[JBoss JIRA] (ISPN-7246) Memory leak with LIRS eviction strategy
by ted won (JIRA)
[ https://issues.jboss.org/browse/ISPN-7246?page=com.atlassian.jira.plugin.... ]
ted won updated ISPN-7246:
--------------------------
Description:
There is a memory leak in org.infinispan.commons.util.concurrent.jdk8backported.BoundedEquivalentConcurrentHashMapV8 with LIRS eviction strategy.
There are cache entries in org.infinispan.container.DefaultDataContainer.entries and its size are bigger than cache size after eviction occurred.
DefaultDataContainer.entries retains keys and values for evicted entries in the cache.
After full GC, DefaultDataContainer.entries size wasn't changed.
The following minimum reproducer exhibits a leak. Eviction is supposed to suppress OOME.
was:
There is a memory leak in org.infinispan.commons.util.concurrent.jdk8backported.BoundedEquivalentConcurrentHashMapV8 with LIRS eviction strategy.
There are cache entries in org.infinispan.container.DefaultDataContainer.entries and its size are bigger than cache size after eviction occurred.
The following minimum reproducer exhibits a leak. Eviction is supposed to suppress OOME.
> Memory leak with LIRS eviction strategy
> ---------------------------------------
>
> Key: ISPN-7246
> URL: https://issues.jboss.org/browse/ISPN-7246
> Project: Infinispan
> Issue Type: Bug
> Components: Core
> Affects Versions: 8.2.5.Final
> Environment: * Infinispan 8.2.5.Final Library mode
> * LIRS eviction strategy
> Reporter: ted won
> Labels: core, eviction, leak, lirs, strategy
> Attachments: reproducer.zip
>
>
> There is a memory leak in org.infinispan.commons.util.concurrent.jdk8backported.BoundedEquivalentConcurrentHashMapV8 with LIRS eviction strategy.
> There are cache entries in org.infinispan.container.DefaultDataContainer.entries and its size are bigger than cache size after eviction occurred.
> DefaultDataContainer.entries retains keys and values for evicted entries in the cache.
> After full GC, DefaultDataContainer.entries size wasn't changed.
> The following minimum reproducer exhibits a leak. Eviction is supposed to suppress OOME.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 1 month
[JBoss JIRA] (ISPN-7246) Memory leak with LIRS eviction strategy
by ted won (JIRA)
[ https://issues.jboss.org/browse/ISPN-7246?page=com.atlassian.jira.plugin.... ]
ted won updated ISPN-7246:
--------------------------
Labels: core eviction leak lirs strategy (was: core eviction l leak)
> Memory leak with LIRS eviction strategy
> ---------------------------------------
>
> Key: ISPN-7246
> URL: https://issues.jboss.org/browse/ISPN-7246
> Project: Infinispan
> Issue Type: Bug
> Components: Core
> Affects Versions: 8.2.5.Final
> Environment: * Infinispan 8.2.5.Final Library mode
> * LIRS eviction strategy
> Reporter: ted won
> Labels: core, eviction, leak, lirs, strategy
> Attachments: reproducer.zip
>
>
> There is a memory leak in org.infinispan.commons.util.concurrent.jdk8backported.BoundedEquivalentConcurrentHashMapV8 with LIRS eviction strategy.
> There are cache entries in org.infinispan.container.DefaultDataContainer.entries and its size are bigger than cache size after eviction occurred.
> The following minimum reproducer exhibits a leak. Eviction is supposed to suppress OOME.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 1 month
[JBoss JIRA] (ISPN-7246) Memory leak with LIRS eviction strategy
by ted won (JIRA)
ted won created ISPN-7246:
-----------------------------
Summary: Memory leak with LIRS eviction strategy
Key: ISPN-7246
URL: https://issues.jboss.org/browse/ISPN-7246
Project: Infinispan
Issue Type: Bug
Components: Core
Affects Versions: 8.2.5.Final
Environment: * Infinispan 8.2.5.Final Library mode
* LIRS eviction strategy
Reporter: ted won
Attachments: reproducer.zip
There is a memory leak in org.infinispan.commons.util.concurrent.jdk8backported.BoundedEquivalentConcurrentHashMapV8 with LIRS eviction strategy.
