[
https://issues.apache.org/jira/browse/OAK-3477?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Chetan Mehrotra updated OAK-3477:
---------------------------------
Attachment: OAK-3477.patch
[patch|^OAK-3477.patch] which just removes the {{id}} counter from ObjectName.
Checked the code around usage of JMX Bean and found that in all cases we are
creating unique names anyway. Couple of places where multiple instance of same
MBean interface are registered are
# SessionStatistics - The name is derived from sessionId which ensure name is
unique
# MBeans for ObservationListener - They are also having a unique {{listnerId}}
field in the name
# CacheStats - Each cache stat has a unique name
With this change running an Oak based application did not gave any duplicate
name exception
[~mduerig] Can you review the proposed change
> Make JMX Bean names predictable so they can be used in configurations.
> ----------------------------------------------------------------------
>
> Key: OAK-3477
> URL: https://issues.apache.org/jira/browse/OAK-3477
> Project: Jackrabbit Oak
> Issue Type: Improvement
> Components: core
> Affects Versions: 1.3.7
> Reporter: Ian Boston
> Assignee: Chetan Mehrotra
> Fix For: 1.3.11
>
> Attachments: OAK-3477.patch
>
>
> In Oak there are JMX beans which get names of the form [1], which can be hard
> to use in monitoring tool configuration that dont support regexes as the ID
> will change. Where possible, the ID should be removed or made static. The
> name and type should be sufficient in most cases in Oak to make the JMX
> Object name unique.
> 1
> {code}
> org.apache.jackrabbit.oak:name="Consolidated Cache
> statistics",type="ConsolidatedCacheStats",id=6
> org.apache.jackrabbit.oak:name="CopyOnRead support
> statistics",type="CopyOnReadStats",id=7
> org.apache.jackrabbit.oak:name="Oak Repository
> Statistics",type="RepositoryStats",id=16
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
