Bug 701375
Summary: | [JBossAS4] JVM metrics stop being collected after a restart of EAP server | ||||||
---|---|---|---|---|---|---|---|
Product: | [Other] RHQ Project | Reporter: | Larry O'Leary <loleary> | ||||
Component: | Plugins | Assignee: | Ian Springer <ian.springer> | ||||
Status: | CLOSED CURRENTRELEASE | QA Contact: | Mike Foley <mfoley> | ||||
Severity: | high | Docs Contact: | |||||
Priority: | urgent | ||||||
Version: | 4.2 | CC: | ccrouch, hrupp, ian.springer, rsoares, skondkar | ||||
Target Milestone: | --- | ||||||
Target Release: | JON 3.0.0, RHQ 4.3.0 | ||||||
Hardware: | All | ||||||
OS: | All | ||||||
Whiteboard: | |||||||
Fixed In Version: | 4.3 | Doc Type: | Bug Fix | ||||
Doc Text: | Story Points: | --- | |||||
Clone Of: | Environment: |
JON 2.4.1
EAP 4.3 CP06
EAP started using the following system properties:
-Djboss.platform.mbeanserver
-Dcom.sun.management.jmxremote
-Djavax.management.builder.initial=org.jboss.system.server.jmx.MBeanServerBuilderImpl
|
|||||
Last Closed: | 2012-02-07 19:23:13 UTC | Type: | --- | ||||
Regression: | --- | Mount Type: | --- | ||||
Documentation: | --- | CRM: | |||||
Verified Versions: | Category: | --- | |||||
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
Cloudforms Team: | --- | Target Upstream Version: | |||||
Embargoed: | |||||||
Bug Depends On: | |||||||
Bug Blocks: | 745494 | ||||||
Attachments: |
|
Description
Larry O'Leary
2011-05-02 18:00:40 UTC
I thought we had done work to improve the connection pooling/dropping for EMS This was in a bug in EMS. EMS caches the MBeanInfos for MBeans it manages. When EMS re-established the remote connection to an MBeanServer that was restarted, it was not clearing the now-defunct cached MBeanInfos. When it later tried to make calls on the cached MBeanInfos, they would fail with InstanceNotFoundExceptions, e.g.: Caused by: javax.management.InstanceNotFoundException: java.lang:type=MemoryPool,name=PS Eden Space is not registered. or other exceptions due to the MBeanInfo being defunct. I have fixed the issue in EMS 1.3, which clears all cached MBeanInfos after a reconnect. The attached patch upgrades RHQ to EMS 1.3. I will apply it once RHQ 4.2 has been released and master is open for business. Created attachment 530751 [details]
patch to upgrade RHQ to EMS 1.3
For the record, I reproduced this bug using EAP 4.3.0.GA_CP09 on JDK 6.0u29 and RHQ 4.2 on JDK 6.0u29, and verified my fix on the same. [master 9c437b0] (http://git.fedorahosted.org/git/?p=rhq/rhq.git;a=commitdiff;h=9c437b0) upgrades EMS from 1.2.16 to 1.3 in the root pom, which should fix this issue. Fix cherry-picked to release_jon3.x branch - commit dc4a80e. Verified on build#100 (Version: 4.2.0.JON300-SNAPSHOT Build Number: 5d0d95e) in release_jon3.x branch. Verified on JBoss EAP 4.3.CP06 on JDK 6u18. Followed the steps and navigated to JBossAS Server -> EAP suniltest.usersys.redhat.com:1099 00024643 -> JVM -> Memory Subsystem -> Memory Pool -> PS Eden Space >> Monitor>> Tables Verified that the JVM metrics are collected after restart of EAP server. changing status of VERIFIED BZs for JON 2.4.2 and JON 3.0 to CLOSED/CURRENTRELEASE |