Colocation is a type of hosting where customers pay to have their own servers hosted within a data centre; colocation services are available via some web hosts through the use of their private cages, or you can pay a data centre directly to have your server located in a cage which is run by the datacentre itself. Other equipment such as networking equipment and storage equipment can also be colocated within server racks. Colocation is seen as a more enterprise level alternative to dedicated server hosting since the equipment hosted is owned by the person or business that is paying the lease. More and more businesses are starting to realize the benefits of colocation hosting; the main reason for this being that with regular dedicated server hosting you don’t receive the level of support or technical know how that you do with colocation hosting. Colocation hosting has many advantages above ordinary dedicated server hosting, such as the fact that most data centres provide a ‘remote hands’ service if your server needs a task doing such as an OS reinstall which can only be done from the console; also, the connectivity that your equipment receives will probably be better than that you would receive on a dedicated server, the reason being that your equipment is surrounded by your own equipment meaning that you don’t have any other customers around you that could be a burden on your connection speed. Colocation is often abbreviated as ‘colo’; the colocation centres themselves are often referred to as ‘carrier hotels’ because of the number of internet carries that the host, as well as the number of businesses which have their servers located within a colocation centre.

Standard Features of a Colocation Centre

Colocation centres are often built in a certain way for the maximum benefit of the servers and equipment which they host. Colocation centres are more or less ordinary data centres, and so always have some sort of fire surpression system that makes use of pipes to ensure that the fluid used to put fires out can spread across the building rapidly to ensure that the fire doesn’t damage too much of the building - that is if the fire can be controlled. Regular internet equipment is around 19 inches wide - the same can be said for an ordinary server, this means that both colocation and data centres have a large collection of 19 inch wide data racks to cope with customer demand; most also carry a collection of 23 inch wide data cabinets which is used to hold any telecoms equipment that customers may wish to have hosted at the location concerned. Most 19 and 23 inch wide data cabinets are lockable to ensure the safety and security of the equipment that is contained within; this is perfect for large corporations or business’s whose servers contain highly sensitive information since they are able to leave their equipment there with peace of mind. Because of the nature of both colocation and data centres, they both contain alot of cables (both data and power) to ensure that the equipment that they host is able to be powered and connected to the internet; it is because of this that you will always find overhead cable racks in both to ensure that more cables for more equipment can be easily set out. One thing that you might also come across is the fact that the power for the equipment is contained in another rack - sometimes referred to as the power distribution rack; this is done to ensure that the most can be made from the space contained within both data and telecoms racks. One main feature of either a colocation or a data centre is air conditioning to ensure that the hosted equipment is kept at a cool temperature; the air conditioning within both types of centre is normally pushed through the raised floors that both contain and then released from under the cabinets that hold the equipment to ensure that every piece of equipment receives cool air to keep its temperature down. The amount of cooling that is available in both a colocation and data centre and restrict the amount of servers or how much equipment can be hosted within the facility; this is because if there isn’t enough cooling available then a fire hazard can be created - some think that it is the floor space (sometimes referred to as square footage since it is measure in square foot) that determines how much equipment can be hosted within a facility. Another idea used to ensure that the temperature of equipment is kept to a minimum is to ensure that the data floors used contain little or no windows; this helps since sunlight cannot reach the equipment or floor itself which can help keep the temperature down, especially on hot summer days.

Security

Because of the nature of the equipment that is hosted within both a colocation and data centre, both require that a member of staff escort a customer across the data floor to the cage in which their equipment is hosted; in this case the member of staff is also required to stay with the customer whilst they do what they need to do to their equipment. This is done because in most cases the equipment of other customers is normally contained within the same rack; if a customer has their own rack or cabinet within the colocation or data centre then they are normally provided with some type of access card which they can use to gain access to the data floor, and then use their own set of keys to unlock the cabinet in which their equipment located. Some colocation and data centres used the biometrics of clients as a pass key, as that ensures that person entering the data floor is the customer that they have on file. Both types of centre also contain many CCTV cameras and staff to monitor them to ensure that no unauthorised personnel manage to gain access to the data floor. In terms of technical security, the use of equipment such as firewalls is down to your own specification, since you are more or less hosting your own equipment; in some colocation and data centres where you are sharing a rack with other customers, the host company will in most cases provide the rack with some sort of security device or firewall to ensure the safety of all hosted equipment. However, if you do have your own rack then you will most likely have to provide the firewall or a device of similar standards yourself.

