Google Analytics

Search This Blog

Sunday, April 30, 2006

Having real-time information at your palm

Google has revolutionalized search engine. However, they are not the only one available to do that.
Based on the book "the google story", I have learnt that:

1. Yahoo uses google's search algorithm.
2. Ask Jeeves uses "Teoma" algorithm, developed by several professors at Rutgers university.

So, we got a lot of information out there, but how much of that has benefited us directly rather than creating a reason for web surfing and information research. Has our life already been changed so much so that decisions are based on available informations ? Even if we do that, to what percentage of that has been done on real-time basis ?

There is a saying that the Malaysian economy is based on crude oil prices, where having risen to $100 per barrel would cause it trumbling down. Currently, it stands at $71.88.

Also, the recent publication from The Star about the average rising of temperature to 0.8C for the last 35 years has brought about unusual thunderstorm and lightning which was responsible for many destruction. This is certainly not a plesant news for the country.

Therefore, I predicted that some day, we would be able to have real-time informations delivered to our palm; via mobile technologies. And we should be able to access information using GIS interfaces.

The truth is out there.

Friday, April 28, 2006

Figuring out the objectives

As I was figuring out the ideas which would enable me to become an entrepreneur, and a great one. I went through mental struggles. I know from the fact that to become a worldwide force, the vision has to be a worldwide impact. So much so that it has be somehow related to changing people's life or enhancing life. If we are going to say that we want to build software to target the SME, then it will never go great because because the objectives are so specifically engineered and focused, not taking in considerations for external forces which may bring about unfavourable conditions to meet business expectations.

This world is already at its best. The only left that is important is for every human being to achieve self-realization. The quest for miraculous feats such as cloning, eternity and etc are not required.

Google was started off as a trial-and-error project. Even though its founders are smart and fun, they have no reason to fail because there is no deadlines.

Thus, there is no way we can do anything similarly 9 years later when it was all started. Even if you are given a big venture capitalist funding (which you will not get), you might still not able to make it.

Thus, what is left now, which I have reasons to believe are:
1. Helping people to earn the same amount of money in lesser time. --> Automation
2. Helping people to live the dream lifes in current life. --> VR
3. Helping people to become smarter (Learning a new language and doing translations seem to work for me)
4. Helping the natural environment.

Figuring out the objectives is important.

Malaysia IT spending prospect 2006 - v2

Just yesterday, we were talking about how the rise in interest rate would help spur ICT activities in the country, today the Malaysian ringgit came to a soaring 8 years high, closing 3.6 to the US dollar. This is expected, to the experts of course, because once the interest rate is increased, spending would reduced (to deflate inflation) and there will be less Malaysian Ringgit flowing around the globe. Thus, demand for Ringgit would thus increased inevitably.

All these proves that Malaysian would be richer to a certain percentage and business trade would be decreased as expected reaction. This is actually good for consumerism. In other words, retails profits, suppliers to consumers activities would increase while b2b (business to business) activities would decrease; a natural choice to increase net income per se.

China also increased its benchmark interest rate and caused the oil prices down to US 71.35 from US 73.00. This proves that interest rate hike would give rise to decrease demand for trading; to benefit from higher profit margin.

To conclude this per se, year 2006 would be a great year for ICT spending in Malaysia. Driven by a few key factors and growing supporting factors:

1. The government just announced the 9th malaysian plan
2. Interest rate increased by 0.5% percent which improves the wealth of Malaysian people in the macro point of view
3. Currency gained strenght which also signifies the overal increased of wealth for the people.

Thursday, April 27, 2006

Malaysia IT spending prospect 2006 - v1

With introduction of 9th Malaysian Plan on March 2006 earlier, many industrial players were positive about making impacts to the society.

Either the government has real money or not, we can't be too sure. But it will be certain that government projects will be awarded and money from the central fund will be liquidated over the next five years. So, wealth are up for grap!

It is quite amazing that many good news from within the Malaysian context emerged. In fact it couldn't be too wrong that every major industry player was waiting for years to the good news of th 9th Malaysian Plan. The irony of all these is that you might have to wait for the next five years if you haven't been prepared to take up instance opportunities!

Can we assume that the central bank also was energized by similar trend ? For raising interest rates to 3.5% yesterday. This is seen by many as within expectations because the prospect for the country is rosy for the coming months. Industrial analyst predicted that interest rates could hit 4% by year end 2006, putting inflationary pressure to a lose valve.

To my understanding, the good news would be that entities would be expected to have windfall of 0.5% when year end comes. Thus, even though the hike of interest rates encourages saving, it also prompted business entities to increase banking transactions because they are assured with higher saving interest rates with a bonus of 0.5% by year end. Therefore, IT spending within Malaysia context would be increased by 0.5% from now till year end.

Let's see how the development goes, I hope to give you more micro-analysis when more news surface in the future.

Wednesday, April 26, 2006

Scott McNealy and Sun Microsystem



Scott McNealy of Sun Microsystem has finally resigned from the post of CEO while remaining as chairman.

He started Sun Microsystem from Stanford and Sillicon Valley. Over these 22 years, his had lead the team to achieve the following milestones.

1. Java powers over 3.5 billion devices.
2. Sun Solaris is one widely adopted operating system for server.
3. Niagara Chip is a great technology.
4. SPARC is the most advanced multi-core system. The SPARC architecture is reportedly to have been made open-source recently(2005).

The tag line for Sun Microsystem is "The network is computer", he predicted that software will be free for all in the future with the statement "software will all go free". Putting a permanent rivalry block against Microsoft who is reportedly currently hired around a few thousands Indian programmers to keep the system updated 24 hours around the world.

Anyway, Sun Microsystem is currently showing business losses and his resigned sent the company's share up 9%.

Tuesday, April 25, 2006

Solving RSI with automation

Google blog's recent entry about Avoiding RSI; no-follow, http://googleblog.blogspot.com/2006/04/avoiding-rsi.html; which estimated to cost US $20 billion annually.

Personally, I had encountered some instances of RSI after a full day of typing.

Hence, have anyone thought about contigency plan for RSI ?

My hope is that by the next decade, automation should NOT just be applicable to business processes, but,  it should also be applicable to processes at the personal level, particularly those which help with automating work processes.

One good example is the usage of bulk emailing software;no-follow, http://www.massmailsoftware.com/

I am quite sure that Google Lab is already developing many personalized technology for its user currently.

I am waiting for the day to come when I don't have to study through a 200 pages book, yet is still able to appreciate and comprehend everything as though I have actually read it page by page - as good as it gets...

That is what I called personal automation.

Thursday, April 20, 2006

VEry high level marketing

whose plan was it to have invited China's president Hu Jintao over to Microsoft for touring and dinner...? Must be a great business architect.

China's PC manufacturer recently committed a purchased of $400 million worth of Microsoft Operating system, after an interesting visit to the software powerhouse HQ at Seattle by China's president.

The China president is expected to exchange frank words with President George W. Bush over the problem with trade deficits between both countries amounting up to billions of US dollars; favoring the China side due to its large scale cheap manufacturing capability.

Thus, it seems like China is offering US help.

It is hard to imagine who could have been the first person who initiated this whole event, a great marketing executive instead.

Wednesday, April 19, 2006

RIP, IGRP, EIGRP, OSPF comparison

CCNA Level:

Routing Fundamentals:
- Static Routing: Manually update possible routes.
Dynamic Routing: Only need to update adjacent routes.
Default routing:
+ You can only use default routing on stub networks. It is also useful for summarization of static route.
LabC(config)#ip route 192.168.10.0 255.255.255.0 192.168.40.1
LabC(config)#ip route 192.168.20.0 255.255.255.0 192.168.40.1
LabC(config)#ip route 192.168.30.0 255.255.255.0 192.168.40.1

LabC(config)#ip route 0.0.0.0 0.0.0.0 192.168.40.1 (Just one statement can replace all three above)

Where 192.168.40.1 is known as the gateway of last resort.

Three ways of creating default route.
A#ip route 0.0.0.0 0.0.0.0 192.168.40.1 --> A.D = 1
A#ip route 0.0.0.0 0.0.0.0 s0/0 --> A.D = 0
A#ip default-network 192.168.40.0 --> REMEMBER to use network ID!
When IGP(RIP,IGRP,OSPF) is configured, default-network would be advertised.

Thus, A.D of static route by default = 1 (unless you wish to change it)
Connected interface = 0
static route = 1
EIGRP = 90
IGRP = 100
OSPF = 110
RIP = 120
External EIGRP (AS redistribution) = 170
Unknown =255 (never be used)

- When routing between routers, layer 3 or above is used.
When the transferring packets between interfaces, layer 2 or below is used.
- Routing from host to router.
ICMP(Layer 3) à IP (Layer 3) à Packet (layer 3) --> ARP (Layer 3) à Frame-Encapsulation(layer 2) à FCS-CRC (layer 2) à Packet (layer 3).
Router will then do the same thing by pushing the packet into the next interface
- Rule of thumb:
The destination MAC address will always be the router's interface (if you packets are destined for outside the LAN), never hub/bridge/switches.

EIGRP
- Dynamic routing (support very large network)
- classless protocol (subnet mask is preferred to be specified during IP addressing)
+ VLSM (Variable length subnet mask) / CIDR, not by default - seldom asked in the exams
+ Give rise to discontiguous network
+ Auto summarization (forced summarization)
--> Therefore, when advertised for network, you NO need to provide the mask information.
- concept of Autonomous Aystem (share routing information)
+ neighbor discovery
+ Mechanism:
+ Hello or ACK received
+ AS numbers match
+ Identical metrics (K values)
+ This will create the topology table.
- used only when all routers are Cisco
- hybrid : send distance-vector and link-state packets
+ distance-vector
+ Feasible distance = reported distance + metric to the neighbor reporting the route
+ Reported distance =
- Event-driven (hold-down value = 0)
+ suitable for very large networks. (maximum hop count of 255)
- Supported all routed protocols (IP, IPX, AppleTalk)
+ Due to its protocol-dependent modules (PDMs)
- Communication via RTP (Reliable Transport Protocol)
- Reliability is the focus
- uses multi-cast and unicast (when no reply is received from particular router, 16 times max a.k.a reliable multicast).
- Best path selection via DUAL (Diffusing update Algorithm)
+ Topology table is used as extension to the routing table, for best path analysis. Means tat will be used as input towards DUAL alogorithm
+ Fastest routing protocol (a.k.a fastest route convergence )
+ Each routers having a copy of neighbors' routes (neighbor table)
+ Keep updating the neighbor table
- Load balance of up to 6 links (based on metrics)
+ Metrics is refer to link cost (which is then bandwidth and delay by default), however, EIGRP may use a combination of four.(bandwidth, delay, load, reliability)
- Support "passive interfaces", similar to switch "port-fast"

Others:
- EIGRP routers that belong to different A.S can't share information automatically (they don't become neighbors). This is a good practice to reduce bandwidth.
If you still insist to share information between these routers, then you have to do it manually. This is known as redistribution.
- Internal EIGRP A.D = 90
External EIGRP A.D = 170 (when redistribution[automatic or manual] happens)
+ Automatically redistribution - In an existing IGRP network, putting a router with EIGRP having the same A.S as the existing IGRP will bring about automatic redistribution.