There are cache entries in org.infinispan.container.DefaultDataContainer.entries and its size are bigger than cache size after eviction occurred.
The following minimum reproducer exhibits a leak. Eviction is supposed to suppress OOME.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 1 month
[JBoss JIRA] (ISPN-7246) Memory leak with LIRS eviction strategy
by ted won (JIRA)
[ https://issues.jboss.org/browse/ISPN-7246?page=com.atlassian.jira.plugin.... ]
ted won updated ISPN-7246:
--------------------------
Labels: core eviction l leak (was: )
> Memory leak with LIRS eviction strategy
> ---------------------------------------
>
> Key: ISPN-7246
> URL: https://issues.jboss.org/browse/ISPN-7246
> Project: Infinispan
> Issue Type: Bug
> Components: Core
> Affects Versions: 8.2.5.Final
> Environment: * Infinispan 8.2.5.Final Library mode
> * LIRS eviction strategy
> Reporter: ted won
> Labels: core, eviction, l, leak
> Attachments: reproducer.zip
>
>
> There is a memory leak in org.infinispan.commons.util.concurrent.jdk8backported.BoundedEquivalentConcurrentHashMapV8 with LIRS eviction strategy.
> There are cache entries in org.infinispan.container.DefaultDataContainer.entries and its size are bigger than cache size after eviction occurred.
> The following minimum reproducer exhibits a leak. Eviction is supposed to suppress OOME.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 1 month
[JBoss JIRA] (ISPN-6776) Administration console - push state transfer and cancel state transfer buttons don't work properly
by Vladimir Blagojevic (JIRA)
[ https://issues.jboss.org/browse/ISPN-6776?page=com.atlassian.jira.plugin.... ]
Vladimir Blagojevic commented on ISPN-6776:
-------------------------------------------
Failing on server. Reassigning to Tristan
> Administration console - push state transfer and cancel state transfer buttons don't work properly
> --------------------------------------------------------------------------------------------------
>
> Key: ISPN-6776
> URL: https://issues.jboss.org/browse/ISPN-6776
> Project: Infinispan
> Issue Type: Bug
> Components: JMX, reporting and management
> Reporter: Roman Macor
> Assignee: Vladimir Blagojevic
> Attachments: domain.xml, host.xml
>
>
> Click on cache container -> Actions -> Manage backup sites -> Push state (on BRN site which is configured)
> result: nothing happens, server log:
> [Server:server-one] 15:15:03,752 ERROR [org.jgroups.protocols.relay.RELAY2] (ServerService Thread Pool -- 36) master:server-one: no route to BRN: dropping message
> expected result: Success! The operation has been successfully executed.
> Please note that if we take the site offline and then online again in the "Manage backup sites" dialog, push state button works correctly.
> Cancel state button is behaving the same, except taking site offline and then online again doesn't help.
> I'm attaching server configuration
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 1 month
[JBoss JIRA] (ISPN-6776) Administration console - push state transfer and cancel state transfer buttons don't work properly
by Vladimir Blagojevic (JIRA)
[ https://issues.jboss.org/browse/ISPN-6776?page=com.atlassian.jira.plugin.... ]
Vladimir Blagojevic edited comment on ISPN-6776 at 11/28/16 11:07 AM:
----------------------------------------------------------------------
Failing on server not in console UI. Reassigning to Tristan
was (Author: vblagojevic):
Failing on server. Reassigning to Tristan
> Administration console - push state transfer and cancel state transfer buttons don't work properly
> --------------------------------------------------------------------------------------------------
>
> Key: ISPN-6776
> URL: https://issues.jboss.org/browse/ISPN-6776
> Project: Infinispan
> Issue Type: Bug
> Components: JMX, reporting and management
> Reporter: Roman Macor
> Assignee: Vladimir Blagojevic
> Attachments: domain.xml, host.xml
>
>
> Click on cache container -> Actions -> Manage backup sites -> Push state (on BRN site which is configured)
> result: nothing happens, server log:
> [Server:server-one] 15:15:03,752 ERROR [org.jgroups.protocols.relay.RELAY2] (ServerService Thread Pool -- 36) master:server-one: no route to BRN: dropping message
> expected result: Success! The operation has been successfully executed.
