Unable to see disk metrics in DB dashboard for DBSS agent, the follow message is logged every five minutes by the agent
YYYY-MM-DD HH:MM:SS.SSS ECHO <DB_SQL_Server/126.96.36.1990/DB_SQL_Server/DBSS-TESTSERVER-agent> ERROR [DBSS-TESTSERVER-wmiThreadPool-1-[DBSS_SQL_Server_Host][YYYY-MM-DD HH:MM:SS.SSS]] com.quest.qsi.fason.sqlserver.collection.processor.DBSSPhysicalDiskProcessor - Failed to calculate physical disk utilization. The result is not in the [0,100] range. calculated Disk Utilization : -316.2953082250646
Results from WMI Explorer (query taken several times) for that Perfmon confirms that % Disk Time is significantly higher than 100% over several times taken (e.g. 2623.305%) compared to other servers and workstations
"When Performance Monitor fails to get data it requests, or when it receives uninterpretable data, it displays a zero as the counter value for that request. The data request might have failed because of a malfunction or simply because the computer or application it was monitoring had stopped. Data considered uninterpretable includes negative times, negative performance values, or percentages greater than 100.
However, if Performance Monitor is logging negative time errors, there might be a problem with your HAL DLL. In the past, this has been encountered in some HALs for multiprocessor computers."
It appears that this is a known issue for Windows 2003 SP2:
"We have found a problem with some WMI raw performance data counters on Windows 2003 SP1 and Windows 2003 SP2. We found that some counters decrement, so each recollection has a lower value that the previous value instead of getting bigger so this renders the calculations done after the recollection useless. For example, this happens in Win32_PerfRawData_PerfOS_Processor with the counter PercentProcessorTime, if we use WBemTest to get the values, we see that any new value of the counter is lower than the previous one, so when we try to get a formatted percentage based on the counter values we end getting an invalid negative percentage of processor time.
Other counters affected are in Win32_PerfRawData_PerfDisk_PhysicalDisk:
The performance counters on the monitored host are corrupted (total disk utilization exceeding 100%). The following Microsoft's article explain how to rebuild the counters: https://support.microsoft.com/en-us/kb/300956
As an alternative, you can continue with the counter's present behavior which causes:
1). Disk utilization metric will not be set
2). Errors in the log file
Other than that, the agent will keep working.