- A feasible successor is a backup route and is stored only in the topoloyg table. A successor route is stored in the topology and also the routing table.
Having feasible successor helps with network convergence (thus no need to study the network anymore, just need to update changes)

OSFP:
- Dynamic Routing (large network support)
- classless protocol (subnet mask is preferred to be specified during IP addressing)
+ VLSM/CIDR (not by default)
+ Gives rise to discontiguous network, therefore catering for huge network (reduce bandwidth)
+ Route redistribution - a translation service between routing protocols.
+ No auto-summarization BUT MANUAL summarization.
--> Therefore, when advertised for network, you have to specify the reversed mask. However, you if don't specify the mask information, then it will assume as the default mask.
- Send link-state packets
- Supported only IP routed protocol.
- Minimize routing update traffic (the one with lowest management bandwidth requirement)
--> Because it is event triggered
- Hierarchical network. Because it is using Area (even though everybody has to connect to Area 0). "Area" is actually a subset for Autonomous system (in EIGRP). Typically, OSPF runs inside an autonomous system.
- Topology database is also known as "link-state database". It is used as input to the algorithm for shortest path computation.
- OSPF can work with layer 2 networks of broadcast-multi-access(i.e Ethernet) and Non-broadcast-multi-access(NBMA) (Frame Relay, X.25, ATM), as well as point to point and point to multipoint.
BMA, Point to mulitpoint = DR and BDR needed
NBMA, point to point = DR and BDR not needed

- OSPF process ID:
+ Locally significant only
+ Grouping of commands under specific running process. (Each process maintain separate copy of its topology table)
+ Enable OSPF routing on the router.
- Recommended to use loopback interfaces because it will always be active, and since loopback interfaces' IP will become the router ID, this will also affect selection for DR and BDR.

RIPv2:
- classless protocol
- Auto-summarization (forced summarization)
--> Therefore, when advertised for network, NO need to specify the netmask.
- Supported only IP routed protocol.
- Distance-vector protocol
- Gives rise to pinhole congestion
- Slow convergence time.
+ Gives rise to inconsistent routing tables and routing loops (when update not done simultaneously).
Routing loop counter mechanism:
+ Split horizon = never send back to same direction
+ Route poisoning = set the hop count in routing table to 16

Routing loop counter tool:
+ Holddowns timer
+ Maximum hop count =15 (the number 15 is considered as a loop)
--> Split horizon and route poisoning rely on holddowns timer and maximum hop count for its working. --> Exam question
+ Not suitable for large network.

RIPv1:
- classfull protocol
+ No VLSM
- Auto-summarization (forced summarization)
--> Therefore, when advertised for network, NO need to specify the netmask.
- Supported only IP routed protocol
- Load balance of up to four links (with equal cost links, that is hop count.)
- Distance-vector protocol
- Gives rise to pinhole congestion
- Slow convergence time.
+ Gives rise to inconsistent routing tables and routing loops (when update not done simultaneously).
Routing loop counter mechanism:
+ Split horizon = never send back to same direction
+ Route poisoning = set the hop count in routing table to 16

Routing loop counter tool:
+ Holddowns timer
+ Maximum hop count =15 (the number 15 is considered as a loop)
--> Split horizon and route poisoning rely on holddowns timer and maximum hop count for its working. --> Exam question
+ Not suitable for large network.

IGRP
- classfull protocol
+ No VLSM
- Supported only IP routed protocol
- Distance-vector protocol
- Maximum hop count of 255, however uses bandwidth and delay as the metric instead.

All in one:
- RIPv2 and EIGRP support discontiguous networking, but no by default. OSPF does support discontiguous networking by default because it does not auto summarize classful boundaries as RIP and IGRP do.
-By default, RIP can load balance across four links as well. However, unlike IGRP and EIGRP, they must be equal links (hop count)
-Under RIP, the passive-interface command will prohibit the sending of route updates but allow their receipt. Thus, a RIP router with a passive interface will still learn about the networks advertised by other routers. This is different from EIGRP, where a passive-interface will neither send nor receive updates.
- Multi vendors environment --> RIPv1 (outdated), RIPv2 (smaller network), OSPF (large network)
Cisco only --> IGRP(outdated), EIGRP (large network)
- Static route between two ends has the less bandwidth requirement.
- Link-state protocol has more intelligence than distance-vector protocol. Hybrid is the king.
- RIP & IGRP can only be used on a classful network (with consistent subnet masks)

Tips:
- Always compare OSPF to RIPv2 and RIPv1 because these are open-standard. Thus, within an all cisco-router environment, EIGRP would be the best choice.
- Due to slow convergence nature, RIPv1 and RIPv2 requires too much bandwidth, thus OSPF is preferred.
- classfull protocol means either:
1. Don't think about subnet / ignore subnet
2. Use the most significant octets

For instance: the classful of 172.16.0.0 /24 is 172.16.0.0
not 172.16.10.0

Monday, April 17, 2006

Marketing Strategy - Free for grap

There is a saying that Malaysian cannot resist free thing.. however, this is depending on the nature of the "free" gift.

Free things which I can't resist:

1. Grants/scholarship for study, R & D and business setup.
2. Free movie tickets
3. Free interest rate for credit cards, loans and etc.
4. Free Internet subscription
5. Free phone calls.
6. Free download of movies, software and music.
7. Free petrol.
8. Free parking.

Free things which I will resist:
ANYTHING as long as I need to provide you with details of my home address, home phone and IC number.

Thus, providing something for free has always been a good marketing strategies, but it is really a fair deal for the recipient ? To figure out the answer, we need to evaluate the objective of the strategy...

For instance, AirAsia 2005 crazy 2 million nos. of free air-ticket for grap has set the standard that providing a free "interface" may not be harmful towards revenue generation. Many speculation arised aftermath, trying to reason how AirAsia actually make the profit. Nevertheless, the idea of "everybody can fly now" is so greatly immersed within the heart of many that the cost of traveling from anywhere to KLIA has been considered as negligible.

DIIP Dot Net, a MSC status web solutions company is again providing similar "free" effort, by providing 1000 SMI with free web site solutions for one year only. The fact that Malaysia has 349, 617 nos. of SME/SMI registered proved that it is a good deal, a great marketing campaign to attract the whole market by profiting only less than 1 percent of the group.

This is good because other people such as freelancers can ride on this opportunity to help customers (SME/SMI) to apply for such facility, and should the 1000 free seats are no longer available, then freelancers may be able to charge for the service.

Take a look at another local web solutions company profile.. Web Channels

At another development, hardware giant Intel Corp is also having plans for SME/SMI market, via its PAT technology.

Sunday, April 16, 2006

HP PSC 1402 unable to print

I just got a new printer HP PSC 1402 installed to a Windows XP system.

Initially, it couldn't print out for good. Trying to cancel requested print job was equally suck as hell. Nevertheless, the printer was in good shape and it could perform self-alignment test and other functions such as copying and etc. It just wouldn't print document based on request, not even "print test page". In all it seemed like it couldn't communicate with the PC via USB.

oh hell, what should I do now ? shouldn' HP just work like tat ?

The next thing I did was to figure out who were faulty ? Printer, Windows XP or simply the USB ports (software / hardware)...

I first tested with my laptop.. initially, it seemed like not going as well. Then, suddenly, it worked. This means that the printer is functional per se.

Then, I hooked an old printer Canon iP1000 to the Windows XP and it couldn't work well too. At certain instance, the error message mentioned "USB device is not recognized".

Therefore, this brings a strong case towards problem with USB; hardware or software. The first thing I did was to uninstall each of the USB devices registered to Device manager. For Windows XP, restarting the PC will get them being reinstalled back.

After the restart, problem solved. It was USB ports software corruption.

Tuesday, April 11, 2006

3G is not an option for Digi

First of all, let's find out some basic background about 3G. 3G is a technology concept. The technology behind it is called the IMS (IP multimedia subsystem) .

What is IMS ?

It is actually a technology to cater for an interface between mobile (celular or analogue) and IP (digital). The vertical application is identified as multimedia services. It uses SIP technology.

So, IMS is actually a concept and standard developed by 3GPP; consortium taking care of 3G.

The enabler ?

To know exactly how IMS is being implemented, we may take a look at a company called Kineto Wireless which developed a sub-technology called UMA.

UMA (Unlicensed Mobile Access) technology is a 3G standard for mobile/WiFi convergence (Google also wants to play the WiFi game). Therefore, it is an extra extension for IMS.

Why 3G ?

According to Kineto Wireless, the reason for adopting IMS and hence UMA is due to the following elements:

1. Mobile Market Saturation
2. Looking for new growth while at the same time reducing cost

and with implementation of IMS, the following benefits are expected.

* Grow ARPU by increasing the adoption and use of mobile data/IMS services
* Lower CapEx and OpEx by offloading the radio access network from mobile data/IMS services
* Reduce churn by proving an enriched mobile data/IMS experience where subscribers spend most their time

In other words, 3G is not an option for the telcos, it is a must for them; it is part of the milestones. This is interesting really because in Malaysia, currently there is one celular carrier operator who doesn't have a 3G operating license; that is Digi.

Previously, when the government announced that Digi will not be given the license, Digi counter-reacted by down-playing the importance of 3G; according to the CEO Morten Lundal, 3G is just another option. Well, it may not seem so unless the could make sure that the following actually happen:

1. Market would NOT become saturate
2. New growth is not important as well as cost management.

They then followed-up with announcing capital repayment for its shareholder(April 2006). Similar corporate strategy was demonstrated by Redtone during 2005 when their CEO was wrongly slapped with legal case (which was eventually over-thrown). Redtone followed-up with a big coverage on The Star Biz with interviews on the CEO and future growth of company.

