Chapter 71 OSIAS (OSI/Application Services) Messages

This chapter describes the messages sent by the Open Systems Interconnection/ Application Services (OSI/AS) subsystem. Within this subsystem, both OSI manager and application, presentation, and session (TAPS) service provider processes generate operator messages. The subsystem ID displayed by these messages includes OSIAS as the subsystem name.

Each message begins with the subject—that is, the name of the process, subdevice, or service about which the error is concerned. The subject process-name is the process name of an OSI/AS process; the subject indirect-process-name is an indirect process name that begins with the OSI manager process name. For more information about these two forms of process name, see the OSI/AS SCF Reference Manual.




	NOTE: Negative-numbered messages are common to most subsystems. If you receive a negative-numbered message that is not described in this chapter, see Chapter 15.

-1

process-name, CPU switched from old-cpu to cpu { Operator Request | Takeover | Hardware | Unknown }

`process-name`	specifies the process name of the OSI/AS process associated with the message.
`old-cpu`	specifies the processor where the primary process ran before the switch. It can have a value from 0 through 15.
`cpu`	specifies the new processor where the primary process is running. It can have a value from 0 through 15.

Cause The backup of the specified process has taken over from the primary for one of the following reasons:

An operator requested the change.
A takeover occurred because the primary process stopped or abended, or because the primary processor went down.
Some other hardware problem occurred.
The cause is unknown.

Effect The backup process takes over for the primary process, with the following possible results:

If the old processor is available, the new primary process starts a backup process.
If the switch occurred because of a processor failure, the backup is started after you reload the old processor or designate another backup processor using the Subsystem Control Facility (SCF) ALTER PROCESS command.
If the primary process does not stop normally, all OSI/AS connections are lost and any files being written at the time of the takeover are left in an indeterminate state. The application processes are notified of the condition.
If the cause is unknown, the effect on the subsystem is also unknown.

Recovery If an operator requested the change, check with the other operations personnel at your installation for instructions.

If a hardware problem (other than processor down) is indicated, contact the network administrator at your installation for further analysis of the problem.

If the backup processor is down and you want to create another backup process, first either reload the old processor or stop the new primary process and designate a new backup processor using the SCF ALTER PROCESS command.

Then follow these steps:

Examine files that were being written to see whether they are usable or salvageable.
Delete unusable files.
Perform recovery of files, if possible and as appropriate.
Rerun applications as required.

-3

process-name, STATE switched from old-objstate to objstate { Operator Request | Unknown }

`process-name`	specifies the process name of the OSI/AS process associated with the message.
`old-objstate`	specifies the former summary-state of an object following a state change.
`objstate`	specifies the current summary-state of an object following a state change.

Cause This message is caused by a Subsystem Control Facility (SCF) command or an internal error.

Effect The state of the process changes.

Recovery Corrective action depends on the previous and current summary‑state and other messages that precede or follow this one. In some cases, no corrective action is needed.

process-name, INTERNAL ERROR: File: objectfile, Timestamp: bindtimestamp, Procedure: entrypointlabel, P=pregister

`process-name`	specifies the process name of the OSI/AS process associated with the message.
`objectfile`	specifies the program file name of the OSI manager or TAPS process.
`bindtimestamp`	specifies the BIND timestamp of the OSI manager or TAPS object file.
`entrypointlabel`	specifies the internal procedure entry-point label of the trapping procedure.
`pregister`	specifies the contents of the P register at the time of the trap.

Cause OSI/AS has detected an unrecoverable internal error.




	NOTE: This error is designated by the subsystem as critical, because it warrants immediate attention if normal processing is to continue.

Effect The specified process abends; the backup process takes over as the new primary process and attempts to start a new backup process. However, this error indicates a software problem, and the continued running of the software could significantly damage the files being accessed.

Recovery If this error occurs, you need to seek additional help. Follow the standard procedures at your site for contacting the appropriate support personnel. If these procedures involve contacting your HP service provider, provide all relevant information as follows:

Descriptions of the problem and accompanying symptoms
Error message codes
Details from the message or messages generated
Saveabend file of the OSI manager process
Supporting documentation such as Event Management Service (EMS) logs, trace files, and a processor dump, if applicable

If your local operating procedures require contacting the Global Mission Critical Solution Center (GMCSC), supply your system number and the numbers and versions of all related products as well.

