When you are trying to figure out why your Internet does not work you have to understand the basic steps that are involved when it does work and so work out where the problem lies. Of course, in most case, this is the job of the person on tech support asking the right questions to work out what is happening, but it is worth understanding a few of the basics.
1. The actual wires involved: there is normally a copper pair which is carrying phone and broadband. If that is physically screwed up in some way (which could be intermittent interference, broken contacts, poor quality) then that will cause problems. The most severe types - like a simple disconnection - affect the phone line as well, and you have no dial tone - those problems are easy to identify. However various types of interference (which can be audible noise on the phone line, or not) and quality issues can cause problems.
These problems with the wires result in sync problem. Sync (or synchronisation) is the general phrase we use for the operation of the modem (MODulator/DEModulator) which translates data (1's and 0's) in to tones (radio frequencies) on the wire and back again. If the line is broken in some way the modem may not get sync at all (normally shown by a light on the box). It could get sync, but at a very low speed. It could be that it only works when interleaving is enabled and a slow speed. It could be that there is sync, but lots of errors on the line (which the modem can detect).
All of these symptoms relate to the actual wire. They could be anywhere from, and including, the modem (usually inside a router) at the premises, all the way to the modem at the other end (exchange or cabinet). To fix the problem you usually want to start by trying a new modem, and new cables to the telephone socket. Always remember that a power supply fault can cause a modem to be faulty as well so remember to change that. You should try to eliminate the wiring in the premises by trying from the master socket. This usually has a removable front plate which disconnects all extension wiring and has another socket behind.
If none of that works, it could be the wires, in which case an engineer is needed. It could even be the equipment at the other end - but an engineer can confirm that and arrange a lift and shift (moving to other equipment at the exchange end).
It is a very important step to work out if the problem lies with the computer, the wires (modem issue), the back-haul, or the ISP (or even somewhere in the Internet), as this affects who can fix the problem, and how. Getting this initial diagnostic wrong wastes a lot of time and effort for everyone, which is why it is one of the key parts of the training for our technical support staff.
2. If the modem is OK, and in sync, and not showing errors or slow speed or disconnections, then the problem must lie somewhere else. In these cases there is no point in sending an engineer, the problem lies with the data being sent on the wire, not the wire itself (or the modems).
These cases can usually be diagnosed in various ways - a common type of problem is simply getting a username or password wrong (an authentication problem). Tests on the broadband service and at the ISP can usually tell if this is the case. The router will normally say if there appears to be an authentication issue (e.g. on a web interface). These can however be everything from a simple typo by the end user to problems in the back-haul system and finally problems at the ISP. A good test here is to try a test login, such as a test user or speed test user which does not involve the ISP. If that does not work then there are issues in the back-haul system and possible the BRAS (Broadband Remote Access Server) configuration. In some cases the ISP can do the same test from the exchange using test systems - where that shows problems the ISP knows it is the back-haul.
If authentication works, then that means there is a working connection from the router all the way to the ISP. Well, normally that is what it means. There is a special case of a default accept which is quite easy to detect. This is where the authentication has been faked because the ISP cannot be contacted. But generally, if you can authenticate then you are connected to the ISP and you and the ISP can see that.
3. If you connect to the ISP then the back-haul provider is not normally involved in sorting issues. Occasionally there can be issues like mis-configured BRAS causing slow speeds, or some packet loss or latency in the back-haul network. These are usually things the ISP can measure and detect though.
4. Of course the problem may be in the Internet, with the server you are trying to reach. Testing other routes (e.g. other web sites) helps identify if you have a working Internet connection generally.
But, of all of these tests, checking if the problem lies with the wires or backup-haul or ISP is a crucial first step.
So, why the Broadband 101? Well, because our favorite telco are, once again, insisting that they send an engineer to check the sync on an FTTC line that is not working. Yet the line has sync, and we even see two way traffic to the BRAS. The issue is BRAS config (not authenticating and passing to us as the ISP). It is a simple problem with config and a rebuild of the config on the BRAS (which takes a few seconds) would fix it. Sadly their staff have no clue and insist over and over again on sending more and more engineers, all of which confirm there is sync and go away without fixing the issue. Maybe this blog could be a useful reference for them in future...
People will know I am a slight Stargate fan (!), and I like making PCBs. So, well... Latest is... LEDs First off, the LEDs. There is a very ...
Broadband services are a wonderful innovation of our time, using multiple frequency bands (hence the name) to carry signals over wires (us...
For many years I used a small stand-alone air-conditioning unit in my study (the box room in the house) and I even had a hole in the wall fo...
It seems there is something of a standard test string for anti virus ( wikipedia has more on this). The idea is that systems that look fo...