[JBoss JIRA] (ISPN-3773) State transfer thread can stop even though there are pending transfer tasks
by Dan Berindei (JIRA)
[ https://issues.jboss.org/browse/ISPN-3773?page=com.atlassian.jira.plugin.... ]
Dan Berindei updated ISPN-3773:
-------------------------------
Fix Version/s: 6.0.1.Final
7.0.0.Alpha1
> State transfer thread can stop even though there are pending transfer tasks
> ---------------------------------------------------------------------------
>
> Key: ISPN-3773
> URL: https://issues.jboss.org/browse/ISPN-3773
> Project: Infinispan
> Issue Type: Bug
> Components: State transfer
> Affects Versions: 6.0.0.Final
> Reporter: Dan Berindei
> Assignee: Dan Berindei
> Priority: Critical
> Labels: nbst
> Fix For: 6.0.1.Final, 7.0.0.Final, 7.0.0.Alpha1
>
>
> Noticed in NonTxOriginatorBecomingPrimaryOwnerTest. The state transfer thread finished the last inbound transfer task, but just before stopping another task is started. The new task doesn't prevent the state transfer thread from stopping, and the node will never request those segments (thus blocking the rebalance from ending).
> {noformat}
> 15:28:31,033 TRACE (asyncTransportThread-1,NodeC:) [InboundTransferTask] Successfully requested segments [33, 6, 7, 8, 9, 11, 13, 50, 54, 20, 52, 22, 59, 25, 24, 27, 26, 29, 28, 31] of cache ___defaultcache from node NodeA-49040
> 15:28:31,264 TRACE (remote-thread-1,NodeC:___defaultcache) [StateConsumerImpl] Adding transfer from NodeA-49040 for segments [32, 5, 6, 7, 8, 10, 12, 51, 49, 19, 21, 53, 23, 59, 25, 24, 27, 26, 28, 30]
> 15:28:31,264 TRACE (remote-thread-1,NodeC:___defaultcache) [StateConsumerImpl] Starting transfer thread: false
> 15:28:31,264 DEBUG (remote-thread-1,NodeC:___defaultcache) [StateConsumerImpl] Finished adding inbound state transfer for segments [5, 6, 7, 8, 10, 12, 19, 21, 23, 25, 24, 27, 26, 28, 30, 32, 51, 49, 53, 59] of cache ___defaultcache
> 15:28:31,264 TRACE (remote-thread-1,NodeC:___defaultcache) [StateTransferLockImpl] Signalling transaction data received for topology 41
> 15:28:31,264 TRACE (asyncTransportThread-1,NodeC:) [StateConsumerImpl] Stopping state transfer thread
> {noformat}
--
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
11 years
[JBoss JIRA] (ISPN-3780) CLONE - InvalidateL1Command during ST should not cancel writing the entry by ST in nontx
by Dan Berindei (JIRA)
[ https://issues.jboss.org/browse/ISPN-3780?page=com.atlassian.jira.plugin.... ]
Dan Berindei updated ISPN-3780:
-------------------------------
Fix Version/s: 7.0.0.Final
7.0.0.Alpha1
> CLONE - InvalidateL1Command during ST should not cancel writing the entry by ST in nontx
> ----------------------------------------------------------------------------------------
>
> Key: ISPN-3780
> URL: https://issues.jboss.org/browse/ISPN-3780
> Project: Infinispan
> Issue Type: Bug
> Components: Distributed Cache
> Affects Versions: 5.3.0.Final
> Reporter: Radim Vansa
> Assignee: William Burns
> Priority: Critical
> Labels: 620
> Fix For: 6.0.1.Final, 7.0.0.Final, 7.0.0.Alpha1
>
>
> When a node which is about to receive the entries for some segment receives InvalidateL1Command, this puts the key into StateConsumer.updatedKeys. After the entries for ST are received, the entry which was invalidated is not updated -> this result in losing the entry.
> Btw., in EntryWrappingInterceptor.visitInvalidateL1Command the trace logs all looked up entries for each entry - this causes some performance problems when tracing is on and there are many invalidated entries. Please, do this only once.
--
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
11 years
[JBoss JIRA] (ISPN-3750) Configuration.toProperties
by Dan Berindei (JIRA)
[ https://issues.jboss.org/browse/ISPN-3750?page=com.atlassian.jira.plugin.... ]
Dan Berindei updated ISPN-3750:
-------------------------------
Fix Version/s: 7.0.0.Final
7.0.0.Alpha1
> Configuration.toProperties
> --------------------------
>
> Key: ISPN-3750
> URL: https://issues.jboss.org/browse/ISPN-3750
> Project: Infinispan
> Issue Type: Feature Request
> Components: Configuration
> Affects Versions: 6.0.0.Final
> Reporter: Michal Linhard
> Assignee: Michal Linhard
> Fix For: 6.0.1.Final, 7.0.0.Final, 7.0.0.Alpha1
>
>
> I'd like to add methods
> Configuration.toProperties() and
> GlobalConfiguration.toProperties()
> that would return current configuration values in flat key value structure (e.g. java.util.Properties) where property keys would reflect names of configuration fields and structure would be reflected by extending the key prefix and dividing by dot. e.g. clustering.hash.numOwners=2
> The least intrusive and maintenance demanding implementation is via Reflection.
> The flat configuration would be exposed via JMX objects, e.g.
> jboss.infinispan:type=Cache,name="testCache(dist_sync)",manager="default",component=Cache
> attribute "configurationProperties"
> jboss.infinispan:type=CacheManager,name="default",component=CacheManager
> attribute "globalConfigurationProperties"
> This is a diagnostic output feature and doesn't affect the way how Infinispan is configured.
--
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
11 years
[JBoss JIRA] (ISPN-3422) In non-tx caches, write operations may not be atomic during rebalance
by Dan Berindei (JIRA)
[ https://issues.jboss.org/browse/ISPN-3422?page=com.atlassian.jira.plugin.... ]
Dan Berindei updated ISPN-3422:
-------------------------------
Fix Version/s: 7.0.0.Final
7.0.0.Alpha1
Affects Version/s: 6.0.0.Final
Component/s: State transfer
> In non-tx caches, write operations may not be atomic during rebalance
> ---------------------------------------------------------------------
>
> Key: ISPN-3422
> URL: https://issues.jboss.org/browse/ISPN-3422
> Project: Infinispan
> Issue Type: Bug
> Components: State transfer
> Affects Versions: 6.0.0.Final
> Reporter: Dan Berindei
> Assignee: Dan Berindei
> Priority: Critical
> Labels: 620, nbst
> Fix For: 6.0.1.Final, 7.0.0.Final, 7.0.0.Alpha1
>
>
> If the cache topology changes while a write command is running and before it has actually committed the entry to the data container, we retry the command (see ISPN-3366 and ISPN-3357). But before we detect the topology change, one or more of the backup owners may have already applied the modification.
> Retrying the command re-acquires the key lock on the primary owner (even if the primary owner didn't change). That means another command could have modified the same key in the meantime, but the retried command is going to ignore any changes and is going to return the value before the first attempt. Obviously, the command is not retried if the first attempt is not successful, but scenarios like this are possible:
> {code}
> thread 1: putIfAbsent(k, v1) -> null
> thread 2: putIfAbsent(k, v2) -> null
> {code}
--
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
11 years