none
HCK测试 NVMe Queue Utilization Test (LOGO) 失败，错误信息： Failed to Send IOCTL to SHIM, GLE = 0x2 RRS feed

 > 

  • 问题

  • Question
    登录进行投票
    0
    登录进行投票

    该测试项有时候能通过，但是最近一直无法通过，查看日志如下：

    <o:p></o:p>

    Machine: Build=9200.00 
             Platform=AMD64 
             OS=Windows Server 2012 R2 Standard Evaluation 
             ServicePack="" 
             Version=6.3 
             BuildLab="winblue_rtm" 
             BuildDate=130821-1623 
             Language=Chinese (Simplified) (REDMOND) 
             Config=n/a 

    Start: Parse command line arguments, TUID=
    End: Pass, Parse command line arguments, TUID=, Repro=nvmeshim.exe  /DriveNumber 4 /DriveLetter D: /operation QueueUtilization /scenario All /verbosity 4 /Duration 300

    Start: Creating Operation, TUID=
    Obtained scenario handle
    HWID: SCSI\DiskNVMe____________0235G6PF2.30
    End: Pass, Creating Operation, TUID=, Repro=nvmeshim.exe  /DriveNumber 4 /DriveLetter D: /operation QueueUtilization /scenario All /verbosity 4 /Duration 300

    Start: IO Workload with Queue Statistics, TUID=
    Description: Run multithreaded IO workload, check for Queue utilization
    Expectation: Device should be able to utilize all the queue slots for IO
    #Workload instances: 1
    Disk Size in bytes =  1d1c0e98200
    Waiting for drive to be cleared up...
    Instantiating 1 workload threads
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 1676GB /v /uniquelogname /f 
    Launching workload threads...
    Output file: Scenario1WorkloadInstance0.out
    Set affinity of workload threads...
    Error: 0x0, Error 0x00000000
             Failed to send IOCTL to SHIM, GLE = 0x2
             File=testsrc\driverstest\storage\wdk\nvme\queue.cpp Line=446
    End: Fail, IO Workload with Queue Statistics, TUID=, Repro=nvmeshim.exe  /DriveNumber 4 /DriveLetter D: /operation QueueUtilization /scenario All /verbosity 4 /Duration 300

    Start: IO Workload with Queue Statistics, TUID=
    Description: Run multithreaded IO workload, check for Queue utilization
    Expectation: Device should be able to utilize all the queue slots for IO
    #Workload instances: 16
    Disk Size in bytes =  1d1c0e98200
    Waiting for drive to be cleared up...
    Instantiating 16 workload threads
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Workload app string: diskio.exe /d D: /c sw;rw /b 512KB /t 300 /q 130 /s 104GB /v /uniquelogname /f 
    Launching workload threads...
    Output file: Scenario2WorkloadInstance0.out
    Output file: Scenario2WorkloadInstance1.out
    Output file: Scenario2WorkloadInstance2.out
    Output file: Scenario2WorkloadInstance3.out
    Output file: Scenario2WorkloadInstance4.out
    Output file: Scenario2WorkloadInstance5.out
    Output file: Scenario2WorkloadInstance6.out
    Output file: Scenario2WorkloadInstance7.out
    Output file: Scenario2WorkloadInstance8.out
    Output file: Scenario2WorkloadInstance9.out
    Output file: Scenario2WorkloadInstance10.out
    Output file: Scenario2WorkloadInstance11.out
    Output file: Scenario2WorkloadInstance12.out
    Output file: Scenario2WorkloadInstance13.out
    Output file: Scenario2WorkloadInstance14.out
    Output file: Scenario2WorkloadInstance15.out
    Set affinity of workload threads...
    Error: 0x0, Error 0x00000000
             Failed to send IOCTL to SHIM, GLE = 0x2
             File=testsrc\driverstest\storage\wdk\nvme\queue.cpp Line=446
    End: Fail, IO Workload with Queue Statistics, TUID=, Repro=nvmeshim.exe  /DriveNumber 4 /DriveLetter D: /operation QueueUtilization /scenario All /verbosity 4 /Duration 300

    Summary: Total=4, Passed=2, Failed=2, Blocked=0, Warned=0, Skipped=0

    另外，在HCK Manager的Job Monitor的Task Execution Status里无法查看日志，如下图：

    特别是经常有以前成功现在失败，多试几次可能成功的情况的发生，现在试了很多次都是失败，请帮忙看看如何解决？

    <o:p>谢谢！</o:p>


    2017年3月1日 1:50
    |

答案