Icon

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

Overview

 

Name and Version

Measure Availability Monitor Plugin, version 0.9.3

Compatible with

dynaTrace 5.5+

Author

Eugene Turetsky (Eugene.Turetsky@dynatrace.com)

License

dynaTrace BSD

Support Level

Not Supported

Download

com.dynatrace.diagnostics.plugin.MeasureAvailabilityMonitor_0.9.3.jar

Description

Measure Availability Monitor plugin monitors availability of measures which are passed to the plugin as parameters

Installation

Import the Plugin into the dynaTrace Server. For details how to do this please refer to the Online Documentation on Plugin Management.

Usage

Plugin monitors availability of measures which are passed to the plugin as parameters. For each monitored measure it dynamically creates availability measure which could be used to setup incidents that will be triggered when monitored measure becomes unavailable.

Configuration

 The following screenshot shows example of plugin's configuration parameters:

Parameters

Plugin has the following parameters: 

  • measures - a semicolon separated list of measures which plugin will be monitoring;
  • prefix - text which will prepend monitored measure
  • suffix - text which will append monitored measure
  • available value - value which is used to show that monitored measure is available. Default value is 1.
  • unavailable value - value which is used to show that monitored measure is not available. Default value is 0.
  • REST port - port of the dynaTrace server REST interface. Default value is 8020.
  • dashboard - name of the dashboard which contains monitored measures. Default value is "myDashboard".
  • user - user id which is used to connect to the dynaTrace REST interface. Default value is "admin".
  • password - password which is used to connect to the dynaTrace REST interface. Default value is "admin".

Note

The Measure Availability plugin internally is using the count attribute of the measure tag. Starting with DT 6.5 the count attribute is hidden in the report by default. To make it visible, the visible property of the Count column needs to be marked in the Counts tab of the dashlet properties. See this screenshot for details.

Examples 

Next screenshot shows availability of sites located in Europe, North America, South America, Australia, Asia, Arctic, and Antarctica. Available value is set to 200, unavailable value is set to 500. North America, South America, and Asia sites are up while other sites are down.

Next screenshot depicts the Measure Availability Monitor configuration screen:

Next screenshot depicts example of measures that are produced by the Measure Availability plugin when the following measures are not available:

  1. MA - BusinessBackend_easyTravel;
  2. MA - CustomerFrontend_easyTravel_8080;
  3. MA - CreditCardAuthorization_easyTravel;
  4. MA - dotNetBackend_easyTravel;
  5. MA - dotNetFrontend_easyTravel

Next screenshot depicts example of measures that are produced by the Measure Availability plugin when the following measures are available:

  1. MA - BusinessBackend_easyTravel;
  2. MA - CustomerFrontend_easyTravel_8080;
  3. MA - CreditCardAuthorization_easyTravel;
  4. MA - dotNetBackend_easyTravel;
  5. MA - dotNetFrontend_easyTravel

Next screenshot shows chat with the following metrics gathered over some time frame:

Next link contains example of the dashboard referenced in the Measure Availability Monitor plugin configuration screen: example of a dashboard.

Feedback

Please provide feedback on this plugin either by commenting on this page or by comments on the Community Plugins and Extensions

 

  • No labels
  1. Anonymous (login to see details)

    Does the plugin automatically detect the server/s or do you have to manually configure them and add them?

     

    1. Anonymous (login to see details)

      Hi Katlego,

      You need to create a dashboard which captures some measure from each server which you are planning to monitor. Server is up when measure is present in the dashboard and down when measure cannot be retrieved from the server. Plugin will create dynamic availability measures which you can use for incident settings, charting, etc.

      Please let me know if you need more details. I will help you to setup this plugin.

      Eugene.

       

       

  2. Anonymous (login to see details)

    Hi Eugene,

    That would be great if you could assist. Lets start with the dashboard firstly then we can look at the plugin and how to configure it. How should my Dashboard look and what properties must it have?

    Regarding the plugin should I configure it before the I create the dashboard and what variables should be in the "Value" Measure box? Should it be a location or system that is being utilized or does it mandatory it be a location?

    Katlego

     

    1. Anonymous (login to see details)

      Hi Katlego,

      There are different ways to implement dashboard which will be used by the plugin. Before we started could you please let me know how many servers are you planning to monitor, are dynaTrace agents installed there, what OSes they are running (UNIX-based, Windows-based, etc.). Are all servers which you are planning to monitor reachable from the dynaTrace Collector which hosts Measure Availability Plugin?

