Sar command for dual cpu's

Hello to everyone!
i am using solaris 8 and the "sar -u" command to get the Cpu % Utilization..
Is "sar" command reliable for dual cpu or gives reliable values if you have one cpu only.
I mean if the first cpu has 50% utilization then the "sar" command gives % idle time 50 % or 25% (1/4) considering we have 2 cpu's
thanx in advance for your time and your help

You've got it all wrong- this conversation is not about threads. SAR is reporting on the cpu-utilization.
Lets take an example:
Lets say you got five running programs. One of these programs occupies about 100% of the processor while the others are idle. Then SAR will report 0% idle - becouse the processor is NOT idle. SAR does not care if there are one or a thousand idle processes.
If this was done on a dual-cpu machine where the active program was to occupy 100% of one processor and the other four programs still idling. Then sar would report 50% since one processor is fully used while the other processor is idle.
I hope this has cleared it up a bit for you.
Regards,
Jon, developer, Sweden

Similar Messages

DI Sever License for Dual CPU Processor

How about a license can be calculated in case of using dual CPU processor?
Is it required 2 licenses of DI Server?
And if it is required 2 licenses for above case but customer purchase only one license of DI Server, does happen some of technical block for CPU processing?

Hi Seung,
You probatly better contact the Sales department in your country for it: <a href="https://www.sap.com/contactsap/directory/korea.epx">https://www.sap.com/contactsap/directory/korea.epx</a>
(Found a Korean flasg in your Business Card, if you need to contact Sales in another country, see this page: [url]https://www.sap.com/contactsap/directory/index.epx[/url])
Hope this helps,
Rowdy

Sar command to find out total percentage of memory utilisation for a day.

Hi,
I am in a process of taking last 10 days average percentage of memory, CPU and Disk utilisation in solaris8.
Can any one help me regarding in this.
I found sar command and I got the details of CPU by using #sar -u -f sa23 command.
For memory If I try sar command with �r option , it gives only free memory details only.
#sar -r -f sa23
00:00:01 freemem freeswap
01:00:00 173009 14301230
Average 172860 14302592
Is there any way to find out total percentage of memory used?
And also
Total percentage of disk utilisation also in solaris8?

vvsha wrote:
# sar -r -f sa25
00:00:00 freemem freeswap
01:00:00 172847 14303687
02:00:00 172816 14301232
22:00:00 172680 14304828
23:00:01 172676 14304467
Average   172674 14302768
Average value of freemem is 172674.The average amount of free memory over the course of a day doesn't seem to be that useful. I'd want to look for the low points (see if we're close to running out during some periods).
Is this value in Megabytes?
# man sar
     -r    Report unused memory pages and disk blocks:
           freemem
                 average pages available to user processes.So it's in pages. You can use 'pagesize' to verify your page size. On modern SPARC systems, it will be 8KB.
from this value can we calculate total memory utilized for this day? like totalmem - freemem = usedmem Depending on what you mean by utilized, yes. Memory can be used for a disk cache, but that use is not shown by decreasing freemem. Uses that are counted are kernel reservations, runtime pages for applications, and direct memory allocations (malloc, etc.)
Darren

CPU fan recommendations for dual 800 Quicksilver

It never rains but it pours.
Just a few hours after posting about HD heat issues my CPU fan has gotten really noisy with a slight rattling noise. Yes, it is the CPU fan. Maybe it's the hot weather we are having causing something to expand, or maybe the cleaning I gave it yesterday with compressed air pushed it over the edge. Anyway, I removed the fan, lubricated it with one drop 3M oil under the sticker, replaced it. All dust and dirt is off the blades. I have poked and prodded it to see if there were other items vibrating (e.g., the famous screws touching something), but nothing changes. The sporadic rattling (sometimes disappears for half a minute or so but usually there) really suggests to me that a bearing might be going. At this stage I think it is time to go shopping for a new fan.
This is a dual processor machine of the early QS series so it's like a toaster inside and I'm not sure if any of the standard QS fan recommendations apply to a dual. Of course I want everything - high CFM and silence. I found a lengthy discussion from last December talking about a single processor replacement but I thought I'd solicit current recommendations, especially for dual processor.
Thanks

