An abnormal end (abend) or program check has occurred in module modname. This implies that there may be an error in CICS code. Alternatively, unexpected data has been input, or storage has been overwritten.
The code aaa/bbbb is a three digit hexadecimal MVS code (if applicable), followed by a four digit alphanumeric CICS code. The MVS code is a system completion code (for example, 0C1 or D37). If an MVS code is not applicable, this field is filled with three hyphens. The CICS code is an abend code or a number referring to a CICS message (for example, AKEA is a CICS abend code; 1310 refers to message DFHTS1310).
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
An exception entry is made in the trace table. A system dump is taken, unless you have specifically suppressed dumps in the dump table.
CICS continues unless you have specified in the dump table that CICS should terminate.
Message DFHME0116 is normally produced containing the symptom string for this problem.
If CICS is still running, it is necessary to decide whether to terminate CICS.
Look up the MVS code, if there is one, in OS/390 MVS System Codes.
Next, look up the CICS alphanumeric code in this manual. This tells you, for example, whether the error was a program check, an abend, or a runaway, and may give you some guidance concerning user response.
If module modname is not crucial to the running of your CICS system, you may decide to continue to run and bring CICS down at a convenient time to resolve the problem.
If you cannot run without the full use of module modname you should bring CICS down in a controlled shutdown.
You need further assistance from IBM to resolve this problem. See CICS® Problem Determination Guide for guidance on how to proceed.
Console
DFHKEDD, DFHKEDS, DFHKEGD, DFHKETI
XMEOUT Parameters: applid, aaa/bbbb, X'offset', modname
An error has been detected in module modname. The code code is the exception trace point id which uniquely identifies what the error is and where the error was detected.
For further information about CICS exception trace entries, see the CICS Problem Determination Guide.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
An exception entry (code X'code' in the message) is made in the trace table. A system dump is taken, unless you have specifically suppressed dumps in the dump table.
CICS will continue unless you have specified in the dump table that CICS should terminate. If appropriate, an error return code is sent to the caller of this domain. In this case, CICS could be terminated by the caller (for example, the domain manager, DFHDMDM). A message will be issued to this effect.
Message DFHME0116 is normally produced containing the symptom string for this problem.
This indicates a possible error in CICS code. The severity of its impact depends on the importance of the function being executed at the time of the error.
CICS may not have been terminated.
If the message occurs once and module modname is not crucial to the running of your CICS system, you may decide to continue to run and bring CICS down at a convenient time to resolve the problem.
If the message recurs or if you cannot run without the full use of module modname you should bring CICS down in a controlled shutdown.
You need further assistance from IBM to resolve this problem. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKEDD, DFHKEGD, DFHKEIN
XMEOUT Parameters: applid, X'code',modname
A CICS function is taking more time to process than CICS expects. A possible loop has been detected in module modname at offset X'offset'. This is the offset of the instruction which was executing at the time the error was detected.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
An exception entry is made in the trace table. A system dump is taken, unless you have specifically suppressed dumps in the dump table.
CICS continues unless you have specified in the dump table that CICS should terminate.
Message DFHME0116 is normally produced containing the symptom string for this problem.
Notify the system programmer. If CICS has not been terminated, it is necessary to decide whether the problem is serious enough to bring CICS down.
Since some CICS functions can use a lot of CPU time, this message may have been caused by a long-running function. So there may not be an error here. Usually, CICS purges a CICS function which exceeds the runaway task time interval which you have specified in the SIT (this is the ICVR which is measured in milliseconds). This means that module modname is terminated and CICS continues.
But if you have declared ICVR=0 in the SIT and you consider that module modname has gone into a loop, you have to terminate CICS in order to terminate the runaway function.
If CICS has terminated module modname and you consider that it was not a runaway, you should increase the ICVR time interval in the SIT. You have to bring CICS down at a suitable time to do this permanently. However, you can change the ICVR time interval temporarily online, using the CEMT transaction.
If raising the ICVR time does not solve the problem, you will need further assistance from IBM. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKEDD, DFHKEDS, DFHKEGD, DFHKETI
XMEOUT Parameters: applid, X'offset', modname
A hardware error has occurred during the running of module modname. The MVS store clock facility is the timing mechanism for the operating system.
The code code is the exception trace point ID which uniquely identifies the place where the error was detected.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
An exception entry (code code in the message) is made in the trace table. A system dump is taken, unless you have specifically suppressed dumps in the dump table.
