[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference microw::acmsxp

Title:ACMSxp product questions and comments
Notice:Refer to notes 1 through 11 for conference information
Moderator:DUCAT::ROSCOE
Created:Tue Oct 05 1993
Last Modified:Thu Jun 05 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:282
Total number of notes:1134

276.0. "TXN-FAILURE 103 ?" by UTRTSC::DORLAND (The Wizard of Odz2) Wed Apr 23 1997 10:10

    Trying to get a simple example running I constantly
    get the message 
    
    TXN-FAILURE-ERROR, Error code 103
    
    When running the client program.
    
    I have a feeling that the real Server code is never called,
    that is, when invoking the processing server it immediately
    returns this error. There is nothing logged in the
    processing server log file. Only the default log shows the error,
    see below.
    
    ACMSXP 3.0, RDB 6.1, Alpha VMS 7.1. All code (clients, servers, DCE)
    runs on the same machine.
    
    Any ideas?
    
    Thanks in advance,
    
    Ton.
    
    
    
    
    Acms List log employee_default.log 
    
    TIME:  1997-04-23-11:33:06.059
POSTED BY:  Server Process
PID:  24e0244f
TYPE:  fault
EXECUTION CONTEXT INFORMATION:  
    TPSystem:  /.:/tdorl/employee
    Server:  employee_task_grp
    Interface:  employee_task_grp
    Device:  _UTURBO$VTA111:
    Request ID:  974c898b-bbbc-11d0-8747-08002be627da
    PRINCIPAL:  /.../utrtsc_cell/tdorl
    Procedure:  EMPLOYEE_INFO_ADD_TASK
SEVERITY:  error

vms MESSAGE TEXT:  system service failure exception, status=!xl, pc=!xh, ps=!xl
tps MESSAGE TEXT:  %TPS-F-TXFAILURE, Fatal error during transaction processing
T.RTitleUserPersonal
Name		DateLines
276.1CAMINO::ROSCOEWed Apr 23 1997 23:0817
    The way to find out how far a call has gone is to enable tracing for
    both the task and processing servers.  You do this by modify the 
    event filter attribute for the servers to include tracing.  for example
    acmsadmin> set tpsys /.:/tdor1/employee server employee_task_grp event
    filter = (fault,security,status,trace).  Do the same thing for the
    processing server as well.   You can then re-run the test, you do not
    have to restart the servers for the new tracing options to activate.
    
    I would guess that the problem is being caused by either decdtm,tcp/ip
    or decnet.  Check and make sure that all are functioning properly.  
    
    If everything checks out then issue the acms list log command again
    on the tpsystem default log file.  Before doing so define the logical
    acms_source_trace to true.  This logical will tell us the line number
    of the source code module that is failing. 
    
    Rich
276.2Problem in TPS_TXN_TM_SERVICESUTRTSC::DORLANDThe Wizard of Odz2Thu Apr 24 1997 07:5663
    	Thanks for your reply, this indeed produces more info.
    	However, the problem seems to be in a file TPS_TXN_TM_SERVICES.C,
    	but I cannot find this file.
    
    	Any suggestions?
    
    	Ton
    
********************************************************************************
USER:  DORLAND
TIME:  1997-04-24-09:25:26.810
POSTED BY:  Server Process
PID:  24e0271b
TYPE:  trace
EXECUTION CONTEXT INFORMATION:  
    TPSystem:  /.:/tdorl/employee
    Server:  employee_task_grp
    Interface:  employee_task_grp
    Device:  _UTURBO$VTA111:
    Request ID:  ec744170-bc73-11d0-93eb-08002be627da
    PRINCIPAL:  /.../utrtsc_cell/tdorl
    Procedure:  EMPLOYEE_INFO_ADD_TASK

Entered Function EMPLOYEE_INFO_ADD_TASK in Group employee_task_grp.

********************************************************************************
USER:  DORLAND
TIME:  1997-04-24-09:25:26.871
POSTED BY:  Server Process
PID:  24e0271b
TYPE:  fault
EXECUTION CONTEXT INFORMATION:  
    TPSystem:  /.:/tdorl/employee
    Server:  employee_task_grp
    Interface:  employee_task_grp
    Device:  _UTURBO$VTA111:
    Request ID:  ec744170-bc73-11d0-93eb-08002be627da
    PRINCIPAL:  /.../utrtsc_cell/tdorl
    Procedure:  EMPLOYEE_INFO_ADD_TASK
SEVERITY:  error
SOURCE LINE:  2590
SOURCE FILE:  TPS_TXN_TM_SERVICES.C

vms MESSAGE TEXT:  system service failure exception, status=!xl, pc=!xh, ps=!xl
tps MESSAGE TEXT:  %TPS-F-TXFAILURE, Fatal error during transaction processing

********************************************************************************
USER:  DORLAND
TIME:  1997-04-24-09:25:26.879
POSTED BY:  Server Process
PID:  24e0271b
TYPE:  trace
EXECUTION CONTEXT INFORMATION:  
    TPSystem:  /.:/tdorl/employee
    Server:  employee_task_grp
    Interface:  employee_task_grp
    Device:  _UTURBO$VTA111:
    Request ID:  ec744170-bc73-11d0-93eb-08002be627da
    PRINCIPAL:  /.../utrtsc_cell/tdorl
    Procedure:  EMPLOYEE_INFO_ADD_TASK

Exited from Function EMPLOYEE_INFO_ADD_TASK in Group employee_task_grp.
276.3CAMINO::ROSCOEThu Apr 24 1997 12:0513
    Hi,
    
    The source code module is one specific to ACMSxp the product.  The
    problem is happening on a DECdtm system service call. I suspect that 
    the problem is in the start transaction or join call.  The call is 
    failing before we get to the processing server.  Post the value of
    the logical sys$node also post the value of scssystemid and scsnode.
    Can you do a set host 0 on your box? 
    
    Rich
    
    ps.  I 'm in a class today, so I will not be able to reply again until
    tonite.
276.4Will try DecDTM...UTRTSC::DORLANDThe Wizard of Odz2Thu Apr 24 1997 14:4411
    Looks as if you are right. The processing server code is never called.
    SYS$NODE and SET H 0 are no problem. (We are running DecNet Ph 5.).
    However, looking with LMCP there is no transaction logged. TPSERVER
    is up and running. 
    
    I think I'll try an example with SYS$Transaction calls, maybe this will
    help us ...
    
    If I have more answers/questions I'll post them here. 
    
    	Rgds, Ton