Ensure continuity of PI system operation
In this demo, we will guide you through the main features of the application. Below you will find scenarios of common PI system issues that PI Monitoring will help you solve efficiently.
Analyzing various system components allows us to detect critical issues.
See how we solved them:
Monitoring of interfaces and data sources
Challenge
There was a spontaneous change of certificates on the data source side and consequently an incompatibility of certificates at the OPC server and OPC client (Connector PI).
Solution
Monitoring indicated a problem with the data source’s availability (OPC UA server), as data transmission was stopped. This suggests several potential failures causes to the Administrator. The system immediately notified The Administrator of the data availability problem.
Effect
Instant alerts allow him to respond immediately and re-enable the transmission of often critical production/business data. Ultimately, the solution to the problem was to replace the certificate in the Connector configuration.
Challenge
The interface saves “frozen” values to the archive, but at first glance, everything looks ok. All status points of the interfaces indicate correct operation. Data-source diagnostic points also suggest that the source system is active (e.g., heartbeat). The “I/O Rate” metric shows that data are continuously saving in the archive.
The diagnostic points of the PI interfaces suggest that the entire acquisition and archiving process is functioning correctly.
Solution
PI Monitoring provides a mechanism that examines a pool of selected process points for value changes (points that fluctuate frequently).
Effect
If the indicated points’ values do not change within the set time interval, the system administrator is notified of a data transmission problem.
Challenge
There was a problem with the SQL Job responsible for uploading data to the PI Data Archive.
Solution
The monitoring indicated and notified the Administrator of the incorrect status of the last job called.
Effect
The Administrator could react quickly to the notifications and manually run the task of loading the missing data into the PI system.
Monitoring of PI Data Archive services
Challenge
The PI Archive Subsystem was overloaded after a disk failure on the PI server. The excessive number of events originating from the PI interface buffers combined with the client applications’ excessive activity caused the archive files to be blocked. Consequently, it was impossible to save data to the PI server.
Solution
Monitoring indicated a problem with the queue files and with data saving by the PI Archive Subsystem service. In a detailed view, the system directly stated a problem with the queue files’ size and sent notifications to the Administrator.
Effect
The Administrator could either react immediately or observe the behaviour of the infrastructure for some time. The solution was a controlled emptying of the queue files. It relieved the workload of PI Data Archive and other dependent components.
Challenge
Removing too many events at once from the archive disrupted the PI Data Archive server.
Solution
The Administrator received notifications of a steady increase in memory usage on the PI Data Archive server, which seriously threatened the PI System’s stability.
Effect
The solution was to prepare batch mechanisms for deleting data that do not overload the PI DA server. Thanks to the fast response, the problem did not affect the work of system users.
SQL Server Monitoring
Challenge
The operation time in the PI system Explorer tool increased. This indicated performance issues with the AF server operation.
Solution
Monitoring of the PIFD database (FilliFactor parameter) showed to the Administrator a performance risk to the database on SQL Server.
Effect
The PIFD database required maintenance work. The solution was to rebuild and update the indexes of the AF configuration database.
Monitoring of related business applications
Challenge
The Administrator wanted to prevent incomplete data in the reports systemically.
Solution
The continuous monitoring of interfaces, data sources, and the quality of the data transmitted to the PI System make it possible to control and react to process data completeness problems. PI Monitoring also makes it possible to automate the back-filling of data in the event of interface or source failures.
Effect
In the event of problems with the PI system interface, data sources or the quality of the transmitted data, the system administrator is notified immediately and can take action. This approach ensures that business users always receive complete information and do not have to contact the system administrator to clarify the situation.