Alert settings

Watching a wall of green dots is reassuring, but you cannot stare at it all day. The Alerts tab lets Ping Monitor tap you on the shoulder only when something is genuinely wrong – and then tell you about it however suits you, whether that is a desktop pop-up, an e-mail to your phone, or a script that opens a ticket.

Alerting has two halves: the conditions that decide when a host is in trouble, and the channels that decide how you are told. You set them up once, globally, and they apply to every host you monitor.

Conditions: when an alert fires

Rather than reacting to a single dropped ping – which on a busy network happens all the time – Ping Monitor looks at a host’s recent behaviour over a rolling time window and alerts only when a problem is sustained. This keeps a momentary blip from waking you at 3 a.m. while still catching real outages quickly. Each condition fires once when it first becomes true, not repeatedly every second.

High latency

The host is still answering, but slowly. The alert fires when latency stays above a chosen number of milliseconds for at least a given share of the samples within the time window. For example: latency above 200 ms for more than 50% of the last 60 seconds.

Packet loss

The host is failing to answer at all. The alert fires when timed-out pings make up at least the chosen share of the samples within the time window. For example: more than 10% loss over the last 60 seconds. This counts non-responses only, so it is a clean signal that a host is dropping off rather than merely running slow.

Tick either condition (or both) and set its figures. Because the test is based on a share of recent samples, it adapts automatically to your ping interval – a slow ping rate will not silently disable alerting.

Channels: how you are told

Tick any combination of the three delivery channels. Each has a Configure… button for its details, and a channel only counts as active once it is actually configured, so a half-filled channel will not quietly swallow your alerts.

Show desktop notification

Pops up a notification toast on the desktop when a condition fires. Click Configure… to optionally attach a sound – pick any WAV file and it will play together with the toast (leave it blank for a silent notification). Use the Test button in the dialog to hear it.

Send email

Sends an e-mail describing which hosts are affected – ideal for forwarding to a phone or a shared mailbox. Click Configure… to set it up.

The quickest option is Use automatic settings, which sends through SoftPerfect’s own mail relay with nothing to configure – just enter a recipient address. If you would rather use your own server, untick it and fill in the SMTP details:

SMTP host & port

The address of your mail server. You can type host:port directly, or set the port separately.

Encryption

Choose the security your server expects: a plain connection, STARTTLS on the regular port, or a direct TLS connection on a dedicated port.

Username (sender) & password

The credentials for your mail account, if the server requires authentication. The username is also used as the sender address.

Recipients

Where the alert e-mail is sent. The dialog keeps OK and Test disabled until the address looks valid and the server details are complete.

Click Test to send a sample message and confirm everything works before you rely on it.

Execute program

Runs a program or script when a condition fires – the building block for almost any custom action, from raising a ticket to flashing a smart bulb. Click Configure… to choose the program and any command-line arguments. Ping Monitor passes the alert details to the program as environment variables:

PINGMON_HOSTS

The affected host names.

PINGMON_KIND

The kind of problem: latency, loss or mixed.

PINGMON_COUNT

How many distinct hosts are affected.

On Windows, .bat and .cmd files are run through the command interpreter automatically.

Cooldown period

When a whole subnet drops at once you want one alert, not a hundred. The Cooldown period (min) is a single global throttle: after an alert is sent, no further alert goes out until the cooldown has elapsed. Any hosts that run into trouble during the lull are not forgotten – they are counted, and the next alert appends a “+N more since last alert” line so a spreading outage is never hidden.

Testing your setup

The Test button on the Alerts tab fires the enabled desktop, sound and execute-program channels straight away with sample data, so you can confirm they behave as expected. E-mail is tested from its own Configure… dialog, which reports success or failure directly.