CICS continues unless you have specified in the dump table that CICS should terminate.
If CICS is still running, it is necessary to decide whether to terminate CICS. This is probably a hardware error and you should first investigate the MVS store clock and find out whether it is working properly. If this is the cause, you should take the appropriate action to have it repaired or replaced.
In the unlikely event that this is not a hardware problem, you will need further assistance from IBM. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKETI
An MVS GETMAIN was issued by module modname but there was insufficient storage available to satisfy the request.
The code X'code' is the exception trace point ID which uniquely identifies the place where the error was detected.
The code mvscode is the MVS GETMAIN return code.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
CICS will terminate with a system dump. An exception entry is made in the trace table (code code in the message).
Message DFHME0116 is normally produced containing the symptom string for this problem.
Inform the system programmer.
You can get diagnostic information about the MVS return code by consulting the relevant MVS codes manual which is listed in the book list at the front of this book.
Try decreasing the overall size limit of the DSAs or EDSAs. Or, try increasing the size of the whole region, if it is not already at maximum size. If CICS is not already terminated, you will need to bring CICS down to do this. See the CICS System Definition Guide or the CICS Performance Guide for further information on CICS storage.
Console
DFHKEIN
XMEOUT Parameters: applid, X'code',modname, mvscode
Part of CICS initialization must be done in an APF-authorized state. The kernel has detected that DFHSIP is not APF-authorized.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
CICS will terminate. The CICS job step terminates with return code 12.
All libraries concatenated in the STEPLIB concatenation should be APF-authorized, and DFHSIP should be link-edited with an authorization code of 1.
Console
DFHKESIP
A domain has failed to pre-initialize and as a result the system will terminate.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
Since this problem has occurred so early in CICS initialization, possible causes include a severe lack of storage or corruption of the local catalog.
CICS terminates.
Examine all earlier messages sent to the console to look for any obvious cause of the domain pre-initialisation failure.
If you cannot find the cause of the failure from the messages, inform the
system programmer. If a dump is taken, investigate this problem
using the exception trace which is issued by the failing domain.
You may need further assistance to resolve this problem. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKETCB.
The kernel has issued an MVS IDENTIFY which has failed.
The code mvscode is the MVS IDENTIFY return code.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
CICS terminates. The CICS job step terminates with return code 12.
Inform the system programmer.
To resolve the problem, use the MVS IDENTIFY return code mvscode and the MVS/XA Supervisor Services and Macro Instructions manual, (GC28-1154), to determine why the IDENTIFY failed.
Console
DFHKETCB.
CICS has validated the SVC number svcno, but it does not correspond to the correct CICS Type 3 SVC for this release of CICS. CICS cannot function without the correct CICS SVC.
The SVC number svcno has been specified in the SIT, or as an override, by the CICSSVC= parameter.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
Control is returned to the parameter manager for interaction with the operator. Further action depends upon which PARMERR= parameter has been specified.
The CICS Type 3 SVC is defined to MVS in SYS1.PARMLIB member IEASVCxx. SVC svcno must be defined as a Type 3 SVC with an entry point equal to the entry point name specified when the SVC module was installed into SYS1.LPALIB. Ensure that this is the case.
Console
DFHKEGD.
The kernel has detected that the release level of the operating system is earlier than the pre-requisite release level required to run CICS.
CICS will terminate. The CICS job step terminates with return code 12.
Refer to the CICS Program Directory and install the pre-requisite operating system release level or higher.
Console
DFHKESIP
An error during pre-initialization or termination, possibly signalled by a previous message, has caused the kernel domain to take a dump, which will issue this message immediately before calling the MVS SDUMP facility.
The dump code dumpcode is the 8-character dump code ‘KERNDUMP’.
The dumpid dumpid is the string ‘0/0000’.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
When the dump is complete, message number DFHKE0202 is issued.
Inform the system programmer. See the associated dump and error messages for further guidance.
Console
DFHKEDD, DFHKEDS, DFHKEGD, DFHKEIN, DFHKESTX
This message is issued on successful completion of an SDUMP.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
CICS will terminate.
Print off the system dump if required. A previous MVS message will identify in which SYS1.DUMP data set this dump can be found.
Console
DFHKEDD, DFHKEDS, DFHKEGD, DFHKEIN, DFHKESTX
At the time of the MVS SDUMP request issued by CICS, another address space in the same MVS system was in the process of taking an SDUMP. This causes MVS to reject the new request. A nonzero value for the DURETRY parameter on the SIT means that CICS waits for five seconds before reissuing the SDUMP request.
