Layer7 API Management

  • Private Community
 View Only

  • 1.  Dynatrace Service-level reporting

    Posted Nov 16, 2022 09:29 AM
    Has anyone had any success viewing services provided by a Layer7 Cluster in Dynatrace?

    ------------------------------
    Philip Mead
    ------------------------------


  • 2.  RE: Dynatrace Service-level reporting

    Posted Dec 06, 2022 10:02 PM
    Yes. What are you trying to achieve?
    Original Message


  • 3.  RE: Dynatrace Service-level reporting

    Posted Dec 07, 2022 10:42 AM

    I have about 140 services I am delivering.

    I would like to get dynatrace to report on the individual services - things like total counts, response time, backend response time.



    ------------------------------
    Philip Mead
    ------------------------------

    Original Message


  • 4.  RE: Dynatrace Service-level reporting

    Broadcom Employee
    Posted Dec 07, 2022 02:09 PM
    I don't think Dynatrace, which looks into the JVM itself if memory serves, can deliver that information. That is why PAPIM (Precision API Monitoring) was created. PAPIM is designed to work with the APM (ex Wiley) monitoring system, but its architecture can be easily modified to deliver the stats to pretty much anything that can present an API for feeding the data to it.

    ------------------------------
    Jay MacDonald - Adoption Architect - Broadcom API Management (Layer 7)
    ------------------------------

    Original Message


  • 5.  RE: Dynatrace Service-level reporting

    Posted Dec 08, 2022 07:38 AM
    We feed the traffic log into Splunk, that allows us to report on anything in the traffic log, including response time, routing time, etc.
    Original Message


  • 6.  RE: Dynatrace Service-level reporting

    Posted Dec 23, 2022 07:27 AM
    What we have done is we created few business transactions in Dynatrace. Then we pie chart them in the Layer7 dashboards. Mind you that total count value contains the /ssg/ping and other internal gateway API services calls. You can create a filter or just leave it. If you decide to leave the numbers then remember that 5-10% requests are ignorable as they /ssg/ping, load balancer probes and internal gateway calls.

    To install dynatrace, depending upon if you are using APM or one-agent, you can either inject the dynatrace string in ssgruntimedefs.sh file or run one-agent script.
    Original Message