Icon

Comments have been closed on this page. Please use AppMon & UEM Open Q & A forum for questions about this plugin.

Overview

This plugin will capture messaging-related metrics from TIBCO EMS.  The measures are composed of broker level measures and destination level measures

The plugin also allows for splitting by topic/queue name when charting the destination level measures

Plugin Details

Plug-In Files

com.dynatrace.plugin.TibcoEmsPlugin_1.0.0.jar

Author

Erwin Pader

dynaTrace Versions

5.6+

License

dynaTrace BSD

Support

Not Supported

Release History

Dec. 4 - v1.0.0 - Release

 

 

Screenshot of the measures screen

 

Configuration

The following properties need to be defined when setting up this monitor

Property NameTypeDefault ValueNotes

Host

String

localhost

hostname or IP address
PortString7222 
Username

String

admin

Username credential to authenticate into TIBCO EMS server.
PasswordString Password credential to authenticate into TIBCO EMS server

Screenshot of the Configuration page

Screenshot of the Destination Level Measure Configuration page

Note the destination key which requires a topic or queue name.  Wildcards can also be used to capture measures for more than one queue or topic.

Installation

Import the Plugin into the dynaTrace Server via the dynaTrace Server Settings menu -> Plugins -> Install Plugin. For details how to do this please refer to the dynaTrace documentation:

Unknown macro: {doclinks} [DOCDT55:Plugin Management]

Contribution

Feel free to contribute any changes on Github

 

  1. Anonymous (login to see details)

    Is this Plugin still valuable and valid in 6.0 or 6.1?

  2. Anonymous (login to see details)

    hi Kenneth,

    it should work for 6.0.  i have not tested against 6.1 but i'd be surprised if it does not work (smile)

     

  3. Anonymous (login to see details)

    Quick question how would I go about using the wildcard mentioned above as all it returns is a 0.0 value but if i specify the specifc destination it works properly?

     

    Great plugin by the way!

    Richard

  4. Anonymous (login to see details)

    hi Richard,

    tibco uses wildcard notations like so QUEUE.> or QUEUE.*

    is this how you constructed yours?

     

  5. Anonymous (login to see details)

    Thanks the > wildcard works the * wildcard doesn't appear to be working.

     

    Thanks for the help!

  6. Anonymous (login to see details)

    Hi, We have multiple que's running on the same host but they need to be grouped differently and have different thresholds assigned to them. Is there any way to have the Destination: used for the measures to make use of a regex to tell it what queues to gather metrics from instead of th defailt starts with and a wildcard?

     

    Also does anyone know if it's possible to assign thresholds on dynamic measures which in this case is the individual queues? We have different requirments in terms of thresholds per queue per measure.

     

    Thanks the plugin has been very useful so far!

     

    Kind Regards

    Richard

    1. Anonymous (login to see details)

      Hi Richard,

      Are you able to create thresholds  for individual queue using ems plugin. We are also using ems plugin where there are 200 queues are running and we want to setup individual threshold for each queue. Is there a way you figured out, if so can you please share that will be helpful.

      Thanks,

      Tarun

  7. Anonymous (login to see details)

    Hello! My monitor is returning "null" for all destination level stats (according to the Retrieved Measurements in the details view of the plugin). Any idea what could be causing that? Could TIBCO be configured not to expose those metrics, or have changed their APIs? EDIT: all broker and server level stats are returning normal values.

  8. Anonymous (login to see details)

    Hi Jacob,

    you will need to put a specific queue/topic (or a wildcard) name if you want your destination level measure to return a value. thnx

     

    1. Anonymous (login to see details)

      Hi Erwin,

      I figured out that using QUEUE.> for my destination wildcard as recommended above didn't work because their queue names were lowercase. "queue.>" worked. My mistake, I didn't realize they were case sensitive. Many thanks for your help!

  9. Anonymous (login to see details)

    Hi Richard,

    you might want to create a custom measure then (one for each queue/topic) so that it allows you to set specific thresholds.  thnx

  10. Anonymous (login to see details)

    Is there some automatic process that triggers the creation of new measures, based on available queues?  Because I manually created some yesterday – copies of InboundMessageRate, OutboundMessageRate, and PendingMessageCount – and when I look today, some of those are gone (local-queue.PendingMessageCount and local-queue.OutboundMessageRate), but the local-queue.InboundMessageRate ones are there ... but there are also some similarly named ones relating to the local queue names, ones that I didn't create...?!

    1. Anonymous (login to see details)

      Hi Mark,

      Dynatrace doesn't automatically create new measures based on available queues. It would just add a new "Dest Name" splitting to the overall measure. It WILL add or remove splittings based on what queues it sees though, so the list of splittings and the list of dynamic measures you see when you chart a measure might change.

