[JBoss JIRA] (ISPN-5351) read()ing protected Attributes should support user-supplied copying methods
by Tristan Tarrant (JIRA)
[ https://issues.jboss.org/browse/ISPN-5351?page=com.atlassian.jira.plugin.... ]
Tristan Tarrant updated ISPN-5351:
----------------------------------
Status: Open (was: New)
> read()ing protected Attributes should support user-supplied copying methods
> ---------------------------------------------------------------------------
>
> Key: ISPN-5351
> URL: https://issues.jboss.org/browse/ISPN-5351
> Project: Infinispan
> Issue Type: Bug
> Reporter: Tristan Tarrant
>
> Cloning a configuration attribute sometimes is not enough. For example when dealing with collections and immutable types we need to overcome the following:
> Caused by: java.lang.UnsupportedOperationException
> at org.infinispan.commons.util.Immutables$ImmutableTypedProperties.put(Immutables.java:622)
> at org.infinispan.configuration.cache.IndexingConfigurationBuilder.setProperty(IndexingConfigurationBuilder.java:130)
> at org.infinispan.configuration.cache.IndexingConfigurationBuilder.addProperty(IndexingConfigurationBuilder.java:107)
>
--
This message was sent by Atlassian JIRA
(v6.3.11#6341)
10 years, 6 months
[JBoss JIRA] (ISPN-5159) Make concurrent startup smooth
by Dan Berindei (JIRA)
[ https://issues.jboss.org/browse/ISPN-5159?page=com.atlassian.jira.plugin.... ]
Dan Berindei commented on ISPN-5159:
------------------------------------
[~rvansa] I wasn't thinking about requiring changes to the JGroups configuration either, my thinking was that the default configuration should be good enough to avoid merges during startup.
I don't think all the internal caches can be recreated on the fly: e.g. the cluster registry cache is created by core and then used by other components like query, we'd need both a way to notify the cluster registry to rebuild the cache, and a way for the cluster registry to tell all the components that they need to rebuild their slice of the cluster registry cache (assuming that those components do have the necessary information).
> Make concurrent startup smooth
> ------------------------------
>
> Key: ISPN-5159
> URL: https://issues.jboss.org/browse/ISPN-5159
> Project: Infinispan
> Issue Type: Enhancement
> Components: Core
> Affects Versions: 7.1.0.Beta1
> Reporter: Radim Vansa
> Assignee: Dan Berindei
>
> When starting many instances in parallel, it often happens that the node does not detect its neighborhood very well and this results in many subclusters, merging views etc.
> Merging two available partitions has undefined results (AFAIK). While we can expect that there are no requests to the cluster from the application ^1^, Infinispan itself uses some caches to store internal information (HotRod routing, Protobuf etc...). It would be better if the available-available merge would provide hooks for rebuilding this info.
> ^1^) Being able to start the cluster with reads/writes disabled and enable them only when the cache has expected number of members would be convenient, too.
--
This message was sent by Atlassian JIRA
(v6.3.11#6341)
10 years, 6 months
[JBoss JIRA] (ISPN-5159) Make concurrent startup smooth
by Dan Berindei (JIRA)
[ https://issues.jboss.org/browse/ISPN-5159?page=com.atlassian.jira.plugin.... ]
Dan Berindei commented on ISPN-5159:
------------------------------------
One more question... I was expecting merges during startup to be much less of a problem since JGRP-1899 was fixed. [~rvansa] [~mcimbora] are you still seeing merges during startup with JGroups 3.6.1+?
> Make concurrent startup smooth
> ------------------------------
>
> Key: ISPN-5159
> URL: https://issues.jboss.org/browse/ISPN-5159
> Project: Infinispan
> Issue Type: Enhancement
> Components: Core
> Affects Versions: 7.1.0.Beta1
> Reporter: Radim Vansa
> Assignee: Dan Berindei
>
> When starting many instances in parallel, it often happens that the node does not detect its neighborhood very well and this results in many subclusters, merging views etc.
> Merging two available partitions has undefined results (AFAIK). While we can expect that there are no requests to the cluster from the application ^1^, Infinispan itself uses some caches to store internal information (HotRod routing, Protobuf etc...). It would be better if the available-available merge would provide hooks for rebuilding this info.
> ^1^) Being able to start the cluster with reads/writes disabled and enable them only when the cache has expected number of members would be convenient, too.
--
This message was sent by Atlassian JIRA
(v6.3.11#6341)
10 years, 6 months