OSI/MHS Configuration and Management Manual

Troubleshooting Your OSI/MHS Subsystem
OSI/MHS Configuration and Management Manual—424827-003
9-52
Interconnection Problems With Gateways
Interconnection Problems With Gateways
Several types of problems can arise when you attempt to connect OSI/MHS with a
gateway or during the subsequent operation of a gateway. Some of these problems
result from application programming errors or incompatibilities; those types of errors,
and the means for correcting them, are described in the OSI/MHS Gateway
Programmatic Interface (GPI) Reference Manual and the OSI/MHS Gateway
Programmatic Interface (GPI) Programming Guide. Other types of errors result from
configuration problems, resource shortages, or process failures; this section gives a
few examples.
The following are three common gateway interconnection problems:
•
The gateway name in the association request does not match the expected value
(that is, the name specified in the OSI/MHS configuration). In this case, an error is
returned to the gateway application. To solve the problem, you check the
GATEWAY object definition and ensure that the name matches the name in the
request.
•
A necessary process or set of processes, such as the MHSMGR, an MR group, or
an entry or wait manager, is not running. Either the process was never started, or
it failed and was not restarted, or the process that was started did not have the
same name specified in the OSI/MHS configuration. An example below shows
how to recognize and solve this problem.
•
The amount of memory requested by the gateway application, or the amount of
space available on the designated swap volume, was insufficient for the number
and size of messages to be transferred across the gateway. An example below
shows how to recognize and solve this problem.
Problem: Missing Entry or Wait Manager Process
The absence of a required process is reported to a GPI application through status
codes and to the operations environment through an event message. If the process
name was specified in the OSI/MHS configuration but is not currently running, the
status value returned by a GPI library call is OM_RC_PERMANENT_ERROR, and the
primary and secondary codes returned by the call to GPI_STATUS_ are
primary code = ZGPI_RC_OPEN_ERROR (if an open call), or
= ZGPI_RC_WRITEREADX_ERROR (during session)
secondary code = 14 (file system error:device does not exist)
The event message is
ZMHS-EVT-IO-ERR - event message 12
The file name in the message identifies the process that the Gateway Interface
Process (GIP) tried to use. The name is the one specified in your OSI/MHS
configuration. Most likely, the process was defined in the configuration but was not
started, or was stopped or suspended after having been opened. To solve this