Recovery manager codes (X'D9')

If a recovery manager reason code occurs that is not listed here, an internal error has occurred. Collect the following diagnostic items and contact your IBM® support center.

Diagnostics

A description of the actions that led to the error or, if applicable, either a listing of the application program or the input string to a utility program that was being run at the time of the error.
The console output for the period leading up to the error.
The queue manager job log.
The system dump resulting from the error.
A printout of SYS1.LOGREC.
If you are using CICS®, the CICS transaction dump output.
Appropriate IBM MQ, z/OS®, Db2®, CICS, and IMS service levels.

00D90000

A recovery manager module received control from its FRR for retry and found an invalid retry point identifier. The name of the module in which the error occurred appears in the SYS1.LOGREC entry showing this reason code in register 15.

System action

Standard diagnostic information is provided. The error is recorded in SYS1.LOGREC, an SVC dump is scheduled, and queue manager termination is requested. The termination reason code reflects the function for which retry was unsuccessfully attempted.

System programmer response

This is a secondary error. Obtain a copy of SYS1.LOGREC and the SVC dump for this error and for the original problem that resulted in the retry attempt. Examine the SYS1.LOGREC information and the dumps from both the original and the secondary error to determine if the recovery parameter area was damaged or if retry incorrectly restored registers for the mainline module.

Restart the queue manager.

00D90002

The recovery manager startup notification routine received an error return code from the recovery log manager when attempting to read a recovery manager status table (RMST) record from the bootstrap data set (BSDS) in one of the following cases:

When reading the record containing the RMST header. The first copy was successfully read, but the second copy could not be found.
When reading records containing the RMST entries. A no record found condition was encountered before all entries were read.
When reading either a header record or an entry record. The record exceeded its expected length.

This is an IBM MQ error.

System action

The recovery manager has no functional recovery routine (FRR) in place when this error occurs. It relies on its invoker, the facility startup function, to perform SYS1.LOGREC recording and to request a dump. The queue manager terminates with a X'00E80100' reason code.

System programmer response

The queue manager determined that the BSDS that it was reading has been corrupted. If you are running in a dual BSDS environment, determine which BSDS is corrupt, and follow the procedures described in Recovering the BSDS to recover it from the valid BSDS.

Similarly, if you are running in a single BSDS environment, refer to Recovering the BSDS, which describes the procedures needed to recover your BSDS from an archived BSDS.

00D92001

The checkpoint/restart serial controller FRR invoked queue manager termination, because an unrecoverable error was detected while processing a request.

This is a queue manager termination reason code.

System action

Queue manager termination is initiated. Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the associated error.

System programmer response

Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error, and follow the instructions associated with it.

Restart the queue manager.

00D92003

The restart request servicer FRR invoked queue manager termination, because an unrecoverable error was detected while processing a restart request.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of SYS1.LOGREC and the SVC dump for the original error and follow the instructions associated with it.

Restart the queue manager.

00D92004

The shutdown checkpoint controller FRR invoked queue manager termination, because an unrecoverable error was detected while processing a shutdown checkpoint request.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of SYS1.LOGREC and the SVC dump for the original error and follow the instructions associated with it.

Restart the queue manager.

00D92011

An internal error has occurred.

System action

The checkpoint process will end abnormally to prevent a damaged URE from being written out to the log, and the queue manager will be terminated. This is to prevent the loss or incorrect processing of an IBM MQ unit of recovery (UR). Restart will use the previous checkpoint and apply all the IBM MQ log records up to the point of the problem. Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is scheduled.

System programmer response

Restart the queue manager.

Collect the items listed in Diagnostics and contact your IBM support center.

00D92012

An internal error has occurred.

System action

The checkpoint process will end abnormally to prevent a damaged RURE from being written out to the log, and the queue manager will be terminated. This is to prevent the loss or incorrect processing of an IBM MQ unit of recovery. Restart will use the previous checkpoint and apply all the IBM MQ log records up to the point of the problem. Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is scheduled.

System programmer response

Restart the queue manager.

Collect the items listed in Diagnostics and contact your IBM support center.

