Check the Transfer Times

When you have checked that the Main and Standby servers are synchronized, you need to check the performance of the servers by examining the transfer times. For an efficient system, the servers have to be able to transfer the necessary amount of data quickly.

To check the transfer times:

  1. Log on to the Main server.
  2. Display the Server Status Tool by right-clicking on the server icon and selecting the Status option.
  3. Display the Standby>Counters status information. The status information consists of three lines of entries for each connection to a Standby server:
    • Line 1 (First)—Provides information about the first pass.
    • Line 2 (Total)—Provides total statistics for the data transfers.
    • Line 3 (Last)—Provides information about the latest pass.
  4. You need to focus on Line 2, in particular, the Count and Time status values.
  5. For each connection, divide the Time status value by the Count status value. The result of this calculation is the average amount of time spent transferring data from the Main server to the Standby server.
  6. If the average transfer time value is greater than or equal to the Update Interval for the Main server (defined in the Partners settings, see Define the Transfer Interval and Transfer Timeouts), and then there is a problem with the amount of data being transferred. The cause of the problem is likely to be a slow network (low bandwidth) or an excessive amount of updates being reported.
  7. Examine the various count statuses for each server connection. The count statuses indicate the amount and type of data being sent to the Standby server(s). You should pay look closely at these statuses:
    • Data—The number of data changes that have been transferred from the Main to the Standby server.
    • Evt—The number of event records that have been transferred from the Main to the Standby server.
    • His Rec—The number of historic records that have been transferred from the Main to the Standby server.
  8. If the Data and Evt counts appear to be high i.e. there appear to be more updates per transfer than you would expect, and then there may be a configuration problem. Typically, this type of problem is experienced when values are being reported too frequently, for example, fluctuating point values are being reported because insufficient deadbands have been configured. To resolve this, you need to examine the Event Journal entries and use the event information to determine which database items are reporting new values too frequently (proceed to step 6).
  9. If the His Rec count is higher than expected, there may be too many historic records being reported. You need to investigate this further by using the Historic Transfer status information (proceed to step 10).
  10. If the Data, Evt and His Rec counts show values within the expected range for your system, it means that the link itself is incapable of transferring the data as efficiently as required. To improve the speed of data transfer, you will need to upgrade the connection. As you have identified the connection hardware as the cause of the problem, you do not need to continue with the remaining steps in this section.
  11. Log on to one of the Main server’s clients and access the Events List.
  12. Use the Events List to determine which database items are reporting their values too frequently.
  13. Access the Forms for the database items and use them to configure appropriate value limits and deadbands. For more information, see Defining Digital Point Details in the Geo SCADA Expert Guide to Core Point Configuration and see Defining Analog Point Details in the Geo SCADA Expert Guide to Core Point Configuration.
  14. Having made the configuration changes, log on to the Main server and check the Counters status information. If the database items have been configured with suitable limits, the Data and Evt statuses should indicate that a reduced amount of data is being transferred from Main to Standby.
  15. If the Data, Evt and His Rec status values are as expected, the data transfer should be working to maximum efficiency. If the link is still slow, the link itself is the cause of the problem—you will need to upgrade the connection. As you have identified the connection hardware as the cause of the problem, you do not need to continue with the remaining steps in this section.
  16. If the Data and Evt status values are as expected, but the His Rec status value is higher than expected, there is too much historic data being transferred. Proceed to step 10.
  17. On the Main server, access the Server Status Tool and examine the Historic Transfer status information. This information provides details about the historic data that is transferred from the Main server to the Standby server.
  18. Examine the Curr Data, Curr Upd, and Curr Del statuses.
  19. When new historic data is reported, it is placed in the queue to be transferred from the Main to the Standby server. The amount of data in the queue is shown in the Curr Data column. Similarly, the Curr Upd column shows the number of historic update records pending transfer, and the Curr Del column shows the number of pending historic deletions.
  20. In normal circumstances, the three statuses will show an amount at the beginning of the historic cycle, and by the end of the historic cycle, the amount will have reduced. Over subsequent historic cycles, the status amounts should eventually reach 0.
  21. If the status amounts increase over several historic cycles, it means that there is too much historic data waiting to be transferred—either there is an excessive amount of historic data being generated or the link is insufficient for the amount of data it needs to transfer. You can either reduce the amount of historic data being generated (by disabling the Historic feature on database items) or you can upgrade the link.

Disclaimer

Geo SCADA Expert 2019