CICS issues an MVS STIMERM macro which causes CICS to stop for five seconds. The request is reissued when the delay interval has expired. CICS delays and retries every five seconds for a total time equal to the number of seconds specified on the DURETRY system initialization parameter.
None.
Console
DFHKEDD, DFHKEDS, DFHKEGD, DFHKEIN, DFHKESTX
At the time of the MVS SDUMP request issued by CICS, another address space in the same MVS system was in the process of taking an SDUMP. This caused MVS to reject the new request. CICS has waited for five seconds (as indicated by message DFHKE0208) and is now reissuing the SDUMP request.
CICS reissues the SDUMP request.
None.
Console
DFHKEDD, DFHKEDS, DFHKEGD, DFHKEIN, DFHKESTX
An MVS SDUMP request from CICS signalled by message DFHKE0201 has failed to complete successfully. The possible reasons reason for the failure are detailed below.
If a nonzero value has been specified for the DURETRY SIT parameter, CICS will have retried the SDUMP request every five seconds for the specified period. This message is issued if SDUMP is still busy after the final retry.
During initialization, CICS does not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
CICS proceeds as if the dump had been successful.
The user response depends on the reason for the failure. For:
Notify the system programmer. If CICS is still running, it will be necessary to decide whether to terminate CICS.
You may decide to continue to run and bring CICS down at a convenient time to resolve the problem.
To resolve the problem, collect any dumps and any relevant messages and contact your IBM Support Center.
Further guidance on how to prepare information for IBM support is given in the CICS Problem Determination Guide If you are not familiar with this process, refer to the guide before contacting IBM support.
Console
DFHKEDD, DFHKEDS, DFHKEGD, DFHKEIN, DFHKESTX
The kernel (KE) domain has issued an MVS GETMAIN for kernel stack storage, but there was insufficient storage available to satisfy the request.
The code mvscode is the MVS GETMAIN return code.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
CICS terminates abnormally with a system dump. No exception entry is made in the trace table since a call to the trace (TR) domain would itself require kernel stack storage.
Inform the system programmer.
You can get diagnostic information about the MVS return code by consulting the relevant MVS codes manual which is listed in the book list at the front of this book. Also look at the kernel domain section of the system dump to see how the kernel stack storage has been used up.
Try decreasing the size limits of the DSAs or EDSAs. Or, try increasing the size of the whole region, if it is not already at maximum size. See the CICS System Definition Guide or the CICS Performance Guide for further information on CICS storage.
Console
DFHKESGM
The kernel (KE) domain has issued an MVS FREEMAIN for kernel stack storage, but a bad return code was returned.
The code mvscode is the MVS FREEMAIN return code.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
CICS terminates abnormally with a system dump. No exception entry is made in the trace table since a call to the trace (TR) domain would itself require kernel linkage.
Inform the system programmer.
You can get diagnostic information about the MVS return code by consulting the OS/390 MVS System Codes manual.
Console
DFHKEDS
The kernel (KE) domain has detected that the current task is recursively abending while attempting to recover from an abend.
CICS terminates abnormally with a system dump. No exception entry is written to the trace table because the trace (TR) domain may be the cause of the loop.
Use the dump provided to investigate the kernel error table to diagnose the earlier abends for the failing task.
Console
DFHKERRI
An attempt to invoke a REGISTER request against the MVS automatic restart manager (ARM) failed.
The codes resp, reason are the hexadecimal response and reason codes from ARM.
A system dump is taken.
CICS continues, but cannot subsequently be restarted by ARM.
It is necessary to decide whether to terminate CICS.
For problem diagnosis look up the return codes from the IXCARM
macro in the OS/390 MVS Sysplex Services Reference manual.
Further information about how to use ARM can also be found in OS/390 MVS Setting Up a Sysplex.
You need further assistance from IBM to resolve this problem. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKEAR
An attempt to invoke a DEREGISTER request against the MVS automatic restart manager (ARM) failed.
The codes resp, reason are the hexadecimal response and reason codes from ARM.
If the request failed during startup, a dump is taken and CICS continues.
If the request failed during shutdown, an exception entry is made in the trace table, and a system dump is taken unless you have specifically suppressed dumps in the dump table. CICS continues to shut down unless you have specified in the dump table that CICS should terminate. The DEREGISTER failed so a subsequent failure of CICS or an IMMEDIATE shutdown may result in ARM restarting CICS.
