The Real (and Virtual) Adventures of Nathan the IT Guy » Windows Server 2008

vSphere Performance tab issues

Nathan Simon — Mon, 25 Jul 2011 01:09:39 +0000

If you run into an issue with the “Overview” page under the Performance Tab in vCenter, then follow these steps to resolve.

Log into the vCenter server
Browse to the following locations:
C:\Program Files (x86)\VMware\Infrastructure\tomcat\webapps\statsreport\WEB-INF folder
Open the web.xml file in notepad
Once the file is open, and search (CTRL+F) for vc.webservice.timeout
Note: The entry should be near the top of the file, about 15 lines down from the top.

Change the value on the following line from 1 to 5

Restart the machine or the vCenter Web Services and you should be good to go…

Good Luck!

-NS

iLO 2 Update 2.06 – 06/04/2011

Nathan Simon — Wed, 29 Jun 2011 02:06:01 +0000

One of the key things I keep on top of is firmware and software updates. Ilo has been updates at the beginning of the month and a lot of people over look it. So I am here to remind you all of the updates. Below you will find the list of fixes, additions, and enhancements.

Problems fixed since last release:
Disabled SSL/TLS renegotiation to protect iLO 2 against SSL/TLS plaintext injection attacks (CVE-2009-3555).
Fixed security vulnerabilities mentioned in CVE-2008-7270 and CVE-2010-4180.
Fixed an issue with ROM randomly fails to communicate to iLO 2 during POST causing some Mezzanine cards on blades to get misconfigured.
Fixed an issue where the iLO 2 login page could be requested and displayed in HTTP mode instead of HTTPS secure mode.
Fixed an issue with Set_Pwreg_Alert_Threshold.xml doesn’t validate parameters correctly.
Fixed an issue with Set_Pwreg_Alert_Threshold.xml gives false message when server is in aux power.
Fixed an issue with iLO 2 accepting invalid value for warning_duration.
Fixed an issue where user cannot exit from serial CLI when login prompt is shown.
Fixed an issue where pressing ESC + SHIFT 9 key sequence and ESC + SHIFT Q key sequence multiple times can cause serial CLI to stop responding.
Fixed an issue where the Linux console via iLO 2 Virtual Serial Port could no longer work after iLO 2 reboots.
Fixed an issue where mounting Scripting Virtual Media using an invalid hostname in the URL can cause iLO 2 network to stop responding.
Enhancements/New Features

Added Brown-out Recovery Feature for ML/DL servers.
Added Custom SSL Certificate Signing Request (CSR).
Added Support for 2048-bit RSA key in SSL CSR.
Added XML command to delete SSH keys.
Added option to disable the Virtual Media Access port.
Added Support for Java 64-bit JVM/JRE.
Added Support for Integrated Remote Console on Internet Explorer 64-bit Editions.
Added support for Internet Explorer 9, dropped support for Internet Explorer 6.
Improved performance of the iLO 2 Virtual Serial Port when connected via SSH.
Known issues:

After importing a SSL certificate containing a 2048-bit RSA key length, the iLO 2 webGUI could become noticeable slower to some users. This is due to the fact that SSL handshakes with 2048-bit keys take several times more CPU cycles than doing SSL handshakes with 1024-bit keys. The iLO 2 login page could take an extra of 5 to 10 seconds to show up and clicking on the webGUI after the websession is left idle for more than two minutes, could cause the old SSL sessions to expire and force iLO 2 to do new SSL handshakes.
After iLO 2 v2.06 firmware is flashed and running for the first time, iLO 2 will automatically begin to pre-generate 1024-bit and 2048-bit RSA key pairs that are required for the Certificate Signing Request (CSR). This process takes around 20 minutes to complete. During this time, the Remote Console needs to remain closed to allow iLO 2 to finish pre-generating these keys and store them in the NVRAM.

Download the executable here. If you are running Windows Server 2008 directly on hardware you can run the executable and update via Windows, if you are using an ESX Host, you can extract the files and then update ILO from a management server via the browser.

-NS

GoDaddy unveils cloud computing

Nathan Simon — Fri, 17 Jun 2011 11:56:54 +0000

Cloud Computing has become so prevalent over the last couple years and GoDaddy is finally on the bandwagon.

It looks like web hosting giant GoDaddy is now in the cloud computing business with a new service called Data Center On Demand, which could potentially make a dent in the market share of providers such as Amazon Web Services or Rackspace. The service is currently in a limited-release phase and is expected to launch in July.

They have many operating systems to choose from like Windows(of course), CentOS, Ubuntu, and Fedora. Prices Range from 30$ a month to over 300$ a month, 1GB of Ram Virtual Machine to 8GB of Ram dedicated machine.