The specs aren't too hard to find; I just searched: " Sunon PMD1206PTB1". Google brought me to a Sunon site.
http://www.tmcscsi.com/Sunon.shtml
the above link doesn't seem to work as a link, but when I type it into my address bar, it works fine. Go figure!
which tells all. The KDE1206PTV1 fan is a different series and rated at 23.5cfm and 33.5 dBA, which I think we knew; it also runs somewhat slower and draws about half the current. More importantly, there are probably half a dozen 60X60X25mm fans of either the PMD or KDE series that would fit, and would push more or less air at more or less noise. What it doesn't tell is what your ear would call an acceptable compromise. In general, of course, more cfm, more dBA. It also doesn't tell how these stats compare to other brands; I ran it all down at one time, and it's a rather laborious process, but it can be done. I didn't have any frame of reference, so it was all just numbers to me; I set the whole matter aside until just now with your inquiry.
I wouldn't worry about the connectors too much; I fit QS fans to the DA model and wind up connecting them with Molex plugs and no problems. The replaced SilenX fan had three wires, of which only two were used, and it worked. Most PCs use a third wire to turn the fan off or on, or up and down, or something - which Apple doesn't. N.B. If you use the Molex connector, you will note that the red (5v) wire plugs into the yellow (12v) socket; not to worry - it spozed to. The PMD fan I got came with two wires, bared at the ends, but no connector(s). I wound up jury-rigging the connexion. PC accessory houses that sell fans also sell wire adapters of dozens of configurations, just search. My two favourite such houses are Xoxide.com and pcToys.com. I think I got the Sunon fan(s) from Allied Electric (alliedelec.com) for about $12 and $12 s&h, which is a minimum figure. You can ship multiple fans for that.
Sunon says its fans work on the Mag-Lev principle, so lubrication wouldn't help much. What would hurt would be an unbalanced fan, so it might help to clean the crud off the blades - evenly, if possible.
I keep wondering what a slot fan would do to cooling and noise, but that may be a thought better postponed. First should be getting the CPU acceptably cool. I also understand it helps cooling if you clip some of the grillwork in front of or behind the fans (power supply fans, especially).
Just a few thoughts ..... Jon

Printer go off when we give command for print

when ever we give command for print of adobe file it shows printer not attach to computer

Welcome to the Adobe Forums.
The more information you supply about your situation, the better equipped other community members will be to answer. Please supply the following Information where applicable:
    •    Adobe product and version number
    •    Operating system and version number
    •    Browser and version number (where applicable)
    •    The full text of any error message(s)
    •    What you were doing when the problem occurred
    •    Screenshots of the problem (if possible)
    •    Computer hardware, such as CPU; GPU; amount of RAM; etc.

Is Logic 8 not optimized for dual core processing duties ?

Running MacBook Pro 2.4 ghz, 2 gig Ram 200 gig HD external eSATA or FW 800. No other apps running.
I am consistently seeing Logic 8 use ALL of the first core and barely tough the other - sometimes freezing up - saying that the CPU(s?) can not handle the audio. Why do I not see the other core "bailing water" for the first core ? Isn't this the purpose of dual core ?. It is like one guy is lifting a 150lb sac and his partner is sitting there eating lunch - and the "team"'s efforts fail.
Is Logic itself not optimized for dual core processing duties ?

I get beachballs just logging in. No apps - then Safari is also very buggy. I have had to "pull the battery out" more times than I would like to recount.
That sounds very suspect, like there is something wrong with the hardware or your OS install. I have a MBP 2.33 and have none of the Logic or other system problems you describe. Have you looked at your System Profiler to make sure both sticks of Ram are being recognized? If one stick is bad and you're only utilizing 1 gig, that would cause slow downs and spiny beachball with Leopard and with Logic. Try running hardware test from your system disks, and also try running Rember, a small app that will run a series of read/write tests on the Ram, incase one stick has a bad sector.
If that comes up with nothing, I'd try setting up a new user, logging in to that user, and seeing if you're having the same slow down problems with Leopard, Safari, Logic, etc. If not, then you have some corrupt files in your User system somewhere. Whether that works or not, the next step would be to back up all of your data and do an Erase and Install.
Finally, if you have specific problems that you can recreate, you could also take the machine to Apple if you still have a warranty, and perhaps they can figure out your performance issues.
Good Luck!

