Archive for March, 2009

I was flicking around the VMware website when I come across something which made me rather excited…. “Access to your Virtual Center features from your mobile phone”

Below Ive done a copy and paste from the origionl VMTN blog which covers this, if you want to have a read you can find it here.

vCMA allows you to monitor and manage VMware Infrastructure from your mobile phone with an interface that is optimized for such devices. Specifically, it allows you to:

  • Search for virtual machines in your data center
  • Migrate virtual machines from one host to another using vMotion
  • Execute recovery plans using VMware Site Recovery Manager
  • Access Scheduled Tasks, Alarms and Events
  • And much more…

In order for you to access vCMA, you will need to deploy a virtual appliance; call it the vCMA server.  The vCMA server must be connected to VMware vCenter or any of the ESX servers that you want to manage.  Once the server component is set up, you can manage your datacenter from the convenience of your mobile phone.

For me the ability to perform tasks such as VMotion from the mobile is pretty cool, but personaly its the VMware Site Recovery intergration that interests me the most, this is really good stuff !

Anyone whos new to VMware, I highly recomemnd you check out the forums and VMTN blogs.

This content is password protected. To view it please enter your password below:

The other night I was up installing the new Celerra SRM failback plug-in and I come across this hideous error “Installation Error – 1155 instmsi30.exe not found” when I tried to install the plug-in.

The virtual machine I installed this on was built from scratch rather than a template which I use at our production site so I figured it would be something to do with the Windows installer package. After upgrading to Windows Installer version 3.1  everything installed perfectly.

You can download the Windows Install 3.1 package here

Hope this helps.

For those of you who are familiar with MMINFO you’ll know how powerful it is and how much information you can get from the media database. Some people like to use the built in reporting functionality built into Networker NMC but personally I find it unreliable and more times that not I find the numbers just don’t add up.

So this week I was importing some mminfo data into Excel to sort when I thought that it might be something worth writing a post on, below you’ll see how to get the MMINFO fields into separate columns.

The command below will ouput  results in standard MMINFO  format.

 C:\>mminfo -avot -r “client,name,ssid,cloneid” -q -t “1 day ago” >report.txt

Now lets take a look at whats in the txt file.

 client    name                           ssid          clone id     VSS SYSTEM BOOT:\              280633504   1236934816     VSS SYSTEM BOOT:\              280633504   1236934817     VSS SYSTEM SERVICES:\          247079075   1236934818     VSS SYSTEM SERVICES:\          247079075   1236934819     D:\                            4240056601  1236934932     D:\                            4240056601  1236934933     VSS USER DATA:\                4038730332  1236935213     VSS USER DATA:\                4038730332  1236935214     VSS OTHER:\                    4005175900  1236935224     VSS OTHER:\                    4005175900  1236935225     VSS ASR DISK:\                 3787072135  1236935293     VSS ASR DISK:\                 3787072135  1236935294

Ok now lets add a -xc, to the command and we should see every is seperated by a comma. You could how ever replace the comma with any symbol or character.

C:\>mminfo -avot -xc, -r “client,name,ssid,cloneid” -q -t “1 day ago” >report.txt

client,name,ssid,clone-id,C:\,2931347332,1236865688,VSS SYSTEM BOOT:\,2964901764,1236865688,D:\,2948124548,1236865688,VSS SYSTEM SERVICES:\,2897792906,1236865688,VSS USER DATA:\,2830684090,1236865688,VSS OTHER:\,2595803177,1236865688,VSS SYSTEM FILESET:\,2428031191,1236865688,VSS ASR DISK:\,2444808367,1236865688

Now lets rename the report.txt file to report.csv and open it in Excel.



 There we go, each field in its own colum. 


There have been a number of improvements for Networker VCB integration in the new 7.4.x and 7.5 builds.

For anyone who decides to integrate VCB into Networker I highly recommend you start with the basics. Install the VCB framework on the proxy server and start with some vcbmounter commands to make sure you can successfully perform a full image export and a file system mount. Once you get this working you can then start to integrate VCB into Networker.

If you’ve already jumped ahead and configured some VCB clients in Networker but your finding the backup fails, here’s a tip which might help identify the problem.

From the Networker backup server, browse to the Networker install path and in nsr\tmp\sg\<groupname> youll find a small file called sso.000002 which contains all the output from the vcbmounter commands running in the background. This is why I recommend getting things working from the command line first as the output you would see and the failures are identical with the information you’ll find in this log.

Here’s an example of an error I received while doing some testing.

