ROOTVG

Hello,
i have P520 SF240_338 and HMC 6.1.3 and ...
I would like to know why when i stop partition AIX or Linux the current available value of managed system for proc and mem don't display the real available value realease after partition shutdown.
the only value who change is the run_proc_units of partition properties.

hscroot@HMC-p520:~> lssyscfg -r lpar -m 9111-520 -F name,state
LPAR_5_redhat,Running
... etc ...

hscroot@HMC-p520:~> lshwres -r mem -m 9111-520 --level sys
configurable_sys_mem=8192,curr_avail_sys_mem=1024,pend_avail_sys_mem=1024,installed_sys_mem=8192,max_capacity_sys_mem=8192,deconfig_sys_mem=0,sys_firmware_mem=512,mem_region_size=32

hscroot@HMC-p520:~> lshwres -r proc -m 9111-520 --level sys
configurable_sys_proc_units=2.0,curr_avail_sys_proc_units=0.6,pend_avail_sys_proc_units=0.6,installed_sys_proc_units=2.0,max_capacity_sys_proc_units=2.0,deconfig_sys_proc_units=0.0,min_proc_units_per_virtual_proc=0.1,max_virtual_procs_per_lpar=64,max_shared_proc_pools=1

hscroot@HMC-p520:~> lshwres -r proc -m 9111-520 --level lpar --filter "lpar_names=LPAR_5_redhat"
lpar_name=LPAR_5_redhat,lpar_id=8,curr_shared_proc_pool_id=0,curr_proc_mode=shared,curr_min_proc_units=0.1,curr_proc_units=0.2,curr_max_proc_units=0.2,curr_min_procs=1,curr_procs=2,curr_max_procs=2,curr_sharing_mode=cap,curr_uncap_weight=0,pend_shared_proc_pool_id=0,pend_proc_mode=shared,pend_min_proc_units=0.1,pend_proc_units=0.2,pend_max_proc_units=0.2,pend_min_procs=1,pend_procs=2,pend_max_procs=2,pend_sharing_mode=cap,pend_uncap_weight=0,run_proc_units=0.2,run_procs=2,run_uncap_weight=0

hscroot@HMC-p520:~> lssyscfg -r lpar -m 9111-520 -F name,state
LPAR_5_redhat,Not Activated
... etc ...

hscroot@HMC-p520:~> lshwres -r proc -m 9111-520 --level lpar --filter "lpar_names=LPAR_5_redhat"
lpar_name=LPAR_5_redhat,lpar_id=8,curr_shared_proc_pool_id=0,curr_proc_mode=shared,curr_min_proc_units=0.1,curr_proc_units=0.2,curr_max_proc_units=0.2,curr_min_procs=1,curr_procs=2,curr_max_procs=2,curr_sharing_mode=cap,curr_uncap_weight=0,pend_shared_proc_pool_id=0,pend_proc_mode=shared,pend_min_proc_units=0.1,pend_proc_units=0.2,pend_max_proc_units=0.2,pend_min_procs=1,pend_procs=2,pend_max_procs=2,pend_sharing_mode=cap,pend_uncap_weight=0,run_proc_units=0.0,run_procs=0,run_uncap_weight=0

hscroot@HMC-p520:~> lshwres -r proc -m 9111-520 --level sys
configurable_sys_proc_units=2.0,curr_avail_sys_proc_units=0.6,pend_avail_sys_proc_units=0.6,installed_sys_proc_units=2.0,max_capacity_sys_proc_units=2.0,deconfig_sys_proc_units=0.0,min_proc_units_per_virtual_proc=0.1,max_virtual_procs_per_lpar=64,max_shared_proc_pools=1

hscroot@HMC-p520:~> lshwres -r mem -m 9111-520 --level sys
configurable_sys_mem=8192,curr_avail_sys_mem=1024,pend_avail_sys_mem=1024,installed_sys_mem=8192,max_capacity_sys_mem=8192,deconfig_sys_mem=0,sys_firmware_mem=512,mem_region_size=32

Thank's in advance.

I cannot answer your question "why"  :(. However, before reporting a call with IBM make sure you have also patched your HMC code (you probably already have).

From IBM's POWER code matrix page on POWER5 (https://www14.software.ibm.com/webapp/set2/sas/f/power5cm/power5.html) I read:

From memory of other systems I have the impression I have seen this myself - but I do not administer those systems and I know they are not patched to latest levels.