      Dashboard needs to be created first because it is one of the plugin's parameters. Dashboard also will give us list of measures which monitor needs for the "measures" parameter. 

       Eugene.

      1. Anonymous (login to see details)

        Hi Eugene,

        There are about 2000 servers which are placed in Agent Groups and are still growing by the month and 70% are Windows 2003+ and the rest are UNIX-based servers. We have configured all the servers to be reachable to a DT Collector and they can host the Measure Availability Plugin.

        I have created a custom dashboard board, I am still debating with my colleagues as to which which measure should be used to track availability:CPU Utilization, Agent Availability, Connected Agents, Application Free Memory or JVM Heap Usage?

        Katlego

        1. Anonymous (login to see details)

          Hi Katlego,

          We have found good solution for your request (thanks to Andi Grabner!). There is the Availability Monitor Plugin on our community portal site which should match exactly your requirements. Consider using hostgroups or sites to minimize plugin's configuration effort. Please use it and let me know if you have any issues or questions.

          Eugene.

  3. Anonymous (login to see details)

    Hi,

     

    I am trying to get this set up to accurately monitor JVM availability as the "Connected Agents" measure is not reporting correctly as it doesn't appear to produce an incident when a JVM gets gracefully shut down.

     

    My config on the plugin:

    measures : Free Memory

    Prefix : -

    Suffix : Availability

    Available Value : 1

    Unavailable Value : 0

    Rest Port : 8023      (Default HTTP Port)

    Dashboard : Availability Test   (Dashboard that was created to test this)

     

    If I am understanding the plugin correctly I should be seeing a new measure called "Free MemoryAvailability" but I don't?

     

     

    1. Anonymous (login to see details)

      Hi Richard,

      You should see the "Free MemoryAvailability" measure. In order to understand why you don't see it, could you please set plugin's log level to FINER, gracefully shut down JVM, execute plugin, and send me plugin's log file. Could you also send me dashboard configuration file? The best way to gather this information is to create dynaTrace support archive and send it to me.

      Eugene.  

      1. Anonymous (login to see details)

        Hi Eugene,

        Thanks for the quick response!

        I will enable the finer logging and try get a log file to you, just a question though do I need a dashboard or can I just use the measure it creates and add it to an incident rule?

        1. Anonymous (login to see details)

          Hi Richard,

          You do need to use the "Free MemoryAvailability" measure created by the plugin in the incident rule to monitor when your JVM is not available. You also need to provide a dashboard for the plugin. I'd like to look into dashboard's xml file to see name of the measure that plugin will be monitoring.

          Eugene.

  4. Anonymous (login to see details)

    Hi Eugene, how can I send you the details your requested? Can you give me your e-mail or send me a mail richardm@discovery.co.za then I will reply.

     

    Thanks

  5. Anonymous (login to see details)

    Hi
    I've installed the plugin and configured it correctly from reading the above info

    Measure: Thread Count

    prefix: Availability-

    available value: 1

    unavailable value: 0

    REST port: 8020 (default http port)
    Dashboard: <correct dashboard name>

     

    I've created a new dashboard for the measuring with the correct dashboard name and defined with measure to use for the monitoring "Thread Count"

    I've also set the logging to FINER as a just in case

    First question - should the agent automatically create a log file? I see nothing in the system information

    Secondly - I assume the way it works is once the plug in is created it will check for that measure in the dashboard and report back on availability by creating a new measure called "Availability-Thread Count"

    SO far I see no sign of logs and no sign of the new Availability-Thread Count measure.
    Just wondering if I may be doing something wrong or if there may be other things I need to do

     

    Thanks

     

    1. Anonymous (login to see details)

      Hi Jason

      The monitor plugin is not executed by an agent but is executed on a collector. When you are looking for the log files you therefore need to look at the logfiles that the collector is writing.

      But in general you are correct: the plugin will be executed on the scheduled interval on your Collector. It will use the REST Interface of dynaTrace to query that dashboard and then checks whether the measure you defined actually has a value. You need to make sure that the Collector that runs your plugin has access to the dynaTrace Server REST Endpoint. The default port is 8020!