00D92021

An internal error has occurred.

System action

The restart processing ends abnormally, which terminates the queue manager. This is to prevent the loss or incorrect processing of a IBM MQ unit of recovery.

System programmer response

Do not attempt to restart the queue manager until the error is resolved.

The log has become corrupted. If you are running with dual logging, try to start the queue manager from the undamaged log.

If you are unable to do achieve this, use the following procedure (you will lose all updates since your last back up):

Restore your page sets from the last set of full backups. The queue manager must have been shut down cleanly before taking the backup copies of the page sets.
Clear the logs by following the procedure detailed in Resetting the queue manager's log.

See CSQUTIL utility for information about restarting the queue manager from one log when using dual logging, and using the CSQUTIL utility. If you are unable to resolve the problem, contact your IBM support center.

00D92022

An internal error has occurred.

System action

The restart processing ends abnormally, which terminates the queue manager. This is to prevent the loss or incorrect processing of a IBM MQ unit of recovery.

System programmer response

Do not attempt to restart the queue manager until the error is resolved.

The log has become corrupted. If you are running with dual logging, try to start the queue manager from the undamaged log.

If you are unable to do achieve this, use the following procedure (you will lose all updates since your last back up):

Restore your page sets from the last set of full backups. The queue manager must have been shut down cleanly before taking the backup copies of the page sets.
Clear the logs by following the procedure detailed in Resetting the queue manager's log.

00D92023

During queue manager restart in 6 byte log RBA mode, a log record has been encountered that is written with an 8 byte log RBA.

System action

The restart processing ends abnormally, which terminates the queue manager. This is to prevent the loss, or incorrect processing, of a IBM MQ unit of recovery.

System programmer response

Do not attempt to restart the queue manager until the error is resolved.

The queue manager might have been started with an incorrect log or BSDS. Ensure that the queue manager is started with the correct log and BSDS data sets. If this was not the cause of the problem, the log or BSDS has become corrupted.

To recover from a corrupted log or BSDS, if you are running with dual logging or dual BSDS, try to start the queue manager from the undamaged log.

If you are unable to do achieve this, use the following procedure. Note, that by carrying out this procedure, you will lose all updates since your last back up:

Restore your page sets from the last set of full backups. The queue manager must have been shut down cleanly before taking the backup copies of the page sets.
Clear the logs by following the procedure detailed in Resetting the queue manager's log.

If necessary, when clearing the logs ensure that you convert the BSDS to the previous version, using the BSDS conversion utility CSQJUCNV.

See CSQUTIL utility for information about using the CSQUTIL utility. If you are unable to resolve the problem, contact your IBM support center.

00D93001

The commit/backout FRR invoked queue manager termination, because an unrecoverable error was detected during 'must-complete' processing for phase 2 of a commit-UR request.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of SYS1.LOGREC and the SVC dump for the original error and follow the instructions associated with it.

Restart the queue manager.

00D93011

A subcomponent of IBM MQ invoked commit when the agent state was invalid for commit-UR invocation. Commit-UR was requested for an agent that was modifying data. Either commit-UR or backout-UR was already in process, or the recovery structure (URE) was damaged.

System action

Abnormal termination of the agent results, including backing out (backout-UR) of its activity to the previous point of consistency. This releases all locks held by the agent for its resources.

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is scheduled. Additional information, identified in the SDWA variable recording area (VRA) by reason code X'00D9CCCC', is added to the VRA.

If the agent was in a 'must-complete' state (in-commit or in-backout), the queue manager is also terminated with reason code X'00D93001'. When the queue manager is next restarted, recoverable activity for this agent (such as an ensure-backout or ensure-commit UR) is handled to complete the commit or backout process.

System programmer response

This is an IBM MQ error. Examine the SYS1.LOGREC data and the dump to establish whether either commit-UR was invoked incorrectly or the control structure that reflects the state was damaged.

00D93012

A subcomponent of IBM MQ invoked commit when the agent state was invalid for commit-UR invocation. Commit-UR was invoked for an agent that was only retrieving data. Either commit-UR or backout-UR was already in process, or the ACE progress state field was damaged.