For additional information on gathering support documentation, see the chapter on troubleshooting in the OSI/AS Configuration and Management Manual.

process-name, TRAP #trapnumber: File: objectfile, Timestamp: bindtimestamp, Procedure: entrypointlabel, signalnumber: signalnumber

`process-name`	specifies the process name of the OSI/AS process associated with the message.
`trapnumber`	specifies the trap number.
`objectfile`	specifies the program file name of the OSI manager or TAPS process.
`bindtimestamp`	specifies the BIND timestamp of the OSI manager or TAPS object file.
`entrypointlabel`	specifies the internal procedure entry-point label of the trapping procedure.
`signalnumber`	specifies the signal number.

Cause OSI/AS detected a trap caused by a hardware or software error.




	NOTE: This error is designated by the subsystem as critical, because it warrants immediate attention if normal processing is to continue.

Descriptions of the problem and accompanying symptoms
Error message codes
Details from the message or messages generated
Saveabend file of the OSI manager process
Supporting documentation such as Event Management Service (EMS) logs, trace files, and a processor dump, if applicable

If your local operating procedures require contacting the Global Mission Critical Solution Center (GMCSC), supply your system number and the numbers and versions of all related products as well.

For additional information on gathering support documentation, see the chapter on troubleshooting in the OSI/AS Configuration and Management Manual.

process-name, BACKUP UP in CPU cpu.

`process-name`	specifies the process name of the OSI/AS process associated with the message.
`cpu`	specifies the processor number in which the backup process is running.

Cause The primary process has successfully started the backup in the specified processor.

Effect Processing continues normally.

Recovery Informational message only; no corrective action is needed.

process-name, BACKUP Down: Reason: { Backup Stopped | Backup Abended | Backup CPU cpu Down | Unknown }

`process-name`	specifies the process name of the OSI/AS process associated with the message.
`cpu`	specifies the backup processor number that went down.

Cause The primary process detected that its backup process was not present for one of the following reasons:

The backup process stopped or abended.
The backup processor went down.
The cause is unknown.
NOTE: This error is designated by the subsystem as critical because it warrants immediate attention if normal processing is to continue.

Effect The primary process tries to create a new backup process unless the designated backup processor is unavailable. If the backup processor is down, the primary process waits until the processor is reloaded or until another backup processor is designated. If the cause is unknown, the effect on the subsystem is also unknown.

Recovery If the backup processor is down and you want to create another backup process, either reload the old processor or stop the primary process and designate a new backup processor using the Subsystem Control Facility (SCF) ALTER PROCESS command.

process-name, Unable to Create BACKUP Process, PROCESS_CREATE_ Error=%error, File=objectfile, Swapvol=swapvol, Priority=priority, CPU=cpu, Name: processname

`process-name`	specifies the process name of the OSI/AS process associated with the message.
`error`	specifies the error that occurred.
`objectfile`	specifies the program file name of the OSI manager or TAPS process.
`swapvol`	specifies the data swap-vol name requested for the backup process.
`priority`	specifies the priority requested for the backup process.
`cpu`	specifies the processor number requested for the backup process.
`processname`	specifies the new process name.

Cause The specified primary process was unable to create a backup process because of a process creation error, possibly caused by the backup processor going down. The message displays the error returned by the PROCESS_CREATE_ procedure, followed by the object-file name, swap-volume name, priority, processor number, and process name that were passed to PROCESS_CREATE_.




	NOTE: This error is designated by the subsystem as critical because it warrants immediate attention if normal processing is to continue.

Effect If the backup processor is down, the primary process tries again when the processor is reloaded. Otherwise, the message indicates that the primary process has exhausted the retry count for starting the backup process.

Recovery For information on the specified error, see Appendix D. Perform the appropriate corrective actions for that error. If the backup processor is down and you want to create another backup process, either reload the old processor or stop the primary process and designate a new backup processor using the Subsystem Control Facility (SCF) ALTER PROCESS command.

process-name, CHECKPOINT FAILURE, Error=proc-error

`process-name`	specifies the process name of the OSI/AS process associated with the message.
`proc-error`	specifies the file-system procedure that reported the error.

Cause A checkpoint I/O operation from the primary process returned a file-system error because of insufficient resources or an internal logic error. If this error occurs frequently, it is probably caused by an internal logic error.