Personally, I feel bad for Digi because the government already stated explicitly that the award of 3G license to MiTV and Time Dotcom is politically driven. However, other analyst have other opinions.

Let's see what Jeff Ooi says about this? --> Blind Spots

So how now ? We will wait and see how Digi react further (If you don't research on it, you will never know). This will be something interesting.

So far, Digi have carried out the following counter strategies:
1. Digi brand refreshes
2. New packages offering
3. Market Segmentation

Many years ago, I also recalled that some "telco" experts predicted that Malaysia will never ever having 3G implementation because it is so costly for the Telcos (Maxis, Celcom, Digi). Looking back, it looks absolutely ludicrous now. The accountants must have misquoted the "expects". Any way, even from Yahoo news, others claim that it will take longer time for UMA to be fully implemented.

But would this affect FDI(foreign direct investment) to Malaysia ? I hope not.

Saturday, April 08, 2006

MSC Rebranding Exercise...

Following similar efforts like

Telekom Malaysia and Intel, MSC (Multimedia Super Corridor) has completed its rebranding exercise, shortly after the government announcement of the 9th Malaysian plan (where over RM 7 billion would be allocated for Ministry of Science and Development.)

MSC is now know as "MSC Malaysia", following similar effort by IT-Sideways: Tech Blog rebranding to "IT-Sideways: Tech Blog Malaysia".

I assumed that TM rebranding exercise was merely to boost corporate ego (the fact that they reportedly spent RM 9 million on it. Perhaps there was a way get tax benefit)
while for Intel was to steer a new direction in facing competitions. For MSC, I guess it is meant to cater for make-over, after what many considered it is time for MSC to buff up. Simply put, when the CEO questions marketing manager of weak performances, the latter might suggest rebranding exercise due to personal instinct. Effectiveness of the instinct will be dependent on the message carried through the media with respect to the exercise. Therefore, may we analyze that ....

MSC Malaysia is supposed to carry the message of :

  • Revitalising the MSC brand and image to enable every Malaysian citizen to enjoy the benefits;

  • MSC Malaysia as a key driver and enabler of the local ICT sector to conquer the local and global markets;


  • Make use of ICT to accelerate the transformation of other sectors of the economy, including manufacturing, services and agriculture; and


  • Creative multimedia content industry as a new growth resource under the Ninth Malaysia Plan.

  • Thus branding is a good corporate tool. When used correctly, it produces cutting edges.

    In Malaysia, there is this thing called the Branding Association of Malaysia (BAM) which provides a brand assessment tool; Brand Report Card.

    "It is designed as an audit tool. The underlying importance to build brands is to know where the brand failed and how much it has failed to live up to its expectations" The Sun Weekend, Jan 28-29, 2006. For more details, please check out..

    Thus MSC Malaysia is a brand carried out by MDec (a new brand for MDC) the company. We hope it will succeed in ...

  • Revitalising the MSC brand and image to enable every Malaysian citizen to enjoy the benefits;


  • MSC Malaysia as a key driver and enabler of the local ICT sector to conquer the local and global markets;


  • Make use of ICT to accelerate the transformation of other sectors of the economy, including manufacturing, services and agriculture; and


  • Creative multimedia content industry as a new growth resource under the Ninth Malaysia Plan.

  • Check out the website for sub-categories.

    Publisher Strategy by CJ

    I received a newsletter from Commission Junction... about publisher strategy.. I find it that it is quite a helpful theory for blogger to improve their blog... It encompasses things like SEO and etc. It can be a good reference for knowing what is to be done in the marco manner, meaning to say that based on these strategies, blogger may be able explore the following issues:

    1. What you want to achieve ?
    2. Where are your now ?
    3. How do you plan to get there ?
    4. What price you are willing to pay for to get there ?

    To keepalive as what Cisco router would do...

    The following are the strategy shared for free..

    Based on the strategy, it is also understood that the biggest price to pay for is commitment. You just got to keep providing rapport.

    Traffic - Driving It, Keeping It, Converting It
    As long as you are not sitting in it, traffic is a welcome site to any publisher. But how can you best monetize those wandering online eyes? In this issue of the CJ Wire, we look at traffic and focus on how to best find it, keep it, and, most importantly, convert it.

    Step One - Drive Traffic
    You might be a great publisher, but are you a good advertiser? It is important for you to attract quality traffic to your site to help increase conversions that, in turn, increase your revenues. By taking some tips from our advertising partners, you can increase the quantity and quality of your traffic. Check to make sure you are adhering to the best practices to get people to visit your site. Here are some important lessons you can learn from your advertising partners:

    • Hide and Seek - Look for new customers on chat rooms, blogs, complementary sites, associations, and groups of people that would be interested in your site.
    • Channel Surf - Can you take advantage of e-mail campaigns or traditional online banner ads? Do some research to see if you need a multi-channel approach to driving traffic.
    • Cover the Basics - Are you listed on basic search and Web directories? Are you looking beyond Google and Yahoo!? Don't get tunnel vision. Second- and third-tier search engines are great options to help drive more traffic to your site and can be very cost-effective.
    • Search High and Low - Probably the most important way to drive quality traffic is to have a comprehensive search program. Make sure your content is current to enhance your natural search listings. To increase your ranking within the search engines' natural search rankings, you need to generate as many links to your site as possible. Get some buzz going with press releases, blogs, or a unique set of features that allow people to link to from their sites.
    • Click Up Your Heels - Investigate a pay-per-click program to see if it is a viable option.
    • You can find more tips about search engine optimization here.

    Step Two - Driving in the Right Direction - Quality Traffic

    It is easy to just think in terms of the number of potential customers visiting your site, but you should change your thinking and look into the quality of those visitors. We define quality traffic as someone who is genuinely interested in your content or product. Driving people with no connection or excitement to your site wastes everyone's time and energy. This is why we encourage our publishers to value quality over quantity. So, how can you drive quality consumers?
    • Price Comparison Sites - Many online shopping sites, such as PriceRunner, will list product information and pricing for free. Set up product catalogs and feeds with sites like PriceRunner to ensure you are in front of consumers at critical decision-making times.
    • Be Specific - Are your key words too broad? Make sure you have investigated all ways for someone to find your products - right down to SKU numbers and product detail. This will help attract customers as they are ready to make a purchase.
    • Think Locally - 70% of online households use the internet to search for local services and businesses. Take advantage of the geographical targeting capabilities of the major search engines to create localized campaigns. Increasing the relevancy of your campaign will increase conversions.
    • Think Globally - Consider promoting some of the many international programs that are available for US-based publishers. When launching an international campaign, make sure that you take into account spelling variations and different regional search habits. Consider hosting your pages in the country in which you are trying to increase their ranking. Also be sure to have as many local sites from that county linking to your site, which will increase your rank.
    • Put into Context - Have you considered contextual search? By having an ad or text link next to and matched with a relevant article or other non search engine-generated content, you can increase the likelihood of receiving higher quality traffic.
    • Best Behavior - Behavioral advertising is like a close cousin of contextual search; behavioral advertising is based on a user's historical behavior on the Web, which helps you focus your ad delivery on your target audience.
    Step Three - Keeping the Traffic There

    While we are not saying you should have "gridlock" on your site, there are some ways for you to make sure visitors stick around awhile to help convert eyeballs to customers. Remember, your goal is to build relationships with your visitors - you want to turn them into lifelong customers. Here are some tips to help you keep your visitors on site:

    • Freshen Up - Keep your content and creative fresh. Nothing is worse than a customer thinking "been there, done that" and leaving to find something new. Change is good.
    • Make it Special - If possible, create a promotion or contest to not only keep people on the site, but to get them to come back to check results.
    • Learn from the Best - Look at the top-ranked pages within your category, as well as other unrelated categories, and try to determine what makes them unique. What do they have that encourages people to visit and remain on their site? Try applying some of those principles to your site.
    • Don't Back Up - One factor that contributes to a lower search ranking in the search engines are users hitting the 'back' button after clicking through your search engine listing. The engines realize that your site is relevant for a particular search term; however, this is not a good user experience so as a result, your site ranking goes down. Do what you can to keep people on your site after they first click through.
    • Tool Time - Consider adding a unique free tool or "widget" to your site to get visitors to remain on your site, as well as revisiting your site. Encouraging customers to link directly to the "widget" will also increase your natural search ranking.
    • Brush Up on Technology - From podcasts to RSS feeds, there are a lot of new ways to keep your customers interested. For more information, please click here.

    Step Four - Conversion - Driving in the Fast Lane

    Above all, your first and foremost thoughts should be on delivering value to those that come to your site - this will help convert shoppers into consumers. Here are some tips on how to bring the best to your visitors:

    • Soft Landing - Build landing pages for each keyword you are targeting instead of directing everyone to the home page. With custom landing pages, consumers are one step closer to making a purchase.
    • Solid Foundation - A well-designed site that is easy to navigate is one of your best tools to convert traffic. Also, make content easy to find. Use call-to-action buttons (such as "Buy It Now!") to cut down on the number of clicks and make sure that the button is placed "above the fold" on the page to increase conversions.
    • Show Off - Use Commission Junction's reporting tools to show offers for best sellers or hot items. Keep successful promotions front and center.
    • Give them a Push - Offers, such as free shipping, sometimes give that little extra something a consumer is looking for. Everything from limited time offers to discounts can go a long way.
    • Don't Overreach - Are you offering what you promised? Nothing will turn away a customer like a misleading ad. Make sure your content is relevant to your search terms to give them what they want. Remember, quality is more important than quantity when it comes to traffic.

    Friday, April 07, 2006

    Cisco - Frame Relay Fundamentals

    Frame Relay (Theory):

    1. Talking about how having a physical interface to pass through a frame relay cloud (a switch). In real life, a router is used as frame relay switch (router can be configured into a switching mode.

    2. Two modes of frame Relay
    2a. Point to point (For a different subnet environment)
    2b. Multipoint (For same subnet environment)

    4. Frame relay makes your medium of connectivity transparent (you can't see it)

    5. Virtual circuits creation out of a single medium (FR switch)

    6. Frame relay (layer 2 protocol) does not contain IP addresses.
    + Apply DLCI number (because frame has no MAC address)
    + DLCI number are given by your ISP.