How to clear DCDIAG Errors

Nathan Simon — Tue, 14 Jun 2011 15:01:04 +0000

What is DCDIAG you ask?

This command-line tool analyzes the state of one or all domain controllers in a forest and reports any problems to assist in troubleshooting. DCDiag.exe consists of a variety of tests that can be run individually or as part of a suite to verify domain controller health.

Let’s get back to clearing the errors, one might say that this is a difficult task. Not so much.

Just open up Eventvwr and clear the logs, re-run dcdiag and the errors should go away.

You could also try running dcdiag /fix to let it apply safe fixes.

For a full list of dcdiag commands, go here.

-NS

Changed Block Tracking and Veeam Backup 5.0.2

Nathan Simon — Wed, 06 Apr 2011 02:49:48 +0000

If you have issues with your backups because of a Changed Block Tracking Issue, it could because you created the backup job with a ESXi user that does not have permissions to enable the “ctkEnabled” (Changed Block Tracking) to true. Either assign the proper permissions and rerun the job, or use vCenter or ViClient to modify the flag manually. The VM must be off. Then Edit the VM’s settings, click on the options tab, then on “General” in the list, then click on “Configuration” then scroll down the Configuration Parameters until you see “ctkEnabled” type “true” for every instance you see. Either solution should fix your issues.

-NS

Server 2008 R2 Robocopy New Features

Nathan Simon — Thu, 24 Mar 2011 11:32:26 +0000

Robocopy has been around for a long time, but there are 2 switches that you want to start incorporating into your scripts, they are /ZB and /DCOPY:T

/ZB :: use restartable mode; if access denied use Backup mode. When you use the /ZB option, this command uses Restart mode. If access is denied, this command uses Backup mode. Therefore, a user account in the Backup Operators group on a computer can run the command successfully.

/DCOPY:T :: COPY Directory Timestamps.This command will make sure that all timestamps for folders is retained, having archive folders with dates that do not make sense, will most definitely confuse the average end-user.

Adding these two switches to your existing scripts will help you migrate files and folders with confidence.

Good Luck!

-NS

License Issues when Upgrading Veeam 4.1 to 5.01

Nathan Simon — Sat, 19 Mar 2011 16:49:50 +0000

What do you do if you upgrade from Veeam 4.x to 5.01 and all your jobs are failing, you’ve applied the proper licensing, you can see that there are licenses available, but you cannot get your hosts licensed. Well i figured out a workaround, basically what is happening is that licensing is bound to the backup job, so to reset this you have to get creative. Lets start with creating a new VM, it can be a plain VM with no disk allocated to it. Now create a new job in Veeam Backup and Replication 5.01, choose that “shell” of a VM and run the backup. This will add the ESX host that the VM is being run by to the licensed servers. After the job has been completed, Veeam will recognize all hosts and you will then be able to resume your backups successfully. Please leave a post or comment if this helped!

-Cheers

-NS

Exchange Management Console fails to Launch

Nathan Simon — Sat, 12 Mar 2011 21:03:27 +0000

You get the following error message.

‘The following error occurred when searching for On-Premises Exchange server:[myservername.local] Processing data from remote server failed with the following error message: The user “domain\username” isn’t assigned to any management roles. For more…. It was running “Discover-ExchangeServer – USeWIA $true -SupressError $true’

Now the reason this would happen, is if you are not a member of “Organization Management”, and this happens mainly if you were not the person that did the Exchange 2010 install. ie. Exchange 2010 was installed with a different userid.

All you have to do is add the proper userid’s to the “Organization Management” Group and you will be good to go.

Delete Failed DCs from Active Directory

Nathan Simon — Fri, 19 Nov 2010 14:26:26 +0000

What do you do if a Domain Controller dies and cannot be revived? Just deleting it from Active Directory won’t do it.

There is only one thing you can do, remove it forcibly.

Petri has an article that can help you do this.

When you try to remove a domain controller from your Active Directory domain by using Dcpromo.exe and fail, or when you began to promote a member server to be a Domain Controller and failed (the reasons for your failure are not important for the scope of this article), you will be left with remains of the DCs object in the Active Directory. As part of a successful demotion process, the Dcpromo wizard removes the configuration data for the domain controller from Active Directory, but as noted above, a failed Dcpromo attempt might leave these objects in place.

Good Luck!

Task Scheduler “Enable All Tasks History”

Nathan Simon — Thu, 23 Sep 2010 15:03:43 +0000

When you check the properties of a recently created task, you notice that “History” is disabled, enabling this can be a little conveluted, but easy when you know how. Close the tasks properties, then highlight “Task Scheduler” and then click on “Action” above, you will then see “Enable All Tasks History”. Once this option has been clicked, you will then be able to see history of any and all tasks created.

-NS