EC2: Image Manifest differs from s3

I tried to start one of my Amazon EC2 AMI’s a few day’s ago, and received the error:

Insufficient capacity

Which I found was very weird. I think that’s just we’re cloud computing stands out from traditional computing, the possibility to add capacity on demand. I found out that I had to AMI’s with the same source (the manifest).

I de-registered one of them, but still received the same error.
Today I tried again (yes, I am stubborn) by starting the one AMI that was left. This resulted in a completely different error:

registered machine image manifest for ami-******* and manifest in S3 differ. Has the image been rebundled but not re-registered.

Start Instance Wizard

I solved it by firing up the EC2 API Command Line Tools, and executing the command:

ec2-register <bucket>/<name.manifest.xml> -n <Name>

For the parameter Name I opened up CloudBerry Explorer for Amazon S3 and opened the manifest.xml file that was causing the problem. Inside this file there is a xml tag named <name>, Use this one to register your AMI.

Cloudshare Pro

Yesterday I came by a blog post which introduced Cloudshare to me ( a service that allows companies to demo software in the cloud).
Cloudshare was introducing Cloudshare Pro, CloudShare Pro is intended for individuals and small teams who want to run complete, networked systems in the cloud and share copies with other colleagues for collaboration.

And the best thing is: While in public Beta, CloudShare Pro is FREE.

After some test with it, I must say “I am impressed”. It is really easy to build your environment. Within a few minutes I created an environment which consisted of a sql server, AD Server, 2008 server and a client machine.

 

image

If you are interested in this, I would suggest “go to their site and sign up”

Note: 2 bad that the Operating Systems used are 32 bits… there goes my Sharepoint 2010 Farm :-(

Update March, 1st:

The CloudShare team had added a windows 2008 X64 server. There is only on problem with it: the server is a bit below it’s normal specs: 1gb RAM and 20GB Disk Space.

BPOS: Announcing Updates to Hosted Blackberry Service

From the Microsoft Online Services Team Blog.

Microsoft Online Services is now a BlackBerry® certified partner and BlackBerry Alliance Elite Member. For the Business Productivity Online Standard Suite, licenses and support are now a part of the Microsoft Online Services Hosted BlackBerry service, which customers buy directly from Microsoft. Previously, customers had to purchase and provide their own licenses and support agreement, purchased through a third party, in order to use the Hosted BlackBerry service.

The Hosted BlackBerry service from Microsoft Online Services offers the most commonly requested BlackBerry capabilities, including wireless access to e-mail, calendar, tasks and contacts with global address list (GAL) integration, and device management such as device wipe and password reset. Details about the Hosted Blackberry Service can be found in the Microsoft Online Services Mobility Solutions Description, which can be downloaded from the Microsoft Download Center. The document includes a full solution description and service pricing.

To learn more or to request a Hosted BlackBerry service agreement, contact your Microsoft representative or a Microsoft Online Services partner. If you do not have a Microsoft partner or representative, you may request the Hosted BlackBerry service agreement through Microsoft Online Services Customer Service.

For additional details please about the Microsoft Online Services mobility offerings, visit www.microsoft.com/online/mobility.

Azure going to be a LaaS platform

According to Prashant Ketkar (marketing director of Azure) the Azure Service will be adding Remote Desktop capabilities a.s.a.p., as well as the ability to load and run Virtual Machine images directly on the platform.

This would be a smart move and something I was hoping for: Azure as a direct competitor of Amazon’s EC2 platform.

The original article where I found this “rumor”can be found here and another one here.

Your cloud needs a Sys Admin

A really nice article written by George Reese (twitter profile here).

As I am a Systems Administrator (which also administrates some cloud instances) and am working with a software development company (amongst others), this article has some interesting quotes like:

  • The cloud, however, has a nasty habit of deluding programmers into thinking they no longer need sys admins.
  • A programmer and a credit card can launch a server in the cloud.
  • The programmer, however, lacks a detailed understanding of ongoing infrastructure management.
  • The reverse is true of the sys admins who fancy themselves programmers. They can craft Perl programs to do just about any task. Those programs, however, ultimately lack the solid architecture that programming skills provide.
  • Programmers Build Bad Machine Images
  • Sys admins live and breathe hardware, the OS, and the network.

You can read the full article here

Disabling End-User Notifications in the Sign-In Application

You can configure the End-User Notification whenever a new version of the Microsoft Online Sign-In application is released. This is useful if you want to control whether you want your end-users to receive a notification or not.

For example: a End-User who has no rights to install anything on his computer doesn’t want to be bothered by any new version release. You can disable this notifications by use of Group Policy’s (preferred by myself) or by editing the registry (in theory the GPO does the same off course :-) ).

Here is a description how to do this: Microsoft Online Services Team Blog – Disabling End-User Notifications in the Sign-In Application

kudos to the Microsoft Online Services Team Blog

Increasing your AWS EC2 root partition (Windows)

As we now have instances which can be booted from EBS (hurray for that), you’re root is also increased to 30gb. But sometimes this still isn’t enough. Here is how you can increase your root partition drive using the AWS Console.

  1. login to  your AWS Console
  2. Go to [Instances], right click your instance and select [create image (EBS AMI)]
    This will make a backup (just in case)
  3. Your instance will reboot, and come up again when the image is created. After that stop your instance (do not terminate it :-) )
  4. Next is to go to [Volumes], you will see a 30Gib volume somewhere which is attached as /dev/sda1 to your instance

    Volume 

    Right click it and select [Create Snapshot from this volume].

  5. Next, go to [Snapshots]. Right click the snapshot that you created in the previous step and select [Create Volume from Snapshot], you can then specify how big this volume has to become (for example 50 Gib).

    Create Volume from Snapshot

  6. After the volume is created go back to volumes again. You will see a volume which is as large as you specified in the previous step. The status of this volume is “available”.
    Now select the volume which is currently attached to your instance. It is the one you located in step 4. Right click it and and select [Detach Volume], it will then become available.
  7. Now select your newly created volume, right click it and select [Attach Volume]. Select your instance (which will have a status of “stopped”). In the textbox “Device” type: /dev/sda1. Now click [Attach]

    Attach Volume

  8. Fire up your instance and log in.
  9. Open up computer management and select [Disk Management] (underneath the option [storage]). You will see that your root partition has some more space behind it.
  10. Right click your root partition and select [Extend Volume], walk through the wizard. Eventually your root partition will be extended.

New Powerful Amazon EC2 Boot Features

Yesterday Amazon announced some cool new feature on their EC2 platform: Boot your instance  from EBS (Amazon Block Store).

Your root partition of your image will be EBS, so this will give you the ability to:

  • fine-grain control of software and data configuration
  • increase the size of the local disk up to 1TB in size
  • boot faster
  • stop an instance and restart it at a later time with the disk configuration intact
  • restart an instance and choose to use a different instance type (e.g., with more memory or CPU), a different operating system (e.g., with new security patches installed), or add new user data

read more @ source