Converting “E:\vcbmnt\\scsi0-0-0-ubuntu704.vmdk” (compact file):
        ********************[2008-11-06 14:23:38.578 ‘blocklist’ 1224 error] Could not renew disk lease. Aborting operation.
[2008-11-06 14:23:38.640 ‘BlockList’ 2824 error]

Now the handy thing about having access to this information is it helps you determine if the problem lies with VMware or Networker. I think from memory this particular error did show up in the savegroup details but there have been many times I had to look here to find out what was really going on.

Disk Alignment

Posted: March 11, 2009 in Celerra, Tips and Tricks, VMware

I wanted to post about something  I recently found made a significant difference in performance testing done on a Celerra NS20.

I found a  this post at Clariion blog central which has an excellent explanation about how and why disk misalignment occurs, defiantly worth a read.

The NS20 I refer to above is our production Celerra we use for dedicated iSCSI storage with our 2 node HA/DRS cluster. So something I had to take into account was disk alignment at ESX/Celerra level as well as Virtual Machine disk alignment.

Looking through VMware documentation I was pleased to read, storage added through Virtual Center is automatically aligned so now I only had to worry about disk alignment for the virtual machines. Below is a step by step walk through of how to add a disk to a virtual machine and align at the 64k boundary.

Step1. Right Click on the Virtual Machine, Edit settings and highlight Hard Disk and select Next.

Step2. Select create a new virtual disk and select next.

Step3. Select a disk size and select to store the new disk with the virtual machine or use the specify a datastore option to place the new disk on a different datastore, select Next.

Step4. Leave the default options under advanced options and select next, and Finish to complete the hardware wizard.

Step5. Navigate to the console tab for the virtual machine and from the desktop right click on My Computer and select Manage, go to Disk Management.

Step6. Right click on Disk Management and select Rescan Disks.\

Step7. Now open a command prompt window and type the following commands.

Invoke Diskpart


Microsoft DiskPart version 5.2.3790.3959
Copyright (C) 1999-2001 Microsoft Corporation.
On computer: SQLDR

type the follow command to show disks

DISKPART> list disk

  Disk ###  Status      Size     Free     Dyn  Gpt
  ——–  ———-  ——-  ——-  —  —
  Disk 0    Online      8189 MB  8033 KB
  Disk 1    Online      8189 MB  8189 MB

You can see Disk 1 is the new disk we just added. Now select Disk 1.

DISKPART> select disk 1

Disk 1 is now the selected disk.

Now we create the partition and set it to the recommended alignment offset.

DISKPART> create partition primary align=64

DiskPart succeeded in creating the specified partition.

Now you can right click on the new volume and format with desired file system type.

Now if your using 2003 enterprise or datacener edition ive noticed a bug where you receive an error when trying to format, the only work around I could find was to assign a drive letter in disk part and then format. <strange>

DISKPART> assign letter=X

DiskPart successfully assigned the drive letter or mount point

 That’s it…. easy aye ! 



Something I see on a regular basis is customers having  issues with a particular client backing up and to stop the group from failing  they tend to remove the client from the group. Often the failure can be something as simple as the machine was rebuilt and the Networker client hasnt been instaled yet.

Now if you remove the client from the group you have to remember to put it back in, and trust me I know all about this because its something Ive done before and its something you only tend to do once.

I’m a firm believer of leaving clients in groups….” LET IT FAIL” when the big boss man comes knocking on your door asking why the backup groups have been failing for the last week you explain its because the administrator of the system hasnt resolved the issue yet.

Also something I see often is people moving clients into “Test” groups because of failures, this is also prone to forgetting to put clients back in groups.

So what options do you have ? (Here are a couple of practical ideas)

1.Use the Networker NMC “copy” option and place the new instance into a test group, change the comments field to show this is a test instance. Once you’ve finished testing you can simply delete it.

2. A single client instance can be placed in multiple groups, rather than removing it from the original group just leave it there and add it to the test group. Once your done just remove it from the test group, if you forget to do that then its probably a good thing you didn’t remove it from the original group because its likely you would have forgotten to put it back.

3. Quite often customers move clients into a different group as they don’t want to run an entire group again just to capture a single client. The best way to achieve this without having to go and move or create clients is to use the command line. The command below runs the client in the specified group only, this will adhere to the schedule assigned to the group so if you have a skip set then just add -l FULL before the -c <client> to run a level full backup.

savegrp -v -c <client> -G <group>

For those of you who are not familiar with the command line side of networker, I highly recommend you have a play.. this is where the real power lies.