	NOTE: This error is designated by the subsystem as critical, because it warrants immediate attention if normal processing is to continue.

Effect The primary process stops the backup process and tries to start a new one.

Recovery See Appendix B, for further information on the specified error. If a recoverable file-system error occurred, perform appropriate actions to correct the error.

If you suspect an internal logic error, you need to seek additional help. Follow the standard procedures at your site for contacting the appropriate support personnel. If these procedures involve contacting your HP service provider, provide all relevant information as follows:

Descriptions of the problem and accompanying symptoms
Error message codes
Details from the message or messages generated
Saveabend file of the OSI manager process
Supporting documentation such as Event Management Service (EMS) logs, trace files, and a processor dump, if applicable

If your local operating procedures require contacting the Global Mission Critical Solution Center (GMCSC), supply your system number and the numbers and versions of all related products as well.

For additional information on gathering support documentation, see the chapter on troubleshooting in the OSI/AS Configuration and Management Manual.

indirect-process-name, Process Creation Error, PROCESS_CREATE_ Error=%error, File=objectfile, Swapvol=swapvol, Priority=priority, CPU=cpu, Name: processname

`indirect-process-name`	specifies the indirect process name of the process whose creation failed.
`error`	specifies the error that occurred.
`objectfile`	specifies the program file name of the OSI manager process.
`swapvol`	specifies the data swap-vol name requested for the process.
`priority`	specifies the priority requested for the process.
`cpu`	specifies the processor number requested for the process.
`processname`	specifies the new process name.

Cause A call to the PROCESS_CREATE_ procedure made by the OSI manager process failed to start one of the subsystem processes because of the specified process creation error. The message displays the error returned by the PROCESS_CREATE_ procedure, followed by the object-file name, swap-volume name, priority, processor number, and process name that were passed to PROCESS_CREATE_.




	NOTE: This error is designated by the subsystem as critical, because it warrants immediate attention if normal processing is to continue.

Effect The OSI manager process continues to function, but the process that could not be created remains in the STOPPED state.

su-name, State Machine Error: SMid=sm-id, State=sm-state, Event=sm-event, Action=sm-action

`su-name`	specifies the name of the subdevice that caused the error.
`sm-id`	specifies an internal event.
`sm-state`	specifies an internal event.
`sm-event`	specifies an internal event.
`sm-action`	specifies an internal event.

Cause OSI/AS detected an internal state-machine error.

Effect The connection abends. Any file being written is left in an unknown state.

Recovery Retry the connection. If a retry is unsuccessful, you need to seek additional help. Follow these steps:

Examine files that were being written to see whether they are usable or salvageable.
Delete unusable files.
Perform recovery of files, if possible and as appropriate.
Follow the standard procedures at your site for contacting the appropriate support personnel. If these procedures involve contacting your HP service provider, provide all relevant information as follows:
- Descriptions of the problem and accompanying symptoms
- Error message codes
- Details from the message or messages generated
- Saveabend file of the OSI manager process
- Supporting documentation such as Event Management Service (EMS) logs, trace files, and a processor dump, if applicable
  If your local operating procedures require contacting the Global Mission Critical Solution Center (GMCSC), supply your system number and the numbers and versions of all related products as well.
  For additional information on gathering support documentation and other troubleshooting techniques and suggestions, see the chapter on troubleshooting in the OSI/AS Configuration and Management Manual.
After the problem has been resolved, rerun applications as required.

su-name, Initial L4 subdevice tsp-su-name access failed: File Error=error

`su-name`	specifies the name of the subdevice that caused the error.
`tsp-su-name`	specifies the name of the transport service provider (TSP) subdevice that caused the event.
`error`	specifies the error that occurred.

Cause The OSI/AS session service could not initiate communication with the specified transport subdevice because of the specified file-system error. This message is caused by one of the following:

The TSP process may not be in the STARTED state.
The TSP process may not be allowing open requests.
There may be resource limitations in the file system; TSP process; or application, presentation, and session (TAPS) process.
There may be internal problems.

Effect No connection is established.

Recovery See Appendix B, for further information on the specified error. Perform appropriate actions to correct the error.

If the TSP process is not started, then start the TSP process and try to reconnect.

If the TSP process is not allowing opens, issue the Subsystem Control Facility (SCF) ALLOWOPENS command and try to reconnect.