For problem diagnosis look up the return codes from the IXCARM
macro in the OS/390 MVS Sysplex Services Reference manual.
Further information about how to use ARM can also be found in the MVS/ESA Setting Up a Sysplex manual.
You need further assistance from IBM to resolve this problem. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKEAR
XMEOUT Parameters: applid, X'resp',X'reason'
An attempt to invoke a WAITPRED request against the MVS automatic restart manager (ARM) failed.
The codes resp, reason are the hexadecimal response and reason codes from ARM.
An exception entry is made in the trace table. A system dump is taken, unless you have specifically suppressed dumps in the dump table.
CICS continues unless you have specified in the dump table that CICS should terminate. The failure of the WAITPRED request may result in other subsystems not being available when CICS initialization completes.
If CICS is still running, it is necessary to decide whether to terminate CICS.
For problem diagnosis look up the return codes from the IXCARM
macro in the OS/390 MVS Sysplex Services Reference manual.
Further information about how to use ARM can also be found in the MVS/ESA Setting Up a Sysplex manual.
You need further assistance from IBM to resolve this problem. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKEAR
XMEOUT Parameters: applid, X'resp',X'reason'
An attempt to invoke a READY request against the MVS automatic restart manager (ARM) failed.
The codes resp, reason are the hexadecimal response and reason codes from ARM.
An exception entry is made in the trace table. A system dump is taken, unless you have specifically suppressed dumps in the dump table.
CICS continues unless you have specified in the dump table that CICS should terminate. Other subsystems which are waiting for CICS are not be informed that CICS is ready for work and continue to wait until timed out.
If CICS is still running, it is necessary to decide whether to terminate CICS.
For problem diagnosis look up the return codes from the IXCARM
macro in the OS/390 MVS Sysplex Services Reference manual.
Further information about how to use ARM can also be found in the MVS/ESA Setting Up a Sysplex manual.
You need further assistance from IBM to resolve this problem. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKEAR
XMEOUT Parameters: applid, X'resp',X'reason'
A WAITPRED request against the MVS automatic restart manager (ARM) timed out.
The codes resp, reason are the hexadecimal response and reason codes from ARM.
An exception entry is made in the trace table. A system dump is taken, unless you have specifically suppressed dumps in the dump table.
CICS continues unless you have specified in the dump table that CICS should terminate. The WAITPRED request time out may result in other subsystems not being available when CICS initialization completes.
If CICS is still running, it is necessary to decide whether to terminate CICS.
For problem diagnosis look up the return codes from the IXCARM
macro in the OS/390 MVS Sysplex Services Reference manual.
Further information about how to use ARM can also be found in the MVS/ESA Setting Up a Sysplex manual.
You need further assistance from IBM to resolve this problem. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKEAR
XMEOUT Parameters: applid, X'resp',X'reason'
CICS is about to invoke a WAITPRED request against the automatic restart manager. This may result in a delay before CICS processing continues.
CICS continues.
None.
Console
DFHKEAR
XMEOUT Parameter: applid
CICS has registered with the MVS automatic restart manager (ARM) after having been restarted but the restart JCL specifies XRF=YES. XRF is incompatible with ARM.
CICS terminates.
Ensure that the XRF=YES option in the restart JCL is correct.
Console
DFHKEAR
An attempt to REGISTER with the MVS automatic restart manager (ARM) has failed when a cold or initial start has been specified in the SIT.
When the CICS region has been restarted with JCL that specifies START=COLD or START=INITIAL, CICS relies on ARM to determine whether to override the start type and change it to AUTO. As the REGISTER has failed, CICS cannot determine whether the region is being restarted by ARM, and so does not know whether to override the start type.
CICS waits until the operator supplies the START type to be used by this region.
If the region is being restarted by ARM, specify AUTO. If the startup type of COLD or INITIAL in the SIT should be preserved, specify ASIS.
See the previously issued message DFHKE0401 for guidance on dealing with the underlying REGISTER failure.
Console Routecodes 1 and 11
DFHKEAR
An attempt to invoke a REGISTER request against the MVS automatic restart manager (ARM) failed because the job type is invalid to ARM. CICS can only register with ARM if it is being run as a started task or a batch job.
CICS continues, but cannot subsequently be restarted by ARM.
None.
Console
DFHKEAR
An attempt to invoke a REGISTER request against the MVS automatic restart manager (ARM) failed because the maximum number of ARM users allowed for in the ARM couple data set has been reached. This response is never given by ARM if ARM is restarting CICS.