    7. Frame routing = Frame switching.

    8. Frame relay signaling (LMI)
    Cisco proprietary =
    ANSI = International standard
    ITU = International standard

    + The purpose of LMI is for keepalive (To figure out whether a link is working or not)

    9. You can apply a particular speed (commited VC speed) onto your virtual circuit (pros)
    This is because it is logical (you can manipulate the bandwidth)

    10. If you didn't manage the bandwidth properly, you might suffer a possible bottleneck. Cisco has this thing called "congestion management"
    The sender has to find a way to figure out whether the path ahead is congested or not.
    BECN = Backward explicit congestion notification
    FECN = Forward explicit congestion notification

    BR = Branch
    FR = Frame Relay

    Path 1 Path 2
    HQ -------------------------------- FR -------------------------------------------BR

    FR would know that if Path 1 is congested by analysing traffic coming to it.
    If path 1 is congested.
    BECN = 0
    FECN = 1 (path 1 congested)

    As traffic reaches BR with encapsulation of BECN = 1 and FECN = 0, BR will then send back traffics to HQ with the following encapsulation:
    BECN = 1 (Path 1 congested)
    FECN = 0

    When HQ receives the traffics (with all encapsulation), it will know that Path 1 is congested and thus reduce traffic my 25%.

    --> This will happen automatically.

    11. In frame relay point-to-point (P2P)...
    Logically all routers are connected in a serial mode (point to point)
    Similar to "slip horizon" of RIP and IGRP, you can only have one way traffic , to overcome loops.
    --> Cater for convergence or collision management. (Refer to the theory of stub network)

    To counter split-horizon problem, you have to create virtual tunnel or link (to cater for bi-directional sort of scenario)

    12. Point to point (mapping) is done automatically because eventually everyone will get it.
    Point to multi-point mapping have to be done manually.

    13. Frame Relay is not outdate, it is meant for the following benefits:
    + Cost benefits (without Frame Relay you probably need to have a few leased lines to connect to branches)
    + More security ( not making use of IP addresses (layer 3) even though still go through the bandwidth)
    + More efficient (faster)

    14. PVC / PVI (In your streamyx)= Frame Relay identifier.
    15. Configuration of Frame Relay will only be covered in CCNP.
    16. In a frame relay network (point to multipoint), the IPs for all devices has to be in the same network. (same subnet)
    17. In the real work, frame relay cloud would be located in the Central Office (CO)

    Ethernet cannot use PPP because of the encapsulation problem (has to use PPPoE). Frame Relay is also another encapsulation (for serial interfaces as well), but it can work on ethernet.

    Frame Relay Configuration (Practical):

    if you are having multiple encapsulations within your network configured to all interfaces, then your serial link will be going up and down.
    Thus, you need to make sure that all serial interfaces having the same encapsulation.

    Point to Point Configuration: (Practical)
    1. Apply frame-relay as encapsulation to serial interfaces.
    2. Create sub-interfaces to serial interfaces.
    3. Apply an IP address to the particular sub interfaces.
    4. Apply DLCI number to sub interfaces.
    5. Ping neighbors nodes to check if it is working.
    6. Enable RIP(version 2) for network. Otherwise, the routing table is not there yet. (Now you have a fully routable frame-relay network)
    7. Advertise for adjacent networks.
    8. Also advertise for ethernet network. (have to wait for 40 seconds)

    Create Multipoint (practical):
    Not purely broadcast (probably just some portion of it)

    1. create only one sub-interfaces and set it to multipoint.
    2. Apply an IP address to the sub interfaces.
    3. Do mapping. (Map local DLCI number to a particular IP address)
    4. Ping your neighbors.
    5. But you are unable to ping yourself (because your frame-relay don't have yourself in the entry, thus can't route back to you) --> why can't this be automatic.

    Thursday, April 06, 2006

    The real google story

    Do you know the real google story, yes i am talking about Google Inc (the most popular search engine in this world)

    let me put it this way...

    Yahoo made its debut on 1997, I attended computing college on 1999 and about a year later I started seeing Google homepage on every PC in the computer lab. Since we would be required to search for information for our course works, it worked fine to me. I started to pay attention to Google's pagerank algorithm when I wanted to get a website I developed for a charity house onto some place appropriate in the search engines. Today, Google is considered a great success of entrepreneurship in modern history. Well, in fact many thought that it only took Larry Page and Sergey Brin 3 years to create something from scratch; which is a common notion during conversations while discussing about setting up possible startup ventures.

    Being a programmer myself, I knew it isn't easy to build another Google or even search engine. Practically, you can't set up a company by hiring 10 programmers to build another search engine from scratch. My reason is that you don't know where to start and where it ends... and you have deadline for it. So, I never actually thought about doing another Google even though I admire the kind of life which Larry Page and Sergey Brin have.

    To prove me right, after reading the first few chapters of "The Google Story" by David A. Wise, I indeed got it right. The idea of setting up Google came all the way back to 1995 (9 years ago) when both Page and Brin were still students at Stanford university. The idea was really simple, one of the founders wanted to download the whole Internet for archiving and eventually found reasons for doing it and then developed softwares and hardware to manipulate those data and improve efficiency. So, one of them is good at maths(software) while the other is good at assembling stuff(hardware). The hardware are know as Googleware. It started off as a research project and finally by year 2000, they got the beta running for the world.

    Therefore, for the record, if any of us wanted to build a $20 billion company, you would need the following:
    1. At least five years of research and development (trials and errors) before going beta.
    2. Having a student life (when survival is not an issue)
    3. Having a post-graduate life. This is important because undergraduates usually feel less secured than post-graduates. Undergraduates tend to worry about livelihood than post-graduates who have had certain achievement in life previously.

    In fact, the real story is that Google was started ever since they were born; in other words, it was started really early. Both (Page and Brin) were borned into families of scientific and mathematics... When I was 10 years old, all I cared about was playing badminton. When I was 15 years old, I never thought about any inventions. I have a lot to catch up indeed.

    Second Conclusion:
    1. Find something meaningful to do (make sure that you have money to survive)
    2. Keep doing it while perfecting it (Just like the Japs) because accumulation of efforts and experiences means everything to success. If you had started something 10 years ago, nobody can catch up with you for the last 10 years.
    3. Do it early (there will never be a better time)

    So many entries been written on Goggle for this blog. To my surprised, they have also invented an ultimate marketing engine for themselves.

    Learning from Cisco

    Cisco wanted to develop internetworking equipment and also to be the market leader to stay competitive.

    I have been attending CCNA courses for the past 8 days and realized something.

    First of all, CCNA course is good for networking fundamentals, despite the fact that it is biased towards Cisco. However, as much as it may, it actually reeducate me a lot about the following fundamentals:

    1. IP, Network mask and subnetting (Very important)
    2. Routing
    3. Switching

    Mind you that other brands such as 3COm and DLink would based their product on these fundamental too. Therefore, I have reason to believe that having attended CCNA would mean many things to 3COm and DLink, as much as it is to Cisco. Besides that, to become networking security expert, fundamentals are important to people like Checkpoint and etc.

    Nevertheless, the best part about sitting in the course full-time is having the opportunity to get insights into internetworking. So having insights is good, having insights is good.

    However, I do notice that, we spent a lot of times doing configurations and 99% of it are done on console or using hyper-terminal. As much as I like to brag about it, I do notice that there might be some element of redundancy for doing Cisco equipment configuration.

    First of all, since Cisco are so damn great, they could possibly create Artificial Intelligence modules which can automate many things to the max. Secondly, most difficult part of configuring Cisco equipments is to parse through all text feedbacks from the equipments itself and thus if they have a way to cater for graphical illustration, things would be much more easier.
    For instance, it would be better to have a graphical illustration of "sh cdp neighbors" or "sh ip route" than going through the text-based feedbacks such as the following:

    1.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
    C 1.0.1.0/30 is directly connected, Serial1
    C 1.0.0.0/24 is directly connected, Serial0
    R 2.0.0.0/8 [120/1] via 1.0.0.1, 00:00:13, Serial0
    R 3.0.0.0/8 [120/1] via 1.0.0.1, 00:00:13, Serial0
    R 4.0.0.0/8 [120/1] via 1.0.0.1, 00:00:13, Serial0
    R 5.0.0.0/8 [120/2] via 1.0.0.1, 00:00:13, Serial0
    R 6.0.0.0/8 [120/2] via 1.0.0.1, 00:00:13, Serial0
    R 172.16.0.0/16 [120/4] via 1.0.0.1, 00:00:13, Serial0 --> THis is the entry.
    R 7.0.0.0/8 [120/1] via 1.0.1.2, 00:00:12, Serial1
    R 8.0.0.0/8 [120/3] via 1.0.0.1, 00:00:13, Serial0
    S* 0.0.0.0/0 [1/0] via 1.0.0.1


    From here, we can derive a few conclusions:
    1. Survival problem - No one, even Microsoft could possibly develop a so wonderful software where after certain period of development, everybody can virtually sit-down and wait for the sales to happen without requirement for support because the software had been done so well.

    Thus, for Cisco to create state-of-the-art automated embedded software for its equipments, it will come in a matter of time. Thus to promote their product as prestige as possible, they would need professional-technical people to use their equipment with challenging interfaces.

    They will promote the idea with the highest regards that it is feasible to become a technie sitting behind a console doing configurations like nobody business.

    2. Business Model and Economical - Perhaps, which I can't be sure of, that the current model of "Equipments + technical-certified-people" will more likely going to make more money to them than other models; which are not proven otherwise. I am sure that this current model really got many fired up, especially those innocent to technical stuff.

    What do you guys think ?

    Cisco - Wan Technologies

    WAN Technology (Theory):
    - Connections between remote sites.
    - What is the limitation of ADSL (Streamyx) ?
    + Topology (need to be 5 km from the nodes)
    Leased LIne = T1, T2, T3
    + Encapsulation: HDLC (High data link control) - Vendor proprietary. HDLC is preferred for leased line because both ends will use similar equipments' brand.
    PPP (point to point protocol) - industry standard
    SLIP (outdated)
    Packet Switched = Like Streamyx. Connect to streamyx and they help you to foward the packets around.
    + Encapsulation: X.25
    Frame Relay
    ATM
    Circuit Switched = PPP (Preferred)
    Menara Telekom, Brickfields, Putrajaya --> Central office for streamyx.
    Demarcation point = Deciding whose responbility is for.
    Customer demarcation point = CPE (customer premises equipment)
    DCE End (from ISP) = Clock rate will be set.
    DTE End (customer side)
    CSU & DSU -
    V.35 faster connector.

