!Warning OlioWeb0.wrf Not All Olio metrics found in this sample , make sure you are using the latest jar files

[Kenneth_hansen]

We have for a long time tried to get Vmmark to work, now we are almost there, and wish to get the TILE SCORE file

But this is saying:

!Warning OlioWeb0.wrf Not All Olio metrics found in this sample , make sure you are using the latest jar files

.

Any suggestions?

[Kenneth_hansen]

Hi Rebecca,

I'm sorry I picked that out.,

Please see atached file

[Kenneth_hansen]

hi rebecca.

Did you have time to look at the file

[RebeccaG]

Hi Kenneth, apologies for the delay.

These workloads are working correctly:

Olio

Standby

vMotion

SVmotion

Deploy

These workloads are not working correctly:

Mailserver

DVD Store 2

I noticed you have a lot of other VMs powered on in your VMmark cluster during your run. You should remove these. VMmark workload VMs should be the only powered-on VMs in your cluster (unless an appliance VM is required by your infrastructure) because they can affect your benchmark results.

Olio is running correctly, but failing response time requirements. This is generally due to resource contention in the underlying environment. Resource contention also might be the root cause of DVD store failures. You should use a dedicated cluster, network, and storage with no other work running on the hardware under test.

I will continue looking at Mailserver for issues and let you know.

[jamesz08]

There seems to be something wrong with your domain on the mailserver.  Can you log into your mailserver VM and run the command dcdiag from the command prompt and post the output?

For ex:

dcdiag > dcdiag.out

[Kenneth_hansen]

Hi Jamesz08

Thanks a lot for your reply.

Please see dcdiag attached

thanks a lot

[jamesz08]

It appears that there are some issues communicating with the domain controller (system log has many errors).

Could you delete the mailserver entry from the hosts file of the mailserver and then reboot it. 

Please verify that the DNS on your mailserver VM is set to 127.0.0.1

Then from cmd prompt run gpupdate

Then redo the dcdiag command and post the output.

[Kenneth_hansen]

I just confirmed that the DNS was 127.0.0.1 and removed the entry from the hostfile

please see attached files

thanks again

[jamesz08]

It looks like the domain problem is fixed, but your exchange transport is failing.  There is an entry about an unexpected shutdown.  Make sure you are using the Windows "Shutdown" or "Restart" options rather than just resetting the VM. 

Restart your mailserver VM and check if the "Microsoft Exchange Transport" service has started.  It might take a few minutes after boot.  If the transport has not started then please follow the "How to Change an Existing Configuration File" resolution step listed in this KB:

https://support.microsoft.com/en-us/kb/944752

Once completed restart again and make sure all the "Automatic" Microsoft Exchange services are running.

[Kenneth_hansen]

Hi Jamesz08

I just fixed the Exchanges services, but it still fails:-(

Can you see anything in the log?

Thanks a lot

Looking forward to hearing from you

[jamesz08]

Can you verify that your client DNS is set to the mailserver IP and it has joined the domain?  If not please change these items and rerun.

Also, you can just try running Loadgen directly on the client until you have fixed the issue instead of using the Vmmark harness. You can use the configuration file found in c:\vclient\mailserver should be named something similar to vmmark2-mailserver0.xml, change the simulation length to 15 minutes and skip the initialization and run immediately. 

[Kenneth_hansen]

Thanks a lot again.

I can confirm that the mailserver is domain joined and the DNS it set to ist own IP. Please see atached photo.

I just rebooted all servers and re-run. Please see attached files.

Looking forward to hear from you soon

[jamesz08]

Are you able to run loadgen independently of the Vmmark harness?

[Kenneth_hansen]

Can you explain how to do that...... Thanks a lot

[jamesz08]

Instructions are listed in Appendix D of the Vmmark 2 Benchmarking Guide.  pg 175 has "Manually Run Loadgen" instructions.

The results and log files after the run will be in C:\Program Files\Exchange Load Generator

[Kenneth_hansen]

Hi Jamez08, sorry for the late reply but just got back from vacation.

I'm able to run loadgen manually, please see attached photo.

I just run VMMARK again and it looks a little better, do you have any ideas what to try next

Thanks a lot

[jamesz08]

Can you try removing the mailserver FQDN entry from your client hosts file.  Just have an entry for mailserver0.

Otherwise I am at a loss.  Might try rebuilding your client,

[Kenneth_hansen]

I tried to move the entry. I looks like it make a diffrence, but still errors

What do you think?

[jamesz08]

Remove the entry from the Client0 hosts file.  It looks like you only removed it from primeclient.

[Kenneth_hansen]

I just removed the entry from the client0 as well, but it still not work.

If anyone need to see log etc, we can arrange webex or teamviewer

Looking forward to hearing from you soon

[jpschnee]

You don't appear to have resolved the Mailserver issues.  No results were recorded. At this point I would suggest you go through the documentation again and make sure you haven't accidentally skipped a step or possibly installed an unsupported version of the software stack.  When you ran mailserver manually did you make sure not to change anything or modify any fields?

As far as your DS2 issues, it appears you may have either a network or naming issue. If you look at both DS2WebB and DS2WebC wrf files the last iteration is good, all threads connected and produced results.  The previous iteration attempts failed.  I can only think that you either updated something (like the DS2-DB hostname) on the two DS2WebB/C systems or you may have more than one system responding to that ip address.  I would rerun this portion and see if this behavior is repeatable.