HP-UX vPars and Integrity VM V6.3 Administrator Guide

A.5.2 Integrity VM and vPar CLI commands experience poor performance when
there are numerous devices on the VSP
The commands like vparmodify, hpvmmodify, hpvmcreate, and hpvmclone, (commands
used to modify the vPar or VM configuration), experience slow performance when there are
numerous devices available on the VSP, or configured in the vPar and/or VM configurations.
When you have a large number of devices, it is more than likely that the majority of those devices
are storage devices. If storage devices are being exposed to the VSP from a SAN and then
individually mapped to vPar/VM configurations, alternatively, you can map SAN-based LUNs
directly to the VMs or vPars using NPIV. Replacing individually mapped SAN-based LUNs with
one or more virtual HBAs using NPIV ports, reduces the number of devices that need to be managed,
and thus improves the CLI performance.
A.5.3 I/Os take long to complete under heavy I/O conditions on vPars or VMs with
large NPIV LUN configuration
In a large LUN configuration, spread NPIV HBAs across multiple physical HBA ports at the VSP
level.
A.6 CLI
A.6.1 hpvmhwmgmt(1M) reports DIO resources are in use by VSP
If hpvmhwmgmt -p dio -a hwpath fails due to a resource being in use by the VSP, check the
/var/adm/cra.log file for additional information on the resources in use by the VSP. The
following example shows the type of error you might see in this case:
system# hpvmhwmgmt -p dio -a 0/0/0/4/0/0/0
hpvmhwmgmt: ERROR - Resources for: '0/0/0/4/0/0/0'
are in use by your host. hpvmhwmgmt: ERROR - could not reserve hwpath: '0/0/0/4/0/0/0'
for DIO. hpvmhwmgmt: ERROR - failed to reserve and store hw path: '0/0/0/4/0/0/0'
for DIO. hpvmhwmgmt: Unable to manage dio pool resource.
A.6.2 hpvmmodify(1M) may fail with the message “intent failed Can’t get the
resource maxima”
The hpvmodify command invoked on a running VM might fail when it should succeed. When
failing, the following error message is displayed:
resource
intent failed 'Can't get the resource maxima.' vPar/VM vm_name configuration problems: Error 1: Internal
error -1 when attempting to use the ragent 'intent' interface
hpvmmodify: Unable to modify the vPar or VM.
This failure might happen only when the same processor is used by several VMs.
A.7 Miscellaneous
A.7.1 While booting a vPar or VM guest the message “WARNING: VCPU0 not
scheduled” is displayed
In V6.2, messages similar to the following are occasionally seen during the initial boot of a vPar
or VM:
WARNING: VCPU0 not scheduled for NNNNN ms" messages in hpvm_mon_log
They can occur during the early portion of booting the vPar/VM before HP-UX is launched into the
vPar/VM, when either the EFI layer or the boot loader is running.
You can safely ignore these messages.
A.7.2 When a vPar is terminated by a TC command from its console, a corresponding
vm.core is not always generated on the VSP
When a vPar is started up, it begins execution as a special application program. A TC command
issued during early stage of starting up produces a vm.core on the VSP.
A.6 CLI 273