    Point to Point (Theory) :
    - By default, CISCO equipment use HDLC.
    - PPP:
    1. Builds a tunnel.
    1a. LCP (Link-control protocol) - subset of PPP (layer 2).
    + Both ends' LCP must be opened. --> Bring up the data-link (layer 2)
    1b. NCP (network control protocol) - (layer 3)
    + Both ends' NCP must be able to communicate

    2. PPP offers the following which HDLC doesn't provide.
    2a. Authentication
    2b. Compression --> Discussed more in BCRAN
    2c. Error Detection
    2d. Multi-link
    - Logical bundling. Take two links (64K) and combine into one link. --> Become a faster link (128K)

    3. PPP establishment steps.
    3a. Link establishment. --> LCP phase
    3b. Authentication phase (optional)
    PAP (only for CCNP) - must have username and password registered in the router.
    CHAP (Challenge handshake authentication protocol) -
    + using Hash
    + Mandatory for both side to authenticate.
    3c. Network communication --> NCP phase

    CHAP (Practical):
    1. Find out what encapsulation u r using ?
    HDLC and PPP cannot be established on Ethernet interface; only on serial interface.
    "sh int serial [number]
    Serial0 is up, line protocol is up
    Hardware is HD64570
    Internet address is 1.0.0.254/24
    MTU 1500 bytes, BW 1544 Kbit, DLY 20000 usec,
    reliability 255/255, txload 1/255, rxload 1/255
    Encapsulation HDLC, loopback not set -------> HDLC by default.

    2. Select a serial interface.

    3. Set to "PPP" Encapsulation

    4. Three reaons for unstable link (going up an down)
    4a. Clock rate not set
    4b. Encapsulation is wrong
    4c. Keep-alive.

    5. "sh int serial 0"
    Serial0 is up, line protocol is down
    Hardware is HD64570
    Internet address is 1.0.0.254/24
    MTU 1500 bytes, BW 1544 Kbit, DLY 20000 usec,
    reliability 255/255, txload 1/255, rxload 1/255
    Encapsulation PPP, loopback not set
    Keepalive set (10 sec)
    LCP REQsent ---> Sending request.

    6. Change all serial interfaces to PPP. Then all serial interfaces would be "up" and thus LCP will be working.
    Serial0 is up, line protocol is up
    Hardware is HD64570
    Internet address is 1.0.0.254/24
    MTU 1500 bytes, BW 1544 Kbit, DLY 20000 usec,
    reliability 255/255, txload 1/255, rxload 1/255
    Encapsulation PPP, loopback not set
    Keepalive set (10 sec)
    LCP Open
    Open: IPCP, CDPCP

    7. Apply authentication to PPP.
    callback Authenticate remote on callback only
    + someone call me via PPP. I can disconnect it and then callback.
    callin Authenticate remote on incoming call only
    + allow someone to call in.
    callout Authenticate remote on outgoing call only
    + only allow someone to call out but not call in.
    ms-chap Microsoft Challenge Handshake Authentication Protocol (MS-CHAP)
    optional Allow peer to refuse to authenticate
    pap Password Authentication Protocol (PAP)
    ISDN = can also run PPP.

    8. All serial interfaces will come down again. Because there is no password applied to it.
    LabD#sh ip int brie
    Interface IP-Address OK? Method Status Protocol
    BRI0 unassigned YES NVRAM administratively down down
    BRI0:1 unassigned YES unset administratively down down
    BRI0:2 unassigned YES unset administratively down down
    Ethernet0 unassigned YES NVRAM administratively down down
    Serial0 1.0.0.254 YES NVRAM up down --> Already down.
    Serial1 1.0.1.1 YES NVRAM up down --> Already down.

    9. Create username(s) and password(s) for your neighbors to log into your routers.

    10. But you don't know who your neighbors are because you can't do "sh cdp neighbors".
    The only way is to guess the "terminal name", such as LabD.

    i.e
    LabD#conf t
    Enter configuration commands, one per line. End with CNTL/Z.
    LabD(config)#encapsulation ppp
    LabD(config)#int serial 0
    LabD(config-if)#encapsulation ppp
    LabD(config-if)#ppp authentication chap --> apply for all serial links
    LabD(config)#username labA password cisco --> Create authentication for labA
    LabD(config)#username labG password cisco
    LabD#debug ppp authentication --> turn on events driven debugging features
    Authentication messages for debug:
    06:06:35: Se1 CHAP: O CHALLENGE id 232 len 25 from "LabD" ---> Challenging
    06:06:35: Se1 CHAP: I CHALLENGE id 228 len 25 from "LabG" ---> Challenging
    06:06:35: Se1 CHAP: O RESPONSE id 228 len 25 from "LabD"
    06:06:35: Se1 CHAP: I RESPONSE id 232 len 25 from "LabG"
    06:06:35: Se1 CHAP: O SUCCESS id 232 len 4
    06:06:35: Se1 CHAP: I SUCCESS id 228 len 4

    If you remove your authentication, it will stop sending the keepalive with authentication informaiton.
    Serial0 is up, line protocol is up
    Hardware is HD64570
    Internet address is 1.0.0.254/24
    MTU 1500 bytes, BW 1544 Kbit, DLY 20000 usec,
    reliability 255/255, txload 1/255, rxload 1/255
    Encapsulation PPP, loopback not set
    Keepalive set (10 sec) --> Keepalive here!

    Sample Questions:
    LabO:
    + username = LabJ
    + password = s3cUr0
    LabJ
    + username = LabO
    + password = s3CUr0
    Why authentication fails ?
    --> Because password has to be case sensitive.

    Difference between router and switches:
    Router (for OSPF) always look for the lowest priority in selection for DR and BDR.
    Switches always look for the lowest priority in selection for root bridge.

    Wednesday, April 05, 2006

    Cisco - Routing Continues - IGRP, EIGRP, OSPF

    Routing- Continues:
    ----------------------------
    RIP metric is based on "hops count".
    IGRP metric is based on "bandwidth and Load".
    RIP has administrative distance of "120"
    IGRP has admistrative distance of "100"
    RIP version 2:
    - classless protocol.
    Thus when setting up RIP version 2, only need to advertise for the most significant octet with no requirement for subnet mask.
    LabD(config)#router rip
    LabD(config-router)#version 2
    LabD(config-router)#network 1.0.0.0


    EIGRP (Theory):
    ------------------------
    - Classless protocol --> only need to advertise the most significant octets coupled with reversed mask.
    - Events driven--> hold down value is zero (because things got updated instantly)
    --> CHek this out form "sh ip protocols"
    - Incremental updates of informations (routing tables and etc) --> reduce bandwidth requirements.




    Support all protocols:
    ------------------------------
    1. IP
    2. Appletalk
    3. IPX
    As compared to RIP and IGRP, which only supports IP protocols.

    Similiarity to IGRP:
    ---------------------------
    1. Autonomous system.
    2. K-Values.

    Compatible with IGRP:
    --------------------------------
    EIGRP = IGRP x 256 bytes
    Network A = EGRP
    Network B = IEGRP
    When Network A receives packets from Network B, they will be divided by 256 bytes. When network B receives packets from Network A, it will be multiply by 256 bytes.

    Troubleshooting:
    ------------------------
    When someone advertises the wrong network (wrong IP or wrong reversed mask), the link will then be down. Thus you need to know your topology map to troubleshoot things out.

    Setup EIGRP (Practical) :
    -----------------------------------
    1. configure terminal
    2. router eigrp [number]
    3. advertise the classless network with reversed mask.
    4. Set "no auto-summary" to EIGRP object.
    5. clear route table --> "clear ip route *"
    6. show ip route --> "sh ip route"

    i.e
    LabD(config)#router eigrp 100
    LabD(config-router)#network 1.0.1.0 0.0.0.3
    LabD(config-router)#network 1.0.0.0 0.0.0.255
    LabD(config-router)#no auto-summary
    LabD(config-router)#clear ip route *
    LabD#sh ip route

    Summarized IP route table:
    1.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
    C 1.0.1.0/30 is directly connected, Serial1
    C 1.0.0.0/24 is directly connected, Serial0
    R 2.0.0.0/8 [120/1] via 1.0.0.1, 00:00:28, Serial0
    R 3.0.0.0/8 [120/1] via 1.0.0.1, 00:00:28, Serial0

    Non-summarized IP route table:
    1.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
    C 1.0.1.0/30 is directly connected, Serial1
    C 1.0.0.0/24 is directly connected, Serial0
    2.0.0.0/8 is variably subnetted, 7 subnets, 5 masks
    R 2.0.0.0/26 [120/1] via 1.0.0.1, 00:00:17, Serial0
    R 2.0.0.0/8 [120/1] via 1.0.0.1, 00:03:02, Serial0
    R 2.0.0.64/27 [120/2] via 1.0.0.1, 00:00:17, Serial0
    R 2.0.0.96/28 [120/3] via 1.0.0.1, 00:00:17, Serial0
    R 2.0.0.112/28 [120/4] via 1.0.0.1, 00:00:18, Serial0
    R 2.0.0.248/30 [120/5] via 1.0.0.1, 00:00:18, Serial0
    R 2.0.0.252/30 [120/5] via 1.0.0.1, 00:00:18, Serial0
    3.0.0.0/8 is variably subnetted, 3 subnets, 3 masks
    R 3.0.1.0/25 [120/2] via 1.0.0.1, 00:00:18, Serial0

    setup Ethernet Link to EIGRP:
    1. Setup ethernet interface --> "no shutdown"
    2. Add ip address to it.
    3. Goto EIGRP object.
    4. Advertise ethernet interface by specifying the classless octets with reversed mask.
    5. Check ip route --> "sh ip route"

    Successors and feasible successors:
    Since EIGRP has the features of successors and feasible successors, thus there is no downtime for it.
    P 1.0.1.0/30, 1 successors, FD is 2169856
    via Connected, Serial1
    P 1.0.0.0/8, 1 successors, FD is 2169856
    via Summary (2169856/0), Null0
    P 1.0.0.0/24, 1 successors, FD is 2169856
    via Connected, Serial0
    P 2.0.0.0/8, 3 successors, FD is 2195456
    via 172.16.70.3 (2195456/2169856), Ethernet0
    via 172.16.70.5 (2195456/2169856), Ethernet0
    via 172.16.70.8 (2195456/2169856), Ethernet0
    P 2.0.0.0/26, 1 successors, FD is 3734016
    via 172.16.70.9 (3734016/3708416), Ethernet0
    P 3.0.1.0/25, 1 successors, FD is 2707456
    via 172.16.70.1 (2707456/2681856), Ethernet0
    via 1.0.0.1 (3193856/2681856), Serial0
    P 3.0.0.0/8, 2 successors, FD is 2195456
    via 172.16.70.6 (2195456/2169856), Ethernet0
    via 172.16.70.2 (2195456/2169856), Ethernet0

    Take the example of ..
    P 3.0.1.0/25, 1 successors, FD is 2707456
    via 172.16.70.1 (2707456/2681856), Ethernet0
    via 1.0.0.1 (3193856/2681856), Serial0

    The successor for it is:
    172.16.70.1
    Ethernet link will always be chosen as successor over others for reason that it will be faster.
    Serial0 is considered to be feasible successor when..
    2681856(advertised)

    RTP protocols:
    1. Send "multicast" to neighbors.
    2. if no reply from someone. Send "Unicast" for at least 16 times.
    3. If still no reply from that someone, then it means dead.

