOperation.java:52)
>         at org.infinispan.client.hotrod.impl.operations.GetWithMetadataOperation.executeOperation(GetWithMetadataOperation.java:35)
>         at org.infinispan.client.hotrod.impl.operations.GetWithMetadataOperation.executeOperation(GetWithMetadataOperation.java:23)
>         at org.infinispan.client.hotrod.impl.operations.RetryOnFailureOperation.execute(RetryOnFailureOperation.java:46)
>         at org.infinispan.client.hotrod.impl.RemoteCacheImpl.getWithMetadata(RemoteCacheImpl.java:145){code}
> Works with isolation="READ_COMMITED"/"NONE" or with no <locking> at all.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira