|
|
Problems that can occur while installing, configuring, and
running an SNA link — and how to deal with them — are
described here. Guidelines for submitting a change request (CR)
are also given.
Common Problems
Some problems are common to many situations — invalid
software installation, version incompatibilities, insufficient MPE
resources, corrupt configuration files, and file system errors.
In most cases, you can resolve these problems yourself; the resolutions
to these problems are described here. However, in some cases you
should submit an CR. To enable Hewlett-Packard to solve your problem
in an efficient manner, you should follow certain guidelines when
submitting an CR; these guidelines are listed later in this chapter.
Invalid Software Installation
A software installation might be invalid. Run NMMAINT to get
a listing of all software modules and their version IDs. Check that
all required modules are present, and that the first five characters
of the version IDs (the v.uu.ff fields) of these modules are
the same within each
subsystem. Finally, check the version IDs with the SYSTEM STATUS
BULLETIN or other HP source to be sure the versions
of software you have are supported by the version of MPE you have.
If a problem is found with any of these checks, you have an incorrect software
version installed. Locate a known valid version of the suspect software
(perhaps from one of your system backup tapes) and install it.
Version Incompatibilities
Version incompatibilities between NMS software and other subsystem software
are possible. Run NMMAINT to get a listing of version IDs for NMS
and for all of the NMS-dependent subsystems. Locate the overall version
IDs for NMS and for each subsystem. Compare the first five characters
of these version IDs (the v.uu.ff fields) with those listed
as compatible with each
other in the SYSTEM STATUS BULLETIN or other
HP source. If a discrepancy is found, locate a known set of compatible
software (perhaps from one of your system backup tapes) and install it.
Insufficient MPE Resources
The system might have insufficient MPE resources such as configured table
sizes. Use HP Glance/XL (HP 50734A) to check
MPE table utilization. Reconfigure MPE to fix any problems found
and restart the system.
Corrupt Configuration File
The configuration file might be corrupt. Validate the configuration
file to check for possible corruption. If the error persists, use
NMMGR to manually check the configuration file (if possible). If
bad records seem to be localized to a particular item, delete that
item and reconfigure it. If necessary, RESTORE a backup copy of the file.
File System Errors
A file system error might have occurred while trying to access
the configuration file. Try to access the configuration file, under
the same userid, using NMMGR. Use the NMMGR error screen to find
out what the underlying FSERR is. Correct the problem and retry.
Configuration Problems
Configuration problems can occur on both the HP e3000 and the host.
HP e3000 Configuration Problems
Listed below are some configuration problems that can occur
on the HP e3000.
PSI problems.
When you run the PUs concurrently, each PSI requires a unique PU and
each PU requires a unique PSI.
Also, if you are using an SDLC link, ensure that you correctly
configured the Physical Path field of the "SNA Node
Configuration: SDLC Link Data" screen (see
Chapter 2 "SNA Node and Link Configuration"
of this manual).
Configuration file misnamed.
Remember, when configuring a node, the configuration file can be
named anything you prefer. However, when you start the SNA link, it
looks for a configuration file named NMCONFIG.PUB.SYS.
Host Configuration Problems
Listed below are some configuration problems that can occur
on the host.
Incorrect device specified.
Too many or the wrong type of device might have been configured for a
particular line.
Incorrect macro definition.
Macro operands might have been incorrectly placed within the
hierarchical structure. Although the operands for lower-level macros
can be defined in a higher-level macro, this structure must be
followed: GROUP, LINE, SERVICE, PU,
and LU.
A macro could be missing a required operand or an invalid operand
might have been specified for a particular macro.
Installation Problems
Some common problems that might occur during installation,
or as a result of improper or unsuccessful installation, are described below.
Inactive VTAM node.
This is probably the most common problem that occurs on the host.
Ensure that the proper activation commands have been issued. Setting
the ISTATUS operand of the LU macro to
ACTIVE will help, especially if it is moved up to either
the GROUP or LINE macro. (The ACTIVE
setting sifts down to all lower-level macros.) This problem
most frequently occurs in a switched line configuration when the
major node containing the VBUILD macro was never activated.
In this case, an inactive macro causes the host to fail to find a
match for the remote's Exchange Station Identification (XID) number.
The only way you will be aware that this XID problem has happened is
a message at the host operator's console. To avoid these problems
remember this rule: When in doubt, activate.
Switched lines.
Be sure the VBUILD macros are in a separate member in the
VTAM parameter library (SYS1.VTAMLST).
Hardware problems.
Ensure that all hardware is installed properly. Many problems
are related to such things as incompatible modems, bad phone lines
and cables, etc.
Ensure that the host is up, the line is active (being polled), and
the host telecommunications subsystem (for example, VTAM) is active.
You might need to use a line monitor, such as an HP 4953, to check
whether the line is being polled and to determine the addresses being
polled. You also can use one of the other HP 495x Protocol Analyzers
in addition to the standard HP 4953.
Submitting an CR
Listed below are some guidelines for submitting a change request
(CR). Some of the guidelines refer to Node Management Services (NMS) programs
(NMDUMP, NMMAINT, and NMMGR). Using NMDUMP and NMMAINT for an SNA link is
explained earlier in this chapter. Using NMMGR for an SNA link is explained in
Chapter 2 "SNA Node and Link Configuration" of this
manual. For more NMS information, see Using the Node Management
Services Utilities.
