What could happen if I install all server roles on Windows Azure Pack: Web Sites on one machine? - server

At work, I'm in the process of installing Windows Azure Pack: Web Sites in a VMWare ESXi lab environment. I have little available RAM and hard drive space on the ESXi.
I originally thought I would be able to do this without spending too much resources. The Azure Pack Express variant is advertised as if it only requires one machine with 8 GBs of RAM. However, after completing the first installation, I discovered that the Azure Pack: Web Sites extension requires no less than 7 different server roles installed on 7 different machines, each with Windows Server 2012 R2. I need a separate Cloud Controller, Cloud Management Server, Cloud Front End, Cloud Publisher, Cloud Shared Web Worker, Cloud Reserved Web Worker and Cloud File Server.
I have no way of freeing up that much resources. In the installation guide for Windows Azure Pack, they "advice" me to use separate VMs for each role, but they don't say explicitly that it won't work. Is it because multiple server roles on one machine will strain resources, or is it because the roles are incompatible and will make the system malfunction? In my case, the Azure Pack will only be used for penetration testing by a single user, so I imagine resources should not be a problem.
I'm not a web administrator, and I'm in over my head on this task. If anyone could give me some advice before I proceed on this, that would be much appreciated.
TLDR: Will there be a critical conflict if I install seven server roles on one machine, or will there just be a strain of resources?

Related

Should XAMPP be installed on an actual physical Server?

Is XAMPP just meant for testing and setting up virtual servers ?(cause that's what wiki say)
Can it be installed on an actual physical server? Do developers actually do that?
I'm a little confused cause if it were true, why would anyone install a virtual server on a physical server? It's like trying to run Excel on VirtualBox.
XAMPP simulates a typical stack used for web development on a local machine. If you have access to an actual physical server, you would typically install things like the web server (such as Apache) and MySQL on the server itself. The developers of XAMPP consider it more of a development tool due to certain features being disabled to make dev easier.
Virtualisation in servers is used because the actual physical machines are very powerful and so are idling a large amount of time. Putting those resources to use by creating two virtual servers on top of the host reduces cost and increases operational throughput.
Virtual server and Docker can be used to test with different environments at the same time, or test beta software for future releases. On Maschines that have 6 or 8 cores and running 3.6 Millions instructions per second, there are plenty of resources to have more than 1 maschine virtual or as a docker file, so that you can uses for example different databases, with out them interfering.
Besides phiscal Hard cost mony to buy and to maintain.
Last virtualisation and docker are only files, that you can simply copy to have a backup. A real maschine is a little more work, to make a backup.
But don't use XAMPP as real maschien that is exposed to the world. There much to many security risks ind teh standard configuration.

OS Supporting Bluemix

If I were to setup IBM Bluemix local, what are the operating systems that can support the Bluemix software?
Been trying to look for answers in the internet but always end up frustrated because I can't find any answer. Even Bluemix site did not mention what is the OS requirement in setting up Bluemix.
Bluemix Local is not a software product that you can install in your own operating system. It runs in your environment but it is still a subscription.
The Bluemix Local install is only performed by IBM and it is an automated deployment of many virtual servers. It is also a managed offering, meaning that you don't have access to the operating system on those virtual machines. All the maintenance tasks are executed by IBM through a secure connection and a mechanism called Relay.
I guess you probably saw this link before, but the infrastructure requirements are described here: https://console.ng.bluemix.net/docs/local/index.html#localinfra
I think you may need Bluemix Support team to help you build the Bluemix local environment, Bluemix local is a cloud platform as a service that need to build on your infrastructure as a service.

Best practice deploying windows service

I'm looking for best practice in continuous delivery of windows services.
Currently we hava a set of powershell scripts that unintall, reboot, install updates but error handling is tricky. We are reviewing System center but are there any other options available for deploying a windows service?
We've been using Presto since Dec 2011, and have done over 1,000 deployments. Most of what we deploy are Windows services.
What's nice is that we set up our apps and servers in Presto, then we can repeatedly deploy, to any server (or multiple servers at once), by just hitting a button. Presto will copy our official release binaries, update all of the items in our app config files, create and start the service, etc...
So, if you have an application that has 30 manual steps to deploying it, you can enter these steps in Presto, then it's done automatically for you after that.
It's worth a look: http://presto.codeplex.com/
Your most basic and generally accepted best option comes from this thread, which basically links to a Microsoft support article on creating an installer for the windows service.