System action

Abnormal termination of the agent results, including backing out (backout-UR) of its activity to the previous point of consistency. This releases all locks held by the agent for its resources.

System programmer response

This is an IBM MQ error. Examine the SYS1.LOGREC data and the dump to establish whether either commit-UR was invoked incorrectly or the control structure was damaged.

00D93100

This reason code indicates that an IBM MQ allied agent does not need to participate in the Phase-2 (Continue Commit) call, because all required work has been accomplished during the Phase-1 (Prepare) call.

This reason code is generated by the recovery manager when it is determined that an IBM MQ allied agent has not updated any IBM MQ resource since its last commit processing occurred.

System action

The 'yes' vote is registered with the commit coordinator.

System programmer response

None should be required because this is not an error reason code. This reason code is used for communication between components of IBM MQ.

00D94001

The commit/backout FRR invoked queue manager termination, because an unrecoverable error was detected during 'must-complete' processing for a backout-UR request.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of SYS1.LOGREC and the SVC dump for the original error and follow the instructions associated with it.

Restart the queue manager.

00D94011

A subcomponent of IBM MQ invoked backout at a point when the agent state is invalid for invoking the function that backs out units of recovery. Either backout-UR or commit-UR phase-2 was already in process, or the agent structure was damaged.

System action

Abnormal termination of the agent results and, because the agent is in a 'must-complete' state, the queue manager is terminated with reason code X'00D94001'. When the queue manager is restarted, recoverable activity for this agent is handled to complete the commit or backout process.

System programmer response

This is an IBM MQ error. Examine the SYS1.LOGREC data and the dump to establish whether commit-UR was invoked incorrectly or the control structure was damaged.

00D94012

During backout, the end of the log was read before all the expected log ranges had been processed. The error is accompanied by an abnormal termination with reason code X'00D94001'.

This could be because the queue manager has been started with a system parameter load module that specifies OFFLOAD=NO rather than OFFLOAD=YES.

System action

The agent is abnormally terminated with completion code X'5C6'. Because the agent is in a must-complete state, the queue manager is terminated with reason code X'00D94001' and message CSQV086E.

Standard diagnostic information is recorded in SYS1.LOGREC. and an SVC dump is requested.

System programmer response

Run the print log map utility to print the content of both BSDSs. Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error. At the time of the error, registers 3 and 4 contain the 8-byte relative byte address (RBA) of the beginning of this unit of recovery. IBM MQ must read the log back to this point to complete the backout of this unit of recovery.

To restart the queue manager, you must add the missing archive log data sets back to the BSDS with the change log inventory utility, and increase the MAXARCH parameter in the CSQ6LOGP macro (the system parameter module log initialization macro) to complete the backout.

If the missing archive log is not available, or if archiving was not active, the queue manager cannot be restarted unless the log data sets and page sets are all reinitialized or restored from backup copies. Data will be lost as a result of this recovery action.

00D95001

The recovery manager's common FRR invoked queue manager termination, because an unrecoverable error was detected during checkpoint processing.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of SYS1.LOGREC and the SVC dump for the original error and follow the instructions associated with it.

Restart the queue manager.

00D95011

The recovery manager checkpoint FRR invoked queue manager termination, because an unrecoverable error was detected while performing its checkpoint functions.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error and follow the instructions associated with it.

Restart the queue manager.

00D96001

The recovery manager's restart FRR invoked queue manager termination, because an unrecoverable error was detected during the restart processor processing.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error and follow the instructions associated with it.

Restart the queue manager.

00D96011

The restart participation FRR invoked queue manager termination, because an unrecoverable error was detected while processing log records during restart.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error and follow the instructions associated with it.

Restart the queue manager when the problem has been corrected.

00D96021

The queue manager was terminated during restart because an error occurred while attempting to read the log forward MODE(DIRECT). It is accompanied by a recovery log manager error X'5C6' with a reason code describing the specific error.