"Design" view as a separate panel for dual-display

I just started using CS6 and was really hoping to see a new workflow enhancement for dual-display production with the "Design" view. I code all of my markup by hand in the code view. I would love to see the "Design" view in a separate panel that I could drag over to my second display by itself. Sort of like designers using "Code Inspector" as a panel on their second display. But for me, to maximize efficiency, I need the code to be closest to all of the menu bars and panels. I just want to make a code edit, click the DW "refresh" button, and glance at the layout in a "Design" without having to do additional clicking or moving pointer from display to display, ya know? Using browsers I have to save the file(s), upload the file(s), wait for the transfer to complete, move my pointer to the second display, click the refresh button, THEN look at the layout. Ugh. Been doing this for years, and really am tired of it. I have looked through the commands and online material hoping to find an update. Is this still a pipe dream?
Thanks for any info,
- DK

Apart from using the Code Inspector, I don't believe this is possible. It sounds like a sensible suggestion, though. You can file a feature request through the form at http://adobe.ly/DWBug.

How to add another operating system for Dual boot in Windows 7 boot configuration

Hi,
I have dell laptop with excellent configuration. On which i had windows 7 Home premium on C Drive and Windows server 2008 on D Drive.
Now i upgraded my Windows 7 Home Premium 32 bit to Windows 7 Ultimate 64 bit. But after fresh installation of Win 7 Ultimate my dual boot screen has gone and Win 7 is getting start directly.
Now i want to add my Windows server 2008 again for dual boot which is still on D Drive.
Can anyone please tell me how to add windows 2008 boot option in boot configuration to active my dual boot options.
Thanks

1) Using bcdedit.exe /enum
To enable the computer to also boot Windows 7 you again use bcdedit.exe with the following command which will copy the current Windows
Boot Loader details for Windows Server 2008 to Windows 7.
bcdedit.exe /copy
{current} /d "Microsoft Windows 7"
bcdedit.exe will
respond with something like the following.
The
entry was successfully copied to {................................}.
Using
bcdedit.exe /enum to again enumerate through the current entries within the BCD store you'll see the newly added entry.
Before you can use the newly created Windows boot loader configuration, you'll need to change the partition for Windows 7 using
the following two bcdedit.exe commands.
bcdedit.exe /set {.................................} device partition=D:
bcdedit.exe /set {.................................} osdevice partition=D:
If you now reboot the system you should now be able to boot into either Windows Server 2008 or Windows 7.

N95, N96 dual CPU, N97 single? Dual CPU s60 5th ed...

I've been playing around with Nokia's new QT development stuff. It's all pretty neat. It would be even better if I could use it to develop for Symbian under Linux or Mac, but that asides..
I came across a list of platform types:
Installer for all S60 5.0 (5th Edition) devices with a single CPU, such as the N97 and 5800 XpressMusic
Seems reasonable, but I also saw:Installer for all S60 5.0 (5th Edition) devices with dual CPUs
Hmmmm.. dual CPU S60 5th edition?
It lists some S60 3rd edition dual cpu phones too: N95 and N96
So, why exactly is the N95/N96 dual cpu, but the N97 isn't? Sure that was a step backwards?
Interesting though.. I wonder what the first dual CPU S605th phone will be? Presumably, something faaar better than the somewhat dull N97! I mean, honestly.. who wouldn't put a decent GPU in their flagship phone? presumably the same people who would put OS storage in double digit MB.. grrrr ..

I aggree, however I'm not sure these devices are frog leaping to that extent.
In the case of the N95, the dual core refers to core CPU and an additional DSP and GPU
N95: OMAP2420 - 330 MHz ARM1136 + 220 MHz C55x DSP + PowerVR MBX GPU
The N97, however,is of the same architecture (ARMv6) but with a higher clock rate (434MHz) It isn't a new generation chip per se. It's just running faster.. and without the GPU..
The N96 is a bit different.. Apparently the difference between the N95 v N96:
CPU: N96 has dual ARM9 264MHz with no floating point instructions (N95 has dual ARM11 332Mhz with vector floating point or graphics accelerator)
So, the N97 has the speed of the ARM11 from the N95, but all of the same missing features as the N96.. more or less..
This makes the prospect of a dual cpu s60 5th all the more intriguing
(thanks to wikipedia)

Quad Core - Dual CPU's

Does all G5 Quad Core (Dual CPU's) support all MAC OS and all MAC software?
For example, could I run MAC OS 10.x and Final Cut Studio on a Quad Core MAC without any configuration or problems?
Dave

Hi Dave;
I can vouch for what the Innovator says about running with a lower version of OS X. When I migrated from my PM G5 MDD to my Quad, I had forgotten to upgrade my maintenance disk to 10.4.3. I tried toboot from it on Sunday. The maintenance disk was at 10.4.2 so I thought that should be close enough. WRONG! The reboot failled big time. I had to down the 10.4.3 upgrade and apply it to that disk before it would boot.
Allan

Creative Cloud Requirement: Does Intel Pentium Dual CPU E2200 @ 2.2GHZ 2.20GHZ work?

