Support / Add targets & link sites

Add targets & link sites

A probe measures what you point it at. Give it targets — the destinations you depend on — and link two of your own sites to watch the path between them, measured from both ends.

Targets and site links

There are two things a probe can measure, and they answer different questions:

Type	What it measures	Answers
Target	The path from one of your sites out to a destination — a SaaS platform, an API, a data centre, a public host.	“Is the way we reach that getting worse, and where?”
Site link	The path between two of your own sites, each running a probe, measured in both directions.	“Is the connection between our offices healthy, and which way is the problem?”

Add a target

Open the Targets screen on your dashboard.
Choose which site should measure it — the probe at that location becomes the vantage point.
Enter the destination as a hostname (preferred, so DNS drift is caught too) or an IP address.
Give it a label you will recognise, and save.

The probe begins tracing to that destination on its next cycle. Within a short while the target carries a reading against its own baseline, hop by hop, with the network on either side of each handoff named.

Link two sites

When two of your locations each run a probe, you can link them and measure the path between them — not a single ping, but the route each way, since the internet is often asymmetric and the outbound path is not always the return path.

Confirm both sites are checked in (see Install a probe).
On the Sites or Links screen, choose Link sites and pick the two.
Save. Each end starts measuring toward the other.

Both directions, separately. A link shows the path A → B and B → A as their own traces. When only one direction degrades, that alone tells you which leg — and which carrier — to chase.

What to point a probe at

The destinations your business runs on — the SaaS apps, payment or API endpoints, and data centres your users and staff actually depend on.
The links between your own sites — office-to-office, office-to-datacentre, branch-to-hub.
A known-good reference or two — a large, stable destination, so you can tell a problem on your path from weather on the wider internet.

You do not need to point at everything. A handful of destinations that matter, measured continuously against their own baseline, is worth more than a long list you never read.

How soon readings appear

A new target or link produces its first reading within a cycle or two. A baseline — the “normal” the colour scale compares against — forms as measurement accumulates, and firms up over the first day or so. Until there is enough history, the panel reads unknown rather than guessing; an unknown is never shown as healthy or as a fault.

Turn on alerting

Once a target or link is measuring, you can have SquidGrid tell you the moment it leaves its baseline — in Slack or by webhook (email coming soon). See Configuring alerts.

Next: Configuring alerts · Reading your dashboard · back to all guides