      Also - depending on what you want to achieve with this plugin. If you want to Alert on the non-existence of a Measure you can now also do this built-in with dynaTrace 6. in dynaTrace 6 Incident Conditions can be based on Measures that do not provide a value. Here are the steps

      a) Create a new Incident, e.g: "No Web Requests Coming in for URL XYZ"

      b) As Condition select or create a Web Request Count Measure for URL XYZ -> this will count how often this URL is requested. As lower severe threshold specify 0

      c) IN the list of conditions select "Count" in the Aggregation Column.

      This Incident will trigger in case Web Request Count Measure doesnt provide ANY measures in the specified Evaluation Timeframe. Thats what the "Count" aggregate is for as it looks at the number of measure values this measure actually returned in the specified time interval. This count is then checked against your lower severl threshold of 0. Therefore - if you have no URL Requests coming in this measure will create 0 (zero) measure values which will trigger this incident

      hope this helps

      1. Anonymous (login to see details)

        Thanks - got it mostly working but I do have one final question - in creating an incident rule based on that measure
        Execution Target is the Collector

        For the Hosts value - is it the host the meaure being checked is located on or other?

        Thanks

        1. Anonymous (login to see details)

          Are you refereing to the list of Hosts you need to define for the Availability Monitor Plugin? In that case it should be the host of the dynaTrace Server. I believe the plugin uses the Host to build the full REST URL to the dynaTrace Server REST INterface

          1. Anonymous (login to see details)

            Thanks again

            Working perfectly now all background network issues resolved

  6. Anonymous (login to see details)

    Hi - Trying to use this plugin but experiencing an issue.

    I installed the plugin, setup the monitor on a collector, but I only get one entry in the plugin log there:

    2014-10-16 10:51:35 SEVERE [MeasureAvailabilityExecutor@mobile22 MeasureAvailabilityMonitor_0] execute method: null

    We do not use http for the REST - we only have it on HTTPS port 8021 - could this be posing an issue here?

    Any comments or suggestions?

    Thank you

    1. Anonymous (login to see details)

      Well since it's so easy to enable non SSL in test - I tried it - and answered my own question.

      It does indeed work when I use the HTTP port.  Unfortunately we will not be enabling this in Production so I cannot use this plugin as is.


       

      1. Anonymous (login to see details)

        Hi Brad,

        It is easy to add SSL support to the plugin if you have dynaTrace non-SSL port  disabled in production. I'm on vacation now but will add SSL support when I come back. Also check if you are on dynaTrace 6.0 you may use its Incident Conditions instead (see this post above).

         Eugene.

        1. Anonymous (login to see details)

          Thanks Eugene - I was already looking at the java files for the mod from http to https but didn't get it figured out yet.

          We have confirmed that 6.0 addresses our issues, however, we have a need in 5.6 until we migrate for this monitoring.

          Thanks

  7. Anonymous (login to see details)

    Hey Eugene - I was able to make the modification myself - I had to do two things - update the contstants file with https vs http and also 'borrow' the disable certificate check logic from the license plugin since our ssl cert doesn't match our name.

    1. Anonymous (login to see details)

      Hi Brad,

      This is good. Actually I have already made changes to the plugin this morning by adding SSL indicator parameter (isSSL). I will publish new version of the plugin (0.9.3) later today when I complete full cycle of tests.

      Please note, that you should avoid using this plugin as soon as you upgrade to dynaTrace 6.0.

      Eugene.

        

  8. Anonymous (login to see details)

    Eugene - It seems like this plugin may not work as I initially expected.

    I have a dashboard with three measures for example - one is reporting data in the dashboard timeframe, and 2 are not.

    I setup the plugin to 'scrape' this xml dashboard page, however, its finding the 'blank' measures as valid and therefore markign them as '1' in the dynamic measure.

    It seems to me you are just counting on the XML containing the tag and attribute 'measure' which exists even when the measure contains no measurements.  This is always resulting in '1' value of every dynamic measure as long as the measure is configured on the dashboard.  It seems like what is needed is to look for the existence of the measurement tag within the measure tag.

    Please advise based on your understanding.