> Please note that if we take the site offline and then online again in the "Manage backup sites" dialog, push state button works correctly.
> Cancel state button is behaving the same, except taking site offline and then online again doesn't help.
> I'm attaching server configuration
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 1 month
[JBoss JIRA] (ISPN-6776) Administration console - push state transfer and cancel state transfer buttons don't work properly
by Vladimir Blagojevic (JIRA)
[ https://issues.jboss.org/browse/ISPN-6776?page=com.atlassian.jira.plugin.... ]
Vladimir Blagojevic reassigned ISPN-6776:
-----------------------------------------
Assignee: Tristan Tarrant (was: Vladimir Blagojevic)
> Administration console - push state transfer and cancel state transfer buttons don't work properly
> --------------------------------------------------------------------------------------------------
>
> Key: ISPN-6776
> URL: https://issues.jboss.org/browse/ISPN-6776
> Project: Infinispan
> Issue Type: Bug
> Components: JMX, reporting and management
> Reporter: Roman Macor
> Assignee: Tristan Tarrant
> Attachments: domain.xml, host.xml
>
>
> Click on cache container -> Actions -> Manage backup sites -> Push state (on BRN site which is configured)
> result: nothing happens, server log:
> [Server:server-one] 15:15:03,752 ERROR [org.jgroups.protocols.relay.RELAY2] (ServerService Thread Pool -- 36) master:server-one: no route to BRN: dropping message
> expected result: Success! The operation has been successfully executed.
> Please note that if we take the site offline and then online again in the "Manage backup sites" dialog, push state button works correctly.
> Cancel state button is behaving the same, except taking site offline and then online again doesn't help.
> I'm attaching server configuration
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 1 month
[JBoss JIRA] (ISPN-7219) Administration console - creating template from existing template fails in some cases
by Vladimir Blagojevic (JIRA)
[ https://issues.jboss.org/browse/ISPN-7219?page=com.atlassian.jira.plugin.... ]
Vladimir Blagojevic reassigned ISPN-7219:
-----------------------------------------
Assignee: Ryan Emerson (was: Vladimir Blagojevic)
> Administration console - creating template from existing template fails in some cases
> -------------------------------------------------------------------------------------
>
> Key: ISPN-7219
> URL: https://issues.jboss.org/browse/ISPN-7219
> Project: Infinispan
> Issue Type: Bug
> Components: JMX, reporting and management
> Affects Versions: 9.0.0.Alpha4
> Reporter: Roman Macor
> Assignee: Ryan Emerson
>
> Click on cache container -> Configuration -> Templates -> create new template -> fill in template name: newTemplate, base configuration: persistent-file-store-write-behind (template with file store and write-behind configured) -> next -> create
> Result:
> Pop up with error message:
> {"WFLYCTL0062: Composite operation failed and was rolled back. Steps that failed:":{"Operation step-2":"WFLYCTL0216: Management resource '[\n (\"profile\" => \"clustered\"),\n (\"subsystem\" => \"datagrid-infinispan\"),\n (\"cache-container\" => \"clustered\"),\n (\"configurations\" => \"CONFIGURATIONS\"),\n (\"distributed-cache-configuration\" => \"new\"),\n (\"file-store\" => \"FILE_STORE\"),\n (\"write-behind\" => \"WRITE_BEHIND\")\n]' not found"}}
> The new template is created, but configuration is not copied.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 1 month