3 Replies Latest reply on May 15, 2014 5:06 PM by dsbalcau

    real time monitoring of storage array and switches

    kaushikpathak

      Real time monitoring of storage arrays, NAS and switches are required for availability or fatal performance degradation. how can we do this in STM and if there is any impact on performance of STM?

        • Re: real time monitoring of storage array and switches
          dsbalcau

          Can you please clarify your question, I'm not sure exactly what you're asking? If you are in an eval, I recommend you direct your questions to your Sales Representative as they will be able to provide you a full answer to your inquiries.

           

          Alternatively - you can reference some of the material here for Storage Manager and see if it might be of use in answering your questions: Storage Resource Monitor (SRM) - Updated May 5, 2017

            • Re: real time monitoring of storage array and switches
              kaushikpathak

              Hi Balki, STM agents poll for data like storage, assets and performance in specific interval. Now if there is any hardware issue in the storage hardware and if this hardware comes under storage or asset category, it will report this failure in may be one hour or 24 hour as per polling frequency set. As a monitoring tool, it will be helpful if STM can monitor some essential components in real time so that it can send traps for alerts immediately when it happens.

               

              Beside this i was trying to setup a health check scheduled report for all devices like Storage, FC Switches, NAS Datamovers / Headers etc... found that for some we are able to set this for have to wait for certain frequency, and for some like if a FC switch gets faulted / down, or a storage processor gets faulted, NAS datamover / filer get faulted we dont have option to send trigger for that.

               

              I've raised query for these two observations.

                • Re: real time monitoring of storage array and switches
                  dsbalcau

                  Hi kaushikpathak,

                   

                  This kind of monitoring with STM is not possible.

                   

                  1. This is a vendor limitation for the most part.
                  2. You might not like the results if you really did this. This kind of monitoring can significantly affect the performance of the system being monitored. I'm assuming you want to do this because these systems are mission critical. Thus you've degraded the performance of these systems in order to get quicker notification of degraded performance.

                   

                  You can increase polling intervals with STM, but this can increase load on the STM server as well. Also, you might configure a NOC system like Orion as a trap receiver for hardware generated traps.