The switch that died at 9 am
A switch has run quietly for nine years. On Monday morning it fails. Forty staff cannot work, and the spare is in a shop that opens at ten.
A plan to look after kit before it fails would have cost far less. This lesson is about that plan: changes, monitoring, and replacing parts at the right time.
Key words
| Word | What it means |
|---|---|
| Maintenance | the work of keeping a system running well |
| Predictive maintenance | replacing a part before it is likely to fail, using its expected life |
| Reactive maintenance | fixing a part after it has failed |
| Run to failure | using a part until it fails, on purpose, because failure is cheap and safe |
| Benchmark | a measured figure that shows how well a system should perform |
| Packet analyser | a tool that captures and reads the data on a link |
| Change log | a list of every change made to hardware and software |
| Stakeholder | a person who is affected by a change |
| Network | devices joined so they can share data |
| Firewall | a device that allows or blocks traffic by rules |
| Vulnerability | a weak spot that an attacker could use |
The cells students fill in are left blank in this preview.
By the end of this lesson
Note
I can explain how a network is kept in good order through changes, monitoring and planned replacement.
Starter: faults
Have a go
Starter: faults
3 questionsYour answers are marked here and are not kept. A college account keeps every student’s marks.
Installing and configuring changes
Changes keep a system safe and current. Apply fixes for known weak spots. Check each change still works with the rest of the system: compatibility.
Log every change, hardware and software. Tell every stakeholder, such as users and managers, what changed and when.
Monitoring performance
Watch how people use the system. Watch traffic and load. Install a monitoring system, such as a packet analyser or a firewall that logs what passes.
Record benchmarks. When a figure drifts from its benchmark, something has changed. That is how you find a fault before a user does.
What to watch and why
| What to watch | What it shows | What you do |
|---|---|---|
| User activity | Who uses what, and when | Spot misuse and plan capacity |
| Traffic and load | Busy links and times | Add capacity before it is full |
| Benchmarks | Speed and errors against a good day | Investigate any drift |
| Change log | What changed before a problem began | Link a fault to a change |
The cells students fill in are left blank in this preview.
Predictive, reactive and run to failure
Predictive maintenance replaces a part before it is likely to fail. It uses the part's expected life. Reactive maintenance fixes a part after it fails.
Run to failure means using a part until it fails, on purpose. It suits a cheap part that is easy to swap, such as a patch cord. Keep parts in use until they fail naturally or are due for upgrade.