    After morning break.
    -------------------------------
    1. IGRP does not understand subnet mask. thus when you advertise, no need to put subnet mask
    LabD#conf t
    LabD(config)#router igrp 100
    LabD(config-router)#network 1.0.0.0
    LabD(config-router)#network 1.0.1.0
    Refer to Advertise_EIGRP_Diagram.jpg
    2. IGRP doesn't understand the EIGRP (classless).
    3. IGRP doesn't support "auto-summary"
    -> Because it can't provide summarization.
    Summarization:
    172.16.10.0 /28
    172.16.10.16 /28
    172.16.10.32 /28
    172.16.10.48 /26
    .0 = 00 000000
    .16 = 00 010000
    .32 = 00 100000
    Thus, the first two bits are similar. Therefore, the subnet mask is thus 11000000 = /26
    can be summarized as 172.16.10.0 /26
    Using RIP, no need to enter all routes manually, just need to advertise adjacent networks.

    OSPF:
    1. open standard, not proprietary to Cisco.
    2. Based on the fastest link (link speed)
    3. Similar with EIGRP (events driven), holddown value = 0. --> check with "sh ip protocols"
    4. Uses "auto-summary"
    Communication:
    1. Using link-state advertisement.
    2. Instead of having a topology table, it has the link-state table.
    EIGRP:
    1. Neighbors table
    2. Topology table
    3. Routing table

    OSPF:
    1. Neighbors table
    2. Linked-state table
    3. Routing table
    Single Area OSPF - Everybody will be in Area 0.

    Based on the rule of OSPF:
    1. Irregardless of any conditions, any other Area(s) has to be connected directly to Area 0 in other to communicate with others. This is for CCNA level. Otherwise, you may connect to non-Area 0 and create a virtual link.

    The metric for your OSPF is cost calculation.
    EIGRP = 90
    IGRP = 100
    OSPF A.D = 110
    RIP = 120

    setup OSPF (Practical)
    1. LabD(config)#router ospf [any number from the list] --> setup OSPF
    i.e LabD(config)#router ospf 666
    2. Add adjacent networks with reversed mask to area 0.
    3. network [adjacent network ID] [reversed mask] [area]
    "sh ip route"
    1.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
    C 1.0.1.0/30 is directly connected, Serial1
    C 1.0.0.0/24 is directly connected, Serial0
    2.0.0.0/8 is variably subnetted, 6 subnets, 4 masks
    O 2.0.0.0/26 [110/933] via 1.0.0.1, 00:01:03, Serial0
    O 2.0.0.64/27 [110/997] via 1.0.0.1, 00:01:03, Serial0
    O 2.0.0.96/28 [110/1061] via 1.0.0.1, 00:01:03, Serial0
    O 2.0.0.112/28 [110/1071] via 1.0.0.1, 00:01:03, Serial0
    O 2.0.0.248/30 [110/1119] via 1.0.0.1, 00:01:04, Serial0
    O 2.0.0.252/30 [110/1119] via 1.0.0.1, 00:01:04, Serial0
    3.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
    O 3.0.1.0/25 [110/192] via 1.0.0.1, 00:01:04, Serial0
    O 3.0.0.0/24 [110/128] via 1.0.0.1, 00:01:04, Serial0

    Example illustration:
    2.0.0.96/28 [110/1061] via 1.0.0.1, 00:01:03, Serial0
    110 = A.D (administrative difference)
    1061 = Cost
    Check ethernet bandwidth
    --> sh int ethernet [number]
    Check serial bandwidth
    --> sh int serial [number]
    --> BW 1544 Kbit (1.5 Mbit)
    Check neighbors table (for OSPF)..
    "LabD#sh ip ospf neighbor"
    Check topology / link-state table (for OSPF)
    "LabD#sh ip ospf database"
    --> Notice that you are on the verge of setting a multi-access network around a switch (switch connected to multiple routers)
    To counter linkage-redundancy within a multi acess network:
    DR - Designated router
    BDR - Backup designated router
    With the concept of DR and BDW, everybody will only be connected to DR and BDR only.
    LabD#sh ip ospf neighbor
    Neighbor ID Pri State Dead Time Address Interface
    88.0.0.9 1 2WAY/DROTHER 00:00:36 88.0.0.9 Ethernet0
    88.0.0.8 1 2WAY/DROTHER 00:00:36 88.0.0.8 Ethernet0
    88.0.0.10 1 FULL/BDR 00:00:36 88.0.0.10 Ethernet0
    88.0.0.11 1 EXSTART/DR 00:00:36 88.0.0.11 Ethernet0
    88.0.0.6 1 2WAY/DROTHER 00:00:37 88.0.0.6 Ethernet0
    88.0.0.5 1 2WAY/DROTHER 00:00:37 88.0.0.5 Ethernet0
    88.0.0.2 1 2WAY/DROTHER 00:00:37 88.0.0.2 Ethernet0
    88.0.0.3 1 2WAY/DROTHER 00:00:36 88.0.0.3 Ethernet0
    88.0.0.1 1 2WAY/DROTHER 00:00:36 88.0.0.1 Ethernet0
    1.0.1.2 1 2WAY/DROTHER 00:00:36 88.0.0.7 Ethernet0
    88.0.0.7 1 FULL/ - 00:00:39 1.0.1.2 Serial1
    88.0.0.1 1 FULL/ - 00:00:39 1.0.0.1 Serial0
    To have a fair election of who will become DR and BDW, everybody will do a clear ospf process.
    LabD#clear ip ospf process
    LabD#Reset ALL OSPF processes? [no]: y
    Election is based on router ID (IP address of the ethernet interface)
    When you reload your router, only it will pick your new router ID.
    The router ID will be based on selection of all the IP addressess between all the interfaces available, such as ethernets, serials and BRIs.
    DR = the one with the highest router ID.
    BDR = the one with the lowest router ID.
    Dead Time = If the router who is currently a DR or BDR not responding to the network within 40 seconds... new DR and BDR will be elected. The following event messages will be expected.
    01:06:25: %OSPF-5-ADJCHG: Process 666, Nbr 88.0.0.11 on Ethernet0 from FULL to D
    OWN, Neighbor Down: Dead timer expired
    LabD#
    01:06:31: %OSPF-5-ADJCHG: Process 666, Nbr 88.0.0.9 on Ethernet0 from LOADING to
    FULL, Loading Done
    In order to have a standardized for your router's Router ID. The router ID will always based on loopback interface's IP. This is because loopback interface can never be down as compared to physical interfaces.
    --> Create a loopback interface (logical interface on your router).
    LabD#conf t
    Enter configuration commands, one per line. End with CNTL/Z.
    LabD(config)#int loopback 0
    LabD(config-if)#ip address 5.0.0.251 255.0.0.0
    LabD(config-if)#no shutdown
    LabD(config-if)#
    01:12:59: %LINEPROTO-5-UPDOWN: Line protocol on Interface Loopback0, changed state to up
    Priority values: The higher the priority, the better the chance to be DR.
    The lowest value for priroty = 1
    Maximum priority value = 255
    Value 0 = (disqualified to become a DR)

    Thus election for selection of DR or BDR rules:
    1. Router ID comparison.
    2. Priority
    3. Multi-access network (switch or hub)
    Down-Time:
    EIGRP don't have don't time but not OSPF because the former has successor and feasible successor.
    --> Check from EIGRP "sh ip eigrp topology"
    Write Erase:
    LabD#wr erase

    1.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
    C 1.0.1.0/30 is directly connected, Serial1
    C 1.0.0.0/24 is directly connected, Serial0
    D EX 4.0.0.0/8 [170/2195456] via 1.0.0.1, 00:05:43, Serial0
    D EX 5.0.0.0/8 [170/2707456] via 1.0.0.1, 00:05:43, Serial0
    D EX 6.0.0.0/8 [170/2707456] via 1.0.0.1, 00:05:43, Serial0
    Notice that for LabD, it will not be able to detect network 2 and network 3. However, it is able to detect IGRP network via "D EX"
    Now, you need to do "Registration"
    1. Can register a few things to you RIP (Version 2)
    Take RIP and pump it out as eigrp or igrp as some metric [number]
    2. Thus, goto each router object (RIP, IGRP, EIGRP, OSPF) and redistribute them accordingly to each other protocols according to appropriate metric.
    --> This will be for CCNP.
    After you done the distribution, you will see the following from "sh ip route"
    1.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
    C 1.0.1.0/30 is directly connected, Serial1
    C 1.0.0.0/24 is directly connected, Serial0
    2.0.0.0/8 is variably subnetted, 3 subnets, 3 masks
    D EX 2.0.0.0/26 [170/2681856] via 1.0.0.1, 00:03:26, Serial0
    D EX 2.0.0.64/27 [170/2681856] via 1.0.0.1, 00:03:26, Serial0
    D EX 2.0.0.96/28 [170/2681856] via 1.0.0.1, 00:03:26, Serial0
    3.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
    D EX 3.0.1.0/25 [170/2681856] via 1.0.0.1, 00:03:21, Serial0
    D EX 3.0.0.0/24 [170/2681856] via 1.0.0.1, 00:03:21, Serial0
    D EX 4.0.0.0/8 [170/2195456] via 1.0.0.1, 00:18:14, Serial0
    D EX 5.0.0.0/8 [170/2707456] via 1.0.0.1, 00:18:14, Serial0
    D EX 6.0.0.0/8 [170/2707456] via 1.0.0.1, 00:18:14, Serial0
    --> All networks are visible, this is because we have setup a "redistributing" router.