CICS continues, but cannot subsequently be restarted by ARM.
None.
Console
DFHKEAR
A WAITPRED request against the MVS automatic restart manager (ARM) has completed.
Further information about how to use ARM can also be found in the MVS/ESA Setting Up a Sysplex manual.
You need further assistance from IBM to resolve this problem. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKEAR
XMEOUT Parameter:
applid
An attempt to invoke a REGISTER request against the MVS automatic restart manager (ARM) failed in module DFHKESVC.
The code resp is the hexadecimal response from DFHKESVC and its meanings are:
CICS continues, but cannot subsequently be restarted by ARM.
It is necessary to decide whether to terminate CICS.
Further information about how to use ARM can also be found in OS/390 MVS Setting Up a Sysplex.
You need further assistance from IBM to resolve this problem. See CICS Problem Determination Guide for guidance on how to proceed.
Console
DFHKEAR
An attempt to invoke the MVS service CSRL16J (Load 16 and Jump) has returned to CICS with a non-zero return code. This service is called from the CICS Kernel 'Reset Address' function.
CICS continues by deliberately executing a privileged instruction which causes a program exception with code 0C2 which leads to an ASRA abend. Register 4 has been loaded with the return code from the CSRL16J service.
Inform the systems programmer.
Console
DFHKERET
The Kernel issued a call to the service-routine callable service, and received a response of EMVSINITIAL with a reason code of X'0012', indicating that the number of processes for the current CICS region userid has been exceeded.
CICS initialization continues.
If this error occurs frequently, consider increasing the MAXPROCUSER value in the BPXPRMxx member of SYS1.PARMLIB.
Console
DFHKETCB
XMEOUT Parameters:
applid,
service-routine
Unix System Services has returned a non-zero return code/reason code to a service-routine call made by the Kernel during CICS initialization.
CICS initialization continues since it is too early to tell whether Unix System Services will be required later in the CICS run.
Determine the reason for this response. The return code and reason code included in the message text should be described in the z/OS UNIX System Services Messages and Codes manual.
Console
DFHKETCB
XMEOUT Parameters:
applid,
X'rvalue',
X'
rcode',
X'
rreason',
uss
MVS has made a call to the CICS ESTAE-type recovery routine DFHKESTX, but it supplied no system diagnostic work area (SDWA). DFHKESTX is unable to continue with the recovery.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
Module DFHKESTX produces a system dump and percolates the error to the next ESTAE routine. This is potentially a serious error. CICS continues processing pending the result of the error percolation.
The abend code X'code' is the reason the CICS ESTAE was called. You need to find out which product has produced the abend. Typically it is an MVS system completion code, for example D37. However the abend may have been issued by CICS, for example abend 1596, or another product such as IMS.
Since there is little further diagnostic information in this case, look for any messages that may indicate the reason for the abend. The entry in the appropriate manual for the abend code gives user guidance regarding the error, and may also give some guidance concerning the appropriate user response.
The reason why no SDWA was passed and subsequently no recovery was attempted is probably a shortage of storage. This storage shortage may also be an influencing factor in the abend itself.
Console
DFHKESTX
During an immediate shutdown of CICS the file owning TCB is detached before VSAM is able to close all data sets normally.
The detaching of this TCB can cause abends during shutdown. At CICS restart VSAM may need to perform a VERIFY for any data set that is open for update when the immediate shutdown is performed, and this can cause a long delay.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If this is the case CICS uses the default applid value DBDCCICS.
The immediate shutdown of CICS continues.
None.
If this message occurs frequently, you may wish to issue a CEMT SET FILE ALL CLOSED before immediate shutdown is performed.
Console
DFHKESIP
This message is issued when CICS has terminated.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
Control is given back to the operating system.
None.
Console
DFHKESIP
CICS issues this message when it terminates abnormally.
During initialization, CICS may not have access to the user’s applid coded in the SIT. If CICS produces this message in these circumstances, it uses the default applid value DBDCCICS.
The abnormal termination of CICS continues. The kernel returns control to the operating system by issuing a user 1800 abend.
The original error which caused the abnormal termination may also have produced a dump. No specific dump is produced to accompany this message.
If a dump is produced, check the dump to determine the cause of the error. Use the CICS Problem Determination Guide to assist you to determine the problem.
If no dump is produced, check for other CICS and MVS messages and abend codes to help you to determine the cause of the problem.
Console
DFHKESIP