I don't see that OR clause on Geode develop. What version of Geode are you
using?
In any event, the JMX MBeans are registered using a ResourceEventsListener
called ManagementListener. That listener is invoked by handleResourceEvent
calls in GemFireCacheImpl for a bunch of different events including region
creation. There isn't any other property that could be set to create the
MBean for that region.
Rather than test the region name in that OR clause, a slightly better way
to implement that would be:
if (!region.isInternalRegion()
|| region.isUsedForParallelGatewaySenderQueue()
|| region.isUsedForSerialGatewaySenderQueue()) {
btw - the MBean should be removed if the region is destroyed. The
LocalRegion recursiveDestroyRegion and basicDestroyRegion methods contain
code like this that should also be modified like above:
if (!region.isInternalRegion()) {
InternalDistributedSystem system =
region.cache.getInternalDistributedSystem();
system.handleResourceEvent(ResourceEvent.REGION_REMOVE, region);
}
Thanks,
Barry Oglesby
On Tue, Jun 20, 2017 at 6:33 AM, Dinesh Akhand <dines...@amdocs.com> wrote:
> Hi Barry,
>
>
>
> Looks like things are already implemented in geode:
>
>
>
> Class : GemFireCacheImpl
>
>
>
> if (!rgn.isInternalRegion() ||
> rgn.getFullPath().contains("GATEWAY_SENDER_QUEUE"))
> {
>
> system.handleResourceEvent(ResourceEvent.REGION_CREATE, rgn);
>
> }
>
>
>
> Above code solved my problem . Async Queue Partition region visible in
> JMX.
>
>
>
> But above code looks like hack in existing code. Is this fix can we take
> and go ahead because in our requirement we need to show all internal
> partition region.
>
>
>
> Or can we define some property in .properties files to enable stats in jmx
> for internal region?.
>
>
>
> Thanks,
>
> Dinesh Akhand
>
>
>
>
>
> -----Original Message-----
> From: Barry Oglesby [mailto:bogle...@pivotal.io]
> Sent: Saturday, June 17, 2017 2:14 AM
> To: dev@geode.apache.org
> Subject: Re: Query for JMX "Async Event Queue properties"
>
>
>
> The AsynceEventQueueMBean is monitoring the AsyncEventQueueStats not the
> PartitionedRegionStats. And, actually, the only stat being monitored by
> that MBean is the eventQueueSize stat. The parallel AsyncEventQueue creates
> a colocated PartitionedRegion under the covers, and the stats for that PR
> are the PartitionedRegionStats you referenced. There isn't currently any
> JMX support for those PartitionedRegionStats.
>
>
>
> Are you looking to add support for those PartitionedRegionStats to the
> AsynceEventQueueMBean? If so, I guess there are a few ways to do. The first
> would be to find the underlying PartitionedRegion for the parallel
> AsyncEventQueue, add its PartitionedRegionStats to the
> AsyncEventQueueMBeanBridge's monitor and reference the appropriate one(s)
> (in your case dataStoreBytesInUse). Another would be to actually create an
> MBean for that internal PartitionedRegion.
>
>
>
> Both of these would only be applicable for a parallel AsyncEventQueue.
>
> Serial AsyncEventQueues don't create a PartitionedRegion.
>
>
>
>
>
> Thanks,
>
> Barry Oglesby
>
>
>
>
>
> On Fri, Jun 16, 2017 at 5:38 AM, Dinesh Akhand <dines...@amdocs.com
> <mailto:dines...@amdocs.com>> wrote:
>
>
>
> > Hi Team
>
> >
>
> > Regarding the following metrics:
>
> >
>
> > /AsyncEventQueue_XYZ_PARALLEL_GATEWAY_SENDER_QUEUE:
> PartitionedRegionStats.
>
> > dataStoreEntryCount
>
> > /AsyncEventQueue_XYZ_PARALLEL_GATEWAY_SENDER_QUEUE:
> PartitionedRegionStats.
>
> > dataStoreBytesInUse
>
> >
>
> >
>
> > I found only count for async queue in JMX , it called here
>
> > EventQueueSize = PartitionedRegionStats.dataStoreEntryCount
>
> >
>
> > /** AsyncEventQueue Stats **/
>
> > public static final String ASYNCEVENTQUEUE_EVENTS_QUEUE_SIZE =
>
> > "eventQueueSize"
>
> >
>
> > we having Missing PartitionedRegionStats.dataStoreBytesInUse in JMX
>
> > for Async Qeueue.
>
> >
>
> > Can you help me , which property I need to map for it.
>
> >
>
> > Thanks,
>
> > Dinesh Akhand
>
> >
>
> >
>
> > This message and the information contained herein is proprietary and
>
> > confidential and subject to the Amdocs policy statement,
>
> >
>
> > you may review at https://www.amdocs.com/about/email-disclaimer <
>
> > https://www.amdocs.com/about/email-disclaimer>
>
> >
> This message and the information contained herein is proprietary and
> confidential and subject to the Amdocs policy statement,
>
> you may review at https://www.amdocs.com/about/email-disclaimer <
> https://www.amdocs.com/about/email-disclaimer>
>