Power

Power sources within a modern day colocation or data centre are normally redundant, meaning that if some power supply gives way then there is another source available to keep the hosted equipment running. The method used by most facilities is to deploy diesel generators to supply the power to the servers and other equipment that may be hosted in the event of a mains power black out; some facilities employ UPS (uninteruptable power supply) between the event of a blackout and the diesel generators - a UPS in a simple context is a backup battery which stores power which can be used in the event of a blackout, which can then be recharged when the power is restored. In some cases, large corporations with their own equipment racks or cages deploy their own UPS amongst their equipment to compliment those power services of the hosting facility as well as to ensure the best redundancy for their equipment can be achieved in the event of a blackout. For additional redundancy, most colocation and data centres have multiple connections to different locations within the local power grid to ensure that if one feed does go down, there is at least one other available to power the equipment; this method is good since it means that the diesel generators or UPS supply banks don’t have to be relied on.

Connections

Most colocation and data centres have multiple feeds to different bandwidth carriers to ensure the best redundancy; this means that if one connection was to fail then there would be at least one other connection to the internet or service provider, to ensure that the hosted equipment was still connected and online. Some facilities contain rooms known as ‘meet me rooms’; this is because most peer points are contained with in data centres and colocation facilities to ensure that corporate customers are able to achieve the best possible connections. The idea of a ‘meet me room’ is to enable all carriers at the data or colocation centre to efficiently transfer data. In some cases, there may be a large internet exchange hosted within the facility where customers are able to link up for peering.

PHP and ASP.NET are the two main scripting languages used by web developers. PHP is a multi-platform open-source language whilst ASP.NET is a single platform language built and owned by Microsoft. Both of these languages are packed full of features and add-ons for both are available so that you are able to extend their framework to build feature rich applications. Although PHP is open source, it has a large active community which is there to support users of the language; ASP.NET is a product of Microsoft meaning that there is a large amount of online documentation and other community forums where you can get support.

PHP

PHP is a language which has been around since 1984 and powers many of the world’s websites. It is free and is known as ‘open source’ software meaning that its source code is available for download, at no cost. It widely used in conjunction with the MySQL database engine, but can be used with others such as Oracle and SQLite. PHP is generally easier to code in than ASP.NET, since it has more functions built in than its rival; however, ASP.NET has built-in AJAX functions which is something that PHP lacks. Since PHP is open source, there is a number of applications that are available for free based on it, most of these applications are commonly used and are very powerful; also, there is a greater number of community forums and blogs where you are able to find information regarding PHP since it is open source.

PHP is a dynamic language since there are several different types of database that you are able to connect to, including: MySQL, Oracle, PostgreSQL and in some cases, Microsoft SQL Server. You are also able to edit the source of the language if you wish to speed it up or install custom modifications so that it is able to run as per your requirements; again, this is because it is open source - although ASP.NET is a free language, it isn’t open source meaning that you are unable to edit its source.

ASP.NET 

ASP.NET is a language based on classic ASP - both languages are owned and were created by Microsoft. Although it is a free language, it isn’t open source meaning that you are unable to view or edit the source code of it. Also, it is very limited in what database types you can use with it - Microsoft only invented it to work with Microsoft Access and Microsoft SQL Server databases, although developers can use it with MySQL databases with the help of third party components. Classic ASP, ASP.NET’s forefather, was created back in the mid-1990s and is the base which ASP.NET has been built on. One thing which annoys many ASP.NET developers is the fact that Microsoft is forever releasing updates to the system, and in its course, changing things along the way meaning that in some cases, developers have to keep on learning the changes as well as the new components and functions which are being added.

ActiveX

This is a control commonly used within Microsoft’s Internet Explorer web browser. They are very similar to applets in the way they behave as well as their functionalities; however, they give developers greater access to a client’s system - applets only provide developers with very limited access.