    Access List:
    Used to control your traffice.
    - Manage traffic as it grows.
    - Filter out certain packets. (check packets IP only)
    + compared to firewall, it will check the contents of the packets.
    Functions:
    1. Stop people from passing into your network.
    2. Stop people from telneting into your network.
    Features:
    1. Traffic Shaping --> CCNP
    - Certain traffic will be allowed to pass through. (based on priority)
    2. Dial-on-demand routing --> CCNA.
    Two type of access list:
    1. Standard access list.
    + Only look at your source. (deny everything)
    2. Extended access list.
    + Check your source, your destination and what you plan to do (FTP or Telnet) ?

    Tips:
    When you put the "standard access list".. make sure don't overkill.
    - To do this, put your "standard access list" close to your destination. (the most relevant router)
    When you put the "Extended access list"... make sure don't over-eat (consume unnecessary bandwidth.)
    - To do this, put your "extended access list" as close to your source. (the most relevant router)
    - It will read every entry from a logical manner.

    Inbound or outbound Access list:
    ------------------------------------------------
    "inbound' attribute is set when the "access list" is loaded to the entering gateway (interfaces) of the most relevant router.
    "outbound" attribute is set when the "access list" is loaded to the exiting gateway (interfaces) of the most relevant router.
    "permit any"
    "implicit deny all"

    Deny:
    1. Deny one by one.
    2. Permit everyone except those who are supposed to be denied.
    3. Deny in a single statement.
    To deny 7.0.0.2 and 7.0.0.3, we can use 7.0.0.0 / 30
    / 30 = 111111 00
    .1 = 00000001
    .3 = 00000011
    LabF(config)#access-list 1 deny [network ID] [wildcard bits]
    --> The reversed mask is now known as [wildcard bits]
    Thus "reversed mask" is supposed to mean "all" while wildcard bits means "selected"
    i.e
    LabF(config)#access-list 1 deny 7.0.0.0 0.0.0.3
    LabF#sh access-list
    Standard IP access list 1
    deny 7.0.0.0, wildcard bits 0.0.0.3
    permit any
    LabF#
    Take your access list and apply to an interface.
    LabF(config-if)#int ethernet 0
    LabF(config-if)#ip access-group 1 out
    Summary:
    1. create access list
    2. put access list to an interface
    i.e
    LabF(config)#access-list 1 deny 7.0.0.0 0.0.0.3
    LabF(config)#access-list permit any
    LabF(config-if)#ip access-group 1 out
    Standard Access List also blocks out "reply" packets.
    Let's assume that there are 20 nodes with ip 7.0.0.1 to 7.0.0.20
    To deny 7.0.0.1 to 7.0.0.3 and 7.0.0.9 to 7.0.0.15
    How to write the shortest access-list command?
    access list 1 deny 7.0.0.0 /30 (0.0.03)
    access list 1 permit 7.0.0.8
    access list 1 permit 7.0.0.8 /29
    access list 1 permit any

    Example:
    7.0.0.32 /29
    .32 = 00100 000
    .39 = 00100 111

    In other words, whether it is "reversed mask" (when used for EIGRP) or "wildcard bits"(when used for access list), it means "the otherwise IP(s)"

    Tuesday, April 04, 2006

    Cisco - Routing Fundamentals

    Routing Theory:

    1. To enable inter-VLAN routing.
    2. Identify the best path.
    - Know the destination
    - Identify sources for more information
    - Discover possible routes
    - Determine the best path.
    3. Router must learn the destinations which are not directly connected.
    4. Static Route
    - Manual.
    - Router doesn't have to learn by itself. Thus, CPU resources are not required (Pros)
    - If you don't put a route to a particular network, nothing can pass through it. Thus it is always secured.
    - You have to enter all possible routes manually. (cons)

    5. Dynamic Route
    - No administrative overhead. (Pros)
    - Router will not be making error. (Pros)
    - More CPU resources required (cons)
    - More bandwidth required; but yet negligible(cons)

    RIP - Sends out the whole routing table (1000 lines long)
    Then you may have things like incremental routing update.

    To configure two routers for connectivity, make sure that both have to be configured properly so that traffic can travel both directions.

    "Complete route" -
    ip route [network ID] [subnet mask] [next hop IP]
    ip route 7.0.0.0 255.0.0.0 8.0.0.1

    "Default routes" - only used on "stub network" (only physical interface for point of entry and exit) Means enter from one place and have to exit from the same place.
    --> ip route 0.0.0.0 0.0.0.0 [adjacent hop]


    Configure static route (Practical):
    "sh ip route"
    C 1.0.0.0/8 is directly connected, Serial0
    S 2.0.0.0/8 [1/0] via 1.0.0.1
    S 3.0.0.0/8 [1/0] via 1.0.0.1
    C 4.0.0.0/8 is directly connected, Serial1
    S 5.0.0.0/8 [1/0] via 1.0.0.1
    [1/0] via 2.0.0.1
    S 6.0.0.0/8 [1/0] via 3.0.0.1
    S 7.0.0.0/8 [1/0] via 1.0.0.1
    [1/0] via 2.0.0.1
    S 8.0.0.0/8 [1/0] via 2.0.0.1
    S 9.0.0.0/8 [1/0] via 2.0.0.1
    S 11.0.0.0/8 [1/0] via 2.0.0.1

    [1/0] = [adminstrative distance/matrix]
    matrix = hops count

    "sh ip route static" --> show only those manually entered route.

    Routing protocol and routed protocol
    - how to move from routing protocol to routed protocol.

    - routing protocol - RIP, IGRP (interior gateway protocol)
    - routed protocol - IP, Appletalk

    "Autonomous System" = Domain of administration.
    Jaring and Streamxy communicate via EGP (a BGP is an EGP) - two different autonomous system to communicate together.

    For CCNA level, we will only look at IGP(Internal Gateway protocol).

    Distance vector (RIP, IGRP)
    - Find the fatest path
    - Classfull routing protocol (with the same subnet)

    Link state (OSPF - Open shortest path first)
    - Find the shortest path.
    - Classless routing protocol

    Hybrid protocol
    - Both (classless and classful)

    Classfull routing protocol
    - Do not send across network mask information (becoz it assumes that everbody are having the same subnet mask.)
    - It will assume that everybody are in the same subnet.

    Ranking routes based on "administrative distance". The lower the administrative distance, the better the preference.

    Logical Interfaces (or sub interfaces):

    If there is no encapsulation provided for creating logical vlan trunk(logical interfaces), then it will be regarded as default trunk - so called "management" trunk.

    you can have up to 4.2 billion logical interfaces.

    Router don't have the luxury of having many ports to cater for vlan as switches have.

    Distance Vector Routing:
    RIP - uses hop count (how many jumps to get to the destination)
    IGRP - Composite matrix (bandwidth, delay, load, reliability, link)

    Routing Loop Problems:
    Looping will happen when there are inconsistency in the countdown timers (for updating routing tables). Before one routing table is informed that a particular network is down, the hop count will keep mounting up until it reaches infinity. This is a theorized bug.

    - Use maximum hop count - RIP has a maximum hop count of 16
    - Split horizon - you cannot send back information from where it came from.
    - Route poisoning - Immediately set the hop count to 16 (in case of RIP). A matter of tagging information.
    - Holddown timers - Give your network some time to actually converge.
    - Triggered updates - If network down, I inform you that the network is down.

    Link-state routing protocols:
    Link-state and hybrid.

    For CCNA, Area 0 will only be of the concern.
    Area 0 - Transit Area.

    - Routers know the topoloy on real time basis. (and not based on "outdated" neighbors)
    - A lot of resources(memory and CPU) are "eaten" (cons)

    Balanced Hybrid Routing:- can find the fastest and shortest path.

    RIP:
    - based on hop count. (it assumes the shorter the hop count, the faster it is) it will not consider the line speed which then renders the conclusion to be inaccurate.
    - Since there is a maximum of 16 hop counts, thus you can't have a topoloy with > 15 routers.

    Summarization:
    172.16.10.0 /28
    172.16.10.16 /28
    172.16.10.32 /28
    172.16.10.48 /26

    .0 = 00 000000
    .16 = 00 010000
    .32 = 00 100000

    Thus, the first two bits are similar. Therefore, the subnet mask is thus 11000000 = /26

    can be summarized as 172.16.10.0 /26

    Using RIP, no need to enter all routes manually, just need to advertise adjacent networks.

    Enabling RIP (practical):

    1. how to enable RIP.
    1a. start RIP.
    "router RIP"
    1b. Advertise your adjacent network.
    "network [network ID of which you are directly connected]"

    2. view status.
    "show ip protocols" --> Opposite of static route.

    You will notice --> Routing Protocol is "rip"

    3. Remove your static route.

    4. "sh ip routes"

    C 1.0.0.0/8 is directly connected, Serial0
    R 2.0.0.0/8 [120/1] via 1.0.0.1, 00:00:02, Serial0
    R 3.0.0.0/8 [120/1] via 1.0.0.1, 00:00:02, Serial0
    C 4.0.0.0/8 is directly connected, Serial1
    S 5.0.0.0/8 [1/0] via 2.0.0.1
    R 6.0.0.0/8 [120/2] via 1.0.0.1, 00:00:02, Serial0
    S 7.0.0.0/8 [1/0] via 1.0.0.1
    R 8.0.0.0/8 [120/4] via 1.0.0.1, 00:00:02, Serial0
    R 9.0.0.0/8 [120/5] via 1.0.0.1, 00:00:02, Serial0
    R 11.0.0.0/8 [120/5] via 1.0.0.1, 00:00:02, Serial0

    C= directly connected
    R= RIP enabled.
    "RIP" uses 120 for "administrative distance".

    5. "debug ip rip"

    5a. how to see "routing poisoning" at work.

