Question

1. A company wants to create highly available datacenters. Describe two full implementations that will allow the company to continue to maintain an Internet presence at all sites in the event that a WAN circuit at one site goes down. Describe also two known implementations that will not work and show why.

2. Network segmentation provides high availability through redundancy. Using appropriate illustrations discuss three full implementations (integrated) describing how this can be achieved

3. A network technician is diligent about maintaining all system servers at the most current service pack level available. After performing upgrades, users experience issues with server-based applications. Discuss four server applications that could be affected and why, showing how they are affected. Finally Describe four standard implementations that should be adopted to prevent such issues in the future

Answer 1

a)

two full implementations that will allow the company to continue to maintain an Internet presence at all sites in the event that a WAN circuit at one site goes down.

A collection of networks that fall within the same administrative domain is called an autonomous system (AS). In this question, each datacenter will be an autonomous system.
The routers within an AS use an interior gateway protocol, such as the Routing Information Protocol (RIP) or the Open Shortest Path First (OSPF) protocol, to exchange routing information among themselves. At the edges of an AS are routers that communicate with the other AS’s on the Internet, using an exterior gateway protocol such as the Border Gateway Protocol (BGP).
If a WAN link goes down, BGP will route data through another WAN link if redundant WAN links are available.

Routing Information Protocol (RIP) is a protocol that routers can use to exchange network topology information. It is characterized as an interior gateway protocol, and is typically used in small to medium-sized networks. A router running RIP sends the contents of its routing table to each of its adjacent routers every 30 seconds. When a route is removed from the routing table, it is flagged as unusable by the receiving routers after 180 seconds, and removed from their tables after an additional 120 seconds.

There are two versions of RIP (the managed switch supports both):

• RIPv1 defined in RFC 1058.
• • Routes are specified by IP destination network and hop count.
  • The routing table is broadcast to all stations on the attached network.
• RIPv2 defined in RFC 1723.
• • Route specification also includes subnet mask and gateway.
  • The routing table is sent to a multicast address, reducing network traffic.
  • Authentication is used for security.

You can configure a given port to do the following:

• Receive packets in either or both formats.
• Send packets formatted for RIPv1 or RIPv2, or send RIPv2 packets to the RIPv1 broadcast address.
• Prevent any RIP packets from being received.
• Prevent any RIP packets from being sent.

Open Shortest Path First (OSPF) is a link-state routing protocol that is used to find the best path between the source and the destination router using its own Shortest Path First). OSPF is developed by Internet Engineering Task Force (IETF) as one of the Interior Gateway Protocol (IGP), i.e, the protocol which aims at moving the packet within a large autonomous system or routing domain. It is a network layer protocol which works on the protocol number 89 and uses AD value 110. OSPF uses multicast address 224.0.0.5 for normal communication and 224.0.0.6 for update to designated router(DR)/Backup Designated Router (BDR)

b)

The Fragmentation-Redundancy-Scattering (FRS) is a technique that can be used to achieve security and fault-tolerance [1]. It consists in fragmenting the condential data and scattering the resulting fragments across several archive sites of the distributed system. Fragmentation is performed so that any isolated fragment contains no signicant information. Scattering is performed in such a way that each archive contains just a subset of (unrelated) fragments and that each fragment is stored in more than one site. The technique provides intrusion-tolerance: if a node is compromised, the intruder has no access to relevant information (and if compromised fragments are deleted or altered, they can be recovered from other nodes).

Redundancy:

The underlying concept of redundant networks is simple. Without any backup systems in place, all it takes is one point of failure in a network to disrupt or bring down an entire system. Network redundancy is the process of adding additional instances of network devices and lines of communication to help ensure network availability and decrease the risk of failure along the critical data path.

Generally speaking, there are two forms of redundancy that data centers use to ensure systems will stay up and running:

• Fault Tolerance: A fault-tolerant redundant system provides full hardware redundancy, mirroring applications across two or more identical systems that run in tandem. Should anything go wrong with the primary system, the mirrored backup system will take over with no loss of service. Ideal for any operations in which any amount of downtime is unacceptable (such as industrial or healthcare applications), fault-tolerance redundant systems are complex and often expensive to implement.
• High Availability: A software-based redundant system, high-availability uses clusters of servers that monitor one another and have failover protocols in place. If something goes wrong with one server, the backup servers take over and restart applications that were running on the failed server. This approach to network redundancy is less infrastructure intensive, but it does tolerate a certain amount of downtime in that there is a brief loss of service while the backup servers boot up applications.