Creative Cloud requirements ask for Intel® Pentium® 4 or AMD Athlon® 64 processor (2 GHz or faster). I have an Intel Pentium Dual CPU E2200 @ 2.2GHZ 2.20GHZ, which supposedly is better than a Pentium 4. Can I install the Creative Cloud apps on my computer as long as I have Windows 7 64 bit version?

By the way, I called support since no-one was answering this. And they said "yes, the requirements are a minimum. If you have something better (in this case I do) it will be fine".

Quicktime on dual CPU-PCs

Hello.
I have got big trouble getting quicktime for windows running on my PC. Everytime I try to export or sometimes even import a quicktime video, the applications that I use crashes, because of the quicktime component. I never had such problem on my mac computers (older mac pro, macbook pro) but since apple is not able to build powerfull maschines anymore, an the new mac pro is nothing but small, I prefer the much faster configuration I can get with PCs. So quicktime works fine on 2 of my PCs. Only on my biggest dual CPU Workstation it crashes. I searched the internet and found out that quite a view Dual-CPU-PC-Users have the same Problem.
Does anybody know how to fix this? I'm really desperat since i need the quicktime capability on this particulary Dual-CPU-Machine for work. Please Help me.
Thanks!!

>Is it possible to ask LabVIEW to execute all subVI in a sigle thread
>on a dual CPU machine. Our SubVI are calling an API. And this API
>look for the TheadID to recognise which application is doing the call.
>We would like to get advantage of multiprocessing. But having all the
>SubVI call with the same ThreadID?
VI Guy wrote:
> Probably not.
>
> Each executionriority system in LabVIEW gets two threads to service
> it in multithreaded mode.
>
> So even though you can specify the exact same execution system and
> priority for each sub vi, you can not be guarantied that you will get
> the same thread each time.
>
> Good luck.
You should be able to do this. If you go to VI.lib/utility/sysinfo.llb,
you can open a vi called threadconfig.vi
. This will allow you to edit
the number of threads per execution system and priority. So to use VI
Guy's idea, you could set one thread per execution system, then set your
different VI's into different execution systems. Also, you should take
a look at the recent issues of LTR. (www.ltrpub.com) There was an
article there about assigning a specific VI to a specific processor on
multi-processor machines.
Good luck,
Dave
David Thomson 303-499-1973 (voice and fax)
Original Code Consulting [email protected]
www.originalcode.com
National Instruments Alliance Program Member
Research Scientist 303-497-3470 (voice)
NOAA Aeronomy Laboratory 303-497-5373 (fax)
Boulder, Colorado [email protected]
There are 10 kinds of people: those who understand binary,
and those who don't.

Dual CPU G4 panics and applications quit

Hi,
I've had a single CPU G4 (digital-audio) machine running OS 10.4.x for quite some time. It's been remarkably stable.
Now, I've acquired a 533 mhz dual cpu machine, also a digital-audio model. Applications spontaneously quite much more often, more than once per day, mostly web browsers, and the kernel locks up several times per week.
I've run MemTest and repaired permissions and run the disk utility and run DiskWarrior, all to no avail. I also did a archive and fresh reinstall of OS 10.4.2, then a batch update with software update. Didn't help.
I'm wondering if certain dual cpu machines are just more prone to crash than their single cpu counterparts.
Any suggestions about how to make this machine more stable? It's not on a network, except for a conventional PPoE DSL connection.
I know I could try zapping the pram and resetting the PMU. Anything else?
It gets a lot of heavy iTunes use and also Rhapsody use. Is that a clue?
Thanks in advance,
Tim

Hi
dual CPU machines should be just as stable.
As already mentioned the first thing I'd suspect is the RAM. If you still have your old machine and could swap the RAM over (if you haven't already done so) that should prove it one way or the other.
Also dual CPUs generate more heat, especially under heavy load, which can result in kernel panics. I assume the computer has plenty of ventilation and isn't boxed in, the vents are clear of dust and the fans are working etc...

