[JBoss JIRA] (WFLY-13556) Memory leak in FlashScope - expired elements are not cleared
by Brian Stansberry (Jira)
[ https://issues.redhat.com/browse/WFLY-13556?page=com.atlassian.jira.plugi... ]
Brian Stansberry resolved WFLY-13556.
-------------------------------------
Fix Version/s: 20.0.0.Final
Resolution: Done
> Memory leak in FlashScope - expired elements are not cleared
> ------------------------------------------------------------
>
> Key: WFLY-13556
> URL: https://issues.redhat.com/browse/WFLY-13556
> Project: WildFly
> Issue Type: Bug
> Components: JSF
> Reporter: Farah Juma
> Assignee: Bartosz Spyrko-Smietanko
> Priority: Major
> Fix For: 20.0.0.Final
>
>
> Objects in FlashScope is not cleared and keep accumulating even if POST-REDIRECT-BACK is completed.
> If we put a huge object to FlashScope continuously as the following code, Java heap space will be exhausted soon.
> {code:java}
> @Model
> public class Action {
> private static final int oneHundredMB = 1024 * 1024 * 100;
> public String next() {
> Flash flash = FacesContext.getCurrentInstance().getExternalContext().getFlash();
> flash.putNow("huge", new byte[oneHundredMB]);
> return "index.xhtml?faces-redirect=true";
> }
> }
> {code}
> When OutOfMemoryError is happened, flashInnerMap field in com.sun.faces.context.flash.ELFlash is bloated.
> {noformat}
> Class Name | Ref. Objects | Shallow Heap | Ref. Shallow Heap | Retained Heap
> -----------------------------------------------------------------------------------------------------------------------------------------
> java.lang.Thread @ 0xeea24380 default task-1 Thread | 1 | 120 | 64 | 20,800
> '- <Java Local> com.sun.faces.context.flash.ELFlash @ 0xee9e4608 | 1 | 40 | 64 | 64
> '- flashInnerMap java.util.concurrent.ConcurrentHashMap @ 0xee9e4630| 1 | 64 | 64 | 314,574,048
> -----------------------------------------------------------------------------------------------------------------------------------------
> {noformat}
--
This message was sent by Atlassian Jira
(v7.13.8#713008)
4 years, 7 months
[JBoss JIRA] (WFLY-13542) Container does not enforce bean-managed concurrency for timeout methods
by Cheng Fang (Jira)
[ https://issues.redhat.com/browse/WFLY-13542?page=com.atlassian.jira.plugi... ]
Cheng Fang resolved WFLY-13542.
-------------------------------
Resolution: Explained
Mark as resolved. Feel free to reopen it if you think there is more work to be done.
> Container does not enforce bean-managed concurrency for timeout methods
> -----------------------------------------------------------------------
>
> Key: WFLY-13542
> URL: https://issues.redhat.com/browse/WFLY-13542
> Project: WildFly
> Issue Type: Bug
> Components: EJB
> Affects Versions: 19.1.0.Final
> Reporter: Fábio Silva
> Assignee: Cheng Fang
> Priority: Major
>
> Bean-managed concurrency for @Singleton EJBs is not enforced for timeout methods annotated with @Schedule. When a timeout is fired while the previous one is still running, the timeout method is not called. Instead, WildFly logs a warning like the following:
> {code:java}
> 20:58:54,000 WARN [org.jboss.as.ejb3.timer] (EJB default - 2) WFLYEJB0043: A previous execution of timer [id=aa716284-ef7b-4bb3-a356-a8ca187255c4 timedObjectId=myweb.myweb.MySingleton auto-timer?:true persistent?:false timerService=org.jboss.as.ejb3.timerservice.TimerServiceImpl@303bd2d5 previousRun=Sun May 31 20:57:10 BRT 2020 initialExpiration=null intervalDuration(in milli sec)=0 nextExpiration=Sun May 31 20:58:54 BRT 2020 timerState=IN_TIMEOUT info=null] ScheduleExpression [second=0/1;minute=*;hour=*;dayOfMonth=*;month=*;dayOfWeek=*;year=*;timezoneID=null;start=null;end=null] is still in progress, skipping this overlapping scheduled execution at: Sun May 31 20:58:54 BRT 2020.
> {code}
--
This message was sent by Atlassian Jira
(v7.13.8#713008)
4 years, 7 months
[JBoss JIRA] (WFLY-13063) Change default values of id-cache-size or confirmation-window-size for cluster connections
by Radoslav Husar (Jira)
[ https://issues.redhat.com/browse/WFLY-13063?page=com.atlassian.jira.plugi... ]
Radoslav Husar updated WFLY-13063:
----------------------------------
Component/s: JMS
(was: Clustering)
Fixing component. This is not a clustering component issue, but JMS component issue.
[~ehugonnet] FYI.
> Change default values of id-cache-size or confirmation-window-size for cluster connections
> ------------------------------------------------------------------------------------------
>
> Key: WFLY-13063
> URL: https://issues.redhat.com/browse/WFLY-13063
> Project: WildFly
> Issue Type: Bug
> Components: JMS
> Affects Versions: 19.0.0.Beta2
> Reporter: Simon Priadka
> Assignee: Michal Petrov
> Priority: Major
>
> When establishing a new messaging cluster, default values of *id-cache-size* and *confirmation-window-size* are producing a *WARN* with following message:
> {code}
> 2020-02-04 21:35:20,228 WARN [org.apache.activemq.artemis.core.server] (Thread-2 (ActiveMQ-client-global-threads)) AMQ224078: The size of duplicate cache detection (<id_cache-size/>) appears to be too large 20,000. It should be no greater than the number of messages that can be squeezed into confirmation window buffer (<confirmation-window-size/>) 1,048,576.
> {code}
> According to the [method|https://github.com/apache/activemq-artemis/blob/3743bc9d9f39b0731f...] validating these values, this variable combination is "invalid"
--
This message was sent by Atlassian Jira
(v7.13.8#713008)
4 years, 7 months