App fails with 0A-0000E019 and Sftfslh.sys uses 50% of CPU



    I am currently testing SoftGrid for use with some of our company’s internal software on a Windows Vista Enterprise Workstation, but occasionally an app will fail to launch with an Error code: 42012E-16D1380A-0000E019.  The problem is that the Sftfslh.sys+0x38300 thread under the System Process will then use around 50% of the CPU (It is a dual-core machine so it is using all of the first core) and the SoftGrid Client stops responding entirely. When attempting to manually restart the SoftGrid Services an Error 1053 is returned saying the services did not respond.  The only way that I have found to kill the thread is by doing a hard shutdown of the machine, if I attempt to shutdown normally the workstation just sits at the “Shutting Down” screen.


    I have checked all of the items listed in the Microsoft KB article ( and all of the programs have their own unique root directories.  The only thing I can think of is that the client is trying to re-cache the program and it is throwing this error, but I am unsure why it would be doing that or how to prevent it.



    The following is recorded in the sftlog.txt file on the client machine while attempting to restart the services:


    [08/29/2008 09:01:00.534 SRVC INF] {tid=A34}

    The SoftGrid Client Service received a stop request from the Service Control Manager and will now attempt to shut down.


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=A34}

    Stopping listener service...


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=A34}

    Attempting to close all connections


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=A34}

    Asked FEC 1954CA07-0C1A-4413-9A47-8B8BF3B2DD5A to disconnect


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=A34}

    Asked FEC 3B10D1C4-90D5-4629-A14C-803A62FF64EA to disconnect


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=A34}

    Asked DCC 6092 to disconnect


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=A34}

    Waiting for 3 more processes to disconnect


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=17A0}

    Removed DCC 6092 from list of DCCs


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=17A0}

    Got normal shutdown request from DCC 6092 (rc 00000083-00003007), cleaning up


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=17A0}

    DCC 6092 exited, doing cleanup


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=17A0}

    Terminating sfttray.exe (pid 3784) because it uses dead DCC.

    DCC (pid 6092) disconnected with error code rc 00000083-00003007


    [08/29/2008 09:01:00.534 INTF VRB] {tid=17A0}

    SWUnregisterEventHandler() with event filter 0x00000003, app filter 0xFFFFFFFF


    [08/29/2008 09:01:00.534 SRVC VRB] {tid=17A0}

    Terminating mmc.exe (pid 4848) because it uses dead DCC.

    DCC (pid 6092) disconnected with error code rc 00000083-00003007



    Then a little farther down it states.


    [08/29/2008 09:01:04.590 INTF CRT] {tid=17C0}

    Client core could not be initialized (rc 0C800F33-000006CC)




    Here is the log entry for the app error:


    [08/29/2008 10:26:12.911 JGSW ERR] {hap=13:app=Panther}

    The SoftGrid client could not connect to stream URL 'rtsp://pgsoftgrid:554/Panther.' (FS status: 16D1380A-0000E019).


    [08/29/2008 10:26:12.911 SWAP ERR] {hap=13:app=Panther}

    The client was unable to connect to a SoftGrid server (rc 16D1380A-0000E019)


     [08/29/2008 10:26:14.159 TRAY ERR] {tid=16D8:usr=Gass}

    The SoftGrid Client could not launch Panther

    An unexpected error occurred. Please report the following error code to your System Administrator.

    Error code: 42012E-16D1380A-0000E019


    [08/29/2008 10:26:16.187 AMGR VRB] {tid=141C:usr=Gass}


    Does anyone have any ideas? 



    Friday, August 29, 2008 4:33 PM


  • Well I figured out what the problem was.  My folder names when I sequenced the programs were not in the 8.3 filename convention.  The 42012E-16D1380A-0000E019 error that I was receiving makes sense now.  The naming convention I was using was the problem; I have multiple versions of the same apps that are very close versions to each other.  It would shorten the folder names and if I would try to open one of the apps at the same time as another was closing it would throw the error. That is also why it was always re-caching the application.  I would run one and it would overwrite the previous app that was cached.  I am just glad that is solved.  Maybe I can actually relax this weekend.

    • Proposed as answer by znack Monday, August 8, 2011 10:40 AM
    • Marked as answer by Aaron.ParkerModerator Sunday, November 18, 2012 5:19 PM
    Friday, August 29, 2008 9:18 PM