Ingres Processes on VMS

The major Ingres processes are as follows:

• Name Server process (iigcn)
• Communications Server process (iigcc)
• Recovery process (II_IUSV_nnn)
• Archiver process (dmfacp)
• DBMS Server process (iidbms)
• Data Access Server process (iigcd)
• Bridge Server process (iigcb)
• ICE Server process (icesvr)

Remote Command process (rmcmd)

Name Server Errors on VMS

The Name Server (II_GCN) is not running if the following occurs:

• You receive a specific error indicating that the Name Server process failed to start.
• The VMS command show system does not list the II_GCN process, indicating that the Name Server is not running.

  Note: Only a privileged user can start up the Name Server.

Check for Name Server Errors on VMS

If the Name Server does not start, follow these steps:

1. Examine the II_CONFIG:errlog.log file for any additional information.
2. To isolate a running Name Server as the problem on failed local connections:
   1. Define the logical II_DBMS_SERVER to point to the DBMS Server process name or mailbox.
   2. If the connection now succeeds, stop and start the Name Server. For additional information about the iinamu command, see the Command Reference Guide.
3. If the Name Server continues to fail, contact technical support, as described in What You Need Before Contacting Technical Support.

Communications Server Process on VMS

The Communications Server (II_GCC_nnn, where the nnn is a number related to the server's process ID) does not start if any of the following errors occur:

• The II_CONFIG:errlog.log file has a notation regarding the failure of the "net server" (Communications Server) process starting.
• The show system command does not list an II_GCC_nnn process.
• The utility iinamu does not display a process name when you enter the show comsvr command at the IINAMU> prompt.

Check for Communications Server Process Errors on VMS

If the Communications Server does not start, follow these steps:

1. Examine the II_CONFIG:errlog.log file for any information regarding Communications Server startup.
2. Make sure that the network protocol software on which the Communications Server executes is started and functioning before attempting to start Ingres Net processes. By default, an attempt is made to open ports for the DECnet and TCP_DEC protocols.
3. If the Communications Server continues to fail to start, contact Technical Support.

Bridge Server Errors on VMS

The Bridge Server process (II_GCB_nnn, where the nnn is a number related to the server's process ID) does not start if the following errors occur:

• The II_CONFIG:errlog.log file has a notation regarding the failure of the Bridge Server process.
• The show system command does not list a II_GCB_nnn process.
• The utility iinamu does not display a process name when you enter the show bridge command at the IINAMU> prompt.

Check for Bridge Server Process Errors on VMS

If Bridge Server does not start, follow these steps:

1. Verify that you have installed the Ingres Net component.
2. Verify that you have installed the correct protocol drivers and they are functioning properly.
3. Verify the Bridge Server has been properly configured using the Configuration-By-Forms (cbf) utility. If you are still having problems, set the following trace to capture additional diagnostic data before calling technical support:

   define II_GC_TRACE 4

   define II_GC_LOG filename

   ingstart -iigcb

Recovery Process Errors on VMS

The recovery process (II_IUSVxx_nnn for stand-alone Ingres, or DMFRCPnnn-xx for Ingres Cluster Solution) must be running before a DBMS server can be started. Failure of the recovery process to start indicates one of the following:

• Improper installation configuration. Check Ingres Installation on VMS.
• Problems with the log file
• Insufficient system memory (NPAGEDYN sysgen parameter)

Check for Recovery Process Errors on VMS

If the recovery process does not start, perform the following procedure.

1. Examine the following files for specific reasons for the startup failure:

   II_SYSTEM:[INGRES]II_CONFIG:errlog.log
   II_CONFIG:IIRCP.LOG

   (If your system is configured for the Ingres Cluster Solution, the suffix _nodename is added.)

2. If configured for Ingres Cluster Solution, check that DECnet is up among all nodes. Ingres uses DECnet to communicate among nodes. Check that you can set host and log into/from all nodes running Ingres.
3. Be sure that the log file location is correctly defined:
   1. Check the definition of the II_LOG_FILE logical.
   2. Check the II_LOG_FILE location for the existence of the log file:

      II_LOG_FILE:[INGRES.LOG]INGRES_LOG.SYSTEM_nodename

4. If you still cannot start the recovery process, the problem may be a corrupted log file. Contact technical support, as described in What You Need Before Contacting Technical Support.