Each time a portion of the log is skipped, a 'read direct' is used to validate the beginning RBA of the portion that is read.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Run the print log map utility to print the contents of both BSDSs. Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error. Follow instructions for the accompanying recovery log manager error. If possible, remove the cause of original error and restart the queue manager. If you cannot correct the error, contact your IBM support center.

00D96022

The restart FRR invoked abnormal termination, because, while reading the log forward during restart, the end-of-log was read before all recovery log scopes had been processed. It is followed by an abnormal termination with the same reason code (X'00D96022').

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the error before queue manager termination is initiated.

System programmer response

Run the print log map utility to print the contents of both BSDSs. Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error. At the time of the error, registers 2 and 3 (as shown in the dump or in SYS1.LOGREC) contain the relative byte address (RBA) of the last log record that was read before end-of-log was encountered. Follow instructions for the accompanying recovery log manager error. If you cannot correct the error, contact your IBM support center.

00D96031

The restart FRR invoked queue manager termination, because an error occurred while attempting to read the log backward MODE(DIRECT). It is accompanied by a recovery log manager error X'5C6' with a reason code describing the specific error.

Each time a portion of the log is skipped, a 'read direct' is used to validate the beginning RBA of the portion that is read.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Restart the queue manager.

00D96032

During restart, the end of the log was read before all the expected log ranges had been processed. The error is accompanied by an abnormal termination with the same reason code (X'00D96032').

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC. An SVC dump is requested. The queue manager is terminated with message CSQV086E.

System programmer response

Run the print log map utility to print the contents of both BSDSs. See Finding out what the BSDS contains for more information.

Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error. At the time of the error, registers 2 and 3 contain the relative byte address (RBA) of the last log record that was read before end-of-log was encountered. Determine where the log went.

00D97001

The agent concerned was canceled while waiting for the RECOVER-UR service to complete.

System action

The RECOVER-UR function is completed. Abnormal termination of the requesting agent occurs. Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested.

The condition that caused cancellation of the agent was installation initiated (for example, a forced termination of the queue manager).

00D97011

The queue manager was terminated during RECOVER-UR because an unrecoverable error was detected during RECOVER-UR (CSQRRUPR) recovery processing.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested. queue manager terminates with message CSQV086E and return code X'00D97011'.

System programmer response

Determine the original error. If the error is log-related, see Active log problems before restarting the queue manager.

00D97012

The RECOVER-UR request servicer FRR invoked queue manager termination, because an unrecoverable error was detected while attempting to recover a unit of recovery.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error and follow the instructions associated with it.

Restart the queue manager.

00D97021

The RECOVER-UR FRR invoked queue manager termination, because an error occurred while attempting to read the log MODE(DIRECT) during forward processing. It is accompanied by a recovery log manager error X'5C6' with a reason code describing the specific error.

Each time a portion of the log is skipped, a 'read direct' is used to validate the beginning RBA of the portion that is read.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Restart the queue manager.

00D97022

The RECOVER-UR invoked abnormal termination because end-of-log was reached before all ranges had been processed for forward recovery. This error is accompanied by an abnormal termination with the same reason code (X'00D97022').

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Run the print log map utility to print the contents of both BSDSs. Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error. At the time of the error, registers 2 and 3 contain the relative byte address (RBA) of the last log record that was read before end-of-log was encountered. Follow instructions for the accompanying recovery log manager error.

Restart the queue manager.

00D97031

The RECOVER-UR FRR invoked queue manager termination, because an error occurred during an attempt to read the log MODE(DIRECT) while reading the log backward. It is accompanied by a recovery log manager error X'5C6' with a reason code describing the specific error.

Each time a portion of the log is skipped, a 'read direct' is used to validate the begin-scope RBA of the portion that is read.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Run the print log map utility to print the contents of both BSDSs. Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error. See the accompanying error reason code. Follow instructions for the accompanying recovery log manager error.

Restart the queue manager.

00D97032

The RECOVER-UR invoked abnormal termination because end-of-log was reached before all ranges had been processed for backward recovery. This error is accompanied by an abnormal termination with the same reason code (X'00D97032').

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Restart the queue manager.

