Noy shown trend
We reset secandary connectivity server for taking backup with acronis ,after that all of trends on operator console do not shown.whats its cause?
The Service Provider statistics for the two Basic History services indicate 6-12 months runtime (date & time of service going live is shown in EnterServiceTime)
Hence, if you made drive image backups using Acronis, it must have been done "hot" (with Basic History and the Microsoft Windows operating system up and running).
ABB recommends only "cold" drive imaging. Perform at least "Configuration Wizard > Maintenance Stop" before taking image backup, even better if you shut down Windows and boot on alternative (backup agent) media and perform the imaging with the target computer operating system completely shut down.
The MaxReadRequestTime of one of the service providers is too large - it may indicate overload, slowness due to memory leakage, too many or too heavy clients (e.g. reports asking for very large amounts of history data) or other problems reading history data from the storage on disk.
The PercentSynchronized indicates one or more logs not able to synchronize between the two redundant Basic History service provider peers. Please ask local ABB to assist you with finding or rectifying the logs that refuse synchronization. This may involve using expert tools like afwapplogviewer (you may search AKS for this topic; I believe I have laid out the fundamentals in this area in some post/article in the past).
With 6-12 months runtime, potential memory leakages may have accumulated significant amounts of RAM and started to introduce runtime problems for Basic History. Please use Windows Performance Monitor (perfmon.exe) to measure AdvHtHistorySrv.exe's Private and Virtual Bytes. These counters should stay below 1GB most of the time. If you find the "failed" Basic History service provider with considerably larger values than the still working one it is probably time to restart one or more Basic History service provider or the server computer(s) running them. Some memory leakages have been resolved in recent revisions / rollups released for System 800xA. If you are running an older version, an upgrade to the most recent revision and/or rollup may be recommended.
If you cannot solve the problems following these steps, please have a support case reported to your regional ABB support center to obtain professional support.
- Process > % Processor Time
Indicator of amount of stress subjected to Basic History by clients and own duties. Should not be high.
- Process > Virtual Bytes
Virtual memory (should normally not exceed 1GB)
- Process > Private Bytes
- Physical Disk > Average Disk Queue Length
Total value for a disk. Should not constantly be above 0.0 or rise too high during heavy activity as it may indicate an overloaded harddisk, RAID controller or just a too slow disk for the job it has been put to do.
You may ask why Private Bytes increased at restart of process? Hard to tell, but most of 800xA is programmed in C# using managed code with Garbage Collection. Memory will be released when no longer used by a process. This automatic trimming is mostly out of control for a process programmed using managed code - a few minutes later the RAM use shrunk (see second image)
The Windows Performance Monitor can also save values to disk for later recall and analysis. Read more about "Counter Log Set" on Google or Microsoft online help.