[JBoss JIRA] (ISPN-8897) Indexes from different caches are mixed-up in server mode
by Adrian Nistor (JIRA)
[ https://issues.jboss.org/browse/ISPN-8897?page=com.atlassian.jira.plugin.... ]
Adrian Nistor updated ISPN-8897:
--------------------------------
Fix Version/s: (was: 8.2.9.Final)
> Indexes from different caches are mixed-up in server mode
> ---------------------------------------------------------
>
> Key: ISPN-8897
> URL: https://issues.jboss.org/browse/ISPN-8897
> Project: Infinispan
> Issue Type: Bug
> Components: Remote Querying
> Affects Versions: 9.1.6.Final, 9.2.0.Final
> Reporter: Gustavo Fernandes
> Assignee: Adrian Nistor
> Fix For: 9.2.1.Final, 9.1.7.Final
>
>
> When configuring multiple caches as indexed in the server, they'll effectively use the same index at runtime since Infinispan internally uses a single entity {{ProtobufValueWrapper}} to store the values in the index.
> This can cause issues with two caches having the same key stepping in each other's toes. Example, when doing:
> {code:java}
> remoteCache1.put(1, entity1)
> remoteCache2.put(1, entity2)
> {code}
> The second call will replace the first indexed entry since the key is identical and the internal index named "org.infinispan.query.remote.impl.indexing.ProtobufValueWrapper" is shared.
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)
6 years, 1 month
[JBoss JIRA] (ISPN-8897) Indexes from different caches are mixed-up in server mode
by Adrian Nistor (JIRA)
[ https://issues.jboss.org/browse/ISPN-8897?page=com.atlassian.jira.plugin.... ]
Adrian Nistor updated ISPN-8897:
--------------------------------
Git Pull Request: https://github.com/infinispan/infinispan/pull/5805, https://github.com/infinispan/infinispan/pull/5810 (was: https://github.com/infinispan/infinispan/pull/5805, https://github.com/infinispan/infinispan/pull/5810, https://github.com/infinispan/infinispan/pull/5811)
> Indexes from different caches are mixed-up in server mode
> ---------------------------------------------------------
>
> Key: ISPN-8897
> URL: https://issues.jboss.org/browse/ISPN-8897
> Project: Infinispan
> Issue Type: Bug
> Components: Remote Querying
> Affects Versions: 9.1.6.Final, 9.2.0.Final
> Reporter: Gustavo Fernandes
> Assignee: Adrian Nistor
> Fix For: 8.2.9.Final, 9.2.1.Final, 9.1.7.Final
>
>
> When configuring multiple caches as indexed in the server, they'll effectively use the same index at runtime since Infinispan internally uses a single entity {{ProtobufValueWrapper}} to store the values in the index.
> This can cause issues with two caches having the same key stepping in each other's toes. Example, when doing:
> {code:java}
> remoteCache1.put(1, entity1)
> remoteCache2.put(1, entity2)
> {code}
> The second call will replace the first indexed entry since the key is identical and the internal index named "org.infinispan.query.remote.impl.indexing.ProtobufValueWrapper" is shared.
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)
6 years, 1 month
[JBoss JIRA] (ISPN-8900) Sockets might be left open if server unreachable on creation
by Tristan Tarrant (JIRA)
[ https://issues.jboss.org/browse/ISPN-8900?page=com.atlassian.jira.plugin.... ]
Tristan Tarrant updated ISPN-8900:
----------------------------------
Status: Resolved (was: Pull Request Sent)
Resolution: Done
> Sockets might be left open if server unreachable on creation
> ------------------------------------------------------------
>
> Key: ISPN-8900
> URL: https://issues.jboss.org/browse/ISPN-8900
> Project: Infinispan
> Issue Type: Bug
> Components: Remote Protocols
> Affects Versions: 9.1.6.Final
> Reporter: Galder Zamarreño
> Assignee: Galder Zamarreño
> Fix For: 9.1.7.Final
>
>
> Seems to affect pre-netty client but not sure yet whether it affects netty version:
> In 9.1.x and before, TcpTransport.destroy/release not called if TcpTransport constructor fails to connect socket. This might be leaving the socket file descriptor open. If an exception, socket and socketChannel should be closed.
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)
6 years, 1 month
[JBoss JIRA] (ISPN-8809) Create test for state transfer turned off
by Vojtech Juranek (JIRA)
[ https://issues.jboss.org/browse/ISPN-8809?page=com.atlassian.jira.plugin.... ]
Vojtech Juranek commented on ISPN-8809:
---------------------------------------
It seems that when the state transfer is turned off, newly connected server sees only part of the entries - tracked as ISPN-8908
> Create test for state transfer turned off
> -----------------------------------------
>
> Key: ISPN-8809
> URL: https://issues.jboss.org/browse/ISPN-8809
> Project: Infinispan
> Issue Type: Task
> Components: Test Suite - Server
> Affects Versions: 9.2.0.CR1
> Reporter: Vojtech Juranek
> Assignee: Vojtech Juranek
>
> Create IT which would tests that entries are correctly loaded from appropriate nodes when state transfer on the cache is turned off.
> Also check {{StateTransferSuppressIT}} which seems that contrary to its name has state transfer turned on.
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)
6 years, 1 month
[JBoss JIRA] (ISPN-8908) Remote cache fails to get entries when state transfer is turned off
by Vojtech Juranek (JIRA)
Vojtech Juranek created ISPN-8908:
-------------------------------------
Summary: Remote cache fails to get entries when state transfer is turned off
Key: ISPN-8908
URL: https://issues.jboss.org/browse/ISPN-8908
Project: Infinispan
Issue Type: Bug
Components: State Transfer
Affects Versions: 9.2.0.Final
Reporter: Vojtech Juranek
When state transfer is turned off, remote cache fails to obtain cache entries from HR server (in client-server mode) which connects to the cluster. Only about half of the entries can be seen by HR client on newly connected server, both in replicated cache and distributed cache.
Investigate it in embedded mode is TBD.
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)
6 years, 1 month