00D98001

The recovery manager's common FRR invoked queue manager termination, because an unrecoverable error was detected during indoubt-UR processing.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error and follow the instructions associated with it.

Restart the queue manager.

00D98011

The FRR for the resolved-indoubt-UR request servicer invoked queue manager termination, because an unrecoverable error was detected processing a request.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Obtain a copy of the SYS1.LOGREC and the SVC dump for the original error. See the accompanying error reason code.

Restart the queue manager.

00D98021

The resolved indoubt FRR invoked queue manager termination because of an error while attempting to read the log MODE(DIRECT) during forward recovery. It is accompanied by a recovery log manager error X'5C6' with a reason code describing the specific error.

Each time a portion of the log is skipped, a 'read direct' is used to validate the beginning RBA of the portion that is read.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Restart the queue manager.

00D98022

Resolved indoubt invoked abnormal termination when end-of-log was reached before all ranges had been processed for forward recovery. This error is accompanied by abnormal termination with the same reason code (X'00D98022').

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Restart the queue manager.

00D98031

The resolved indoubt FRR invoked queue manager termination, because an error occurred during an attempt to read the log MODE(DIRECT) while reading the log backward. It is accompanied by a recovery log manager error X'5C6' with a reason code describing the specific error.

Each time a portion of the log is skipped, a 'read direct' is used to validate the begin-scope RBA of the portion that is read.

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Restart the queue manager.

00D98032

The resolved indoubt FRR invoked abnormal termination when end-of-log was reached before all ranges had been processed for backward recovery. This error is accompanied by abnormal termination with the same reason code (X'00D98032').

This is a queue manager termination reason code.

System action

Standard diagnostic information is recorded in SYS1.LOGREC, and an SVC dump is requested for the original error before queue manager termination is initiated.

System programmer response

Restart the queue manager.

00D99001: The checkpoint RBA in the conditional restart control record, which is deduced from the end RBA or LRSN value that was specified, is not available. This is probably because the log data sets available for use at restart do not include that end RBA or LRSN.
System action: The queue manager terminates.
System programmer response: See message CSQR015E.

00D99104: Queue manager restart detected that backward migration of messages was required. For backward migration to be possible, there must be no uncommitted units of recovery present at the end of restart. During restart, however, a decision was made not to force commit a detected indoubt unit of work. The decision is based on the response to message CSQR021D, or by the presence of a service parm which prevents the CSQR021D WTOR from being issued.
System action: Queue manager restart is terminated.
System programmer response: Either restart the queue manager with a higher level of code so that backward migration is not required, or, allow indoubt units of work to be force committed during restart.

00D9AAAA: This reason code identifies additional data stored in the system diagnostic work area (SDWA) variable recording area (VRA) following an error during backout-UR.
System action: Data is stored in the field indicated by VRA key 38 following the EBCDIC string 'RMC-COMMIT/BACKOUT'. This information is useful for IBM service personnel.
System programmer response: Quote this code, and the contents of the VRA field indicated by key 38 when contacting your IBM support center.

00D9BBBB: This reason code identifies additional data stored in the system diagnostic work area (SDWA) variable recording area (VRA) following an error during begin-UR.
System action: Data is stored in the field indicated by VRA key 38. This information is useful for IBM service personnel.
System programmer response: Quote this code, and the contents of the VRA field indicated by key 38 when contacting your IBM support center.

00D9CCCC: This reason code identifies additional data stored in the system diagnostic work area (SDWA) variable recording area (VRA) following an error during commit-UR.
System action: Data is stored in the field indicated by VRA key 38 following the EBCDIC string 'RMC-COMMIT/ABORT'. This information is useful for IBM service personnel.
System programmer response: Quote this code, and the contents of the VRA field indicated by key 38 when contacting your IBM support center.

00D9EEEE: This reason code identifies additional data stored in the system diagnostic work area (SDWA) variable recording area (VRA) following an error during end-UR.
System action: Data is stored in the field indicated by VRA key 38. This information is useful for IBM service personnel.
System programmer response: Quote this code, and the contents of the VRA field indicated by key 38 when contacting your IBM support center.