4 Replies Latest reply on Nov 28, 2016 9:40 PM by fabiomariotto

    PI OPC Server frozen data




      I have an OPC-client collecting data from PI OPC-Server.

      This PI has a secondary server as a redundancy.

      One of the servers has frozen/wrong data, the other one is running well.


      My OPC-Client is collecting data from the frozen server.


      I've tried manually changing the IP address that my OPC client connects to, but it didn't change the output.


      Any idea on how to get data from the running server (regardless of the frozen data problem)?



        • Re: PI OPC Server frozen data

          Hi Fabio,

          Is this a custom built OPC client? If this is a custom client? If so, are these servers part of a collective? Are your OPC ItemIds fully qualified? If this is not a custom client you may want to move this to the 'All Things PI' forum

            • Re: PI OPC Server frozen data

              Its a custom client developed with C#.

              They are fully qualified.

              The client was working ok, extracting data periodicaly. No changes in configuration were made.

              Error only occurred when some of the PI TAGS got frozen on PI primary server.


              There is a primary server with some tags frozen, and some good tags (Same good/froozen tag value is fetched, for any tag, whereas using by PI data link or my opc-client).

              And then there is a secondary server where all tags are good and updating correctly (I can see this with data link for example).


              But the thing is, while I can change the server being target from primary to secondary, in tools like data link, I can't change it in my opc-client application. Even if i change the IP the cleint is targeting, it will still get the frozen data from primary server.

                • Re: PI OPC Server frozen data

                  Hello Fabio,


                  If I recall correctly, PI OPC Server might be using PI API to read data from PI. Because PI API is not collective aware, I assume there's some logic to identify what PI Data Archive node is used for the data retrieval. I would assume it's using the PI Collective member with the highest priority. This is defined in the Known Servers table which is a client configuration of available connections and some settings like connection and data access timeouts.


                  However, it appears to me as if fixing the issues of "frozen" data items on the primary should be highest priority. Addressing issues should always be prioritized upon working around the effects.

                  In case there's archive corruption or similar on your Primary, you should consider this a high priority issue. If the issue is with PI Buffer Subsystem configuration on the data collector node, this should as well be treated high priority. The redundancy introduced with a PI Collective is supposed to reduce the risk of data loss ..


                  In case your issue is still current, I recommend you to contact Technical Support.

                  1 of 1 people found this helpful