I have a number of LPARs on one P520. All LPARs are running 5.3 and I observe the following:
On some LPARs the number of CPUs found do not match between topas and mpstat.
Server 1:
$ mpstat
System configuration: lcpu=4 ent=0.2 mode=Uncapped
cpu min maj mpc int cs ics rq mig lpa sysc us sy wt id pc %ec lcs
0 3236207143 10334920 18148 3129232930 4185678899 2113273478 1 52924493 100 367388385328 53 45 0 1 0.02 8.0 1128243307
1 603958753 5039979 16124 1524069311 1859429164 970562064 0 41186584 100 222067273272 49 49 0 2 0.01 4.9 68661856
2 1710507793 3119307 4364 1114068729 1623216950 830873642 3 34012743 100 206322132851 48 50 0 1 0.01 4.8 1775110222
3 250548546 2479568 3764 1040035340 1334942079 712336992 0 23302581 100 159082003333 48 50 0 1 0.01 3.5 1588140807
U - - - - - - - - - - - - 3 76 0.16 78.8 -
ALL 5801222235 20973774 42400 6807406310 9003267092 4627046176 4 151426401 100 954859794784 11 10 3 76 0.04 21.2 4560156192
$ topas -c 4 -i 2
Topas Monitor for host: anker-prod EVENTS/QUEUES FILE/TTY
Wed Jan 11 00:07:56 2012 Interval: 2 Cswitch 1404 Readch 1191.2K
Syscall 81019 Writech 1592.3K
CPU User% Kern% Wait% Idle% Physc Reads 7208 Rawin 0
0 71.4 27.3 0.1 1.2 0.51 Writes 9568 Ttyout 448
1 64.4 28.9 0.0 6.6 0.05 Forks 1 Igets 0
2 0.0 36.3 0.0 63.7 0.00 Execs 1 Namei 499
Runqueue 0.3 Dirblk 0
Topas only shows 3 CPUs where mpstat shows 4 CPUs.
On Server2 both commands agree on the number of CPUs
$ mpstat
System configuration: lcpu=4 ent=0.2 mode=Uncapped
cpu min maj mpc int cs ics rq mig lpa sysc us sy wt id pc %ec lcs
0 9860614934 2732104 7298 6735808867 9419381532 4702171407 3 4684492 100 65615358106 32 57 0 10 0.01 3.0 418549046
1 1973963245 223280 2636 859857037 781266777 458989902 0 12072469 100 21592902586 34 44 0 22 0.00 0.4 1140935784
2 67906433 112022 2726 251438391 16705237 8752709 0 124434 100 498800985 17 39 0 44 0.00 0.0 253852298
3 2206185 58927 2548 246896346 5519713 2819540 0 183719 100 23307562 4 36 0 60 0.00 0.0 247932362
U - - - - - - - - - - - - 0 96 0.19 96.5 -
ALL 11904690797 3126333 15208 8094000641 10222873259 5172733558 3 17065114 100 87730369239 1 2 0 97 0.01 3.5 2061269490
$ topas -c 4 -i 2
Topas Monitor for host: kamstra-test EVENTS/QUEUES FILE/TTY
Wed Jan 11 00:11:15 2012 Interval: 2 Cswitch 211 Readch 223.7K
Syscall 1226 Writech 3298
CPU User% Kern% Wait% Idle% Physc Reads 109 Rawin 1
0 12.4 74.1 0.1 13.4 0.01 Writes 9 Ttyout 1207
1 0.0 34.1 0.0 65.9 0.00 Forks 1 Igets 0
2 0.0 43.0 0.0 57.0 0.00 Execs 1 Namei 263
3 0.0 43.5 0.0 56.5 0.00 Runqueue 1.5 Dirblk 0
Both LPARs are configured on the same P520.
I hope someone can explain to me why the commands do not match up on some LPARs.
Greetings,
Peter