Common Information
For any CR, include this common information, where applicable:
A characterization of the problem. Describe the events leading up
to and including the problem. Try to describe the source of the
problem. Describe the symptoms of the problem.
Your characterization should include MPE commands, communication
subsystem commands, job streams, result codes and messages, and data
that can reproduce the problem.
List the context and environment in which the message occurred.
Prepare copies of the HP e3000 system console information.
Give a brief history of your site, including when the SNA link product
was installed, the circumstances that usually produce problems,
other data communications products installed, and any other products
you were using when the problem arose.
Obtain the version, update, and fix information for all software
by running NMMAINT. This allows Hewlett-Packard to determine whether
the problem is already known and if the correct software is installed
at your site.
Record all result messages and numbers that appear at the user
terminal and the system console.
Run NMDUMP to format the NM log file
(NMLGnnnn.PUB.SYS) that was active when the problem
occurred. You might need to issue the MPE command SWITCHNMLOG
to free the NM log file (see chapter 6 of this manual). Inspect the
formatted output and try to locate errors. Prepare the formatted
output and a copy of the log file for your Hewlett-Packard
support contact to analyze.
Prepare a listing of the configuration file and the MPE
configuration you are using for your Hewlett-Packard support contact
to analyze. Inspect the output and try to locate errors.
Try to determine the general area, within the software, where you
think the problem exists. If you think the problem is caused by an
SNA service, refer to the service's Node Manager's Guide and
follow the guidelines there for gathering information for problems.
Document your interim, or "workaround," solution. The cause of the
problem can sometimes be found by comparing the circumstances in
which it occurs with the circumstances in which it does not occur.
If a system failure occurs, a full memory dump must be taken.
Save copies of all active or in-use trace files for analysis by
your Hewlett-Packard support contact.
Save copies of all active or in-use log files for analysis by your
Hewlett-Packard support contact.
Always obtain a copy of the generation of the host system you are
using for ACF/VTAM and ACF/NCP. For the host system to be compatible
with the SNA link, Hewlett-Packard sets certain operands and
parameters to required values (see Chapter 2
"SNA Node and Link Configuration" of this manual). Note any
differences between the actual system generation and your
expectations. Keep this information for your Hewlett-Packard support
contact. This is very important.
Give a copy of NMCONFIG.PUB.SYS to your Hewlett-Packard
support contact.
SNA Transport Information
For problems that involve SNA Transport, include the following additional
information where applicable:
Always issue the MPE command
SNACONTROL STATUS;NODE=NodeName
to identify the state of the PU and the LUs in the problem node.
Inspect the output to try to locate an identification of the problem.
Keep a copy of the output from this command for your Hewlett-Packard
support contact.
The SNACONTROL commands are described in Chapter 6
"MPE Commands" of this manual.
SDLC Communications Link Information
For problems that involve the SDLC communications link, include
this additional information where applicable:
Linkname = LinkName Event Type = EventType
Log Class = LogClass Location = Location
Sublocation = Sublocation Error Code = ErrorCode
State = State Cause Status = CauseStatus
Error Actions = ErrorActions
Be sure to include this error message in the CR exactly as it
appears on the operator's console.
If a PSI failure has taken place, give the PSI dump file
NETDMPnn.PUB.SYS (where nn is a
number from 00 through 99) to your Hewlett-Packard support contact
for analysis.
If the problem is easily reproducible, and link-level tracing
was inactive when the problem took place, turn on link-level tracing
with the MPE command SNACONTROL START or SNACONTROL
TRACEON (see Chapter 6 "MPE Commands"
of this manual). When the problem has been reproduced, turn off the
tracing and give the PSI dump file and this trace file to your
Hewlett-Packard support contact for analysis.
 |
NOTE: It is important to give a copy of the PSI dump that
took place while link-level tracing was enabled to your Hewlett-Packard
support contact for analysis.
|
If a PSI failure has not taken place, turn on link-level tracing
with the MPE command SNACONTROL START or
SNACONTROL TRACEON (see Chapter 6
"MPE Commands" of this manual). This allows your Hewlett-Packard
support contact to study the line activity.
If link trace logging is enabled, use the SWITCHLOG command
to close the current system log file (LOGnnnn.PUB.SYS,
where nnnn is a number from 0000 through 9999) and
open a new one. Then, give a copy of the system log file you just
closed to your Hewlett-Packard support contact to analyze.
If link trace logging is not enabled, enable it through NMMGR so
that the information will be available if the problem can be
reproduced.
X.25 Communications Link Information
For problems that involve the X.25 communications link, include
this additional information where applicable:
If a fatal X.25 link error has occurred, messages in these formats
might be logged:
Link connect failure on node NodeName
Subsystem=SubsysID, Error Code=ErrorCode
Link failure, shutting down the node NodeName
subsystem=SubsysID, procedure=ProcNum, error num=ErrorNum
SOCKERR nnn
Link failure, shutting down the node NodeName
subsystem=SubsysID, procedure=ProcNum, error num=ErrorNum
Be sure to include these error messages in the CR exactly
as they appear on the operator's console.
|