In the event of a failure, redundancy allows your network to remain in service by providing alternative data paths or backup equipment. Network redundancy is introduced to improve reliability and ensure availability. The basic concept that if one device fails, another can automatically take over.

patching:

A typical corporate network will include servers and user endpoints running a large number of different applications on top of different operating systems. These applications probably include software developed in-house, proprietary commercial software packages, and open-source applications. Most software will have been approved by the IT department, but most corporate networks will also include an element of unauthorized consumer-oriented software installed by end users as well.

The goal of patch management is to ensure that all applications running on the network are patched to ensure that they are secure and stable, and achieving this goal involves:

• software inventory management
• monitoring the availability of security and other patches for all operating systems and applications in the inventory
• detecting any software which is has not been appropriately patched
• deciding which patches need to be applied to which systems
• testing the patches to ensure that compatibility with other software and systems is maintained
• installing the patches in the correct order to ensure that one patch does not undo another
• ensuring that patches are installed in a timely fashion, or within a target timeframe that is commensurate with the organization’s overall risk management approach
• final testing after installation to ensure that the installation is successful and that the software continues to operate as expected

c)

configure the test lab for updates:

To prevent the service pack issues make sure, before going ahead and applying a new Service Pack in your production environment, to validate them in a test/lab environment first.

server applications that could be affected are:

1. DNS issues and network connectivity
2. Slow servers and loading time
3. Poorly written code
4. Lack of load balancing
5. Traffic spikes
6. Specific HTML title tags
7. Failing to optimize bandwidth Usage

1) DNS issues and network connectivity

An essential element of successful web traffic management is DNS queries, which is why an issue with these systems can result in a plethora of issues. Without the proper protection, faulty DNS queries can prevent visitors from reaching your website, while also causing errors, 404s, and incorrect pathways. In a similar vein, network connectivity and an efficient firewall are key parts of your site’s access and productivity.

The best way to tackle these issues is by implementing DNS monitoring safeguards to identify what’s causing them. You should also check your switches, VLAN tags, and distribute tasks between servers.

2) Slow servers and loading time

If your servers are particularly slow, they could be hosted using a shared account, which means that your site is sharing the server with hundreds, possibly thousands of other websites. You can address this common roadblock by checking with your hosting company to determine whether or not the site is hosted on a dedicated server.

If you’re hoping to see just how slow your site is, go to Google and use its PageSpeed Insights tool. All you have to do is enter your domain name and click Analyze. The tool looks at the contents of the site and identifies the elements that are making it run slower. The tool churns out suggestions that will help your website run faster.

3) Poorly written code

Another web application performance problem that many face is with poorly written code, which could refer to inefficient code, memory leaks, or synchronization issues. Your application could also deadlock due to ineffectual algorithms, as well as the performance degradation of a web application. Old versions of software or integrated legacy systems can also take a toll on your website’s performance.

You can tackle this issue by ensuring that your developers are using the optimal coding practices, as well as some automated tools such as profilers and code reviews.

4) Lack of load balancing

Slow response times can also be caused by poor load distribution. When new site visitors are assigned incorrectly, it can drown out your servers even if the system is under capacity. Such an issue can cause a slow response time, especially if your site is receiving too many requests.

Tools such as NeoLoad and AppPerfect help you find infrastructural weaknesses that you may be experiencing, while also testing problem areas. You should also work on a cluster of servers instead of simply having a single server that takes all the load. Service-oriented architecture (SOA) can help with scalability issues when more servers are added. This design tool causes application components to provide services to the site’s other components through the communication protocol.

5) Traffic spikes

Spikes happen, especially during a marketing promotion with videos, and a company may not be prepared for the extra traffic. This issue can also cause your servers to slow down, hindering the performance of your site and harming your brand.

One solution is by setting up an early warning system using simulated user monitoring systems such as NeoSense. Doing so will help you see when traffic is impacting transactions before users are affected negatively by the experience.

6) Specific HTML title tags

Even the name of your website can affect its performance as HTML title tags are essential to its success. These tags sum up the entire content of your website or web page to major search engines such as Google. However, a lack of specificity in your domain name can lower its visibility. This is due to the fact that sometimes site owners use the same title throughout their website, which causes search engines to look for duplicate title tags and pares them, causing sites to lose traffic.

You can tackle this issue by doing a name search as “site:yourdomain.com.” Go to Google Search Console (which used to be known as Google Webmaster Tools) to analyze your website. The tool will offer you information on HTML errors such as missing title tags, duplicate meta descriptions, missing descriptions and more.

7) Failing to optimize bandwidth usage

When developing and testing a site, businesses often rely on a local network environment. This may not seem like an issue at first because adding visual, audio, video or other high-volume data may not affect your local network. However, consumers accessing the website at home through their smartphones may face a series of issues you weren’t anticipating.

Make sure you optimize your bandwidth usage for a performance boost. Some of the elements you can include are the minification of JavaScript, the minification of all CSS, server side HTTP compression and optimization of image size and resolution.