EEM for high CPU C4510R with SUP7E

Hi all,
I have a client that experiences high CPU usage on a pair of C4510Rs at random times. The issue causes their HSRP to flap which in turn causes network outages. It only seems to occur for around 10 to 15 minutes and by the time I am called to look into the problem it has passed. I suspect it is some type of traffic that is being CPU punted and is overloading the CPU and I would like to collect some CPU stats and CPU packet captures while the problem is occuring to further troubleshoot.
I used a previous C4510R EEM script (https://supportforums.cisco.com/thread/2065177) as a template and created the following script:
event manager applet cpu_stats
event snmp oid 1.3.6.1.4.1.9.9.109.1.1.1.1.3.1 get-type exact entry-op ge entry-val "70" exit-op lt exit-val "70" poll-interval 5 maxrun 300
action 1.01 syslog msg "------HIGH CPU DETECTED----, CPU: $_snmp_oid_val %"
action 1.02 cli command "enable"
action 1.03 cli command "show clock | append bootflash:cpu_stats"
action 1.04 cli command "show proc cpu sort | append bootflash:cpu_stats"
action 1.05 cli command "show platform health | append bootflash:cpu_stats"
action 1.06 cli command "debug platform packet all count"
action 1.07 cli command "show platform cpu packet statistics | append bootflash:cpu_stats"
action 1.08 cli command "show proc cpu history | append bootflash:cpu_stats"
action 1.09 cli command "debug platform packet all receive buffer"
action 1.10 cli command "show platform cpu packet buffered | append bootflash:cpu_stats"
action 1.11 cli command "show proc cpu sort | append bootflash:cpu_stats"
action 1.12 cli command "show platform cpu packet statistics | append bootflash:cpu_stats"
action 1.13 cli command "Show proc cpu history | append bootflash:cpu_stats"
action 1.14 cli command "show platform cpu packet buffered | append bootflash:cpu_stats"
action 1.15 cli command "show log | append bootflash:cpu_stats"
action 1.16 cli command "undebug all"
end
However when looking at the CPU history after the problem occured again it appears that the SUP7E has one physical CPU that has two cores and one or two of the cores hits 100% when the issue occurs while the system CPU stays low as can be seen below:
C4510#show proc cpu hist
History information for system:
output omitted
    111111111115511111111111111111111111111111111111111111111111111111111111
    600105000112231001211131111017221011110001112010212016101010121012112112
100
90
80
70
60
50            **
40            **
30            #*
20 *    *     ##                *                       *
10 ########################################################################
   0....5....1....1....2....2....3....3....4....4....5....5....6....6....7.
             0    5    0    5    0    5    0    5    0    5    0    5    0
                       CPU% per hour (last 72 hours)
                    * = maximum CPU%   # = average CPU%
History information for core 0:
output omitted
               991
    888787888789968888787877787787888788787888879898888898888788
100            **
90            **
80            **
70            **
60            **
50            **
40            **
30            *#
20            ##*
10 ***********##***********************************************************
   0....5....1....1....2....2....3....3....4....4....5....5....6....6....7.
             0    5    0    5    0    5    0    5    0    5    0    5    0
                       CPU% per hour (last 72 hours)
                    * = maximum CPU%   # = average CPU%
History information for core 1:
output omitted
    211112111119921111111111111113111111111111111111111112111111
    855554544659906476955796556651995456554446767454847468565565
100            **
90            **
80            **
70            **
60            **
50            **
40            **
30 *          #*                *                       *
20 ******* **##** ***************** ****   **** * * * ********************
10 ########################################################################
   0....5....1....1....2....2....3....3....4....4....5....5....6....6....7.
             0    5    0    5    0    5    0    5    0    5    0    5    0
                       CPU% per hour (last 72 hours)
                    * = maximum CPU%   # = average CPU%
So I walked through the SNMP OIDs for the two cores without luck (also came across this article
https://supportforums.cisco.com/thread/2179888 to create SNMP OIDs for the cores by parsing the output of "show proc cpu" with a cron job). An SNMP walk of the CPU MIBs comes back with the following:
.1.3.6.1.4.1.9.9.109.1.1.1.1.3        - cpmCPUTotal5sec.5000
.1.3.6.1.4.1.9.9.109.1.1.1.1.4        - cpmCPUTotal1min.5000
.1.3.6.1.4.1.9.9.109.1.1.1.1.5        - cpmCPUTotal5min.5000
.1.3.6.1.4.1.9.9.109.1.1.1.1.6        - cpmCPUTotal5secRev.5000
.1.3.6.1.4.1.9.9.109.1.1.1.1.7        - cpmCPUTotal1minRev.5000
.1.3.6.1.4.1.9.9.109.1.1.1.1.8        - cpmCPUTotal5minRev.5000
Each of which seems to correspond with the system CPU. So in order to continue without creating an overly complex EEM scripts that parse the show proc cpu output and pass it to another EEM script that checks the value I decided to try a lower entry and exit value of 40%. The next time the problem occured the system CPU went over 40% but did not seem to trigger the script. What is important to note is the first time I did a full SNMP walk of the OID I was able to get the CPU over 50% and the script triggered.
So I moved down to 20% for the trigger value, tried with different SNMP OIDs like .1.3.6.1.4.1.9.9.109.1.1.1.1.6.5000 and doing an SNMP walk of the Cisco MIB without success in getting the script triggered (see attached output). Is there a problem with the script or the SNMP MIBs I am using?
Thanks in advance!
David.