Considerations for developing for a VM deployment

I'm setting up a system that uses SQL Server 2005, several custom Windows Services, Web Services and a few IIS .NET applications. Getting the whole system setup is a somewhat tedious process.
I wondered whether it would be a good idea to settup the whole system in a VM. Could I then just drop the VM onto a new server and get a huge headstart on configuration?
What things should I be aware of if I pursue this approach? Is it a viable option? Is a VM a decent unit of deployment?
If the concept is feasible, I'd certainly appreciate specific suggestions about the VM setup.
I frequently use this approach. I'll set up a VM in VMware Workstation, configure it to my liking, and then use VMware Importer to import my virtual machine into an ESX environment. From there, I can turn the virtual machine into a template that I can use over and over again for deploying clones of my server or just as a starting point when creating new servers.
· Large quantity of virtual machines (one for each customer)
· Less quantity of physical machines
· VM's we are working on has to be up while the others can be down
· Easy backups so, in case of issues we can start working on the same moment we shut down the vm, etc...
· Physical machines has to configured with the last hardware or almost.
· Depending on your develops, a physical machine can keep between 4 and 8 VM.

What kind of servers did you virtualize lately?

I wonder what type of servers for internal usage you virtualize in the last -say- 6 months. Here's what we got virtual so far:
mediawiki
bugtracker (mantis)
subversion
We didn't virtualize spezialized desktop PCs which are running a certain software product, that is only used once in a while. Do you plan to get rid of those old machines any time soon?
And which server products do you use? Vmware ESX, Vmware Server, Xen installations...?
My standard answer to questions like this is, "virtualization is great; be aware of its limitations".
I would never rely on a purely-virtual implementation of anything that's an infrastructure-level service (eg the authoritative DNS server for your site; management and monitoring tools).
I work for a company that provides server and network management tools. We are constantly trying to overcome the marketing chutzpah of virtualization vendors in that infrastructure tools shouldn't live in infrastructure tools.
Virtualization wants to control all of your services. However, there are some things that should always exist on physical hardware.
When something goes wrong with your virtual setup, troubleshooting and recovery can take a long time. If you're still running some of those services you require for your company on physical hardware, you're not dead-in-the-water.
Virtualization also introduces clock lag, disk and network IO lag, and other issues you wouldn't see on physical hardware.
Lastly, the virtualization tool you pick then becomes in charge of all of the resources under its command for its hosted VMs. That translates to the hypervisor - not you - deciding what VM should have priority at any given moment. If you're concerned about any tool, service, or function being guaranteed to have certain resources, it will need to be on physical hardware.
For anything that "doesn't matter", like web, mail, dhcp, ldap, etc - virtualization is great.
Our build machine running FinalBuilder runs on a Windows XP Virtual Machine running in VMWare Server on Linux.
It is very practical to move it and also to backup, we just stop the Virtual Machine and copy the disk image.
Some days ago we needed to change the host pc, it took less than 2 hours to have our builder up and running on another pc.
We migrate to a new SBS 2005 Domain last month. We take the opotunity to create virtual machines for the following servers
Buid Machine
Svn Repository Machine
Bug Traking Machine (FogBugz)
Testing Databases
I recently had to build an internal network for our training division, enabling the classrooms to be networked and have access to various technologies. Because of the lack of hardware and equipment and running in an exclusive cash only environment I decided to go with a virtual solution on the server.
The server itself is running CentOS 5.1 with VMWare 1.0.6 loaded as the virtualisation provider. On top of this we have 4 Windows Server 2003 machines running, making up the Active Directory, Exchange, ISA, Database and Windows/AV updates component. File sharing and internet routing through the corporate network and ADSL is handled via the CentOS platform.
The setup allows us to expand to physical machines at a later stage quickly, and allows the main server to replaced with minimum downtime on the network, as it only requires the moving of the Virtual Machines and starting them up on the new box.
Project Management (dotProject)
Generic Testing Servers (IIS, PHP, etc)
Do you plan to get rid of those old machines any time soon? No
And which server products do you use? MS Virtual Server
We use ESX in our labs and lately we've virtualized our document sharing service (KnowledgeTree), the lab management tools and almost all of our department's internal web servers.
We also virtualized almost all of our QA department's test machines, with the exception of the performance and stability testing hardware.
We aren't going to get rid of the hardware any time soon, it will be used to decrease the budget needs and increase the number of projects that can be handled by one lab.
We use VMware ESX 3.5.x exclusively.
We virtualise a copy of a test client and server, so we can deploy to them before sending the files to the customer. They also gets used to test bug reports.
We find this is the biggest benefit to virtualisation as we can keep lots of per-customer versions around.
We also VM our web server, and corporate division has virtualised everything.