ADSL

ADSL stands for Asymmetric Digital Subscribers Line. It is similar to a DSL line, the only difference being that a number if dialled on connection with an ADSL connection.

Applets

These are very similar to ActiveX controls. Unline ActiveX controls they can be run in most browsers but provide developers with limited access to a client’s system as they are executed within a sandbox. They are normally written in the Java programming language.

ARPANET

ARPANET is a project which began in 1969 and was started by the Advanced Research Projects Agency with the aim of linking up 4 universities. The project became well known for being the start of the internet.

ASCII

ASCII stands for American Standard Code for Information Interchange. It is a data representation code used mainly by microcomputers and minicomputers.

ATM

ATM stands for Asynchronous Transfer Mode. It is a network configuration utilizing a single standard for data, video and audio that transfers information in short, fixed-length packets.

Access

Access is Microsoft’s advanced database creation and management software. It is popular among developers who maintain online databases with Windows NT.

Aliased Nameservers

Aliased nameservers are nameservers which are in the name of the client and not that of the hosting provider.

Apache

Apache is a free open source webserver widely used on the internet.

Application Hosting Service Provider

This is a third party provider which manages software applications.

One server responding to all the incoming HTTP requests for a website might not able to handle all the incoming traffic of a website. This would become more difficult if the website becomes more popular. As a result pages will load slowly and users will have to wait for a long time to view web pages. Due to increase in traffic and connections to website there will be need to upgrade the server and it would be no more cost effective.

To improve the server scalability, more servers are needed to add more scalability and distribute load among the servers which is also called as clustering and load distribution among these servers is called as Load balancing. Load balancing applies to all types of servers including application and database servers.

Load Balancing Mechanism

The load balancing mechanism used for spreading HTTP requests is known as IP Spraying. HTTP traffic needs to be evenly distributed when there are multiple servers in the group. It acts as one server to clients for example an internet browser. “Load dispatcher” and “network dispatcher” are the equipments used for IP spraying. You can simply call them as “Load Balancer”.IP Sprayer redirects HTTP requests to a server in a server cluster. It all depend on the type of IP sprayer involved so that the architecture can provide more scalability, load balancing and overcome all the fail-over requirements.

You can have a look at Load Balanced Servers offered by eUKhost.com for a complete load balancing solution.

Sometimes actions performed by the attacker on the server may affect its functionality. So its always advisable to check server’s security to avoid attack on the server. Always check the resources of server which might be affected.

You can check the CPU usage by firing top command and look for the application or scripts that consume your CPU

For strange processes you can check with ps -awux command.

Check /tmp directory and /var/tmp directory for scripts and binaries copied there.

The attacker might use the server to host IRC bot like psybnc or eggdrop which connects to port 6667 when a server is compromised . You can if any of your applications connect to that port with sockstat:

#sockstat | grep 6667

If there’s not much traffic on your server you could use netstat command to see if suspect connections are made.

#netstat -a

Install and run regularly an rootkit finder application (for e.g /usr/ports/security/rkhunter).

Look for the other open ports that you run other than the ones you use for your running services.

Rsync is a powerful command that is used to synchronize and transfer files and directories. Rsync is widely used for mirroring Web sites. and transfers only the changes in files, using a clever rsync algorithm. This is a very useful method for backups as it is uses minimal bandwidth and it is also commonly used to keep clustered servers working together. The ending slashes are very important while using rsync. If you are not familiar with how it works simply make sure to have an ending slash on both the source and the destination and it should be fine.

The general form is:

rsync source destination

A few examples are:
rsync -vrplogDtH /old/usr/local/apache/conf /usr/local/apache

That will sync the /old/usr/local/apache/conf/ directory to the /usr/local/apache/conf/ directory on the same server. If you want to use rsync to go between servers then use the following:

rsync -ave ssh root@192.168.0.1:/backup/ /backup/

That will take the backup directory on 192.168.0.1 and copy it to the server the command is run from. The command will also accept a remote destination if you adjust the command line accordingly.