I'm a student and i'm making a college work about VMmark, i haven't the hardware described on "VMware VMmark®
Benchmarking Guide (VMmark 2.1 Revised March 10, 2011)". So I need to know if i can execute the test with an inferior hardware in just one wirtual machine with Windows XP, for example. I already have the ESX5 installed on server and the vSphere on a normal machine.
Hi Daniel,
We're glad that you are interested in using VMmark for academic purposes.
A official run of Vmmark (i.e. one that can be published on the VMmark website) is made up 3 application workloads over a minimum of eight different virtual machines running concurrently on an ESX host cluster. Additionally, VMmark creates periodic vMotion, Storage vMotion and VM Deploy operations that stress the infrastructure. So, the VMmark test is not executed in one virtual machine, but made up of at least one set of eight virtual machines plus infrastructure operations.
However, the rules for using VMmark in an academic paper or research differ from those for publishing an official benchmark result.
For example, if you are using VMmark for academic purposes, you are not required to run all workloads. You could choose to enable only one, some, or all workloads at a time. This means that it is likely you can run at least some VMmark workloads on an ESX host that has inferior hardware to what's recommended, because the recommendation assumes that all workloads will be running. I'd recommend starting with only one workload running, then add workloads incrementally to be sure your host can handle the load.
You mention that you have only one ESX host. An official run requires two hosts to support the vMotion operation. However, if you have only one ESX host, you should disable the vMotion operation in the VMMARK2.CONFIG file. You would also need to set DoClusterMapping=0 (in the VMMARK2.CONFIG file) to avoid errors. All other workloads should be fine running on only one host.
In addition to your ESX host and vCenter machine, you will also need a minimum of one other physical machine installed with Windows Server 2003 or 2008 to serve as the benchmark's client.
VMmark's rules for academic and research use are outlined in the VMmark Run and Reporting Rules; be sure to give these a careful read-though. For example, an academic paper should not attempt to make competitive comparisons using VMmark results, and should disclose any way in which the benchmark differs from its official use, e.g. if some workloads have been disabled.
Please feel free to post back here if you have any further questions.
Rebecca
Hi, RebeccaG!
First, I want to thank you for your answer my last question. Your answer really helped me.I have another doubt, this time it would be a problem actually, as I said before I have not a great hardware, so I decided to create a mailserver in a virtual machine created by the book and also the client machine, everything by the book "VMmark Benchmarking Guide" and following the steps, I came in running the "VMmark harness" and noticed that I have a problem on the client machine. Checked the process manager that stafproc is not running, so I tried to reinstall but it does'nt work, it does'nt run, it found that the mailserver is running without the problems, but does not run on the client. Can you give me any tips about it?
Hi Daniel,
That's fine, I am happy to help troubleshoot. I apologize for taking a while to get back to you, I will try to be faster in my responses in the future.
Good, it is important to follow the benchmarking guide exactly. I would just like to double check a couple of things: Did you allow the STAFProc.exe to pass through the firewall? Is STAF configured to start automatically on the client using a Scheduled Task? Lastly STAF may not start if the STAF.cfg file has a misconfiguration.
Could you open a cmd prompt on your client, then type 'staf local ping ping' and let me know what is the output?
Besides your ESX host, you must use at least one other physical machine installed with Windows Server 2003 or 2008 to serve as the benchmark's client. Are you using this one physical machine to run the harness and also act as client0? i.e. it is possible to have the physical machine run the harness, while creating client0...n as VMs.
Rebecca
Hi, as your question I discovered what was wrong, for being my client0 I added the command line "service stax library JSTAF execute C:\STAF\services\stax\STAX.jar OPTION \J2=-Xverify:none parms "numthreads 100 logtcelapsedtime enabled logtcstartstop \enabled"" and with this line the STAFProc.exe not start, I tried the forum and found another topic on this subject which appears in the following command "SERVICE EVENT LIBRARY JSTAF EXECUTE C:\STAF\services\event\STAFEvent.jar SET MAXQUEUESIZE 1000" and also tried this command did not work. My point is: right now I'm getting this "WARNING: The STAX service is not registered on the specified machine." when trying to run the command "java -jar C:\STAF\services\stax\STAXMon.jar".
Hi Daniel,
The text that needs to be present in your staf.cfg file on client0 is:
service stax library JSTAF execute C:\STAF\services\stax\STAX.jar OPTION \
J2=-Xverify:none parms "numthreads 100 logtcelapsedtime enabled logtcstartstop \
enabled"
service event library JSTAF execute C:\STAF\services\stax\STAFEvent.jar
(page 71 of the VMmark Benchmarking Guide)
The first code snippet you posted is missing the last line ("service event library JSTAF execute C:\STAF\services\stax\STAFEvent.jar"). Then, if you replaced the first three lines with the second snippet you posted, that would also cause problems because all four lines are required.