Remote Command Process Errors on VMS

The Remote Command Server (RMCMD_nnn, where the nnn is a number related to the server's process id) does not start, if show system command does not list an RMCMD_nnn process.

Check for Remote Command Server on VMS

Use the following procedure if the RMCMD process did not start:

1. Ensure that the PRCLM for the system administrator is sufficient to start this additional process. This value must be a minimum of 15.
2. Check II_SYSTEM:[ingres.bin]RMCMD.log for errors.

Archiver Process Errors on VMS

The Archiver process (DMFACP or DCMFACPxx) does not start if any of the following conditions occur:

• Unless the recovery process is running; however, an installation runs without an archiver process until the log file fills up. User programs are suspended until the archiver is started.
• When the recovery system is backing out aborted transactions from a system failure. Recovery Process Monitoring can help you determined whether Ingres is in a recovery state.
• Due to improper memory resources, inability to read the transaction log file, or inability to write journal files.
• You receive an error indicating that the Archiver process failed to start.
• The VMS command show system does not list the Archiver process.

Check for Archiver Process Errors on VMS

Use the following procedure to check Archiver process startup problems:

1. Examine the II_CONFIG:errlog.log startup log file for information.
2. Examine the IIACP.LOG_nodename file for possibly specific reasons for the startup failure. Take note of which databases are referenced in the IIACP.LOG_nodename file.
3. Check that the II_JOURNAL location is a valid location:
   1. Check the definition of the II_JOURNAL logical.
   2. Check the II_JOURNAL location. Type the following command at the operating system prompt:

      directory II_JOURNAL:[INGRES.JNL]

4. Check that the disk containing the journal files is not 100% full. This makes it impossible to write journals and stops the Archiver.

   If the journal disk became full as a result of the Archiver writing journal information, an error appears in the errlog.log stating a "device full" condition.

   1. After making disk space available, restart the Archiver using the ingstart -dmfacp command.
   2. Optionally, modify the acpexit script to take whatever measures are necessary to notify you of this situation as soon as it occurs. For example, you can modify the script so that it sends you mail when the Archiver encounters an error.
5. If the journal files have been corrupted, the errlog.log reports any problems with the Archiver writing information to the journal files.

If there are corrupted files, or the Archiver process continues to fail, contact technical support, as described in What You Need Before Contacting Technical Support.

Check for DBMS Server Process Errors on VMS

If the DBMS server (II_DBMS_nnn, where the nnn is a number related to the server's process ID) does not start, follow these steps:

1. Examine the II_CONFIG:errlog.log file for any information regarding DBMS Server startup.

   Contact Technical Support to examine your authorization string.

2. If there is a DBMS Server startup failure reported in the II_CONFIG:errlog.log, but no error reported in the errlog.log:
   1. Check for the existence of any MBAnnn processes on the system that are linked to the IIRUNDBMS.EXE. Examine the MBAnnn process by issuing the following command at the operating system prompt:

      show proc/id = pid /cont

      where:

      pid is the process ID of the MBAnnn process.

   2. If the MBAnnn process is linked to the IIRUNDBMS.EXE, free the IIRUNDBMS.EXE by issuing the following command at the operating system prompt:

      stop /id = pid

      where:

      pid is the process ID of the MBAnnn process.

   3. Check the errlog.log for any new messages relating to DBMS Server startup failure.
3. Try starting a server manually, using the ingstart -iidbms command.
4. If the DBMS Server continues to fail to start, contact Technical Support.

Data Access Server Errors on VMS

The Data Access Server (II_GCD_nnn, where the nnn is a number related to the server's process ID) does not start if any of the following conditions occur:

• The II_CONFIG:errlog.log file has a notation regarding the failure of the Data Access Server process.
• The command show system does not list a II_GCD_nnn process.
• The utility iinamu does not display a process name when you enter the command show dasrv at the IINAMU> prompt.

Check for Data Access Server Errors on VMS

Use the following procedure if the Data Access Server process did not start:

1. Verify that you have installed the Ingres Net component.
2. Verify that you have installed the correct protocol drivers and they are functioning properly.
3. Verify the Data Access Server has been properly configured through the Configuration-By-Forms (cbf) utility.
4. If you are still having problems, set the following trace to capture additional diagnostic data before calling technical support:

   define II_GCD_TRACE 4

   define II_GCD_LOG filename

   ingstart -iigcd