[JBoss JIRA] (ISPN-5513) State Transfer can miss entries that are concurrently activated
by Pedro Ruivo (JIRA)
[ https://issues.jboss.org/browse/ISPN-5513?page=com.atlassian.jira.plugin.... ]
Pedro Ruivo updated ISPN-5513:
------------------------------
Fix Version/s: 8.1.0.Final
(was: 8.0.0.Final)
> State Transfer can miss entries that are concurrently activated
> ---------------------------------------------------------------
>
> Key: ISPN-5513
> URL: https://issues.jboss.org/browse/ISPN-5513
> Project: Infinispan
> Issue Type: Bug
> Components: State Transfer
> Affects Versions: 8.0.0.Alpha1
> Reporter: William Burns
> Fix For: 8.1.0.Final
>
>
> Currently the OutboundTransferTask iterates upon the data container and then runs process for the state loader. However if an entry is activated during or after the data container iteration it is possible this entry is then not seen and subsequently is not present in the store when it is processed.
> EntryRetriever had this same issue and it was required to register a cache listener to listen for activations and then replay the data after finishing with the store.
> This can cause duplicate values as well, however replacing the same exact value is fine and if a non ST write occurs the state is ignored anyways.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
8 years, 8 months
[JBoss JIRA] (ISPN-5510) Provide better Hot Rod client socket timeout and retry defaults
by Pedro Ruivo (JIRA)
[ https://issues.jboss.org/browse/ISPN-5510?page=com.atlassian.jira.plugin.... ]
Pedro Ruivo updated ISPN-5510:
------------------------------
Fix Version/s: 8.1.0.Final
(was: 8.0.0.Final)
> Provide better Hot Rod client socket timeout and retry defaults
> ---------------------------------------------------------------
>
> Key: ISPN-5510
> URL: https://issues.jboss.org/browse/ISPN-5510
> Project: Infinispan
> Issue Type: Enhancement
> Reporter: Galder Zamarreño
> Assignee: Galder Zamarreño
> Fix For: 8.1.0.Final
>
>
> The current defaults are:
> * Socket timeout = 60 seconds
> * Max retries = 10
> As a result of these defaults, if the server hangs an operation, it'd take 10 minutes (60 second timeout x 10 retries) for the operation to finally return an exception to the client, which is way too much.
> So, these default value should change to be more aggressive: 30 second socket timeout and 3 max retries.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
8 years, 8 months
[JBoss JIRA] (ISPN-5623) Retried prepare commands do not wait for backup locks
by Pedro Ruivo (JIRA)
[ https://issues.jboss.org/browse/ISPN-5623?page=com.atlassian.jira.plugin.... ]
Pedro Ruivo updated ISPN-5623:
------------------------------
Fix Version/s: 8.1.0.Final
(was: 8.0.0.Final)
> Retried prepare commands do not wait for backup locks
> -----------------------------------------------------
>
> Key: ISPN-5623
> URL: https://issues.jboss.org/browse/ISPN-5623
> Project: Infinispan
> Issue Type: Bug
> Components: Core
> Affects Versions: 7.2.3.Final, 8.0.0.Beta1
> Reporter: Dan Berindei
> Assignee: Pedro Ruivo
> Fix For: 8.1.0.Final
>
> Attachments: OptimisticPrimaryOwnerCrashDuringPrepareTest.java
>
>
> When the primary owner crashes during prepare, the prepare command is retried on the new primary owner. But because the transaction topology id stays the same, the new primary owner does not check for backup locks owned by other transactions from the previous topology.
> That makes it possible for the retried prepare command to lock a key that was already locked by another transaction on the primary owner.
> This wasn't a problem before the ISPN-4546 fix, because a primary owner crash would have forced the transaction to roll back.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
8 years, 8 months
[JBoss JIRA] (ISPN-5614) Write performance regression after ISPN-5484
by Pedro Ruivo (JIRA)
[ https://issues.jboss.org/browse/ISPN-5614?page=com.atlassian.jira.plugin.... ]
Pedro Ruivo updated ISPN-5614:
------------------------------
Fix Version/s: 8.1.0.Final
(was: 8.0.0.Final)
> Write performance regression after ISPN-5484
> --------------------------------------------
>
> Key: ISPN-5614
> URL: https://issues.jboss.org/browse/ISPN-5614
> Project: Infinispan
> Issue Type: Bug
> Components: Core
> Affects Versions: 8.0.0.Beta1
> Reporter: Dan Berindei
> Assignee: Dan Berindei
> Fix For: 8.1.0.Final
>
>
> Regression test shows a significant drop in throughput in the replicated and distributed write tests.
> This was after adjusting the internal thread pool settings in the JGroups configuration: with the default (min=5, max=20, queue=0), the distributed read test would fail to finish.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
8 years, 8 months