OSDCEQA - FREQUENTLY ASKED QUESTIONS FOR OS/2 DISTRIBUTED COMPUTING ENVIRONMENT (DCE) 1.00.2 01/19/96 ======================================================================= Frequently Asked Questions for OS/2 Distributed Computing Environment (DCE) 1.00.2 ======================================================================= Please Read Entire Document for Full Explanation of Procedures -------------------------------------------------------------- DESCRIPTION This document contains the questions most frequently asked of IBM Technical Support regarding the IBM OS/2 Distributed Computing Environment (DCE) 1.00.2 product. If you require assistance after reviewing the information in this document, please call 1-800-992-4777 for information about obtaining technical support from IBM. RESOLUTION Question 1 ---------- I am having problems compiling the "Greet" sample shown in the IBM DCE for OS/2: Application Developer's Guide. What should I do? Answer ------ The Guide has some errors regarding the Greet sample application. See the following for corrections. Corrections for the Greet Sample -------------------------------- Page 2-6: Make the following corrections for the Greet sample "makefile" file: * The CFLAGS line of code should show the following flags: /C+ /Q+ /Sem /J+ /Gm+ /W3 /O- /Ss+ /Ms /Su4 * The DCEFLAGS line of code should show the following symbols: -DINTEL80x86 -DIBMOS2 -D__STDC__ -DIDL_PROTOTYPES All other flags and symbols are invalid in this makefile. * In the SERVER.EXE and CLIENT.EXE lines of code, make sure you are linking to the DCEOS2 library. * In the IDLFLAGS line, use the -fat flag if you are using a FAT drive. Page 2-10: Add the following to the MANAGER.C file: include Page 2-11: In the GREET.C file, find the line of code that assigns validfamily, and make sure the protocol sequence reads NCADG_IP_UDP instead of CADG_IP_UPD. Question 2 ---------- I receive the EPFIE187 product-specific error message when using the MKDCEPM command to configure to the DCE cell. How can I fix this? Answer ------ Either MPTN device drivers are not specified correctly in the CONFIG.SYS file, or Multi-Protocol Transport Services (MPTS) did not get started with the correct options. MPTS Not Starting Correctly --------------------------- * The CONFIG.SYS file should include these device-driver statements: DEVICE=\MPTN\PROTOCOL\MPTN.SYS DEVICE=\MPTN\PROTOCOL\LIPC.SYS DEVICE=\MPTN\PROTOCOL\INET.SYS DEVICE=\MPTN\PROTOCOL\IFNDIS.SYS * The MPTSTART.CMD file should include these lines: START D:\MPTN\BIN\CNTRL.EXE /P MPTN_OS$ MPTN_IN$ INETWAIT CALL D:\TCPIP\BIN\SETUP.CMD CALL D:\TCPIP\BIN\TCPSTART.CMD Recycle MPTS after making changes to the MPTSTART.CMD file. If you change the CONFIG.SYS file, you need to restart your system. Question 3 ---------- When I am using the MKDCEPM command to configure to the DCE cell, my OS/2 DCE client receives the following message: Cannot authenticate. What should I do? Answer ------ Check the last message of the MKDCE.LOG file in the root directory of the drive where DCE is installed: * If the message is "Password validation failure--Invalid password (dce / sec)," the password is incorrect. Passwords are case-sensitive. * If the message is "Password validation failure--Clock skew too great (dce / krb)," the client's time is not within a 5-minute window with the security server. You also should check the time-zone environment variable on the client and server to make sure they are synchronized. * If the message is "User identification failure--Registry object not found (dce / sec),", the specified cell name is wrong. Cell names are NOT case-sensitive; however the cell name must be exact (for example, dot-qualified). Question 4 ---------- When using the MKDCEPM command to configure to the D-C-E cell, my OS/2 DCE client receives the error message, "Unable to determine local hostname and I-P address," or the window that states "Initializing. Please Wait" locks up. How can I fix this error? Answer ------ * Make sure MPTS is started by verifying that CNTRL.EXE is running. * Make sure the system has a HOSTS or a RESOLV file in the subdirectory assigned to the ETC environment variable. Type SET ETC to check the assignment of this variable. The system looks for the RESOLV file first. It contains the nameserver's domain name and IP address. If the RESOLV file does not exist, the system looks for the HOSTS file. This file contains the IP addresses and host names of all the systems the local system intends to communicate with when not using the nameserver. Make sure there is a Carriage Return after each entry; otherwise, the system cannot interpret the data correctly. If the RESOLV file exists and you are experiencing a system lockup condition, the nameserver could be down. If neither of these files exists, you need to create one of them. * If you have more than one network adapter in your local system, make sure you disable one of them to avoid this error message. Question 5 ---------- When configuring my DCE Windows client to the DCE cell, the following message is received: Unable to create binding handle. What causes this? Answer ------ The RPC daemon did not start up successfully. Select the RPCD icon in the IBM DCE Client for Windows Program group. Question 6 ---------- When trying to start the RPCD daemon, my DCE Windows client receives the following error message: Can't set EPT object type: (90X16C9A03A) invalid object (DCE/RPC). What should I do? Answer ------ Make sure your system's hardware address is specified when configuring your Windows client to the DCE cell. Select PCDCECFG and then Advanced to make sure the Hardware Address field is filled in. Question 7 ---------- When compiling a DCE application, I receive the error message: Object Reference Already in Use. How do I solve this? Answer ------ Make sure that you are using the following DCE compiler flags: -DINTEL80x86 -DIBMOS2 -D__STDC__ -DIDL_PROTOTYPES Also make sure you are using the following C flags: /C+ /Q+ /Sem /J+ /Gm+ /W3 /O- /Ss+ /Ms /Su4 Question 8 ---------- When attempting to start a client application, my OS/2 DCE client receives the error message: An exception raised address of exception object $08X(DCE/THD). How do I fix this error? Answer ------ 1. Make sure that you set your rpc_default_entry environment variable if you are using automatic bindings. 2. Make sure you have the latest version of OS/2 DCE 1.00.2. You can determine this by typing SYSLEVEL and pressing Enter at a command prompt. 3. Make sure you have OS/2 2.11. If you are using OS/2 2.1, you need to order the PJ09481 fix for a DOS thread problem. 4. If your client is using string bindings to connect to the server, make sure you are using the correct port number of the appropriate server. For example, the client issues client ncadg_ip_upd:129.35.44.62{4065} to use string bindings instead of using the CDs to find out where the server has registered its endpoint map. Question 9 ---------- When attempting to start up the server code while using automatic bindings, I receive the following error message: Entry not found (DCE/RPC) How can I fix this? Answer ------ You have not set your rpc_default_entry environment variable. For example, if your entry point is assigned /.:/sample, issue: set RPC_default_ entry=/.:/ sample Note that the entry point is case-sensitive. ______________________________________________________________________ IBM disclaims all warranties, whether express or implied, including without limitation, warranties of fitness and merchantability with respect to the information in this document. By furnishing this document, IBM grants no licenses to any related patents or copyrights. Copyright (c) 1994, 1995 IBM Corporation. Any trademarks and product or brand names referenced in this document are the property of their respective owners. Consult your product manuals for complete trademark information.