    1. Anonymous (login to see details)

      Hi Brad,

      New version of the plugin is located here. Its version is 0.9.3. Besides ability to work with the HTTP/HTTPS protocols and new "Is SSL" configuration parameter, this version of the plugin has slightly modified logic which takes into consideration number of measurements of given measures. It also can be deployed side by side with the old version 0.9.2.

      Please let me know if you have any issues.

      Eugene.

      1. Anonymous (login to see details)

        Thanks so much Eugene - I appreciate your efforts here.

        The plugin seems to work as expected now with the added benefit of SSL support.

        1. Anonymous (login to see details)

          Hi Brad,

          Sounds good. Please keep me posted in case you have any issues.

           Eugene.

  9. Anonymous (login to see details)

    Hi Eudene,

    I have used the Measure Availability Plugin and created Monitors for a Business Transactions - User Action Count measures. I am getting the availability values. But I am seeing two availability values. I am not sure why there are two availability values coming up. When I am using this measure, its creating two alerts in the incident rule since its seeing two availability measures. 

    Sample below: (Check the Retrieved Measured Data)

    Schedule Details
     Name: <dynaTraceServer>
     Status: scheduled
     Description:
     Type: MeasureAvailabilityMonitor
     Last Run: 10:20:58 (CET)
     Last Run Result: ok
     Next Run: 10:21:58 (CET) Duration (last run): < 1s
     Schedule: Every Minute
     Execution Target: <dynaTraceCollector>
     Plugin Active: active
     Result Status: success
     Result: Last polling execution terminated successfully
     Message: Success
     Detailed message: Plugin execution successful
     Retrieved measurements: Availability: 0.0
                                               Availability (Availability Split Measures,XXXXXXXXXXXXXXXXXXXXXXX - User Action Count): 0.0  

     

    1. Anonymous (login to see details)

      HI Srikanth,

      Please attach here Plugin Properties screenshot and dashboard which is set in the Dashboard parameter of the plugin. Please note, that you should avoid using this plugin if you are on dynaTrace 6.0.

      Eugene.

        

      1. Anonymous (login to see details)


        I am not able to attach the docs here. Can I email them to eugene.turetsky@compuware.com?

      2. Anonymous (login to see details)

        Please let me know if you got the files that I have sent you or is there any other way I can send them?

        1. Anonymous (login to see details)

          Hi Srikanth,

          I did get e-mail from you but haven't looked into your issue yet. I'll look into it early next week.

           Eugene.

          1. Anonymous (login to see details)

            Hi Eugene,

            Did you get a chance to look into this request?

            Thanks

            Srikanth

            1. Anonymous (login to see details)

              HI Srikanth,

              I'm sorry, I haven't had a chance to look into your e-mail yet. I'll try to look into it in few days.

               Eugene.

  10. Anonymous (login to see details)

    Hello,

    I am still a rookie when it comes to Dynatrace, so I hope this isn't too stupid of a question.  I am using the MeasureAvailabilityMonitor against an MQ.  In the 'Settings' section, I have the following for Measures:

    Status - esbmqp01pri.serv.dteco.com;Status - esbmqp01bk.serv.dteco.com;Status - esbmqp02bk.serv.dteco.com;Status - esbmqp02pri.serv.dteco.com

    So I have 4 queues (2 Primary and 2 Backup).  I need to configure an incident when a Primary and Backup on a server are both down.  Currently, it is alerting every 5 min due to it always seeing either a Primary or Backup down, as  there will always be one of them down on each server.

    Does anyone have any ideas on how I could achieve this.  Any help would be greatly appreciated.

    Thanks,

    Tom

     

    1. Anonymous (login to see details)

      Hi Tom. Any chance you could repost this question on the discussion forum - https://answers.dynatrace.com/spaces/148/index.html - it is a more generic question and with posting it on the forum you also get more eyes to see it

  11. Anonymous (login to see details)

    Hello,

    I have Dynatrace 6.5. When I configured the dashboard I get the status it failed, due to "execute method: getFileFromUrl method:FAILED : Retrieved dashboard report is not a XML type".

    Does the plugin support release 6.5 as well? Or is the issue caused due to changes in the rest interface of 6.5?

    Kind regards,

    Mark

    1. Anonymous (login to see details)

      Hi Mark,

      Most likely issue that you are observing related to the web server configuration: in DT 6.3.x and DT 6.5.x by default web server is configured using HTTPS protocol and port 8021. Please download and deploy version 0.9.3 of the plugin and configure it to use port 8021 with SSL enabled protocol. Version 0.9.3 is among attachments with timestamp Nov 19, 2014 18:55. I've just checked that plugin is working fine with DT 6.5.