Hi Joseph
I have written the following set of scripts to do the job (good to just post them here if anyone else is having the same problem!), I have tested individual components and they all seem to function as expected, so just waiting now for the high cpu event to occur again:
The first script is just to set the prev_cpu_state to low so the main script can detect the first transition to high and does not spit log messages out every minute because the context variable is not initialised:
event manager applet set_cpu_state
event none
action 1.01 handle-error type ignore
action 1.02 context retrieve key CPU_STATE variable prev_cpu_state
action 1.03 handle-error type exit
action 1.04 set prev_cpu_state "low"
action 1.05 context save key CPU_STATE variable prev_cpu_state
end
The second and main script checks the CPU level of both cores over 5 seconds and takes the highest of the two, if this is over 50% and the cpu state was previously low then it will trigger the third script.
event manager applet dualcore
event timer cron cron-entry "* * * * 0-6"
action 1.00 cli command "en"
action 1.01 cli command "show process cpu | include five"
action 1.02 regexp "Core 0: CPU utilization for five seconds: ([0-9]+)%" $_cli_result result c0cpu5sec
action 1.03 if $_regexp_result eq 1
action 1.04 end
action 1.05 regexp "Core 1: CPU utilization for five seconds: ([0-9]+)%" $_cli_result result c1cpu5sec
action 1.06 if $_regexp_result eq 1
action 1.07 end
action 1.08 handle-error type ignore
action 1.09 context retrieve key CPU_STATE variable prev_cpu_state
action 1.11 handle-error type exit
action 1.12 if $c0cpu5sec ge $c1cpu5sec
action 1.13 set highest_core $c0cpu5sec
action 1.14 else
action 1.15 set highest_core $c1cpu5sec
action 1.16 end
action 1.17 if $highest_core ge 50
action 1.18 if $prev_cpu_state eq low
action 1.19   set prev_cpu_state "high"
action 1.20   syslog msg "----HIGH CPU DETECTED----, CPU: $highest_core %"
action 1.21   cli command "event manager run collect_stats"
action 1.22 end
action 1.23 else
action 1.24 if $prev_cpu_state eq high
action 1.25   set prev_cpu_state "low"
action 1.26   syslog msg "----CPU BACK TO NORMAL----, CPU: $highest_core %"
action 1.27 end
action 1.28 end
action 1.29 context save key CPU_STATE variable prev_cpu_state
The third script just runs all of the require show commands for debugging the issue and appends the output to a file called cpu_stats on the bootflash:
event manager applet collect_stats
event none
action 1.02 cli command "enable"
action 1.03 cli command "show clock | append bootflash:cpu_stats"
action 1.04 cli command "show proc cpu sort | append bootflash:cpu_stats"
action 1.05 cli command "show platform health | append bootflash:cpu_stats"
action 1.06 cli command "debug platform packet all count"
action 1.07 cli command "show platform cpu packet statistics | append bootflash:cpu_stats"
action 1.08 cli command "show proc cpu history | append bootflash:cpu_stats"
action 1.09 cli command "debug platform packet all receive buffer"
action 1.10 cli command "show platform cpu packet buffered | append bootflash:cpu_stats"
action 1.11 cli command "show proc cpu sort | append bootflash:cpu_stats"
action 1.12 cli command "show platform cpu packet statistics | append bootflash:cpu_stats"
action 1.13 cli command "Show proc cpu history | append bootflash:cpu_stats"
action 1.14 cli command "show platform cpu packet buffered | append bootflash:cpu_stats"
action 1.15 cli command "show log | append bootflash:cpu_stats"
action 1.16 cli command "undebug all"
end
Regards,
David.

Msi 9105 dual cpu config

in bios, mps version control for os i have setting 1.1 or 1.4? my system op. is windows xp professional
dual cpu working
1.4 ghz 512k 133 x2
2.1 ghz ram ddr memory

1.4 for windows,some linux versions need older 1.1

Sar command for dual cpu's

Similar Messages

Maybe you are looking for