    03:17:15: RIP: received v1 update from 1.0.0.1 on Serial0
    03:17:15: 2.0.0.0 in 1 hops
    03:17:15: 3.0.0.0 in 1 hops
    03:17:15: 5.0.0.0 in 16 hops (inaccessible)
    03:17:15: 6.0.0.0 in 2 hops
    03:17:15: 7.0.0.0 in 16 hops (inaccessible)
    03:17:15: 8.0.0.0 in 16 hops (inaccessible)
    03:17:15: 9.0.0.0 in 16 hops (inaccessible)
    03:17:15: 11.0.0.0 in 16 hops (inaccessible)

    6. "un all" --> stop debugging.
    By defaut, we are using RIPv1. how to change to version 2, to detect classless protocol.
    "conf t"
    "router rip"
    "version 2"

    However, the default is:
    send (packets) = using version 1
    recieve (packets) = using version 1 or 2
    To enable default, do the following.
    labD(config)#router rip
    labD(config-router)#no version 1
    labD(config-router)#no version 2

    Sample Exam questions:
    I run RIP and IGRP, how come I don't see the RIP routes ?
    --> Because IGRP has lower "administrative distance".

    The tough part is fine-tuning the routes.

    IGRP - Proprietary to Cisco
    - You have to talk about "autonomous system".
    - Compared to RIP, which can only load balance between equal cross-path.
    - Composite metric in order: bandwithd, delay. IGRP also picks the lowest matrix.
    - by default: variance of IGRP = 1
    Variance is useful to open up all other path for evaluations.

    for instance, if you have 3 paths with metrics of 10,30 and 50.
    Thus, with default variance, only the first path with composite metric = 1 will be chosen. Thus, to open up the second path, you need to change the variance to 3 (so that when 3 x 10 = 30), the first two path will be considered valid. To open up the third path, have to change variance to 5.

    This is known as "load balancing"

    IGRP (Practical):

    1. start routing protocol.
    "router igrp [autonomous number]

    autonomous number = globally significant number.
    In order for igrp to work, autonomous number must be the same.

    two criterias:
    1a. autonomous number must be the same.
    1b. k-values must be the same

    IGRP metric weight K1=1, K2=0, K3=1, K4=0, K5=0

    2. advertise your adjacent network.
    "network [adjacent network]"
    ~if you advertise the ethernet port, then you will create another route for your route. For instance, if your ethernet is connected to the network of 172.16.0.0, by advertising it under the a particular igrp. Then you will see a difference from "sh ip route"

    3. "sh ip protocols"
    - check out your k-values
    4. "sh ip route"

    IGRP - is obsolete. Newest router with new IOS will not be supporting IGRP. Thus, only RIP will work for them.

    5.To view "route poisoning".
    5a. "debug ip igrp transactions"
    5b. "debug ip igrp events"

    IGRP metric is a 32 bit= 2^32 = 4294967295

    Troubleshooting:
    1. if you advertised wrongly. the route will be interupted.
    2. if you choose the wrong protocol such as the newer protocol eIGRP.

    Remove RIP and IGRP
    -------------------
    1. "conf t"
    2. "no router rip"
    3. "no router igrp"

    Monday, April 03, 2006

    Cisco - Switching Fundamentals

    Theory part:

    Switching:
    1. A switch is a layer 2 (MAC address) device.
    2. Within a LAN connected by a switch, there is only one broadcast domain.
    3. Switches mode:
    Cut-through (directly foward)
    fragment-free (read only the header - first 64 bytes)
    Store and foward (read all contents)

    4. MAC address table.
    - Address learning capability for broadcast or multicast frames. There is no problem with unicast frame because the destination address will be provided in the frame.

    Problems in Redundant Switched Topologies:
    1. Frames don't arrive at the same time("broadcast storm"). This will gives "MAC database instability". This occurs in a situation where there are two uplinks (cross-cable) between two switches.

    --> Cisco provides "loop avoidance" technology - "spanning tree protocol" (STP).

    2. "Multiple Frame copies" - Nodes receive multiple copies of similar frame over and over again.

    STP (Spanning Tree Protocol):

    RSTP will be covered in CCNP.

    2a. Root bridge (main bridge of the network) - All ports has to be opened up (rule of thumb)

    Rules for STP:
    2b. One root bridge per network.
    2c. One root port per nonroot bridge. (means if given choice, the link connected to the root-bridge must be chosen or opened over others)
    2d. One designated port per segment.
    2e. Nondesignated ports are unused.

    Root bridge selection (make sure there is no loop):
    without knowing your root bridge, you wouldn't know which link to be closed.

    Maximum bridge priority = 2^16
    BPDU(sends every two seconds) contains:
    - MAC address
    - Priority
    - Root bridge (the one with the lowest bridge ID)

    Bridge ID consists of Priority + MAC Address. Thus, when the priority's number are the same, the one with lowest MAC address (numerically) will emerge as root bridge.

    The backup link will come up after 30 + 15 +15 = 50 seconds. (Exam question)

    Spanning Three Path Cost:
    - Based on specialized calculation provided by IEEE.

    In RSTP, instead of having NDP (B), we have the "disable" function. which is much faster.

    Basic Switches configuration:
    STP = auto enabled.
    CDP = auto enabled.
    RSTP = auto enabled.

    - "Show mac-address-table" - Switch MAC address table.
    - Similar to router configuration, you can view all interfaces via "show ip int brief"
    - "show cdp neighbors"
    - "show spanning-tree"
    - "show spanning-tree bridge"
    - "show spanning-tree root"

    How to make the switch be part of the full network ?

    Set DNS:
    ip name-server [ip address]

    set default gateway:
    ip default-gateway [ip address]

    How to change to half to full duplex (by default it is auto):
    - Pick a particular port which is not currently using.

    Apply Port Security:
    Access port = a port which is connected to a host (Nodes or PC) / or carrying one vlan information.
    Trunk port = carrying more than one vlan information.

    1. set the port as "access port".
    2. enable port security to the port.
    3. Mac filtering.

    let's say someone who is not given the MAC access connected to the switch. you can provide there options for these people.
    protect = send an SNMP message to administrator.
    shutdown = shutdown that particular port.

    "terminal monitor" - monitor your terminal activities.

    Switch can't ping to localhost - because of the "firewall" problems. (switch off firewall at the local PC). However, router can access local PC even with firewall on.

    VLAN:
    1. Segmentation (subnet or broadcast domain)
    Access protection.
    2. Flexibility
    Present at other segment vicinity, but still able to access the designated segment.
    3. Security

    Implementing VLAN (for layer 2 switch) requires router. Because VLAN is to separate out ports of the same subnet into segments. Since for same network, ARF will be performed. In other words, you need to create subnets and use router.

    You can't get 6 ports which N.N.10.1, N.N.10.2,N.N.10.3,N.N.10.4,N.N.10.5,N.N.10.6

    and separate into half using router, means separate N.N.10.1,N.N.10.2,N.N.10.3 away from the other 3. Because the router still confused that you are in the same subnet. Thus, the only way out is to create a different subnet. For instance,

    N.N.10.1 /24
    N.N.10.2 /24
    N.N.10.3 /24
    N.N.20.1 /24
    N.N.20.2 /24
    N.N.20.3 /24

    Access link = port connected to nodes (PCs)
    Trunk link = At least a fast ethernet (100 MBits)
    Carries multiple VLAN.

    Encapsulation:
    1. Frame tagging = 802.1Q (modifies your frame)
    2. ISL (Cisco proprietary) = per VLAN Spanning Tree.
    Being phase out, because it is only Cisco proprietary.

    VLAN Modes:
    1. Static mode = manually set a MAC address to a port.
    2. Dynamic VLAN.
    VMPS (VLAN management property server)

    Important of Native VLANs
    Native VLAN = untagged traffic.

    Per VLAN Spanning Tree:
    - by default is not on.
    - Each VLAN would have own path, even though they share the switches. (good for fine-tuning)
    - can only be done on ISL (DLINK, 3COM will not have this.) This is bcoz during ISL encapsulation, besides adding "VLAN ID" to the frame, "BPDU" information is also added in.

    Summary:
    1. In a switch,by default, there is only one broadcast domain.
    2. However, you can create more broadcast domain (VLAN)

    VTP (VLAN Trunking Protocol) :
    1. Server-client architecture.
    2. All switches must have the same VTP domain.
    3. Information are saved in server's (switch) NVRAM. You may also want to have multi-server replication.
    4. "Transparent" - VTP mode. Something like Jackal in card games.
    5. "Revision number" - Updated version number.
    6. VTP Prunning - Uses advertisement.

    Practical part:

    VLAN configuration (practical):
    1. create VLAN. - "vlan 2 name [name]"
    2. "apply" - confirmation.
    3. Have to assign ports to these VLANs.
    4. Move particular port to a VLAN.

    Cat3550(config)#int fa0/1
    Cat3550(config-if)#switchport access vlan 2

    5. Select all ports - use "range"

    Cat3550(config)#int range fa 0/1 - 22
    Cat3550(config-if-range)#switchport access vlan 3

    6. Create trunk.
    6a. Trunking protocol -
    Cisco Catalyst 3550 - has ISL and 802.1Q

    6b. How do you know which link are trunk ?
    "show interfaces trunk"

    ~The purpose making a port to be a trunk is to enable VLAN information to pass through. Thus, either you are using encapsulation "dot1q" or ISL. usually will use dot1q because ISL is only available for Cisco.

    A trunked port is thus considered VLAN-less because it is not supposed to be connected to any nodes.

    7. Change a particular port to "no switchport", it will thus be considered as available port.

    8. Maximum VLAN allowed = 4094 - 1 (VLAN 0)

    VTP:
    1. Check "VTP" information.
    "sh vtp status"
    the most important is "operating mode" and "domain name"

    2. Change domain name.
    Cat3550(config)#vtp domain CCNA

    3. create "vtp password"
    Cat3550(config)#vtp password cisco
    Password already set to cisco

    4. VTP helps to synchronize between two switches.
    Except for the port. For instance when creating a vlan, both switch can see the information, expect for the ports.

    5. change to client mode.
    Can't create VLAN.

    Simulation questions:

    1. IP addresses (first and last).
    have to check out using "sh run" or "sh ip vlan 1"
    2. vtp domain
    3. vtp password.
    Cat3550#sh vtp password
    VTP Password: cisco
    4. server mode.
    5. default gateway.
    --> This type of question will check on you about the usage of "VTP" related command. It doesn't call about subnet or VLAN as well.

    In order for VTP to work, two main criterias.
    1. VTP domain name must be the same.
    2. VTP password.