      Let me know if you have any further questions.

      Eugene. 

  12. Anonymous (login to see details)

    Hi,

    Monitor is getting failed with this error "getFileFromUrl method:FAILED : Retrieved dashboard report is not a XML type". Can you please help me in resolving this issue? 

    I am not sure from where this monitor is trying to pull data and getting failed due to the extension type which is not in XML.

    Attaching logs and configuration setting for the same. Also would like to understand how this monitors works.

    2017-02-23 09:21:26 SEVERE [MeasureAvailabilityExecutor@Real User - OrderConfirmation - Belgium_0] execute method: getFileFromUrl method:FAILED : Retrieved dashboard report is not a XML type

    Measure - Realuser - order
    prefix -
    suffix -
    available Value - 1
    unavailable value - 0
    REST port - 8020
    Is SSL -
    dashobard - "Dashobard name"
    user - admin
    password - XXXX

    Hosts
    "dynatrace server hostname"

     

    Regards,

    Suraj Gupta

     

     

    1. Anonymous (login to see details)

      This error indicates that the URL that the plugin is requesting (which is the URL to retrieve the dashboard you specified) doesnt return the XML version of the dashboard. Most common root causes are

      #a) you have a typo in your dashboard name -> in your message it says "dashobard" -> shouldnt it be "dashboard"?

      #b) the username or password is wrong

      #c) the user you provided is not allowed to access the REST API -> even though unlikely as you use admin

      Andi

      1. Anonymous (login to see details)

        Hi Anders,

        The dashboard name is correct, i didn't disclose here. Even i checked with our admin user id but the monitor is still failing. 

        I've attached snapshot with detailed description in answer.dynatrace.com.

         https://answers.dynatrace.com/questions/171971/view.html

        Please let me know how we check why these monitors are failing.

        Regards,

        Suraj Gupta

        1. Anonymous (login to see details)

          Have you tried accessing the dashboard directly in your browser to see whether it actually comes back correctly? Dynatrace offers either HTTP (8020) or HTTPS (8021) access. By default HTTP is not enabled - so - make sure that this is not the issue.

          I would check the log files created by that plugin and check if the actual URL that the plugin tries to execute is logged. If so - copy/paste it in your browser window and see if it works.

          I havent created this plugin - but this would be my approach for troubleshooting

  13. Anonymous (login to see details)

    Hi Andreas,

    https://dynprdeu08:8021/rest/management/reports/create/Admin%20-%20Measure%20Availability?filter=tf%3ALast15Min&type=HTML&go=Go

    The above URL is accessible but not sure how to check whether dashboard is in which format. Did not find anything in logs as well. Please let me know how to check whether dashboard is in xml or not.

    Attached logs...

    L type
    2017-03-03 04:30:26 SEVERE [MeasureAvailabilityExecutor@Real User - OrderConfirmation - Belgium_0] execute method: getFileFromUrl method:FAILED : Retrieved dashboard report is not a XML type
    2017-03-03 04:31:26 SEVERE [MeasureAvailabilityExecutor@Real User - OrderConfirmation - Belgium_0] getFileFromUrl method:FAILED : Retrieved dashboard report is not a XML type

  14. Anonymous (login to see details)

    Now this looks interesting for me, I've selected SSL connection and used 8021 and monitor is started working fine. 

    Seems like 8020 port is not working properly. Can you please suggest how to resolve issue for 8020?

     

    1. Anonymous (login to see details)

      There is a check box in Server Settings -> Services -> Management that says "accept authenticcation data only with HTTPS". if that is checked it means port 8020 (HTTP) will not give you access to pages that require authentication. I assume this is the problem you run into

  15. Anonymous (login to see details)

    Yes Andreas, you are right (smile) , but if i disable this would not create any other issue? I am suspecting client use 8021 to connect dynatrace server.

    1. Anonymous (login to see details)

      The Client uses a different port all together. This setting only has an impact on accessing the REST API of the Dynatrace Server

  16. Anonymous (login to see details)

    I am sorry, i mean my client use this port for accessing dynatrace server(smile).