      Jake

      1. Anonymous (login to see details)

        Oh, ok – thanks!

        1. Anonymous (login to see details)

          No problem! I would try creating those measures manually again, and make sure that you click Apply and Ok to save them, and then double-check that you can chart with them. If after all that they disappear AGAIN, I would open a support ticket.

  11. Anonymous (login to see details)

    Hi,

    The monitoring plugin on my end has been reported to be keeping multiple connections open to TIBCO. I have also seen that we have a lot of TIBCO queues in the environment - could this be a probable cause?

    Regards

    K

     

  12. Anonymous (login to see details)

    Hi Katlego,

    the plugin opens just one connection every time it executes.  kindly check if there are more than one instances of the plugin configured as that might be the cause.  thnx

  13. Anonymous (login to see details)

    Hi Erwin,

    Yes we do have other queues that have been configured to extract specific measures. This is going to be a bit tricky as we require to trigger alerts for specific queues. Is there a way we can achieve this without establishing these multiple connections. 

    I would also an additional feature to be added, is it possible we can extract the size of the messages that are sent and recieved.

    thanks

    Katlego

  14. Anonymous (login to see details)

    hi Katlego,

     

    there's probably no way around the connection issue other than combining the queues by specifying a wildcard.  this way you only have one instance of the plugin executing. i will look into message size as well and will keep you posted.  thnx 

  15. Anonymous (login to see details)

    Thanks Erwin,

    I have applied the method you mentioned and when using the wild card there is only one connection that is established and we have not seen any TIBCO issues that arise. You can let me know when you added this feature.

     

    thanks

    Katlego

  16. Anonymous (login to see details)

    Hi,

    I have installed this plugin to monitor our TIBCO EMS server. and in destination measures like Pending Message Count, Queue Receiver Count etc. I have used wild card * to capture data related to all queues. But It is not showing data specific to all queues instead it shows 0.0 value for sample.

    For eg. Queue Receiver Count metrics we are getting:

    Queue Receiver Count: null
    Queue Receiver Count (Dest Name,sample): 0.0

     

    Can you help me in troubleshooting the same. What are the pre-requisites for using this Plugin? Does it require Hawk Administration to be Enabled for EMS?

  17. Anonymous (login to see details)

    Hi Bhavin,

    Did you try using the using the > character, I also noticed that with the asterisk this does not capture.

    regards

    Katlego

  18. Anonymous (login to see details)

    Hi

    We are adding a measure to our Tibco EMS Monitor. This measure is (Broker) - Max Msg Mem Allowed. The max memory is 10GB and the thresholds that were assigned for the alerting was 60% of that 10GB for upper warning which was set to 6.442450944E9 bytes & 75% of that 10GB for upper severe which was set to 8.05306368E9 bytes. 

    The alert that was triggered though was well below this 60% threshold. Are they bytes calculation that we have correct? 

    Please help. Thanks!

  19. Anonymous (login to see details)

    Is it possible to modify this plugin to include other methods that are available from the ServerInfo clase, like getState.

    I am attempting to monitor for a dual-active scenario in my EMS clusters.  There are occasional instance of dual-active that I want to alert on.

     

    Get the current state of the server. Can return one of the constants:
    SERVER_FT_STANDBY - server is in fault tolerant standby mode
    SERVER_ACTIVE - active server

    Let me know if this is possible

    1. Anonymous (login to see details)

      Hi Troy,

      We are also having same problem. we want to know if our primary server goes down we want to get some alert.

      Can someone help us whether it will be possible to add.

       

       

  20. Anonymous (login to see details)

    Hi,

    We have TIBCO EMS running on Linux and we seem to have an issue with connecting and retrieving the values. Is there a conflict when using the plugin to execute this on Linux?

    Kind Regards

    Katlego

    1. Anonymous (login to see details)

      Hi Katlego Kau,

      Are you using a Linux or Windows based monitor Collector for this?
      Can you make the plugin's log files available, after changing the log level to FINER and re-running it?

      Regards,
      André 

  21. Anonymous (login to see details)

    Hi,

    We tried to use this plugin in production environment. It seems to open a new connection to EMS each time it executes. It caused performance overhead in some of the EMS servers monitored. I review the source code. The connection is opened in execute() method. I think opening the connection in setup() and closing it in teardown() would be better regarding the connection reuse. There may still be a control in execute(), so that it can open a new connection if the existing one is invalid or stale. What do you think?

    Right now, we have disabled the plugin to prevent the overhead.

    Regards,

    Meltem

  22. Anonymous (login to see details)

    Hi all,

    We have disabled comments on this plugin page.

    Please use the AppMon & UEM Plugins forum for questions about this plugin.

    Sorry to interrupt ongoing discussions. Please re-post your last question in AppMon & UEM Plugins forum.