# Demonstration: Watching a Process Actually Terminate

## OBSERVE

Start a long-running background process and note its PID:

```
$ sleep 300 &
[1] 1421
```

`1421` is the PID (yours will differ every time).

## HYPOTHESIS

Predict: will `kill -0 1421` print anything if the process is still running?

## INSPECT

```
$ kill -0 1421 && echo "still running"
still running
```

`kill -0` sends no real signal — silence (or, here, our own `echo`) means the process exists.

## CHANGE

Now actually terminate it:

```
$ kill 1421
```

## VERIFY

```
$ kill -0 1421 2>/dev/null && echo "still running" || echo "gone"
gone
```

The exact same check now reports the opposite result — real, observable proof the process is
gone, not an assumption.

## Instructor Notes

**Demo goal:** Make "checking before and after" a visible habit, not just narrated — students
should see the SAME command return a different real answer once state actually changed.

**Approximate timing:** 8 minutes live.

**Common mistake to watch for:** reaching for `kill -9` by default "to be safe" — reinforce that
plain `kill` (SIGTERM) is the correct first attempt almost always.