If OSI/AS or OSI Transport Services (OSI/TS) is encountering resource limitations, the problem can be corrected by reconfiguring the subsystem to spread the load across several processes.

If a temporary resource limit is reached, try to reconnect after the crisis has passed.

service-name, ACSE Threshold for { ACPM Rejects Received } { ACPM Rejects Sent } { Assoc P Rejects } { ACPM Aborts Received } { ACPM Aborts Sent } { AP Aborts } Counter reached, Threshold=value, Last Reported=time

`service-name`	specifies the service object associated with the message.
`value`	specifies the predefined threshold value that was reached.
`time`	specifies the time the threshold was last reached and reported.

Cause An Association Control Service Element (ACSE) network-management counter has reached a predefined threshold. ACSE network-management counters are the modulo threshold values specified in the Subsystem Control Facility (SCF) or Subsystem Programmatic Interface (SPI) ALTER SERVICE command for the ACSE service. An ACSE counter is incremented each time the ACSE service detects the occurrence associated with that counter—for instance, the receipt of an ACPM reject protocol data unit (PDU) from the remote entity. This error might indicate a problem, or it might merely indicate that the threshold was set too high.

Effect No direct effect.

Recovery You can monitor the frequency of this message for an indication that a problem exists. You can perform a trace of the application, presentation, and session (TAPS) process using the ACSE and ASN.1 selection options to investigate the PDU structure.

If no problem is indicated, thresholds can be adjusted by the OSI/AS subsystem administrator to increase or decrease the frequency of error messages or to completely suppress the generation of this error message.

service-name, Presentation Threshold for { Temporary CPR PPDUs Received } { Permanent CPR PPDUs Received } { Permanent CPR PPDUs Sent } { Protocol Error ARP PPDUs Sent } Counter reached, Reason code=reason-code, Threshold=value, Last Reported=time

`service-name`	specifies the service object associated with the message.
`reason-code`	specifies the reason code that caused the threshold to be reached.
`value`	specifies the predefined threshold value that was reached.
`time`	specifies the time the threshold was last reached and reported.

Cause A Presentation Layer network-management counter has reached a predefined threshold. Presentation Layer network-management counters are the modulo threshold values specified in the Subsystem Control Facility (SCF) or Subsystem Programmatic Interface (SPI) ALTER SERVICE command for the presentation (L6) service. A Presentation Layer counter is incremented each time the presentation service detects the occurrence associated with that counter—for instance, the sending of a permanent CPR-PPDU from the remote entity. This error might indicate a problem, or it might merely indicate that the threshold was set too high.

The values of the reason code are described in International Standards Organization (ISO) 8823 (1988), Information Processing Systems—Open Systems Interconnection—Connection Oriented Presentation Protocol Specification.

Effect None.

Recovery Check the reason code. You can monitor the frequency of this message for an indication that a problem exists. You can perform a trace of the application, presentation, and session (TAPS) process using the L6 selection option to investigate the protocol data unit (PDU) structure.

service-name, Session Threshold for { Permanent Refuse SPDUs Received } { Permanent Refuse SPDUs Sent } { Temporary Refuse SPDUs Received } { Protocol Error Abort SPDUs Sent } Counter reached, Reason code=reason-code, Threshold=value, Last Reported=time

`service-name`	specifies the service object associated with the message.
`reason-code`	specifies the reason code that caused the threshold to be reached.
`value`	specifies the predefined threshold value that was reached.
`time`	specifies the time the threshold was last reached and reported.

Cause A Session Layer network-management counter has reached a predefined threshold. Session Layer network-management counters are the modulo threshold values specified in the Subsystem Control Facility (SCF) or Subsystem Programmatic Interface (SPI) ALTER SERVICE command for the session (L5) service. A Session Layer counter is incremented each time the session service detects the occurrence associated with that counter—for instance, the receipt of a permanent refuse session protocol data unit (SPDU) from the remote entity. This error might indicate a problem, or it might merely indicate that the threshold was set too high.

The values of the reason code are described in International Standards Organization (ISO) 8327, Information Processing Systems—Open Systems Interconnection—Basic Connection Oriented Session Protocol Specification, Addendum 02—1988.

Effect None.

Recovery Check the reason code. You can monitor the frequency of this message for an indication that a problem exists. You can perform an SCF trace of the application, presentation, and session (TAPS) process using the TRACE command L5 selection option to investigate the protocol data unit (PDU) structure.