A SLEDFILE can be created for a guest system, for the monitor system or for the whole VM2000 system.
SLED for a BS2000 guest system that is not the monitor system
The SLED of the guest system can be created with /START-VM ...,DIAGNOSTIC-IPL=*YES or by an automatic restart after SETS in the guest system or via the SE manager.
 | On SU x86, the SLED of the guest system can also be created using the SVP functions of the KVP console which is assigned to the VM. |
See "Assignment sets, implicit device assignment and release" for details on how to handle implicitly assigned devices in the case of /START-VM.
 | On SU /390, the diagnostic data of the VM2000 hypervisor is also contained in the SLEDFILE of a guest system (dump file section VM2HYPVS). |
 | On SU x86, the SLEDFILE contains no data of the Xen hypervisor or of domain 0 (X2000). |
SLED for the monitor system
The SLEDFILE also contains the diagnostic data of the VM2000 monitor.
 | The SLED of the monitor system can be created when the monitor system is restarted: On SU /390, the diagnostic data of the VM2000 hypervisor is also contained in the SLED of the monitor system (dump file section |
 | The SLED of the monitor system can be created by:
The SLEDFILE contains no data of the Xen hypervisor or of domain 0 (X2000). |
SLED for the whole VM2000 system (SU /390)
The SLED of the entire system can be created by explicitly loading the SLED in the same way as it is loaded in BS2000 operation without VM2000 or at a restart after VM2000 hypervisor termination (see "Automatic restart following VM2000 hypervisor termination (SU /390)").
With the non-standard SLED you can specify the extent of the SLED output using the MODE parameter in response to message NSD3001:
MODE=REAL
The entire main memory is output.MODE=STD
The VM2000 hypervisor main memory and the entire monitor system are output.MODE=ALL
The entire main memory and the complete monitor system are output.
Normally, SLED is output with MODE=STD (response Y or EOT to message NSD3001).
MODE=REAL or MODE=ALL should only be specified if it is clear that the main memory of the other VMs is required to diagnose the error in the VM2000 hypervisor or the monitor VM. This is only the case in exceptional situations.