Thanks for your environment description. It is helpful because your hardware is different from the typical setup.
Daniel,
I wanted to add that if you got that second code snippet out of this thread, that solution is referring to an earlier release of VMmark, VMmark 1.0.
Lastly the second code snippet you posted below should include the path C:\STAF\services\stax\STAFEvent.jar not C:\STAF\services\event\STAFEvent.jar. You can see the difference in the code I posted in the previous post.
Thank you,
Rebecca
Hi, on the last tip that you gave me, actually I forgot to copy a piece of the code, it worked with the part that you gave me and about the other code that I put not really realize it was another version, much thanks for the help.
Hi Daniel, I am happy to help, that is why I am here.
The error you have experienced happens if the trust levels have not been set correctly for the local machine.
In your STAF.cfg, you should have:
# Set default local trust
trust machine local://local level 5
in addition to your other trust settings. The two lines above are part of the default STAF.cfg when you install staf. I believe you deleted this line at some point.
When you modify STAF.cfg, remember to stop the staf service then restart it before you test the harness again. Since the local machine does not have permissions, you will probably have to kill the STAFproc in the Task Manager or reboot the machine. Then start STAF using c:\STAF\startSTAFProc.bat.
Lastly, I noticed your local machine name is "note-daniel.dexter.local". For mailserver to work correctly, you must have the mailserver be the domain controller of a three part domain, e.g. "eng.vmware.com", and client0 is a member of this domain. It looks like your domain is "dexter.local", which is only two parts. Please use a three part domain name. I believe your client will also need the hostname "client0", not "note-daniel".
Thank you,
Rebecca
Hi again, I made the change as recommended domain and it worked, thanks again. I have another doubt as mentioned
Hi,
I'd recommend you take a look at the benchmarking guide and the notes sections of the VMMARK2.config file, searching for WORKLOADLIST. Those sections should help in case you want to make future modifications.
In this case, you'd likely want to set the workload list as the following in your VMMARK2.config:
WORKLOADLIST="MailServer"
Hi this part of WORKLOADLIST was wrong because I neglected at the time to edit the file and not wanting to put any name MAILSERVER capitalized, ie the name was right but it was all in capital letters when it should be lowercased, thanks again, I found another error that the description points to a part of the code and saw vmmark2_main.xml also has a part of him talking about that stretch the "local://local".
Thanks again
You're seeing that error because you have specified "mailserver" instead of "MailServer". The workload names are case sensitive. You should review the VMMARK2.config and follow the provided formats.
Hi good evening, I checked the writing mode and if I put it exactly MailServercomes back to the error "Could not resolve VMmark / OlioDB / SERVERS" but the only item in WORKLOADLIST is MailServer and if I put everything lowercased it returns to the error mentioned above.
As the harness is failing instead of cleaning up after a good run you may have a stale variable for WORKLOADLIST. I'd suggest either restarting staf/stax or rebooting your prime client then re-running with suggested WORKLOADLIST="MailServer".
Hi, thanks for the help but it did'nt work, even restarting the client machine it keeps showing the error "Could not resolve VMmark / OlioDB / SERVERS" Attached is my VMMARK2.CONFIG. I do not know what might be going wrong if the changes already been made as recommended below is a list of virtual machines
Can you try uncommenting all of the other workloads server variables? The harness may be failing due to it expecting variables that aren't there.
Ex:
Hi, I made the change and giving it another error "Could not PING The Following 1 machines: [u'MailServer0] Check STAFProc That is running" I checked the mailserver and the client and are running normally STAFProc, I did a traditional ping test and this also functioned normally, I checked the firewall from both machines and they started or stopped is giving the same error, also noticed that the "VMmark/MailServer/dblocation" pointing to "E:\VMmark" and "VMmark/MailServer/backuplocation" pointing to the F:\ when in fact due to the unavailability of letters had to put "F:\VMmark" to dblocation and "G:\" backuplocation to where I can change the configuration that is already defined?
Hi. About the problem of change in the path "F:\VMmark" to dblocation and "G:\" backuplocation it's ok now, I just keep with the other problem mentioned lately, the normal ping test is responding to any of the machines. Doing a ping test by the STAF MailServer for himself and for the client it is already responding normally. About the ping test client in the STAF, I found a problem trying to ping the MailServer. Doing a ping test to himself, ping works right. But to MailServer has an communication error, I set the hosts to MailServer but it's not working.
Hi,
Can you retry the staf ping test with both firewalls turned off? Also, can you update this thread with the ipaddresses being used and attach the STAF.cfg files from both your prime client and your MailServer VM?